Posts by JDWhale
|
1)
Message boards :
Number crunching :
What if... All future WU's were VLAR ?
(Message 1014224)
Posted 10 Jul 2010 by  JDWhale
Post: From the replies I see here, it seems that the BOINC Manager Preference not to "Use GPU while computer is in use" option is either being ignored by users, or not implemented very well by the application, if at all. Not being a "GPU cruncher" myself, I can't feel the pain by those of you that are. But my experience programming parallel computing devices dating back 25 years (past tense called 'array processors') might lead me to believe that some lessons learned long ago are possibly ignored with the current implementation [possibly at the driver level?]. |
|
2)
Message boards :
Number crunching :
Why did they limit dl's at all?
(Message 1014187)
Posted 10 Jul 2010 by JDWhale
Post: Not everyone is getting downloads as needed.. and people have not yet been able to stock up for next one... so ALMOST brilliant. Plenty of time to 'stock up' as seen last Monday... Apriori is being able to upload and report results after the outage. In the past this might take 24+ hours, now it's almost instantaneous with new work distributed to the hosts that need it most. Without upload/reporting new downloads were impossible... Not anymore. BRILLIANT !!! (Agreed, some fine tuning is desirable... but it's still much, much better than after prior outages.) Did I say BRILLIANT ? |
|
3)
Message boards :
Number crunching :
Why did they limit dl's at all?
(Message 1014181)
Posted 10 Jul 2010 by JDWhale
Post: I said it last week and I'll repeat it again now... Throttling back downloads after the outage is BRILLIANT !!! Uploads went quickly and without a hitch and downloads are available as needed... BRILLIANT !!! |
|
4)
Message boards :
Number crunching :
What if... All future WU's were VLAR ?
(Message 1014156)
Posted 10 Jul 2010 by JDWhale
Post: Please excuse if this topic has already been argued, but here goes... I've got a problem with all the moaning about folks not being able to feed their GPUs at maximum efficiency... complaining the the required AR's for max RAC are not being supplied by the download feeders upon request. Not too long ago, in the CPU only crunching era, this was considered "cherry picking"... The act of cancelling or otherwise thwarting poorer paying WU's to maximize your RAC... your perceived throughput. In the past some crunchers might have realized that different cache sizes or CPU architectures yielded varying performance on different Angle Ranges and might have cancelled poorly rewarding WUs based on this knowledge (or other 'cherry picking' reasons), while the 'pure' cruncher crunched whatever he was fed... even if it tasted/paid like sh*t. How is reschedueling poorer performing WU's from GPU to CPU any different, especially when you might end up with months worth of work queued for the CPU while your GPU sits idle waiting for the cherries to show up? If you are as committed to crunching for the project as many of you say you are, and have spent the cash to build super crunching machines that exclude poorly paying (credit wise) WUs... What would you do if all future WU's were VLAR? Which just might happen as interesting targets are identified and the search narrows to specific points in the sky (I don't know if this is true, just hypothesising :) ? Just a thought... Warm regards, |
|
5)
Message boards :
Number crunching :
This computer has reached a limit on tasks in progress??
(Message 1011145)
Posted 3 Jul 2010 by JDWhale
Post: AFAIK, this project has never asked for more than "spare" CPU cycles to get the job done. If this has changed, please let me know. Building more and more powerful systems then complaining when work is not available to feed them seems against the original concept of the project. It's my opinion that we should be happy when the admimistrators of the project allow us to crunch for them. Isn't that why we all originally joined the project. That was the way it was when I joined back in 1999 :) |
|
6)
Message boards :
Number crunching :
This computer has reached a limit on tasks in progress??
(Message 1011039)
Posted 2 Jul 2010 by JDWhale
Post: This is brilliant! But if the downloads/outstanding WUs are limited to 20 tasks per host after the outage, the download pipe won't stay clogged for 24+ hours, thus blocking uploads as in the past. Bottom line... my host has work shortly after the servers came back online. In the past it might take a full day before my results could be fully uploaded and requests for new work issued. Again... This is brilliant!!! |
|
7)
Message boards :
Number crunching :
This computer has reached a limit on tasks in progress??
(Message 1011016)
Posted 2 Jul 2010 by JDWhale
Post: This is brilliant! Normally after a significant outage I can't upload my results and therefore can't download any new tasks. However this time that's not the case. Uploads were no problem after the outage and a few new tasks (number is 20) were downloaded to the hosts that had run dry. I can hope that this is the intended behavior and more WUs will flow freely as the network/server traffic after the outage eases. |
|
8)
Message boards :
Number crunching :
heat and noncrunching...
(Message 1008804)
Posted 27 Jun 2010 by JDWhale
Post: I'm near Houston and the daily heat index is over 100F... I've considered turning on some A/C, but it's not even July yet... let alone August :) I'm a firm believer in acclimatizing... Long stays in libraries or supermarkets chill me to the bone and I can't wait to get back outside and into the heat. To each their own :) |
|
9)
Message boards :
Number crunching :
Creditnew, Is this Another Failure
(Message 1006076)
Posted 19 Jun 2010 by JDWhale
Post: Here are some of my favorites (CPU only tasks!!!)... 1630332500 621117854 7 Jun 2010 12:14:25 UTC 16 Jun 2010 21:56:14 UTC Completed and validated 16.53 15.19 372.31 Anonymous platform 1630332170 621117712 7 Jun 2010 12:14:03 UTC 16 Jun 2010 18:32:44 UTC Completed and validated 25.42 24.19 372.32 Anonymous platform 1630332163 621117708 7 Jun 2010 12:14:03 UTC 16 Jun 2010 18:32:44 UTC Completed and validated 9,068.67 9,018.16 372.32 Anonymous platform 25 seconds CPU time vs. 9000 seconds CPU time both pay 372 credits!!! No wonder averages are on the rise :( |
|
10)
Message boards :
Number crunching :
Apparent error calculating granted credit
(Message 1004920)
Posted 16 Jun 2010 by JDWhale
Post: 1629194324 5316621 6 Jun 2010 12:39:20 UTC 15 Jun 2010 6:46:59 UTC Completed and validated 7,747.64 7,525.73 84.33 368.43 Anonymous platform 1629194325 5137231 6 Jun 2010 12:39:21 UTC 16 Jun 2010 18:01:52 UTC Completed and validated 540.56 92.96 141.58 368.43 Anonymous platform Yep... strange things are happening... on this one I claimed 84, wingman claims 141... we're granted 368 credits. Also noticed a spike on this hosts average credit a couple days ago, but since I did detatch, that picture is gone :( |
|
11)
Message boards :
Number crunching :
Can Seti run off a ram disk?
(Message 999488)
Posted 29 May 2010 by JDWhale
Post: I've run BoincPE on diskless systems that boot from a 512mB USB flash chip... The chip can be marked as read only and copys the disk image to RAM at boot/system startup. Persistance can be configured to a networked server at scheduled intervals. Remote access is pretty easy to setup using TightVNC. It's been a couple years since I set up the diskless systems and am no longer running them for various reasons. IIRC, the most difficult part was getting the network interface controller (NIC) drivers installed for the different motherboards. It isn't easy to use the various tools required to generate the final disk image on the flash drive, but if you've got enough spare time to research and do the necessary trials, it can be a gratifying experience. Good luck, |
|
12)
Message boards :
Number crunching :
Early SETI contributors - who's still around?
(Message 998742)
Posted 25 May 2010 by JDWhale
Post: 6 April 1999 SETI@home classic workunits 31,813 SETI@home classic CPU time 221,187 hours |
|
13)
Message boards :
Number crunching :
Welcome to the 10 year club
(Message 908584)
Posted 18 Jun 2009 by JDWhale
Post: Dr. C.E.T.I. wrote: [lookin' forward to my celebration on February 29th 2010] LOL |
|
14)
Message boards :
Cafe SETI :
The New Mini-City News
(Message 863662)
Posted 8 Feb 2009 by JDWhale
Post: I have also now definitely shown that the Miniville site only rewards Miniville cities. I disagree with this point, I just pointed my miniville config to carnoustie.myminicity.com and saw visits begin near instantly. I use this city sometimes for "Happy Times" clicking at Yarolds, but not for the past few days. Did you prefix your city name with 'world_', as required for myminicity links? |
|
15)
Message boards :
Cafe SETI :
"The New Moon" Bar
(Message 863627)
Posted 8 Feb 2009 by JDWhale
Post:  Weeks and weeks in the hills of Skofus, tending to the 'still in the hills' has worked up quite a thirst for some good hooch. Bartender, if you please, drinks for all. |
|
16)
Message boards :
Cafe SETI :
FreeRice Donation Challenge - 1,207,980 grains!
(Message 837631)
Posted 7 Dec 2008 by JDWhale
Post: +2000 |
|
17)
Message boards :
Cafe SETI :
Arizona Moon`s coconut posting thread II ;-)
(Message 837591)
Posted 7 Dec 2008 by JDWhale
Post: mmhh.. it must have In the 80's, all of us were 'young'... I was commuting 12,500 miles between work & home back then... 4 days one way (trains, planes, & automobiles) (8 weeks work, 4 weeks off)! |
|
18)
Message boards :
Cafe SETI :
Mini-Cities News
(Message 837588)
Posted 7 Dec 2008 by JDWhale
Post: One problem using Idarionis "Visitor" mode is that some types of sites, like "Dragon Cave" or "Valenth", might require an extra action to credit the subscriber. I've found that using "snap links" add-on (different version depending on FF version) in addition to "flash-blocker" with FireFox allows me to cruise through the Idarionis links faster and perform the extra action faster than using "Visitor" mode and lets me perform extra action necessary to credit those "other" type sites. (right mouse button-drag opens all links inside dashed green box in separate tabs simultaneously... very, very fast! Use <ctrl-w> to close tabs after visiting). Additionally, this method allows you visit only those cities you wish to visit quickly and to leave messages at your favorite cities. "Auto-visit" requires you to make extra visits, which aren't counted, to leave messages. Also, it doesn't help any to have more than one user listing your site... Each mini-city is only allowed one visit per day per IP address... any more than one user listing your site just wastes clicks. This is why I don't show up on many "history logs"... I've already visited the minicity and don't want to "waste" another visit via Idarionis. Happy clicking, JDWhale |
|
19)
Message boards :
Cafe SETI :
FreeRice Donation Challenge - 1,207,980 grains!
(Message 835590)
Posted 30 Nov 2008 by JDWhale
Post: +4000 |
|
20)
Message boards :
Cafe SETI :
FreeRice Donation Challenge - 1,207,980 grains!
(Message 835035)
Posted 28 Nov 2008 by JDWhale
Post: +8800 |
|
21)
Message boards :
Cafe SETI :
Thanksgiving - an explanation for a non US citizen please
(Message 834691)
Posted 27 Nov 2008 by JDWhale
Post: And Thanksgiving always has 2 NFL football games. Always with Detroit Lions first game. Always with Dallas Cowboys second game. And this year there is a 'special' third game featuring Philadelphia Eagles at home... 'Times they are a changing'!!! |
|
22)
Message boards :
Cafe SETI :
FreeRice Donation Challenge - 1,207,980 grains!
(Message 834654)
Posted 26 Nov 2008 by JDWhale
Post: And how long has KenzieB been doing this...daily... Remind me never to get into a war of words with her...LOL +3000 |
|
23)
Message boards :
Cafe SETI :
FreeRice Donation Challenge - 1,207,980 grains!
(Message 834533)
Posted 26 Nov 2008 by JDWhale
Post: +7500 |
|
24)
Message boards :
Cafe SETI :
FreeRice Donation Challenge - 1,207,980 grains!
(Message 834015)
Posted 24 Nov 2008 by JDWhale
Post: added another 5600 grains today. |
|
25)
Message boards :
Cafe SETI :
FreeRice Donation Challenge - 1,207,980 grains!
(Message 833581)
Posted 23 Nov 2008 by JDWhale
Post: Another run... You have now donated 8100 grains of rice. |
|
26)
Message boards :
Cafe SETI :
FreeRice Donation Challenge - 1,207,980 grains!
(Message 833447)
Posted 23 Nov 2008 by JDWhale
Post: +3000 grains of rice [later] make that +5000 grains |
|
27)
Message boards :
Cafe SETI :
Mini-Cities News
(Message 833257)
Posted 22 Nov 2008 by JDWhale
Post: Sirius B has a loose screw :O |
|
28)
Message boards :
Cafe SETI :
"The New Moon" Bar
(Message 832985)
Posted 21 Nov 2008 by JDWhale
Post: I'm ringing the bell.... Round of B-52 for all... <flames optional> BTW... that's a layer of Baileys in the middle :-)  |
|
29)
Message boards :
Cafe SETI :
"The New Moon" Bar
(Message 832326)
Posted 20 Nov 2008 by JDWhale
Post: That looks good..........any chance of adding just a tiny drop of rum? Thast rum the other night has really upset my taste buds.............Damn, I'm going to have to stick my head in a basin for the next 25 years :) No more rum for you Sirius...maybe a dash of rum extract will do :-> |
|
30)
Message boards :
Cafe SETI :
"The New Moon" Bar
(Message 831039)
Posted 16 Nov 2008 by JDWhale
Post: How wonderfull Sirius, sometimes good things happen to nice people. Congrats. JD what is up with Luna or do we have to help ourselves ;) Sorry, I had Luna caddying for me earlier on the golf course... it was chilly and very windy... She had to go for a soak in the hot tub before reporting for work... But here she is, ready to serve all the free drinks in Sirius Celebration... Starting with that Double Rum & Black... 
|
|
31)
Message boards :
Cafe SETI :
Arizona Moon`s coconut posting thread II ;-)
(Message 829846)
Posted 13 Nov 2008 by JDWhale
Post: Hi Moon, I was out late last night looking for your bright shining face... My area of the world is cloudy these couple days, so I couldn't see you.. I did see a glow behind the clouds...so I know you are still there, just "under the weather" ;-) Get well soon. |
|
32)
Message boards :
Cafe SETI :
Arizona Moon`s coconut posting thread II ;-)
(Message 829178)
Posted 11 Nov 2008 by JDWhale
Post: and..what should i do then? Put up a Petition; Sign here? ;-) - leave Hmmm... I forgot what a paycheck looks like ;-) Could you fax me a copy ? |
|
33)
Message boards :
Cafe SETI :
"The New Moon" Bar
(Message 827428)
Posted 6 Nov 2008 by JDWhale
Post:
While I'm in Stavanger later this month to play 'winter golf'... I will not be carrying any "buds" on my journey... |
|
34)
Message boards :
Technical News :
Composite Head (Nov 05 2008)
(Message 827385)
Posted 6 Nov 2008 by JDWhale
Post: I figured, since the network news bungled the past couple presidential elections... I'd watch "Get Smart" DVD and time shifted "Boston Legal" & "30 Rock" rather than the election coverage last night..."Denny Crane" Oh... were there SETI servers issues last night? Sorry, I didn't notice.. |
|
35)
Message boards :
Cafe SETI :
"The New Moon" Bar
(Message 827382)
Posted 6 Nov 2008 by JDWhale
Post: Think that might cover it at least for Bud products. ermmmm... your idea of 'Bud products' might not be the same as others...[cough] |
|
36)
Message boards :
Cafe SETI :
The Last Person To Post Wins # 73 A Rainy Holiday Tuesday....................
(Message 827036)
Posted 4 Nov 2008 by JDWhale
Post: first to winning after outage :))) |
|
37)
Message boards :
Cafe SETI :
pulverized pumpkin guts . . .
(Message 825883)
Posted 2 Nov 2008 by JDWhale
Post: (8) Met a nice young man at the party. We chatted for a bit. Turn’s out he is five years younger than me. Sigh. Welcome to cougar-ville. Don't think you qualify yet... Younger cougars are in 40s... (been there, done that) Was he younger than half your age + 7.. if so they you might be cradle robbing, if not... then go for it baby ;-) |
|
38)
Message boards :
Cafe SETI :
"The New Moon" Bar
(Message 825530)
Posted 1 Nov 2008 by JDWhale
Post: Well... 'twasn't easy...but I ended up strapping the pool table on the back of a goat with a broken ankle... Took 2 days, 10 lbs of carrots, 2 little blue pills, and 7 sheep... but the move was a success. Game of 9-ball anyone ? |
|
39)
Message boards :
Cafe SETI :
"The New Moon" Bar
(Message 825163)
Posted 31 Oct 2008 by JDWhale
Post: Here you go...18 years old... That'll be $20...same as last night ;-) 
|
|
40)
Message boards :
Cafe SETI :
The MOON BAR
(Message 824988)
Posted 30 Oct 2008 by JDWhale
Post: Grabbing the dartboard, BBQ pit, and drink-making flammables before it's too late. Sure gonna miss this place :( |
|
41)
Message boards :
Cafe SETI :
Mini-Cities News
(Message 824947)
Posted 30 Oct 2008 by JDWhale
Post: Calling all kitties.........meeeeeeeeeeeeowwwwwwwwrrrrrrrrrrr. Have you considered team FedEx-Cup offer to host a golf tour event in Seti City ? |
|
42)
Message boards :
Number crunching :
100+ WUs missed validation/back end processing!!
(Message 824480)
Posted 29 Oct 2008 by JDWhale
Post: All but a couple of the 100+ WUs whos reports were Lost in Transition validated today when the original deadlines lapsed. The few unvalidated WUs either have MIA wingmen and were reissued, or are still awaiting deadline. This issue is resolved IMO, and I'm satisfied that BOINC handles the case of lost reports admirably. Regards, JDWhale BTW... a nice bump in RAC when 100 WUs validate instantaneously :) |
|
43)
Message boards :
Number crunching :
Seagate.......grrrrr
(Message 824224)
Posted 28 Oct 2008 by JDWhale
Post: The Fiend wrote: You might notice from my profile I have my location set to Thailand, but I am not Thai and I don't live there, but I do spend a couple of months there a year on holiday. [OT] Ever make it down to Pattaya ? I used to keep a house and spend a few months a year there on Holiday, nearly 20 years ago. Cheap entertainment was showing up at "Marine Bar" to watch Muay Thai (Thai kick boxing). Wondering if the "Green Hut Beer Bar" is still open... a few New Year's party there and the food brought in by the 'hostesses' was always delicious. [/OT] I agree with Mark that his stereotyping comment was made in jest, as was my follow up about 'Ya Baa'. Long live King Rama IX... |
|
44)
Message boards :
Cafe SETI :
The MOON BAR
(Message 823177)
Posted 25 Oct 2008 by JDWhale
Post: @Moon - My prayers are with you during these violent storms... Here's something extra to help keep your nerves under control... LOL...I'd deliver personally, but "Magic Carpet" isn't waterproof... 
|
|
45)
Message boards :
Number crunching :
Seagate.......grrrrr
(Message 822992)
Posted 25 Oct 2008 by JDWhale
Post: Freddie Mercury said it best ... Another One Bites The Dust Ya Baa more likely... |
|
46)
Message boards :
Cafe SETI :
The MOON BAR
(Message 822984)
Posted 25 Oct 2008 by JDWhale
Post: Moon, that sounds very good. My wife and I will have to try it this winter. :) Ehrmmm...if you need to be married to give it a try... I'll volunteer to be your wife for the evening...lol But you better do the cooking... I only know how make flaming drinks :D |
|
47)
Message boards :
Number crunching :
100+ WUs missed validation/back end processing!!
(Message 822453)
Posted 24 Oct 2008 by JDWhale
Post: Thanks for the info, I'll wait 'til after next weeks outage & deadlines have passed to post the rest of the affected WUs. (~50 more) |
|
48)
Message boards :
Number crunching :
100+ WUs missed validation/back end processing!!
(Message 822241)
Posted 23 Oct 2008 by JDWhale
Post: Bump... And a few more of these Lost in Transition WUs... As near as I can tell, the deadline on most these WUs is coming up in ~5 days... Might that trigger validation/backend events? Only time will tell... Since no one has offered any insight, I'm assuming that this is an undocumented bug in some server side process :-( wuid=343473039 wuid=343473045 wuid=343473123 wuid=343473129 wuid=343473135 wuid=343473079 wuid=343473091 wuid=343473139 wuid=343473145 wuid=343473151 wuid=343473106 wuid=343473160 wuid=343473166 wuid=343473172 wuid=343473178 wuid=343473149 wuid=343473155 wuid=343473168 wuid=343473164 wuid=343473163 wuid=343473235 wuid=343473190 wuid=343473215 wuid=343473228 wuid=331551422 wuid=343473249 wuid=331551493 wuid=343473247 wuid=343473274 wuid=343473251 Regards, JDWhale |
|
49)
Message boards :
Cafe SETI :
Mini-Cities News
(Message 821910)
Posted 22 Oct 2008 by JDWhale
Post: Done for the day.........Earlier than normal............ Sorry Sirius, my girlfriend is a hooker and I was showing her how it's done :D Next time I'll try to remember taking off the team Fedex-Cup hat. Stop back by the 19th hole and I'll put some ice on your head and some head on your beer. Cheers, JDWhale |
|
50)
Message boards :
Cafe SETI :
Arizona Moon`s coconut posting thread II ;-)
(Message 821876)
Posted 22 Oct 2008 by JDWhale
Post: What is this ??? A coconut wearing sunglasses..lol Looking good Ms Moon...Are you going 'Hollywood'? :D G'day to all... |
|
51)
Message boards :
Number crunching :
100+ WUs missed validation/back end processing!!
(Message 821518)
Posted 21 Oct 2008 by JDWhale
Post: A random sample of the examples given show the wingpeople reported different dates/times and all before the 15th. So if you're a wingperson to one of these it would be difficult to spot. Correct... if the wingman reported after I did, then the WU validated without incident was removed after 24 hours. It seems that the servers hiccup'd when I reported these results causing backend events to be lost. |
|
52)
Message boards :
Number crunching :
100+ WUs missed validation/back end processing!!
(Message 821467)
Posted 21 Oct 2008 by JDWhale
Post: Bump... My hosts 'stats' chart displays a 10% drop in RAC because of the 100+ WUs (about 1 days worth) lost in validation after last weeks outage. All my lost WUs were reported at 15 Oct 2008 0:01:18 UTC and seem to be limited to one host. |
|
53)
Message boards :
Cafe SETI :
The Last Person To Post Wins # 71 TGIF
(Message 821445)
Posted 21 Oct 2008 by JDWhale
Post: LOL....cobble cobble...... Then there is 'Ethanol Subsidies'... Lets pay farmers to raise the price of food by diverting crops to producing inefficient fuel. Going back to bed and winning. |
|
54)
Message boards :
Number crunching :
One Week Later
(Message 821437)
Posted 21 Oct 2008 by JDWhale
Post: I think the lesson here is... go ahead and mount the HSF with "stock" push-pin mounts... take care and be sure you feel/hear the click on opposite corners. Grab the HS and wiggle it around to see if the push-pins release. If not then you have a secure mount, If some corners release, then you must try again. Follow Jokers Xigmatek HDT-S1283 or Kingwin RVT-12025 cpu cooler Tutorial for yer crunchers thread... use needle nose pliers on rubbers to remove/install the fan to the heatsink.. else you will skin your knuckles when applying proper pressure to push-pins. (You being a 'fine' young woman will not have the same problems I had with my 'fat' old knuckles :)) In my case, it was 3 trys before the push-pins passed the 'wiggle' test and I had a secure mount. The problem had to do with not applying enough pressure on diagonal corners when mounting the HSF. This is not an easy HSF to mount...use Intels instructions of applying pressure to diagonal pins and 'locking' them in before applying pressure to adjacent corners and you should be okay. If you still can't get the HS to 'lock' down... then get yourself a mounting kit. Gluck, JDWhale |
|
55)
Message boards :
Cafe SETI :
. . . SETI@home 10th Anniversary NEWS & LINKS - May 2009
(Message 821405)
Posted 21 Oct 2008 by JDWhale
Post: <snip original post> Are we talking "central party/reunion" or virtual reunion... IMO Berkeley is nice late June-August, but still a bit cold in May... 'Course we could just all gather at Uli's [ducking for cover]. |
|
56)
Message boards :
Number crunching :
100+ WUs missed validation/back end processing!!
(Message 821384)
Posted 21 Oct 2008 by JDWhale
Post: Bump... Maybe I'm the only user noticing this problem... but, IMO, it should be addressed... Here are some of my WUs that seem to have been Lost in Transition to the backend. There are ~76 more in my queue if I need to list them for resolution... wuid=343341103 wuid=343341095 wuid=343341101 wuid=343341113 wuid=343341116 wuid=343341144 wuid=343341130 wuid=331395713 I am 3rd (missing?) wingman wuid=343341156 wuid=343341189 wuid=343341250 wuid=343341245 wuid=343341270 wuid=343341276 wuid=343341276 wuid=343341291 wuid=343341303 wuid=343341315 wuid=343341298 wuid=343341301 wuid=343341307 wuid=343341313 wuid=343341319 wuid=343341317 wuid=343341323 wuid=343341314 wuid=343341320 wuid=343341326 wuid=343454800 wuid=343454806 wuid=343454850 wuid=343454856 wuid=343454836 wuid=343473042 Regards, JDWhale |
|
57)
Message boards :
Cafe SETI :
Yawn. Another boring SETI day...
(Message 821353)
Posted 21 Oct 2008 by JDWhale
Post: Are we there yet? :-) Been there, done that. :D |
|
58)
Message boards :
Number crunching :
One Week Later
(Message 821239)
Posted 21 Oct 2008 by JDWhale
Post: I have noticed in my virtual window shopping that there are dedicated chipset coolers as well. But I don't think that I need to go that far just yet. :) Looking at this review of ASUS Silent Knight II I'd rather stick to something with better cooling properties... I've got both the ThermalRight Ultra Extreme 120 (TRUE) and the Kingwin RVT-12025 (Xigmatek HDT-S1283 equivalent) and can recommend both as having similar cooling capabilities. (Both are mounted on same Gigabyte MB & Q6600 systems) The Kingwin, IMO, wins hands down at half price with fan included, if you can still find one. Mount the HSF so the heatpipes are perpendicular to the bead of AS5 for best cooling and the fan will blow across the northbridge HS and you will not need a northbridge fan at all. Don't get caught up with all the LEDs... choose performance parts over show parts for a cruncher... If you want to stay at stock speed the ASUS HSF is fine and IMO is aimed at teens for show... but it will not help much making Q-Baby go. Regards, JDWhale |
|
59)
Message boards :
Number crunching :
One Week Later
(Message 821034)
Posted 20 Oct 2008 by JDWhale
Post: I am going to let her run ’as is’ for another 2 to 3 weeks but, before tweaking things, I guess I had better think about upgrading the stock cooler first. I'm really happy with KingWin RVT-12025 HSF (about $30 US including fan) Orient so the heatpipes cross the line of CPU cores. A bit difficult to install, but Joker posted a fine tutorial a few months back. (might have referred to XigmaTek equivalent HSF, but they're seemingly identical) You might consider purchasing a mounting backplate, but the pushpin mount works fine for me. Q6600 @ 3.0 Ghz on Gigabyte GA-P35-DS3L Ambient 30C CoreTemp 60-56-52-58C running S@H AK_V8 on 4 cores 24/7 (RAC > 5200, MB only, NO Astropulse) A word of warning...it might take a few practice mounts before you 'get it right'... refering to amount of AS5 and uniform spread/pressure. Yes, the northbridge HS gets hot, but with good airflow through the case I've not had any problems. Good luck... |
|
60)
Message boards :
Number crunching :
Puzzled about short wu's
(Message 819773)
Posted 17 Oct 2008 by JDWhale
Post: Yeah, that sounds effective and meets your needs. Ehrm... since you run a quad... isn't that 4 WUs/hr/core so 1920 WUs must be stored on server for 5 day cache in 'worst case scenario'? At least that is the rate my quads crunch 'shorties'. |
|
61)
Message boards :
Number crunching :
Seagate.......grrrrr
(Message 819721)
Posted 17 Oct 2008 by JDWhale
Post: F'in crap is now made in Thailand............ Didn't Meredith Johnson (Demi Moore) change the specs for the air filtration in the factory... Could be the cause of the recent dependability issues in addition trailing performance as of late... Don't worry...Tom Sanders (Michael Douglas) wins the sexual harassment case and keeps his job... Everything should be back to normal before long =) Disclosure (1994) Oh wait... those were Arcamax drives... Didn't Seagate buy them a few years ago ;-) |
|
62)
Message boards :
Number crunching :
100+ WUs missed validation/back end processing!!
(Message 819567)
Posted 17 Oct 2008 by JDWhale
Post: wuid=331600495 wuid=343506138 wuid=343506230 One of my hosts reported ~150 results after the outage earlier this week. Apparently there was a hiccup doing the report as the messages from the log show that some of the results were then refused... 14-Oct-2008 19:00:11 [SETI@home] Sending scheduler request: To report completed tasks. Requesting 0 seconds of work, reporting 147 completed tasks 14-Oct-2008 19:01:03 [SETI@home] Scheduler request failed: HTTP internal server error 14-Oct-2008 19:02:53 [SETI@home] Computation for task 22au08ag.27142.10297.10.8.91_1 finished 14-Oct-2008 19:02:53 [SETI@home] Starting 22au08ag.27142.10297.10.8.79_1 14-Oct-2008 19:02:53 [SETI@home] Starting task 22au08ag.27142.10297.10.8.79_1 using setiathome_enhanced version 528 14-Oct-2008 19:02:55 [SETI@home] Started upload of 22au08ag.27142.10297.10.8.91_1_0 14-Oct-2008 19:03:00 [SETI@home] Finished upload of 22au08ag.27142.10297.10.8.91_1_0 14-Oct-2008 19:03:23 [SETI@home] Sending scheduler request: To report completed tasks. Requesting 0 seconds of work, reporting 148 completed tasks 14-Oct-2008 19:03:59 [SETI@home] Scheduler request succeeded: got 0 new tasks 14-Oct-2008 19:03:59 [SETI@home] Message from server: Completed result 22au08ag.28785.4162.8.8.244_1 refused: result already reported as success 14-Oct-2008 19:03:59 [SETI@home] Message from server: Completed result 23au08ab.29087.4162.4.8.224_1 refused: result already reported as success 14-Oct-2008 19:03:59 [SETI@home] Message from server: Completed result 22au08ag.28785.4162.8.8.247_0 refused: result already reported as success ... 14-Oct-2008 19:03:59 [SETI@home] Message from server: Completed result 24au08aa.15229.6616.11.8.56_2 refused: result already reported as success ... 14-Oct-2008 19:03:59 [SETI@home] Message from server: Completed result 23au08ad.28372.3344.5.8.224_1 refused: result already reported as success The problem is that it seems that the validator was not invoked for ~108 of these returned results and are now stagnant, waiting for some kind of event to trigger them through the back end. WUs that my wingman reported after I reported have validated without issue... it is just the WUs that I reported completed results after my wingman had reported are "stuck". This host has been up and running BOINC without interruption for at least the past 10 days. Any ideas? |
|
63)
Message boards :
Cafe SETI :
HAPPY BIRTHDAY TO ARIZONAMOON 10/15.....
(Message 818794)
Posted 15 Oct 2008 by JDWhale
Post: No wonder you were soooo very bright early this morning... Sending you a full moon "Happy Birthday" !!! 
|
|
64)
Message boards :
Number crunching :
Panic Mode On (10) Server problems
(Message 818773)
Posted 15 Oct 2008 by JDWhale
Post: wuid=343506710 Backend problems? One of my hosts has many (possibly +100) WUs like this, where both results have been returned, but seem to have gotten lost in the back end ? If this issue has already been addressed, please excuse my oversight. Regards, JDWhale |
|
65)
Message boards :
Number crunching :
Gonna build a new toy. . .desperately need help
(Message 818215)
Posted 13 Oct 2008 by JDWhale
Post: Ambient is about 24 C Are you sure that you installed the AK_V8 in the proper directory ? Looking through your returned results, all I see are WUs crunched with stock 6.03 client. Fantastic job kenzieB... May Q-Baby crunch long and carefree. Cheers, JDWhale |
|
66)
Message boards :
Number crunching :
Panic Mode On (9) Server problems
(Message 817523)
Posted 12 Oct 2008 by JDWhale
Post: The funny thing is, that my xp computer downloaded a couple of more work units, makes me think that this is an early maintenance call Probably not... If you've got more pending uploads than the number of cores (or was it twice the number of cores?) then your host will not request any more work :( |
|
67)
Message boards :
Cafe SETI :
The MOON BAR
(Message 817519)
Posted 12 Oct 2008 by JDWhale
Post: Uli wrote: You know were to send the tab. ermmm, Uli...We never closed out your tab from the ongoing celebration... Since you're 'Astra of the Month' and all, Luna figured the whole month of Oktober was put on your tab :D That Luna... she's not too bright, but she has a wonderful glow...lol I'm off to Las Vegas for some R&R, please don't let The MOON BAR burn down... :) A final round of flaming B-52s on my way out the door....   (Please don't tell AzMoon)
|
|
68)
Message boards :
Number crunching :
Panic Mode On (9) Server problems
(Message 817314)
Posted 12 Oct 2008 by JDWhale
Post:
Haven't seen the clip... but a buddy had the album back in Jr High... "C'mon all you big strong men Uncle Sam needs help again. Got himself in a terrible jam way [over] yonder in [Afganistan]" ...LOL... seems to fit these days, too... Ya.... I saw Country Joe and the Fish a few years later at a university gig in Nor Calif... Back when he was trying to save the whales :D |
|
69)
Message boards :
Number crunching :
Panic Mode On (9) Server problems
(Message 817257)
Posted 11 Oct 2008 by JDWhale
Post:
uploaod ??? |
|
70)
Message boards :
Cafe SETI :
The MOON BAR
(Message 815753)
Posted 8 Oct 2008 by JDWhale
Post: Nice tail ;-) Buy you a drink ? 
|
|
71)
Message boards :
Cafe SETI :
The MOON BAR
(Message 815477)
Posted 7 Oct 2008 by JDWhale
Post: Whiskey... Get me outa here please....  |
|
72)
Message boards :
Cafe SETI :
True music II
(Message 814884)
Posted 5 Oct 2008 by JDWhale
Post:
I'm plugging in that DVD (Death Proof) right now... Excellent Saturday night movie ;-) Here's a link to Death Proof trivia which features 'The complete list of songs contained in the jukebox'... 50 of QTs faves...makes for an interesting playlist.... |
|
73)
Message boards :
Cafe SETI :
True music II
(Message 814876)
Posted 5 Oct 2008 by JDWhale
Post:
Also from Death Proof... Vanessa Ferlito is lap dancing to The Coasters Down in Mexico... [edit]Cleaned it up a bit... Changed link to 'song only' (without lapdance video)[/edit] |
|
74)
Message boards :
Cafe SETI :
True music II
(Message 814865)
Posted 4 Oct 2008 by JDWhale
Post:
Indeed there is... As can be witnessed by the Five Point Palm Exploding Heart Technique. Have you ever noticed how the newer QT movies are 'chick flicks'? lol... ...Jackie Brown, Kill Bill, and especially Grindhouse Something about a chick with a prosthetic leg machine gun ! [edit] oops... that part of 'Grindhouse' was from Planet Terror by Robert Rodriguez [/edit] |
|
75)
Message boards :
Cafe SETI :
True music II
(Message 814648)
Posted 4 Oct 2008 by JDWhale
Post: Kurt Russell getting p****whipped...Chick Habit from Death Proof |
|
76)
Message boards :
Cafe SETI :
The MOON BAR
(Message 814609)
Posted 4 Oct 2008 by JDWhale
Post: Luna... 6 shots of Wild Turkey please... just in case the kitty man shows up ;-) |
|
77)
Message boards :
Number crunching :
Gonna build a new toy. . .desperately need help
(Message 814606)
Posted 4 Oct 2008 by JDWhale
Post: If all the bits haven’t arrived by the Sunday (my next day off of work) I’ll likely break down and do some assembly. :) keepin' my fingers crossed... just remember... a thin bead of AS5 inline with the CPU cores... thinner is better with a slight back & forth motion of the HSF to smear the AS5 a little. that AS5 really spreads out when you push down on the HSF mounting pins. Use 90+% isopropyl alcohol and a coffee filter to remove the factory gloop from the stock HSF and to prepare both HSF & CPU surfaces prior to mounting. g'luck, |
|
78)
Message boards :
Cafe SETI :
Arizona Moon`s coconut posting thread II ;-)
(Message 814120)
Posted 2 Oct 2008 by JDWhale
Post: I don't wear jewelry anymore... not even a watch. But there was a time that the only piece I always wore was a 24kt necklace worth enough that I could sell it at market rates to buy a plane ticket home from anywhere in the world. It wasn't 'white gold' :-( |
|
79)
Message boards :
Cafe SETI :
The MOON BAR
(Message 813789)
Posted 1 Oct 2008 by JDWhale
Post: This Congrats Uli...  My mom used to 'crown' me when I was bad... Sometimes I could see stars for a week ;-) |
|
80)
Message boards :
Cafe SETI :
The MOON BAR
(Message 813569)
Posted 1 Oct 2008 by JDWhale
Post: Hi Steve. Welcome to The MOON BAR... Luna here will be your waitress... Tonight is her first night, too...   Mind if I join you for a frosty ale ? 
|
|
81)
Message boards :
Number crunching :
Gonna build a new toy. . .desperately need help
(Message 813286)
Posted 30 Sep 2008 by JDWhale
Post: My Seagate HD arrived today. (Yeah!) Enjoy the quiet before the storm ;) |
|
82)
Message boards :
Cafe SETI :
The MOON BAR
(Message 813257)
Posted 29 Sep 2008 by JDWhale
Post: Whew... late tee-time made for a hot day (32C) on the golf course... I'm in need for something to cool down and found out that Häagen-Dazs will deliver... Place your orders for...     For a limited time we add extra Baileys to all Baileys drinks ;-) |
|
83)
Message boards :
Cafe SETI :
Smoking cessation......and digital cameras
(Message 812824)
Posted 28 Sep 2008 by JDWhale
Post: Sorry folks, quitting cigarettes are easy... Heroine is much more difficult!!! [edit] Just speaking from personal experience. [/edit] |
|
84)
Message boards :
Cafe SETI :
Paul Newman passes on......
(Message 812818)
Posted 28 Sep 2008 by JDWhale
Post: How many hard boiled eggs can you eat on a bet ? Thanks for the memories Paul Newman !!! |
|
85)
Message boards :
Cafe SETI :
Favorite movie chase scenes............
(Message 812713)
Posted 27 Sep 2008 by JDWhale
Post: One of my faves... Quentin Tarantino's DeathProof '69 Charger vs. '70 Challenger [edit] Zoe Bell, the woman riding on the hood does all her own stunts [/edit] |
|
86)
Message boards :
Cafe SETI :
To my friend Uli...........and all kindred kitty folk
(Message 812525)
Posted 27 Sep 2008 by JDWhale
Post: I want an invitation ! |
|
87)
Message boards :
Number crunching :
Gonna build a new toy. . .desperately need help
(Message 812520)
Posted 27 Sep 2008 by JDWhale
Post: I still have to pick up some incidentals (AS5, wrist strap, jackhammer, booze etc) Need more booze !!! :) The only thing I'm a bit concerned about besides the mobo fitting the case is whether or not you have the 4 pin 12V connector on the Power Supply that plugs into the mobo near the CPU, required to power the CPU... it a small 2x2 pin connector different from the 24 pin mobo connector. Also did I mention more booze :-? |
|
88)
Message boards :
Cafe SETI :
The Last Person To Post Wins # 70 Wacky Wednesday.....:):):)
(Message 812517)
Posted 27 Sep 2008 by JDWhale
Post: just so you know JD.. im the one who wins ;-) Guess I'll go back to answering those Nigerian cash & British Lottery emails I keep getting... Seems to be a better chance of payoff with those than playing here. G'bye P.S. winning here for now ;-) |
|
89)
Message boards :
Cafe SETI :
The Last Person To Post Wins # 70 Wacky Wednesday.....:):):)
(Message 812511)
Posted 27 Sep 2008 by JDWhale
Post: just so you know JD.. im the one who wins ;-) Please excuse my ignorance... What is the prize when you win? Am I wasting my time for nothing other than bragging rights? |
|
90)
Message boards :
Cafe SETI :
The Last Person To Post Wins # 70 Wacky Wednesday.....:):):)
(Message 812498)
Posted 27 Sep 2008 by JDWhale
Post: $3.29/gal today for regular in Sugar Land, TX. Winning on my way to Friday night golf. |
|
91)
Message boards :
Number crunching :
Gonna build a new toy. . .desperately need help
(Message 812428)
Posted 26 Sep 2008 by JDWhale
Post: Yep, I would definitely use the built-in video for a cruncher-only machine. It will work like a champ!!! Also, no fan so it will be nice and quiet :) |
|
92)
Message boards :
Number crunching :
Gonna build a new toy. . .desperately need help
(Message 812418)
Posted 26 Sep 2008 by JDWhale
Post: Yep, I would definitely use the built-in video for a cruncher-only machine. The mobo being used does not have built-in video. @kenzieB - you will need an el-cheapo video card. |
|
93)
Message boards :
Number crunching :
Gonna build a new toy. . .desperately need help
(Message 812405)
Posted 26 Sep 2008 by JDWhale
Post: About $400 in so far and I have, or on order, mobo, Q6600, hd and memory. Don't forget to factor in a fine bottle of wine to enjoy after assembly (can also be used to put out the fire :) Hmmm... no video card yet... The cheapest PCIE card you can find... I like the GEForce 6200LE with passive cooling (no noisy fan). Sometimes free after rebate. |
|
94)
Message boards :
Number crunching :
Gonna build a new toy. . .desperately need help
(Message 812178)
Posted 26 Sep 2008 by JDWhale
Post: Just notice this thread and saw one point missing. Dual channel is the way to go, but also consider that you are installing a quad. That is 4 cores that are all going to need to use the ram. S@H seems to work best if each core has 512K. So I strongly suggest 2 1 gig sticks. As for brand, I like corsair. But basically get the fastest you can afford, it's all pretty cheap these days. A suggestion re mail in rebate... wait until your system has been up and running ~2 weeks before sending in rebate just in case your memory fails early. The stores usually will not exchange once you've cut the UPC from the product forcing you to deal with the manufacturer directly for an exchange thru the mail. Also agree with Mark on hard drive... Seagate. |
|
95)
Message boards :
Cafe SETI :
Texas Crunchers
(Message 812023)
Posted 26 Sep 2008 by JDWhale
Post: Last area of devastation to allow residents to return for a few hours. Yes... The reason given to allow these people to return is "Closure... let these people see there is nothing to come home to... there are no homes left". Here's a photo just north of the bridge allowing folks onto the peninsula...  A single home is left standing among debris from Hurricane Ike September 14, 2008 in Gilchrist, Texas. Floodwaters from Hurricane Ike were reportedly as high as eight feet in some areas causing widespread damage across the coast of Texas. (David J. Phillip-Pool/Getty Images) # Public ferry service to the peninsula is suspended until further notice. Prayers to those affected. |
|
96)
Message boards :
Cafe SETI :
Texas Crunchers
(Message 812014)
Posted 25 Sep 2008 by JDWhale
Post: Last area of devastation to allow residents to return for a few hours. Bolivar peninsula, residents of Crystal Beach, approx. 5000 homes, are being allowed to return to observe their homes & neighborhoods for a few hours, there is a single lane of a heavily damaged bridge onto the peninsula that will be opened between the hours of 6 AM and 2 PM. No one can stay. Still total devastation on the most heavily hit part of the Texas Gulf coast. |
|
97)
Message boards :
Cafe SETI :
To the kitty people VI......
(Message 812007)
Posted 25 Sep 2008 by JDWhale
Post:  A cat looks down from a tree in the flooded West End section of Galveston, Texas, after hurricane Ike hit the area Saturday, Sept. 13, 2008. (AP Photo/LM Otero) # |
|
98)
Message boards :
Number crunching :
Gonna build a new toy. . .desperately need help
(Message 811978)
Posted 25 Sep 2008 by JDWhale
Post:
Yep, you need 2 matched sticks to run 'Dual Channel' which allows twice memory bandwith. Will definitely show improved performance when crunching 4 VHAR 'shorty' WUs. Also.. memory prices & rebates vary day-to-day... keep your eyes on the local discount store ads and pounce when you can get a 2 x 1GB kit for ~$40 after rebate. Also look for PC2-6400 DDR2-800 for best bang for the buck...example deal G.SKILL 2GB (2 x 1GB) |
|
99)
Message boards :
Number crunching :
Gonna build a new toy. . .desperately need help
(Message 811893)
Posted 25 Sep 2008 by JDWhale
Post: Am thinking of 1gB of Patriot memory for now. (about $50). Will add another gB later if needed. I'd suggest buying a "Dual Channel Memory Kit" either 1GB (2 x 512MB) or 2GB (2 x 1GB) for better performance. If you only buy 1 memory stick now, then add another memory stick later, your computer might not (probably will not) operate in "Dual Channel" mode. Look for 2 memory sticks in a kit. Thermakatite cooler. ($40. A good deal?) IMO, you do not need to buy an aftermarket cooler until you decide to OC your CPU. The stock cooler works just fine for running at stock speed... assuming that you are buying a 'retail' CPU that includes the stock heatsink/fan(HSF). OTOH, some aftermarket HSF require you to remove the the motherboard for installation of a mounting plate on the backside of the motherboard. If that's the case then it does make sense to use the aftermarker HSF on first install to save you from later grief. Measure your case to make sure you have clearance for any aftermarket HSF. Also, I have tools (screwdrivers, needle-nose pliers, cutting torch, sledge hammer, and can curse in 3 languages so I have that bit covered!) :) The more hammers the better... And I bet you cuss like a sailor...LOL :D gluck, |
|
100)
Message boards :
Cafe SETI :
Stealth Eagle.. Happy Birthday 25.09 ;-)
(Message 811886)
Posted 25 Sep 2008 by JDWhale
Post: 
|
|
101)
Message boards :
Number crunching :
Gonna build a new toy. . .desperately need help
(Message 811848)
Posted 25 Sep 2008 by JDWhale
Post: So after a bit of agonizing, I finally made a decision. That's a wonderful board. I'm running 2 of it's predecessor 'GA-P35-DS3L' with Q6600 CPUs and I am plenty happy with both :) |
|
102)
Message boards :
Cafe SETI :
The whatever....it is my thread thread..........
(Message 811686)
Posted 24 Sep 2008 by JDWhale
Post: Thanks Mark... For sending down Excel Energy line crews from Wisconsin to help restore electricity to Sugar Land, Texas !!! Saw about a dozen trucks from Excel today when I was out hitting golf balls, love those pretty blue license plates :) |
|
103)
Message boards :
Cafe SETI :
The Last Person To Post Wins # 70 Wacky Wednesday.....:):):)
(Message 811602)
Posted 24 Sep 2008 by JDWhale
Post: Just passing through... It was a gift from the Persian woman who cuts my hair... Says she got it from Baba Yaha on a recent visit back home ;-D @Mike - HDTV is great! |
|
104)
Message boards :
Cafe SETI :
The Last Person To Post Wins # 70 Wacky Wednesday.....:):):)
(Message 811584)
Posted 24 Sep 2008 by JDWhale
Post: Just passing through... on a magic carpet ride... |
|
105)
Message boards :
Number crunching :
Gonna build a new toy. . .desperately need help
(Message 811576)
Posted 24 Sep 2008 by JDWhale
Post: One problem that I've had trying to reuse a Dell case for a new build is that the backplate (where MB USB, E-net, sound, etc connections) was proprietary and was only usable with the Dell motherboard & MB tray. My advise here is to verify that the backplate on your HP case is removable and will allow you to fit a different motherboards backplate. It is easy to assume that since you have a case that any ATX MB will fit, but it's been my experience that most the time it's more work than it's worth when working with a proprietary case. Building your first computer from 'off the shelf' parts is a thrilling experience, I just hate to see the frustration when the case doesn't fit your new motherboard quite right. Just be sure to leave enough $ in your budget to buy a new case if necessary. Maybe take the case you already have with you to the shop when you buy parts and ask them if what you plan will work. Good luck, JDWhale |
|
106)
Message boards :
Number crunching :
Gonna build a new toy. . .desperately need help
(Message 811481)
Posted 24 Sep 2008 by JDWhale
Post: Can you post a picture of the inside of your case, the MB mounting area? It will allow us to see if it is ATX or micro-ATX form factor... Then we can advise compatible motherboard. |
|
107)
Message boards :
Number crunching :
Gonna build a new toy. . .desperately need help
(Message 811473)
Posted 24 Sep 2008 by JDWhale
Post: Check the case you are using will fit an ATX motherboard... since in the future you'll want to OC some and all the best OCing MBs are ATX form factor. I recommend the GIGABYTE GA-EP35-DS3L MB paired w/ Q6600 CPU & cheap Kingston 2x1GB memory sticks @ PC6400 add any hard drive & compatible graphics card. You do not need anything more than a minimal graphics card. Stock CPU cooler is fine until you decide to OC, then switch to Kingwin or Xigmatek. Important step in assembly is not to apply too much thermal grease between CPU & heatsink. Good Luck, JDWhale |
|
108)
Message boards :
Cafe SETI :
The MOON BAR
(Message 810835)
Posted 22 Sep 2008 by JDWhale
Post: G'Day Dirk & all, I'll be good! Please don't send me to Coconut Island... 
|
|
109)
Message boards :
Cafe SETI :
The Last Person To Post Wins #69 Not So Sunny Sunday........
(Message 810741)
Posted 21 Sep 2008 by JDWhale
Post: Stopping by for a USA Ryder Cup win! |
|
110)
Message boards :
Cafe SETI :
The Last Person To Post Wins #69 Not So Sunny Sunday........
(Message 810607)
Posted 21 Sep 2008 by JDWhale
Post: Winning watching Audrey Hepburn in 'How to Steal a Million'. |
|
111)
Message boards :
Cafe SETI :
One to Tee you off
(Message 810596)
Posted 21 Sep 2008 by JDWhale
Post: How many days do they play? Sorry SE, According to Ryder-Cup Schedule Tuesday-Thursday are practice days. Matches begin Friday... Friday : 2 rounds (4 matches of foursome & 4 matches fourball) Saturday : 2 rounds (4 matches of foursome & 4 matches fourball) Sunday : 1 round (12 matches singles play) Each team has 12 players...Some players play all matches, while others are given 'rests' by team captains during first 2 days of matches. All players participate in Sunday singles matches. 1 point awarded for each match win 1/2 point awarded for each match 'halved' (tied) Europe needs 14 points to retain cup USA needs 14 1/2 points to win the cup Should be an exciting day of singles play with the USA beginning the day up 9-7 points. Sunday's Singles Pairings Set (All times Eastern Daylight Time) 12:03 PM: Kim (USA) vs. Garcia (EUR) 12:14 PM: Mahan (USA) vs. Casey (EUR) 12:25 PM: Leonard (USA) vs. Karlsson (EUR) 12:36 PM: Mickelson (USA) vs. Rose (EUR) 12:47 PM: Perry (USA) vs. Stenson (EUR) 12:58 PM: Weekley (USA) vs. Wilson (EUR) 1:09 PM: Holmes (USA) vs. Hansen (EUR) 1:20 PM: Furyk (USA) vs. Jimenez (EUR) 1:31 PM: Cink (USA) vs. McDowell (EUR) 1:42 PM: Stricker (USA) vs. Poulter (EUR) 1:53 PM: Curtis (USA) vs. Westwood (EUR) 2:04 PM: Campbell vs. Harrington (EUR) I'm looking forward in particular to Kim vs. Garcia Furyk vs. Jimenez Stricker vs. Poulter Go USA |
|
112)
Message boards :
Cafe SETI :
The MOON BAR
(Message 810590)
Posted 21 Sep 2008 by JDWhale
Post: G'Day Dirk & all, Hmmmm... Amaretto is one ingredient of Flaming Dr Pepper... Coincidence that both Amaretto & Dr Pepper are posted back-to-back, or do we want to see flaming drinks here in the Moon Bar? |
|
113)
Message boards :
Cafe SETI :
To the kitty people VI......
(Message 810056)
Posted 20 Sep 2008 by JDWhale
Post: This kitty doesn't even know the meaning of 'trouble'... 
|
|
114)
Message boards :
Cafe SETI :
The MOON BAR
(Message 810041)
Posted 20 Sep 2008 by JDWhale
Post: While the Moon is away... the kitties will play... Flattery would get you a Moondance if you were here :) @Mark : Will do... Since Moon's not here tonight... How about one of those Flaming Guinness? |
|
115)
Message boards :
Cafe SETI :
The MOON BAR
(Message 810015)
Posted 20 Sep 2008 by JDWhale
Post: I promised Moon to stay away from flaming drinks & deserts here in the bar... While the Moon is away... the kitties will play...
|
|
116)
Message boards :
Cafe SETI :
The MOON BAR
(Message 809961)
Posted 19 Sep 2008 by JDWhale
Post:
I promised Moon to stay away from flaming drinks & deserts here in the bar... One night when she isn't around I want to give the Flaming Dr. Pepper a try... The guy in the video doesn't quite do it right, IMO... you do not need to light the shots individually... Supposed to let the giant fireball finale do that part for you ;-) |
|
117)
Message boards :
Cafe SETI :
The MOON BAR
(Message 809916)
Posted 19 Sep 2008 by JDWhale
Post:
B-52 is a very tasty treat, but just the thought of floating Wild Turkey on a pint of Guinness turns my stomach... Why ruin both a great stout and a decent bourbon??? Different strokes... @Mark : Bad kitty... Very, very bad kitty!!! |
|
118)
Message boards :
Cafe SETI :
The MOON BAR
(Message 809883)
Posted 19 Sep 2008 by JDWhale
Post:
Is this the weekend we get pictures of Flaming Guinness? |
|
119)
Message boards :
Number crunching :
Geek@Play is Okay...
(Message 809574)
Posted 18 Sep 2008 by JDWhale
Post: I was extremely lucky... only dropped power for 18 hours... very minimal damage. Driving the nearby neighborhoods there are lots of folks without electricity, lots of trees down and some roof damage. Fortunately we didn't get a lot of rain in my area with the storm, just 80 MPH winds. Hundreds of electric utility crews from more than 30 states and Canada are in the Houston area working 16 hour shifts to restore power. Galveston has been proclaimed non-habitable by the Texas Health Comission because of lack of electricity, water and sewer services. Boliver Peninsula to Galveston Island northeast was the hardest hit... I fear it will be months before public services are restored there. Supermarkets in my area are open, limited hours, and gas stations have gasoline. I heard last night on the local news that 50 of 52 Home Depot stores in the greater Houston area have reopened. Best wishes to those less fortunate, JDWhale [edit]Here's Weather Underground link to a station the center of IKE passed over. Note the minimum Barometric Pressure and calm winds at 3:43AM, 13Sep and the 180 degree shift in wind direction. This station is not too far from Geek@Play home. [/edit] |
|
120)
Message boards :
Cafe SETI :
Arizona Moon`s coconut posting thread II ;-)
(Message 809523)
Posted 18 Sep 2008 by JDWhale
Post: I am on my way. Flying in at top speed. of course the butter may be a bit melted by the time I get there. Fly higher... catch the jet stream... I've got the cooker fired up... The lobsters are starting to scream. Wow... you can really make food boy! ;-) Coconut milk based curries are some of my favorites, be careful for the Thai chilies... they can make your face turn red and your whole head to sweat. ;-) |
|
121)
Message boards :
Cafe SETI :
Arizona Moon`s coconut posting thread II ;-)
(Message 809349)
Posted 18 Sep 2008 by JDWhale
Post: Thanks Eagle & Uli... I just got back from Asian market with Thai basil & lemon grass... I hope Moon brings her coconut ;-) For those a little more adventerous... lobster-coconut bisque w/ lemongrass and thai basil  As you can see, there is plenty of lobster for all 
|
|
122)
Message boards :
Cafe SETI :
Arizona Moon`s coconut posting thread II ;-)
(Message 809327)
Posted 18 Sep 2008 by JDWhale
Post: I love lobster.. and when we are all joining up, we can paint the Bring your windows with you... I've got a crew to do drywall and roofing repairs tomorrow... A few more windows shouldn't be a problem :D |
|
123)
Message boards :
Cafe SETI :
The Last Person To Post Wins #69 Not So Sunny Sunday........
(Message 809323)
Posted 18 Sep 2008 by JDWhale
Post: Power is restored, again... Not for me... Like Carl from Slingblade says... "Ain't got no gas"... |
|
124)
Message boards :
Cafe SETI :
The Last Person To Post Wins #69 Not So Sunny Sunday........
(Message 809320)
Posted 18 Sep 2008 by JDWhale
Post: Power is restored, again... Couldn't wait for it to get warm... Ran out... Waiting for you to bring more. :-D |
|
125)
Message boards :
Cafe SETI :
Arizona Moon`s coconut posting thread II ;-)
(Message 809318)
Posted 18 Sep 2008 by JDWhale
Post: Sorry internet shopping... I order washing powder... they deliver lobsters!!! All are invited to lobster dinner... Please RSVP. BYOB... Bring your own butter ;-) |
|
126)
Message boards :
Cafe SETI :
The Last Person To Post Wins #69 Not So Sunny Sunday........
(Message 809314)
Posted 18 Sep 2008 by JDWhale
Post: Power is restored, again... |
|
127)
Message boards :
Number crunching :
Geek@Play is Okay...
(Message 809224)
Posted 17 Sep 2008 by JDWhale
Post: Because Geek@Play is a frequent poster on this board and who lives in direct path of Hurricane IKE near Galveston, Texas... I just received this reply from Geek@Play to PM I sent 3 days ago.... Geek@Play wrote: JDWhale wrote:We're not getting any national news here in Houston, but I realize the center of the eye passed over Eagle Point and I realize this is very close to [your home]. I wish you the best when you can come back home. Best wishes. @Mods : Please excuse my double post here and in Cafe SETI. |
|
128)
Message boards :
Cafe SETI :
Texas Crunchers
(Message 809221)
Posted 17 Sep 2008 by JDWhale
Post: Just received this reply from Geek@Play to PM I sent 3 days ago.... Geek@Play wrote: JDWhale wrote:We're not getting any national news here in Houston, but I realize the center of the eye passed over Eagle Point and I realize this is very close to [your home]. I wish you the best when you can come back home. Best wishes. |
|
129)
Message boards :
Cafe SETI :
Texas Crunchers
(Message 809211)
Posted 17 Sep 2008 by JDWhale
Post: Sad story on the news today... a woman had just purchased a gasoline powered generator.. still in the back of her pickup truck... waited in line 4 hours to buy gasoline and the station ran out of gas as she neared the pumps. She has the geny... but she ain't got gas. On the upside... a cool front passed through the area 2 days ago... overnight temps are dropping below 70F... about 10 degrees lower than normal this time of year. Daily high temps are running low to mid 80s... about 8-10 degrees lower than normal. I got out today and drove a bit larger loop... still a lot of folks without electricity. Gas stations that have electricity the lines are getting shorter, but the gas stations without electricity can not pump gas. Those stations with gas are selling regular 2 week quantity within 12 hours. Here's a link to KPRC Houston NBC-TV affiliate with some interesting information & videos... check out our senior weatherman, Frank Billingsley' Walking Tour of Historic Galveston. He has a home in Galveston that was flooded with 3 ft of water. Best wishes for all, JDWhale |
|
130)
Message boards :
Cafe SETI :
The Last Person To Post Wins #69 Not So Sunny Sunday........
(Message 809206)
Posted 17 Sep 2008 by JDWhale
Post: Headed to the practice range...and winning!!! |
|
131)
Message boards :
Cafe SETI :
The MOON BAR
(Message 809162)
Posted 17 Sep 2008 by JDWhale
Post: Good morning all I thought I heard someone say breakfast was being served here. It was mmm, mmm good, too...  |
|
132)
Message boards :
Cafe SETI :
The Last Person To Post Wins #69 Not So Sunny Sunday........
(Message 808777)
Posted 16 Sep 2008 by JDWhale
Post: My last win before bed. Good night all. I will see you tomorrow. Sorry Eagle, I rolled over and hit the mouse button,,,, I'm winning again :^P |
|
133)
Message boards :
Cafe SETI :
The whatever....it is my thread thread..........
(Message 808759)
Posted 16 Sep 2008 by JDWhale
Post: I will never forget the ride I had in my '69 Buick GS 400 one afternoon in high school............. More than once I had to lock 'em up else I'd have been straight through the T-bone in the road.... Sliding to a stop locked up looking over the cliff out your window... I changed my driving habits after the second time... But the chicks loved it ;-) |
|
134)
Message boards :
Cafe SETI :
The Last Person To Post Wins #69 Not So Sunny Sunday........
(Message 808757)
Posted 16 Sep 2008 by JDWhale
Post: You were JD. Huh? |
|
135)
Message boards :
Cafe SETI :
The whatever....it is my thread thread..........
(Message 808752)
Posted 16 Sep 2008 by JDWhale
Post: I will never forget the ride I had in my '69 Buick GS 400 one afternoon in high school............. Similar story in my 1970 Z28... Had a s-turn stretch alongside a cliffside... at night you had to watch for opposite direction headlights while we were straightning the curves. Not a pretty sight when the guy you were racing against had to steer toward the cliff to avoid the head-on collision. 300 ft down into the river :-( |
|
136)
Message boards :
Cafe SETI :
The Last Person To Post Wins #69 Not So Sunny Sunday........
(Message 808748)
Posted 16 Sep 2008 by JDWhale
Post: 2 minute warning, both Dallas & I'm winning!!! |
|
137)
Message boards :
Cafe SETI :
The Last Person To Post Wins #69 Not So Sunny Sunday........
(Message 808737)
Posted 16 Sep 2008 by JDWhale
Post: Winning for now... Not for long... Dallas Cowboys just scored again against the Philadelphia Eagles... Now I'm winning :-) |
|
138)
Message boards :
Cafe SETI :
The Last Person To Post Wins #69 Not So Sunny Sunday........
(Message 808728)
Posted 16 Sep 2008 by JDWhale
Post: Winning after a terrible storm. It was enough to be considered Hurricane strength at times! Galveston, Texas reported 104MPH winds.... we had 84MPH winds..... not too far off! Just got electric back after a day and a half without it. Most of Ohio, Kentucky, and Indiana are still without power though. Last night when it hit 99% of the state of Ohio and ALL of Northern Kentucky was pitch black..... kinda creepy. Down here in Houston area, we're still not getting any national news... the local news pictures are horrific enough now that the local news helicopters are allowed to fly again. The news reports in Houston are that the most devasted areas are on Bolivar Peninsula and High Island. Yes, Galveston Island will not be habitable for the next 6+ weeks, but the areas to Galveston northeast (Bolivar & High Island) were hit even harder by storm surge and higher winds :-( BTW: I'm winning. |
|
139)
Message boards :
Cafe SETI :
The Last Person To Post Wins #69 Not So Sunny Sunday........
(Message 808717)
Posted 16 Sep 2008 by JDWhale
Post: Dallas Cowboys are winning... Me, too!!! Yes, he was...Dr. Strangelove... is one of my all time favs... BTW: Dallas scores again... |
|
140)
Message boards :
Cafe SETI :
The Last Person To Post Wins #69 Not So Sunny Sunday........
(Message 808712)
Posted 16 Sep 2008 by JDWhale
Post: Dallas Cowboys are winning... Me, too!!! |
|
141)
Message boards :
Cafe SETI :
AriZonaMoon`s web-site + AZM City ,and other relevant inputs ;-)
(Message 808698)
Posted 16 Sep 2008 by JDWhale
Post:
Ice cold beer is just what I need... a few dark ale, please, if available while I'm waiting in 3 hour line to buy petrol :-( |
|
142)
Message boards :
Cafe SETI :
Arizona Moon`s coconut posting thread II ;-)
(Message 808696)
Posted 16 Sep 2008 by JDWhale
Post: I need some music to do the job.. Hey Moon... Is that you I see climbing the coconut tree ???? Bang on the Drum |
|
143)
Message boards :
Cafe SETI :
Texas Crunchers
(Message 808658)
Posted 15 Sep 2008 by JDWhale
Post: A cyclone is a unique type of storm. Having lived on the Texas Gulf coast for more than 30 years I have become somewhat experienced with the characteristics of these horrific storms in my area. The difference of being located 35 miles either side of the center path of the eye of the storm is like the difference between night and day. I live 35 miles southwest of the closest approach of Hurricane IKE' center and am located on the 'clean' side of the storm, we are only having minor inconveniences. Folks located nearer the center and up to 100 miles to the northeast of the path are suffering very much, still being without running water, no electricity and roads that are not passable. Some services are expected to take 2-6 weeks to restore... The surviving residents of Galveston Island and Bolivar Peninsula have been asked to leave their residences and not to return until infractructure can be restored... maybe upto 2 months from now. My heart goes out to the residents that live just 50 miles east and northeast from me... I wish all of them the best. National news is still unavailable over broadcast media, we have enough local news to keep all network broadcasts busy... 24 hours/day! Local news right now is reporting that electricity has been restored to 550,000 customers, yet 1.5 million customers are still without electricity. Those of us that live outside the most devastated areas have been requested to stay home and share our resources with our nearby neighbors that are less fortunate than ourselves. Lots of BarBQs in the neighborhoods with folks meeting some neighbors for the first time. Best Regards to all, JDWhale Postscript : I understand Matt Lauer from NBC 'Today Show' was in the Houston area this morning... Unfortunately those of us in Houston did not get to see the show because it was pre-empted by local news :-( I really would to like to see what news is getting out of our area to the rest of the country. This just in... : Local news reports... Hundreds and hundreds of ultility velicles (electricity and tree choppers) are making way to dispatch area (horse race track in northwest Houston). These vehicles are from 25 states and Canada. Thanks for the help folks !!! Also a family rescue of 6 people feared dead on Bolivar Peninsula. Praise God! |
|
144)
Message boards :
Cafe SETI :
Arizona Moon`s coconut posting thread II ;-)
(Message 808635)
Posted 15 Sep 2008 by JDWhale
Post: Okidok.. Come on Eagle.. We have work to do ;-)) Hey Moon, I'm glad Eagle and Whiskey are offering to help you on this project... Sorry, I can't help paint this week... too much clean up work after the cyclone. I can't use the firewood myself... it doesn't stay cold long enough on the Gulf of Mexico coast to justify keeping firewood around... Adios Amigos... |
|
145)
Message boards :
Cafe SETI :
The Last Person To Post Wins #69 Not So Sunny Sunday........
(Message 808307)
Posted 15 Sep 2008 by JDWhale
Post: Thanks for the halftime show... Chargers-Broncos seemed exciting... Like the Steelers... I'm winning :-) |
|
146)
Message boards :
Cafe SETI :
The Last Person To Post Wins #69 Not So Sunny Sunday........
(Message 808277)
Posted 15 Sep 2008 by JDWhale
Post: Winning again... WooHoo...Local NBC is allowing us to watch Browns vs Steelers... Winning :) |
|
147)
Message boards :
Cafe SETI :
The Last Person To Post Wins #69 Not So Sunny Sunday........
(Message 808222)
Posted 14 Sep 2008 by JDWhale
Post: Aren't there still a few million people without power over there (and no, I am not expecting to hear from any of those people for a while). Haven't heard any figures about electricity today... Last numbers I have were from last night 112,000 customers in lesser damaged areas have been restored, more than 2 million customers still without... One problem is that there is no point even attempting to restore electricity to areas that are flooded. You folks outside the Houston area have probably seen more of the damage than us locals have. Without national news all we're seeing is the local newscasters/reporters... local news choppers are restricted from flying because of the national guard and federal disaster relief. One local news report is that the death toll is being intentionally underreported until next of kin can be notified. People staying behind in mandatory evacuation areas were asked to write SSN on their palms so they could be identified later. Sad, but true. |
|
148)
Message boards :
Cafe SETI :
The Last Person To Post Wins #69 Not So Sunny Sunday........
(Message 808210)
Posted 14 Sep 2008 by JDWhale
Post: Winning whilst the Packers are stomping the Lions...Poor little kitties.... I'm jealous... All the networks in Houston are only showing local news for past 3 days.... No national news, no Saturday Night Live, and today no football :-( |
|
149)
Message boards :
Cafe SETI :
AriZonaMoon`s web-site + AZM City ,and other relevant inputs ;-)
(Message 808155)
Posted 14 Sep 2008 by JDWhale
Post: Eagle's Flight is ranked now ranked 92769 Salute/ |
|
150)
Message boards :
Cafe SETI :
Arizona Moon`s coconut posting thread II ;-)
(Message 808154)
Posted 14 Sep 2008 by JDWhale
Post: Im off for some serious training.. ;-D That MoonWalk should help you on the golf course, too... 
|
|
151)
Message boards :
Cafe SETI :
The whatever....it is my thread thread..........
(Message 808148)
Posted 14 Sep 2008 by JDWhale
Post: I am so hard on them........dropping, kicking, throwing.....cursing....... You can make a manly man keyboard fail by cursing at it? I'm impressed!!! |
|
152)
Message boards :
Cafe SETI :
Texas Crunchers
(Message 808025)
Posted 14 Sep 2008 by JDWhale
Post: Here's Weather Underground link to a station the center of IKE passed over. Note the minimum Barometric Pressure and calm winds at 3:43AM, 13Sep and the 180 degree shift in wind direction. This station is not too far from Geek@Play home. Geek@Play : I wish you the best. |
|
153)
Message boards :
Cafe SETI :
The MOON BAR
(Message 807954)
Posted 14 Sep 2008 by JDWhale
Post:
You're quite welcome. A good caddie is hard to find...in addition to carrying the clubs, a good caddie makes beer runs when necessary :) I've got a feeling you will be kept very busy at the tournament. @Stealth Eagle: No dog paddle in my neighborhood, but I think I lost my kite :-( and there is plenty of fresh firewood laying around. |
|
154)
Message boards :
Cafe SETI :
The MOON BAR
(Message 807922)
Posted 14 Sep 2008 by JDWhale
Post: Just what i need for a nightcap or three.. ;-) Everyone knows that you can't keep a good whale down :) I'll be supplying beverages at the 'Full Moon Golf Tournament' later today... All are welcome to come out and see the new golf course and play. Show up early if you need a quick lesson. |
|
155)
Message boards :
Cafe SETI :
The MOON BAR
(Message 807891)
Posted 13 Sep 2008 by JDWhale
Post:
G@P lives very close to Galveston... I wish him the best when he comes home. Drink of the night... Mango Margarita  And the final product.... 
|
|
156)
Message boards :
Cafe SETI :
Texas Crunchers
(Message 807890)
Posted 13 Sep 2008 by JDWhale
Post: So where is JD? he should have given us a new sign by now....... :p Here I am... electricity went out at midnight local time and just came back on, after 18 hours. Had a walk about the neighborhood... a few of trees and fences down. I'got a leaky roof after losing a few shingles. Cell phone seemed to work most of the time, no loss of water pressure here in Sugar Land. You all probably know more over all story since I've been in the dark without any TV news the past 18 hours. Cheers, JDWhale |
|
157)
Message boards :
Cafe SETI :
The Last Person To Post Wins #69 Not So Sunny Sunday........
(Message 807887)
Posted 13 Sep 2008 by JDWhale
Post: I survived IKE and am winning again ! |
|
158)
Message boards :
Cafe SETI :
The Last Person To Post Wins #69 Not So Sunny Sunday........
(Message 807678)
Posted 13 Sep 2008 by JDWhale
Post: Geny is movie business slang for generator. No Geny here... I'm roughin' it :) The lights are starting to flicker every now and then and I'm hearing about transformers blowing a few miles to the northeast (in Houston)... but my crunchers haven't failed yet... |
|
159)
Message boards :
Cafe SETI :
The MOON BAR
(Message 807661)
Posted 13 Sep 2008 by JDWhale
Post: Hoping that all the Texas crunsher are safe and to thank Scarecrow for his little gift to me all drinks are on the me. Lucky for all, we are running a special on the Hurricane drink tonight... Buy one get one free :) Exchanged PM with Geek@Play, he evacuated to the northwest ~250 miles. I'm baking a pizza at the moment and looking forward to hitting golf balls and a BarBQ tomorrow... though I might not get much sleep tonight. |
|
160)
Message boards :
Cafe SETI :
The Last Person To Post Wins #69 Not So Sunny Sunday........
(Message 807642)
Posted 13 Sep 2008 by JDWhale
Post: Just waiting for IKE... I'm "hunkered down" :) ... and winning! |
|
161)
Message boards :
Cafe SETI :
The Last Person To Post Wins #69 Not So Sunny Sunday........
(Message 807626)
Posted 13 Sep 2008 by JDWhale
Post: Just waiting for IKE... |
|
162)
Message boards :
Cafe SETI :
Arizona Moon`s coconut posting thread II ;-)
(Message 807604)
Posted 12 Sep 2008 by JDWhale
Post: I found ;-D ..and i dont want to mention loud whats in it.. ehrm. Thanks for the recipe... Coconut, Chocolate, Peanut Butter... my mouth is ready to try these cookies :) |
|
163)
Message boards :
Cafe SETI :
Arizona Moon`s coconut posting thread II ;-)
(Message 807506)
Posted 12 Sep 2008 by JDWhale
Post: Geez JD.......all the coast is on edge, and you are off playing with your balls......LOL. Oooohhhh... How thoughtful... My very own Golf Ball cookie jar... Thanks Coconut... I have just the place to put it 
|
|
164)
Message boards :
Cafe SETI :
The MOON BAR
(Message 807500)
Posted 12 Sep 2008 by JDWhale
Post: The drink for today... A Rum imbibe of New Orleans fame. Starts Rum, Simple Syrup and Passion Fruit... Add Orange, Lime and Grenadine... Then shake it up to form a "Hurricane" :) 
|
|
165)
Message boards :
Cafe SETI :
Texas Crunchers
(Message 807490)
Posted 12 Sep 2008 by JDWhale
Post: Thanks for the wishes... Eye of the storm is still ~11 hours from landfall...With the adjustments in predicted path over the past 24 hours, I have moved from a direct hit with the eye crossing over the city of Sugar Land, TX... Now the eye of the storm is predicted to pass 35 miles to my northeast. If you look at a map the current expected path follows Interstate 45 up from Galveston, TX toward Dallas. This places me on the "clean" side of the storm with winds mostly coming from the north, then from the west as the eye of the storm passes nearby. Much better position to be in than if the storm track were to my southwest, putting me on the "dirty" side of the storm, where spawned tornados are expected to occur. Though the gun hasn't actually been fired, I am feeling that I have already dodged the bullet... I might feel different in 12-24 hours. Our friend Geek@Play lives in the direct path and much closer to the coast... I hope he has already evacuated. Later, JDWhale |
|
166)
Message boards :
Cafe SETI :
Arizona Moon`s coconut posting thread II ;-)
(Message 807486)
Posted 12 Sep 2008 by JDWhale
Post: Geez JD.......all the coast is on edge, and you are off playing with your balls......LOL. A day without golf is like a day without sunshine ;-) Then Later... You may better ask; What else is there to do!? hehe.. It is really a nice day still... Temp is 30C, sunshine and a steady breeze from the north... The calm before the storm... so they say :D |
|
167)
Message boards :
Cafe SETI :
Arizona Moon`s coconut posting thread II ;-)
(Message 807421)
Posted 12 Sep 2008 by JDWhale
Post: Now i have to go and eat dinner with the boys.. I made pancakes Yes, I just had smokey bacon for breakfast... Yumm... Your boys know what is good ;-)
The sea water is beginning to cross over Galveston Island. I'll post as long as we have electricity and cable modem has connection... For now I'm going to the golf practice range with my son to hit some golf balls. It's not too often I get to practice in 40km/h winds. I fear all flights have been cancelled to/from Houston... so the paint will have to wait a few days :) |
|
168)
Message boards :
Number crunching :
Crunch this...
(Message 807408)
Posted 12 Sep 2008 by JDWhale
Post: x+y*pi(3.141592653589793238462643383279502884197169399311481966593000574 according to XP x64 sp2)=? He is a Joker... Try this... Why pi + x ? |
|
169)
Message boards :
Cafe SETI :
The MOON BAR
(Message 807402)
Posted 12 Sep 2008 by JDWhale
Post:
Sorry for the wait.... Quite a madhouse here at the moment... Do enjoy... [edit]Oops... ran out of the last stuff... Try this..[/edit]  Time to grab the surf board and head to the beach..."Catch-a-Wave". NOT!!! |
|
170)
Message boards :
Cafe SETI :
Arizona Moon`s coconut posting thread II ;-)
(Message 807399)
Posted 12 Sep 2008 by JDWhale
Post: [quote]... Oh.. and I ment to paint that table.. its just one of the things i never reached.. *sigh*. JD - will you do it? ;-D [quote] I'm all over it...paint brush in hand... paint on nose :o ...thinking color white, to match the chairs... |
|
171)
Message boards :
Cafe SETI :
Texas Crunchers
(Message 807232)
Posted 12 Sep 2008 by JDWhale
Post: Had do pick my son from the airport this evening... The evacuation from Galveston Island went very smoothly earlier today, by this evening there was only light traffic along the evacuation route. The weather reports are looking a bit more favorable for me... Latest models show the eye of the storm path having moved from where I live to about 30 miles to my northeast... straight through downtown Houston. With possible 20 ft storm surge up the Houston ship channel. Also known as "refinery row". Should have a much better idea on path in 8-10 hours. Cheers, JDWhale |
|
172)
Message boards :
Cafe SETI :
The MOON BAR
(Message 807147)
Posted 11 Sep 2008 by JDWhale
Post: I hope that everyone enjoyed the B-52s... But today is another night... Can anyone guess the drink of the day ??? Here's the fixin's...  |
|
173)
Message boards :
Cafe SETI :
Arizona Moon`s coconut posting thread II ;-)
(Message 807067)
Posted 11 Sep 2008 by JDWhale
Post: Ok.. what was I suppose to do..? It was these cabinets..vanilla..aye, what did you think? ;-) Thanks for sharing... your cabinets do look Vanilla... Maybe even Baileys colour :) Yummmmmm... I was mistaken when I offered to paint your kitchen, it's quite lovely already... only thing missing is your coconut radio...  What is that I find on the counter top? Bombay Sapphire... I like mine with a twist of lime... Yes, it seems hurricane IKE is on a direct path to my home... Fortunately I'm ~80km inland so flooding from the storm surge won't be a problem, but the winds expect to be ~160km/h when the center of the storm passes through my city. |
|
174)
Message boards :
Cafe SETI :
The MOON BAR
(Message 807012)
Posted 11 Sep 2008 by JDWhale
Post: Sorry to keep y'all waiting... We ran out of the over-proof rum, no fire for now... Here you go...   Nothing like a b-52 for breakfast... goes great in coffee to help get the day underway :) uli wrote: Laying on the grass, looking up, I noticed a planet. My telescope is a bit out of shape, so I used the binoculars. Venus is plenty bright, sets in the evening... Jupiter is the next brightest and is staying up most of the night and was near the moon last night. With binoculars, you should be able to see 4 or 5 of Jupiters larger moons. |
|
175)
Message boards :
Cafe SETI :
The MOON BAR
(Message 806939)
Posted 11 Sep 2008 by JDWhale
Post: I had posted a while back in a thread to Johnny Guinness......but here you go.... Two or Three shots of 101 Turkey? Are you effin Crazy? I wanna see a picture... Beer on fire... You gotta be NUTZZZ! Nowhere but The MOON BAR!!! |
|
176)
Message boards :
Cafe SETI :
The MOON BAR
(Message 806927)
Posted 11 Sep 2008 by JDWhale
Post: You seem to have thing about open flames............ Flaming Guinness ??? Please enlighten us. B-52 is a wonderful Pousse-Cafe with carefully constructed layers of Kahlua, Baileys, & Grand Marnier... It only becomes flaming B-52 when you pour a thin layer of over-proof rum on top and light it up. Tonight is the trial run and the drinks are free... B-52 night on Fridays ain't gonna be cheap... (You know how expensive insurance is after the last fire ?) Drink up while you can...  I used to drink these non-stop on the Thai Airlines flights from Beijing to Bangkok... Ahhh...Tiananmen tanks for the memories... |
|
177)
Message boards :
Cafe SETI :
AriZonaMoon`s web-site + AZM City ,and other relevant inputs ;-)
(Message 806901)
Posted 11 Sep 2008 by JDWhale
Post: The unemployment is my fault this time. Who let the criminal in, I don't know. Sorry, I've never been that kind of bad... Only a bit mischievous :-D |
|
178)
Message boards :
Number crunching :
Crunch this...
(Message 806897)
Posted 11 Sep 2008 by JDWhale
Post: ...and don't stop! If you code it correctly... there's never overflow... int x; while (true) x++; At least not a detected overflow ;-D [edit/] |
|
179)
Message boards :
Cafe SETI :
The MOON BAR
(Message 806891)
Posted 11 Sep 2008 by JDWhale
Post: You know, for a Bar this one is awful quiet. I think I stepped all over Uli'  in the garden... I am soooo sorry :oCan't remember much, but I did have dirty  :(Get the fire extinguishers to the ready... Because I'm ringing the bell... A round of flaming B-52s on me... There is a bell, isn't there ??? 
|
|
180)
Message boards :
Cafe SETI :
Texas Crunchers
(Message 806868)
Posted 11 Sep 2008 by JDWhale
Post: Thanks all... A lot of these storms tend North/clockwise near landfall... Us near Houston might be looking right down this IKE SOBs throat in 48 hours. It's been 25 years since I was in the eye of a hurricane, Alicia 1983, it's not alot of fun... Cheers, JDWhale |
|
181)
Message boards :
Number crunching :
Panic Mode On (8) Server problems
(Message 806844)
Posted 11 Sep 2008 by JDWhale
Post: Really no PANIC... just an opportunity to cancel the 'undesirables' to help feed those that are hungry ;-) C'mon... we've all been here before... there is no need to panic. |
|
182)
Message boards :
Number crunching :
Crunch this...
(Message 806829)
Posted 11 Sep 2008 by JDWhale
Post: ...and don't stop! int x; x++; |
|
183)
Message boards :
Cafe SETI :
Arizona Moon`s coconut posting thread II ;-)
(Message 806828)
Posted 11 Sep 2008 by JDWhale
Post: Good Morning! ;-) Mmmhh... Thank you! ;-) My house is more lemon colored than orange colored i think.. but its nice you like it anyhow.. I was on comp here looking for a pic of that orange cheese cake..i Know i have one, but it has hided away for some reason..lol.. It probably didnt want to be exposed ;-D Lemon, Orange, Lime... All are delicious with vanilla ice cream.. and cantaloupe  too :-DYes, these tropical storms (IKE) cause a big scare... The last one to come close, "Rita", cost my marriage 3 years ago. Maybe I can play golf again in the morning... I only play Monday-Thursday mornings, so maybe the storm will not be too much problem if it comes Friday night. G'night |
|
184)
Message boards :
Number crunching :
The new Seti@home GFX
(Message 806824)
Posted 10 Sep 2008 by JDWhale
Post: As a returning longtime user, i noticed that the gfx has changed, gone is the statistics over possible finds, and in is a irritating moving gl thingie that's impossible to focus on without getting dizzy. What gfx??? I've never seen any... Are there supposed to be gfx? I've been robbed!!! Call the Police!!! |
|
185)
Message boards :
Number crunching :
Astropulse FAQ
(Message 806817)
Posted 10 Sep 2008 by JDWhale
Post: OK - so I can run AK_V8 for MB data and AP will still work as a seperate application.... Yes, but you will have to manually install/configure Astropulse... But, I recommend waiting a while before starting Astropulse... I think I see an optimized version on the near horizon. Regards, JDWhale |
|
186)
Message boards :
Cafe SETI :
AriZonaMoon`s web-site + AZM City ,and other relevant inputs ;-)
(Message 806803)
Posted 10 Sep 2008 by JDWhale
Post: The unemployment is my fault this time. Who let the criminal in, I don't know. Sorry... I might be the 'criminal' element here... I heard somewhere that women perfer 'bad' boys :-D |
|
187)
Message boards :
Number crunching :
Host hitting 400 quota every day, now what?
(Message 806787)
Posted 10 Sep 2008 by JDWhale
Post: JDWhale, Sorry to hear of your recent losses :-( As you probably know, performance depends on a few factors... Your "Dads" box has higher BOINC benchmarks than Marks Frozen Penny(QX9650), but he recently switched over to DDR3 RAM and he is running at a higher freqency. Additionally, he's not getting the benefit from crunching all the VHAR WUs that you are. Hard to get an apples-to-apples comparison... One thing remains certain, when it comes to bang-for-the-buck it still seems single quad-core boxes rule... Too much overhead and contention for bandwith when running more than 4 cores, IMO. [quote]if (SETI client gets faster) {
you'll be waiting for work
} ;-)
[/quote]Regards,JDWhale |
|
188)
Message boards :
Number crunching :
Host hitting 400 quota every day, now what?
(Message 806761)
Posted 10 Sep 2008 by JDWhale
Post:
Doing the math on a few of the WUs you've returned... You are crunching VHAR WUs in ~9 minutes. If you were only to crunch these you could complete 600+ WUs/day. Figuring that you will receive a mix of Angle Ranges from the project and the speed that you are crunching the MidAR WUs (~0.44 AR), ~32 minutes each, it is conceivable that you will crunch more than the 400WU/day download limit for your CPU on some days. This is mainly due to the proliferation of the shorties (VHARs) currently being distributed. Expect your cache to build slowly if at all. I figure your new host might make the top 10 hosts with RAC > 8000 if the project can keep feeding you enough work. Nice box you've built. Regards, JDWhale |
|
189)
Message boards :
Cafe SETI :
Arizona Moon`s coconut posting thread II ;-)
(Message 806676)
Posted 10 Sep 2008 by JDWhale
Post: :-) I can understand your choise Sir.. Since its up to you, I`ll just sit here and watch it become as you want it.. But of course.. do you need it, i will help out. btw.. do you know what colour my kitchen cabinets have? hehe.. maybe i can find a pic.. :p Oops... I just look at the AZMoon home pictures... Beautiful! Seems your house is nearly "Dreamsicle" color already... Honestly, I didn't know :-D Maybe we could consider "Coconut" color accents for the kitchen ... Off to play final round of golf before Hurricane Ike forces evacuation along the coast and brings the nearby roads to a stand still. (I won't have to evacuate, but I live beside the evacuation road.) |
|
190)
Message boards :
Cafe SETI :
Arizona Moon`s coconut posting thread II ;-)
(Message 806540)
Posted 9 Sep 2008 by JDWhale
Post: What colour should the kitchen have then? Or maybe you allready have bought the painting ;-) keep me updated. Thank you so much for the chocolate... I love the hazelnut/filbert... Delicious :) I'm still thinking about that Orange-Cheescake that you talk about ;-) In that theme... I'm tending toward "Orange Creamsicle" color for the kitchen with vanila color ceiling and cabinets.   Do you have Creamsicle in Norway? It's an orange flavor popsicle with vanila ice-cream inside. Yummy... Do you like? |
|
191)
Message boards :
Number crunching :
AK V8 ported release ap. issues, install, questions etc.
(Message 806498)
Posted 9 Sep 2008 by JDWhale
Post: Update: Your T7500 is a Core2Duo... It should be able to run SSSE3 client for a bit more performance than SSE3. This CPU is 65nm fabrication and thus not SSE4 capable. Regards, JDWhale |
|
192)
Message boards :
Cafe SETI :
Arizona Moon`s coconut posting thread II ;-)
(Message 806341)
Posted 9 Sep 2008 by JDWhale
Post: To the one who help me clean up this place.. When the job is done, you can have Okay... the dishes are washed and put away... I'll need a ladder to wash the upstairs windows, better wait 'til morning, can't see the streaks in the dark... No point making the beds this time of night ;-) But I can still paint the kitchen. |
|
193)
Message boards :
Number crunching :
New error -131 on file upload
(Message 806305)
Posted 8 Sep 2008 by JDWhale
Post: wuid=327251307 Agreed. Beyond the lost science, these WUs are a total waste of project resources... Waste of splitter time and storage resources... Each WU is sent out 6 times, wasting bandwith resources... Each WU is crunched 6 times, wasting volunteer computing resources... Thankfully, we've only seen VHAR WUs up to present. |
|
194)
Message boards :
Number crunching :
Ideas for Seti
(Message 806101)
Posted 8 Sep 2008 by JDWhale
Post: Now with helping to set up a 501c for Seti and all the work envolved, I think I lost it Wish I could help with the paperwork... But I'm just an ex-programmer who likes to play golf ;-( |
|
195)
Message boards :
Number crunching :
New error -131 on file upload
(Message 806098)
Posted 8 Sep 2008 by JDWhale
Post: wuid=327251307 Am I the only person upset crunching these WUs and not getting any credit? Shouldn't we get credit for work completed? |
|
196)
Message boards :
Number crunching :
Claimed credit zero
(Message 806035)
Posted 8 Sep 2008 by JDWhale
Post: Pooh Bear 27 wrote: <core_client_version>4.43</core_client_version> Good time to get all BOINC 4.xx users to upgrade... Question remains... How? |
|
197)
Message boards :
Cafe SETI :
Arizona Moon`s coconut posting thread II ;-)
(Message 806024)
Posted 7 Sep 2008 by JDWhale
Post: I have heard there are Birthdays on the go.. but since im not sure who has birthday when, I just mixed together some cakes, and invite you all in for a nice joining.. ;-) My BD is still a month away, but I can't resist a Lime cheesecake... Please save me a slice or two or three... Pretty please... with Baileys on top ;-) |
|
198)
Message boards :
Cafe SETI :
FEDEX Cup
(Message 806011)
Posted 7 Sep 2008 by JDWhale
Post: Ooops. Sorry. Yer right again. It's the Tour Championship not the TPC. My mistake. Thankfully, they aren't playing in Atlanta next week... I fear Hurricane Ike will be in the area... Hopefully Valhalla in Kentucky will be in perfect condition for the Riders Cup in 2 weeks ;-) Sure hope Ike doesn't head toward Houston. Cheers, JDWhale |
|
199)
Message boards :
Cafe SETI :
FEDEX Cup
(Message 805999)
Posted 7 Sep 2008 by JDWhale
Post:
Isn't the final FedEx Cup tournament at East Lake Golf Club in Atlanta in 3 weeks? Bobby Jones' home course. Maybe VJ is playing TPC Sawgrass next week, but not in any tournament I'm aware of. BTW... Kudos to Camillo Villegas on a fantastic wire-to-wire performance at the BMW!!! |
|
200)
Message boards :
Cafe SETI :
The MOON BAR
(Message 805665)
Posted 7 Sep 2008 by JDWhale
Post: ehrmm.. do we have any helmets?? ;-p This should offer some protection... might not help you sing though...lol  |
|
201)
Message boards :
Cafe SETI :
The MOON BAR
(Message 805389)
Posted 6 Sep 2008 by JDWhale
Post: I was thinking of donating a dart board... maybe right wall in the back... far, far away from the podium, for safety sake ;-) 
|
|
202)
Message boards :
Number crunching :
New error -131 on file upload
(Message 805301)
Posted 5 Sep 2008 by JDWhale
Post: wuid=327061387 Seems Berkeley is still splitting these oversized WUs. |
|
203)
Message boards :
Cafe SETI :
"Simple" Maths Problems II
(Message 805274)
Posted 5 Sep 2008 by JDWhale
Post: Q30 (1 point) : The front wheels of a wagon measure 3.5 feet in diameter. The real wheels measure 4.25 feet in diameter. While the wagon is stopped somebody makes a chalk mark on both a front and rear wheel. How far must the wagon travel before both chalks marks return to their initial position at the same time? A30 : ~186.925 |
|
204)
Message boards :
Number crunching :
Ideas for Seti
(Message 805169)
Posted 5 Sep 2008 by JDWhale
Post: To answer a question posted earlier. Astropulse is a funded project. Huh? Astropulse is funded, Seti@home is not? How can Astropulse ride piggyback on unfunded servers & resources ? Seems backwards to me. I smell rotten fish... Afterthought: Let's get some cash from Astropulse... |
|
205)
Message boards :
Number crunching :
Ideas for Seti
(Message 805094)
Posted 5 Sep 2008 by JDWhale
Post: As far back as I can remember, SETI@home was a volunteer project. "Install our screensaver and supply unused computer cycles to a search for ETI (SETI)". Has that mandate changed ? This "Pay to Crunch" suggestion implies that the S@H project no longer wants our computer cycles unless we pay them ca$h to accept our computing donation. I smell rotten fish... Regards, JDWhale |
|
206)
Message boards :
Cafe SETI :
AriZonaMoon`s web-site + AZM City ,and other relevant inputs ;-)
(Message 805010)
Posted 4 Sep 2008 by JDWhale
Post: What..?? oh no.. for the first time the city has gone down on the list from 113 to 114.. Have anyone here desided to resque the city in front?? hehe.. Time to procreate (make some babies ;-) (must be Tequila Time)... or... Maybe we can use transporter from Star Trek..."Scotty... Beam me up"...LOL |
|
207)
Message boards :
Cafe SETI :
The MOON BAR
(Message 804451)
Posted 3 Sep 2008 by JDWhale
Post: Have one on me...hic...uuuurp... Yummy... Could I get some  hot sauce with these?
|
|
208)
Message boards :
Number crunching :
Astropulse FAQ
(Message 804437)
Posted 3 Sep 2008 by JDWhale
Post: Thank you for the reply Richard. What are you saying... 34% complete after 2 weeks??? Then you are expected 4 weeks more 'til completion? Do not believe the 'remaining time' shown.. calculate the 'true remaining time' yourself from the %complete and elapsed CPU time. This will miss deadline and be reissued with the possibility that you will not receive any credit. Your PIII should not have received any AP WUs because your host cannot complete before deadline!!! |
|
209)
Message boards :
Technical News :
Tuesday Twinge (Sep 02 2008)
(Message 804366)
Posted 2 Sep 2008 by JDWhale
Post: Meanwhile, we're out of space again on the workunit server, and with no fast/easy way to add space. Eric's playing with the splitter mix to reduce the number of Astropulse workunits being generated (they are much larger than SETI@home workunits). Maybe that will help, but not immediately. This is what's mostly causing our headaches today as we can't create enough work to keep up with demand. Maybe if Astropulse paid more, there'd be more demand for its WUs... I'm about to finish up an AP WU sent to me 3 weeks ago, but it's worth less when it was issued to me than now... This will not thrill my new wingman who is claiming more credit than me. In the economy this is call deflation :-( WUs should be worth more over time rather than less... There is no investment value if the claim doesn't increase over time with new wingmen! How can the same WU be worth more credit just because it had to be reissued? Maybe all users should cancel their AP WUs so the WUs will increase in value! |
|
210)
Message boards :
Number crunching :
New Credit Adjustment?
(Message 803546)
Posted 31 Aug 2008 by JDWhale
Post: jreinstedler wrote: Seen the same thing. Can ANYONE with the project please verify if there was an adjustment? 300+ posts to get this across with speculation is BS. From what I have seen, it looks like there's been a 20% drop in RAC. After adding a couple more systems, I was looking at RAC of around 72k, but am barely holding onto 60k when my systems dump their data every other day... Speculation started this thread... Confirmation and an explanation followed a couple days later, also in this thread... Here is a link to a post from Eric Korpela, "SOMEONE with the project"...Message ID 789160 Kind Regards, JDWhale |
|
211)
Message boards :
Cafe SETI :
The MOON BAR
(Message 803513)
Posted 30 Aug 2008 by JDWhale
Post: Looking for coffee, Gota' have coffee OMG!!! That looks like the ALIEN probe in my nightmares... Kinda reminds me of mines protecting SF Bay during WWII... Beware of the 'Death Ray' when 'frothing'...lol |
|
212)
Message boards :
Number crunching :
New error -131 on file upload
(Message 803325)
Posted 30 Aug 2008 by JDWhale
Post: wuid=322221628 Guess Berkeley is still sending these out.... Here's another one where the first 2 hosts returned the -131 error, but one host sucessfully returned result while the other did not. Why the different behavior??? |
|
213)
Message boards :
Number crunching :
RAC dropping fast
(Message 803319)
Posted 30 Aug 2008 by JDWhale
Post: i too have a rac reduction well below the so called "15% correction". before all this credit mess and cp talk my rac was around 5500 to 5600. now it is at 4298. heck of a drop of more than the so called for 15%. nothing else has changed on any of my computers i crunch on. even my pending has been about the same. so something seems fubared by all this....no 15% credit reduction on my end, that's for sure. Maybe that does explain the extra reduction... on my hosts the "shorties" pay the best RAC. Killing them only hurts your RAC more. It's the old 63.98 credit WUs now earning ~52 credit that were recently distributed in excess and caused increased drop in my RAC. Anyway, whatever your reasons... we're sorry to see you go. Regards, JDWhale |
|
214)
Message boards :
Number crunching :
RAC dropping fast
(Message 803307)
Posted 30 Aug 2008 by JDWhale
Post: i too have a rac reduction well below the so called "15% correction". before all this credit mess and cp talk my rac was around 5500 to 5600. now it is at 4298. heck of a drop of more than the so called for 15%. nothing else has changed on any of my computers i crunch on. even my pending has been about the same. so something seems fubared by all this....no 15% credit reduction on my end, that's for sure. There has been a significant number of VLAR WUs (AR 0.008-0.010) recently. The 'pay' for crunching these is much less than Mid AR WUs. This might attribute to the recent increased loss of RAC on some model CPUs running AK V8. I've noticed it on my Intel Core2 based hosts. |
|
215)
Message boards :
Number crunching :
Claimed credit zero
(Message 803189)
Posted 29 Aug 2008 by JDWhale
Post: Just checked the results of my wingman everyone he has done has claimed zero credit, sois something wrong with his computer? With this other piece that web03 mention I have got another one where I got half of the credit that I claimed with this Can't set up shared mem: -1, is this a problem with Linux wuid=319707490 I think this the WU you refer... Another case of your wingman running BOINC 4.43. We know all hosts running BOINC 4.xx underclaim credit with MultiBeam WU... Not just a problem with Linux. |
|
216)
Message boards :
Number crunching :
Claimed credit zero
(Message 803115)
Posted 29 Aug 2008 by JDWhale
Post: I am not a credit hunter what got me mad was this person had been already granted zero credits when mine had not gone up. Can anyone explain this little thing why was it granted zero without me his wingman sending in my report All WU's show "granted credit 0" before validation...
wuid=321702376 |
|
217)
Message boards :
Number crunching :
New error -131 on file upload
(Message 802584)
Posted 27 Aug 2008 by JDWhale
Post: wuid=321342238 Hmmm... This one is strange... 3 hosts returned -131 error, but one host seems successful returning the result regardless of the error. Have adjustments been made to the servers so quickly ? |
|
218)
Message boards :
Cafe SETI :
The MOON BAR
(Message 802444)
Posted 27 Aug 2008 by JDWhale
Post: Installs a podium in the back corner of the bar. Well away from the door... Hopefully... 'How not to set your hair on fire while cooking with Bacardi 151 rum' ;-) [sorry... personal issue (lol++)[wink,wink] :-) peace...out...] |
|
219)
Message boards :
Number crunching :
New error -131 on file upload
(Message 802443)
Posted 27 Aug 2008 by JDWhale
Post: wuid=321214337 Offhand I see ~409 instances of <coordinate_t> totaling ~52KB in header?? @Joe - See PM for link to WU. |
|
220)
Message boards :
Cafe SETI :
Olympic Games 2008 in Beijing
(Message 802437)
Posted 27 Aug 2008 by JDWhale
Post: I hear that 'dog' is back on the menu, now that the Olympics are over. They fed us dog once, but we didn't find out 'til 2 days later... The camel foot tasted better. |
|
221)
Message boards :
Number crunching :
New error -131 on file upload
(Message 802398)
Posted 27 Aug 2008 by JDWhale
Post: wuid=321214337 Here's one I just DL'd... has -131 error by 3 previous crunchers. Noticed WU filesize is 411 KB rather than the standard 367 KB. I'll make a copy of the WU available for anyone interested. Regards, JDWhale |
|
222)
Message boards :
Number crunching :
New error -131 on file upload
(Message 802363)
Posted 27 Aug 2008 by JDWhale
Post: We've all got it for WU 320848817 - three x stock v6.03, and my AK_V8 SSSE3x. Here are a couple that were reported a week ago: Message 799367 wuid=316574417 wuid=312048949 |
|
223)
Message boards :
Number crunching :
I no longer care about Seti.
(Message 802093)
Posted 25 Aug 2008 by JDWhale
Post: A Dollar Bill was gold. Might the 15% degradion in credit be attributed to the release of 6.03 client though never stated as a reason; In casual observation I seem to see ~10% boost in performance over 5.28 client. I'd be interested in CPU times of same AR WUs on your G3 running 5.28 vs. 6.03, post 'em if you got 'em. |
|
224)
Message boards :
Cafe SETI :
Arizona Moon`s coconut posting thread II ;-)
(Message 802015)
Posted 25 Aug 2008 by JDWhale
Post:
And good day to you Ms Moon.. I find my eyebrows are now gone... The bandages come off next week...lol Lesson learned... Don't run when your hair is on fire ;-) |
|
225)
Message boards :
Cafe SETI :
Arizona Moon`s coconut posting thread II ;-)
(Message 801934)
Posted 25 Aug 2008 by JDWhale
Post: JD it depends on where you live. If you live east of me it is morning if you live west of me it is night. This I know... Hiked the Trinity Alps and Siskiyou mountains as a Boy Scout when I lived in Humboldt County, Ca. Stealth Eagle... you live in Gods Country. |
|
226)
Message boards :
Number crunching :
AstroPulse errors - Reporting
(Message 801929)
Posted 25 Aug 2008 by JDWhale
Post: Looks like you got bitten by the same bug as me, I lost an AP unit as well when it was suspended. Okay... I don't see how the numbers are changing... it's going to take you ~300 CPU hours to complete. About 6 more days... post every day if you wish, but this is the fact and we have already been there... Astropulse underclaims credits and does not estimate the correct CPU time. Welcome to the club! Good night, JDWhale |
|
227)
Message boards :
Number crunching :
New Credit Adjustment?
(Message 801925)
Posted 25 Aug 2008 by JDWhale
Post: my rac has decreased from 3800 to 2100 on a e6700 processor machine using the V8 ssee3.1 app. Maybe you chould check your numbers... The way I do the math... it's closer to a 45% drop... |
|
228)
Message boards :
Cafe SETI :
Arizona Moon`s coconut posting thread II ;-)
(Message 801914)
Posted 25 Aug 2008 by JDWhale
Post: Hmmmm... coconuts posting thread.... Kermit the Frog - Lime in the Coconut Is it morning yet? What do I do? "Call me in the morning" |
|
229)
Message boards :
Number crunching :
Upgrade
(Message 801887)
Posted 25 Aug 2008 by JDWhale
Post: Just ordered a gig of Ram from crucial.com for my single core AMD. Will bring it to the max of 3 gig. Will have to watch for improvement. Seti@Home does't require more than 50MB/instance tops even running Astropulse... Maybe your games need more memory... I don't know, I don't play games... I find Compiling/linking programs using MS Visual Studio 2008 while browsing 4 instances of IE7.0, running OpenOffice 2.4 and running 4 cores of S@H on my Q6600 host w/Vista Ultimate... 2 gig is more than enough. Any more would be a waste to me. |
|
230)
Message boards :
Cafe SETI :
Olympic Games 2008 in Beijing
(Message 801855)
Posted 25 Aug 2008 by JDWhale
Post: What is the minimum age to participate in the Olympics? And does that vary year to year or is it set in stone? Currently: Gymnastics is 16 Years in year of Olympic games. Maybe Track & Field is a bit older. Diving is 14 years(Thomas Daley, GBR). I remember seeing Greg Louganis hanging at Narita Airport, Tokyo, with all the young boys of the US Diving Team enroute the 1988 games in Seoul, Korea (I was headed to Beijing). Looked like a kid in a candy store. |
|
231)
Message boards :
Number crunching :
I no longer care about Seti.
(Message 801825)
Posted 25 Aug 2008 by JDWhale
Post: I was really thrilled when Seti started, imagine looking for life outside of Earth. Me, too... When I started running SETI 'Classic' back in 1999, I knew there would not be results in my lifetime, that this was a 'long term' commitment. I am happy that the client is faster now and we are nearly able to keep pace with the gathering of data, but more optimizations will be required in order to crunch increasing data streams. I do not have interest in computing for other projects or in what credit they might claim. It they choose to keep their source code to themselves, how can they have the fastest clients? Maybe they do not have the amount of data to crunch as SETI... Maybe they do not require optimized clients. |
|
232)
Message boards :
Number crunching :
Upgrade
(Message 801813)
Posted 24 Aug 2008 by JDWhale
Post: Just ordered a gig of Ram from crucial.com for my single core AMD. Will bring it to the max of 3 gig. Will have to watch for improvement. Unless your computer is doing other memory intensive work simultaneous while crunching SETI, I would not expect any improvement. And if that is the case, IMO, since it is single core there are other bottlenecks than memory. |
|
233)
Message boards :
Cafe SETI :
The MOON BAR
(Message 801134)
Posted 23 Aug 2008 by JDWhale
Post: I have never had a banana flambay(?), in Europe we used plantains. I will see you outside for a sampler. JD, that looks good. Can I have some for dessert? Clear the way.... Dessert on fire!!! 
|
|
234)
Message boards :
Cafe SETI :
The MOON BAR
(Message 801107)
Posted 23 Aug 2008 by JDWhale
Post: I will second that in celebration of your recent win. BTW, the smoking section is outside on the patio tables, well away from the doors and windows. I was wanting to serve up Bananas Foster for anyone with a sweet tooth... Maybe I should do it outside...  @Rebest : Can I borrow a shot of that Bacardi 151 ? |
|
235)
Message boards :
Number crunching :
Question on Astropulse versus regular WU credit
(Message 801057)
Posted 23 Aug 2008 by JDWhale
Post: [ Which explains why the whale had an interesting and active, though brief existence, and the bowl of petunias simply said "oh, no, not again." Oh yeah, it's been 25 years since I read the books... back in the 'stone age' to some of our younger members ;-) Guess I should 'brush up' since on the history since I'm the 'Whale at the Center of the Galaxy at the Edge of the Universe'... ;-) Cheers, BTW: I still recommend 'Magnolia'... not your average 'chick' flick. |
|
236)
Message boards :
Number crunching :
Question on Astropulse versus regular WU credit
(Message 801044)
Posted 22 Aug 2008 by JDWhale
Post: [ Which explains why the whale had an interesting and active, though brief existence, and the bowl of petunias simply said "oh, no, not again." Have you seen the movie Magnolia ? Opening scene... frogs... Some days... Weird Just Happens ! or something like that ;-) Cheers, |
|
237)
Message boards :
Cafe SETI :
The MOON BAR
(Message 801018)
Posted 22 Aug 2008 by JDWhale
Post: Ok, I found a glass that Moon hasn't thrown at someone. I'll have a Tequila now :) WOW... 'bullet proof glasses'... Why Ms. Moon...You're a GENIUS!!! I told you that drinking tequila makes you smarter ;-))) Barkeep... Another round of tequila shots on me... Whoa... Hold on to your horses Moon ;~[ You were saving that glass to throw at me ? Better make yours a double shot... You ain't smart enough yet ;-D |
|
238)
Message boards :
Number crunching :
Question on Astropulse versus regular WU credit
(Message 801013)
Posted 22 Aug 2008 by JDWhale
Post:
In the case of AP, all days have been "wierd" so far... The slowest hosts haven't had enough elapsed time to report any results yet... So the fleet is looking faster than it actually is. |
|
239)
Message boards :
Technical News :
Blips and Bursts (Aug 07 2008)
(Message 801007)
Posted 22 Aug 2008 by JDWhale
Post: First, I'm sorry for showing my newbie ignorance. Please be kind. Welcome to the message boards. A couple possibilities for Host Averages(RAC) dropping... 1) Recent drop of credit per WU being enacted by the project. 2) Introduction of Astropulse WUs that are currently underclaiming credit. If you make your hosts visible, the forum experts might be able to offer some additional help or suggestions. Lots of good reading over in the Number Crunching forum. Kind Regards, JDWhale |
|
240)
Message boards :
Cafe SETI :
The MOON BAR
(Message 800949)
Posted 22 Aug 2008 by JDWhale
Post: Thanx! yes.. thats what i prefer. ;-) salt and lemon as long as there is..and No problemo...Drinking tequila makes you smarter... In my case... makes me 10 feet tall and bullet-proof... LOL ;-) |
|
241)
Message boards :
Cafe SETI :
The MOON BAR
(Message 800934)
Posted 22 Aug 2008 by JDWhale
Post: but if you live in Texas, you may have been and bought some Tequila in neighbourhood.. if so, can I have some of that? ;-) My Dear AZMoon... Tequila is the 'active' ingredient in the Margarita... Maybe you prefer a shot........... 1) lick the salt 2) drink the shot 3) bite the lime 4) repeat...   Oh... I see that you've done this before.Bartender... Another shot for the lady, please. |
|
242)
Message boards :
Cafe SETI :
The MOON BAR
(Message 800883)
Posted 22 Aug 2008 by JDWhale
Post:
Kissin' frogs ??? Still looking for your prince, eh ? Okay... So 'Frozen Mint Daiquari' isn't your cup o' tea... But it's hot here in Texas... and we like our drinks cold :-D Next in the blender... Strawberry-Mango-Margaritas (aka: Jupiter Red Spot)...   |
|
243)
Message boards :
Number crunching :
Question on Astropulse versus regular WU credit
(Message 800590)
Posted 21 Aug 2008 by JDWhale
Post: Thanks JD what ya got for AMD SSE3? Looks like you're already running it ;-) Cheers, |
|
244)
Message boards :
Number crunching :
Question on Astropulse versus regular WU credit
(Message 800586)
Posted 21 Aug 2008 by JDWhale
Post: DaBrat and DaBear wrote: ...the machine was not optimized as a matter of fact there really is no difference in performance. If you're interested... I've heard that AMD SSE capable hosts perform better with this optimized client...2.4V_Windows_x32_SSE_FFTW Sorry that we were not successful building a faster "AK V8" client for your model AMD :-( Regards, JDWhale P.S. Yes... I know that you didn't optimize 'til afterward... just want you aware of the 'best' optimized client for your host... it might make a difference. |
|
245)
Message boards :
Number crunching :
Nice credit claim.....
(Message 800557)
Posted 21 Aug 2008 by JDWhale
Post: @Mark - Just looked at your result...resultid=943904987 Any particular reason for running a pre-release version ? SSSE3x Win32 RC1 Build 35 , Ported by : Jason G, Raistmer, JDWhale Regards, JDWhale |
|
246)
Message boards :
Number crunching :
linux or windows best for optimised app?
(Message 800346)
Posted 21 Aug 2008 by JDWhale
Post: My wife and I are building our first from the ground up computer, a quad 6600 SETI/Rosetta cruncher. On the cheap... Linux ubuntu 8.04 has a better price :-) than OEM Windows XP or Vista, but does anyone know which one will run the optimized SETI app faster. If the differences are really minor, we'll just use Linux, at least for now. Check the Top Hosts... go ahead and start with Ubuntu...(it's free)... Optimized Linux SETI is about the same speed as on Windows... Just make sure that it's "AK V8" for best results... ;-) Cheers, JDWhale |
|
247)
Message boards :
Number crunching :
4 WU's per day quota limit
(Message 800304)
Posted 21 Aug 2008 by JDWhale
Post: So I just saw that my PC wasn't crunching any WUs But a week or so ago you changed from BOINC 5.10.28 to BOINC 6.2.16... there are known problems that have been posted in this forum... switch back to BOINC 5.10.xx and I bet your problems go away... Cheers, JDWhale |
|
248)
Message boards :
Number crunching :
How to develop an optimized application?
(Message 800293)
Posted 21 Aug 2008 by JDWhale
Post: Hi! Q. How do I develop an optimized application? A. Look at the original code and find a better way to code the algorithms (or start with the already optimized client from lunatics). Q. Are there free compilers? A. Yes... Intel offers 30-day evaluation of latest compilers... Just visit intel.com. But you will also require Microsoft Visual Studio(also free with 90-day evaluation) and a working knowledge of how to use it. Also, fetch an evaluation copy of Intel Performance Primitives (IPP) to use optimized FFT, Convolution, and Transpose functions. Q. How should one test optimized application? A. Download/run 'KnaBench' from lunatics Q: Where to get a test workunit? A. Included with 'KnaBench'. Gluck, JDWhale |
|
249)
Message boards :
Number crunching :
4 WU's per day quota limit
(Message 800281)
Posted 21 Aug 2008 by JDWhale
Post: So I just saw that my PC wasn't crunching any WUs Can't tell anything because your hosts are 'hidden'... If you want to remain 'anonymous' just point us to a workunit... Regards, JDWhale |
|
250)
Message boards :
Cafe SETI :
The MOON BAR
(Message 800278)
Posted 21 Aug 2008 by JDWhale
Post: Much better today... Bartender; A round of drinks on me...    
|
|
251)
Message boards :
Technical News :
Blips and Bursts (Aug 07 2008)
(Message 800274)
Posted 21 Aug 2008 by JDWhale
Post: Towards the end of the afternoon yesterday we put in a new scheduler to fix a bug with "anonymous platforms" and the way they handle Astropulse workunits. This is working fine as far as I know, but at first there were some brief issues with uploads in general (human error when installing new scheduler). I ain't at Berkeley but... There was a rough patch about a week ago... folks worked late and gave up their weekend to make things right. Everything is running better than expected... WUs are being sent out... Results are being received... Everyone is sleeping better at night as a result. Oh yeah... I haven't received an AP WU since 6 Aug... So I figure that the fix worked. BOINC On...On... JDWhale |
|
252)
Message boards :
Number crunching :
4 WU's per day quota limit
(Message 800266)
Posted 21 Aug 2008 by JDWhale
Post: With no WU's on the PC??? IMO... There should not be a penalty for aborting work that is past deadline... you are saving redundency... But what do I know? @tunajunior : Try switching back to BOINC 5.10.45... there are known issues with the version of BOINC that you are using. |
|
253)
Message boards :
Cafe SETI :
The MOON BAR
(Message 800017)
Posted 20 Aug 2008 by JDWhale
Post: Tanks... I needed tthat... Now if it'snot too much trouble... prop me up where I can see the stars... Such a clear night...  |
|
254)
Message boards :
Cafe SETI :
The MOON BAR
(Message 799974)
Posted 19 Aug 2008 by JDWhale
Post: When does the bar open? I need something strong; Gimme whatever the guy passed out on the floor was drinking ! |
|
255)
Message boards :
Number crunching :
SSE2 optimized. Does it really matter?
(Message 799493)
Posted 18 Aug 2008 by JDWhale
Post: Another couple of units completed. The VHAR WU (~17 Credits) comparison is about as expected... But the other comparison is a bit suspect, IMO. Rather than using credit to align compares, Angle Range(AR) is more appropriate basis; Especially with credits for the same WU changing daily. Cheers, JDWhale |
|
256)
Message boards :
Number crunching :
Possible 6.12.14/16 issue?
(Message 799466)
Posted 18 Aug 2008 by JDWhale
Post: I find it hard to believe that a new "Recommended Version" seems to be released every 2 days or so. Think I'll wait until the "Recommended Version" stabilizes... |
|
257)
Message boards :
Number crunching :
Coputation error Exceeds size limit
(Message 799422)
Posted 17 Aug 2008 by JDWhale
Post: I have been getting this error on some files that I process on my PC. Can you list any associated error messages from your log file? (..\BOINC\stdoutdae.txt) Looking at the source code, when that error occurs the actual filesize and its size limit are supposed to printed somewhere. (Sorry I don't know where, I'm not at all familiar with BOINC code.) |
|
258)
Message boards :
Cafe SETI :
. . . 'strange' anomaly Screen Capture
(Message 799338)
Posted 17 Aug 2008 by JDWhale
Post:
I ain't at Berkeley, but can duplicate your 'anomaly' simply by renaming the '..\BOINC\statistics_setiathome.berkeley.edu.xml' file and restarting BOINC. The date/time of 01 Jan 1970 is t-zero for system clocks... system time is kept as seconds since 01 Jan 1970. Since the 'stats' file is missing, there is no valid data to populate the graph with, so it centers around some default/initialization value... t-zero and zero credits. As soon an your host performs an 'update', the statistics data is DL from the project and the 'stats' file is created with valid data and the 'anomaly' goes away. Keep on BOINCin' JDWhale Added: 
|
|
259)
Message boards :
Number crunching :
SSE2 optimized. Does it really matter?
(Message 799181)
Posted 17 Aug 2008 by JDWhale
Post: First WU completed on the V8 client for SSE(P3) WOW... ~23% improvement in CPU time at one of the worst performing ARs (1.08) We were pretty sure that the new client was faster than the prior optimized clients on the P3... but test data was few and with sparse ARs. IIRC the Mid-ARs (0.35-0.45) showed the best improvements. VLAR and VHAR improvements were minimal if any. Since most WUs are in the Mid-ARs, there should be overall improvement. I'll be keeping an eye on this hosts returned work over the next week or so for comparisons. Thanks, JDWhale |
|
260)
Message boards :
Number crunching :
SSE2 optimized. Does it really matter?
(Message 798975)
Posted 16 Aug 2008 by JDWhale
Post: I only have a PIII - so I tried out the SSE optimization for Windows. Doesn't seem to have made a huge difference (like 10% max). I have read other posts that claim SSE3 or higher really make a difference. See the sticky thread "AK V8 ported release ap. issues, install, questions etc." for more details. Which "SSE" optimized clients have you tried ? I see from your history that you have a result from "R-2.4V"... Have you also tried Windows 32 bit AK v8.0 SSE (Pentium 3)? I'd be interrested in seeing some comparisons of CPU times on WUs with similar Angle Range (AR) of both clients. Regards, JDWhale |
|
261)
Message boards :
Number crunching :
Saving $$ In a Tough Time
(Message 798750)
Posted 16 Aug 2008 by JDWhale
Post:
Digital camera comes in handy for "drum" style brakes. Snap a pic before removing any springs. $aving Tip #7 Buy cheaper booze... You will drink less of it... Costs less up front... and it lasts longer... Save the 'good stuff' for special occasions. |
|
262)
Message boards :
Number crunching :
Saving $$ In a Tough Time
(Message 798649)
Posted 16 Aug 2008 by JDWhale
Post: The only thing I won't do is turn down the water bed heater temp for the kitties........ I used to turn off the waterbed heater during hot season... helps keep you cool when you sleep. Turn it on during cool season to keep you warm. If the kitties need to keep warm... just turn off the AC... then they can climb on the waterbed to cool off... and you can save twice. |
|
263)
Message boards :
Number crunching :
It's Friday, The Weekend is Nearly Here
(Message 798407)
Posted 15 Aug 2008 by JDWhale
Post: Noooo, it's 1:17am Saturday Morning... It's after midnight in Beijing... The only sports being contested right now... Glow-in-the-Dark HorseShoes & Pole Dancing ;-) Oh yeah... Can't forget Elbow-Bending over at Franks Place... Visited the original location in the early '90s... Nice place... |
|
264)
Message boards :
Number crunching :
**UPDATED**-Blurf's 3 Day Seti Donation Challenge
(Message 798268)
Posted 15 Aug 2008 by JDWhale
Post: Count me in... |
|
265)
Message boards :
Number crunching :
I suppose there is an excuse for no new work
(Message 798205)
Posted 15 Aug 2008 by JDWhale
Post: About 5 minutes after I wrote this our science database crashed and we ran out of work again. We got it back up (fingers crossed) and we're hanging out at around 20k workunits ready to send. Excellent turnaround on the crashed DB. Kudos to the S@H team at Berkeley. If you need help holding things together, I've got a spare box of Duct Tape & Velcro in the closet (works great on C++ code). Just let me know ;-) |
|
266)
Message boards :
Number crunching :
Astropulse - how to get it when using optimized SETI@home?
(Message 798010)
Posted 14 Aug 2008 by JDWhale
Post: well - i am at 10.122% after 37 hours 18 minutes. so, i think i need to stick to seti units. darn. So you have about 2 weeks CPU time left... Plenty of time to beat the 11 Sep, deadline. Your RAC on that host will be ~48 if only crunching AP WUs(assuming standard 715 credits per AP WU). Quite a drop from present RAC of ~140 :-( |
|
267)
Message boards :
Number crunching :
Saving $$ In a Tough Time
(Message 797669)
Posted 14 Aug 2008 by JDWhale
Post: Andy Worth wrote: I'm afraid the ultimate money saving effort for me will be to shut down some (or all) the SETI Quads over the Winter. You could try moving the farm into the house to help curb heating costs. Later: There you go... bed warmers... |
|
268)
Message boards :
Cafe SETI :
"Simple" Maths Problems II
(Message 797552)
Posted 14 Aug 2008 by JDWhale
Post:
1.720477401 |
|
269)
Message boards :
Number crunching :
AstroPulse errors - Reporting
(Message 797521)
Posted 14 Aug 2008 by JDWhale
Post: magpie2005 wrote:
I'm guessing that your host is a P4 Prescott with hyperthreading(HT)... and that much of the time your first AP WU was crunching, it was the only WU being crunched... Couldn't download new work from the servers for the past 4 days. These HT CPUs perform much better when there isn't competition from a second WU, almost doubling it's performance... There is still better overall throughput crunching 2 WU at a time, but each WU pays a significant penalty when sharing the CPU... You might consider putting the AP WU in suspended state and then resume if/when the servers run out of work.... Just a thought. Happy BOINCing, JDWhale |
|
270)
Message boards :
Number crunching :
Saving $$ In a Tough Time
(Message 797423)
Posted 13 Aug 2008 by JDWhale
Post: Andy Worth wrote: We used to live in caves, times change, we change with them. Only some of us coders still live in caves... Summertime I keep the A/C to kick in at 85F... in the winter I set the heat to 50F. I run a dehumidifier during the hot season, May thru September... 85F feels cool enough with the reduced humidity so the perspiration can evaporate; Jokers technique for keeping cool also helps. ;-) Years living on the edge of the western Gobi Desert and in Thailand without heat or A/C make acclimating in the cave easy. $aving Tip #23... Walk the daily fee golf course rather than hiring an electric cart. Also gives incentive to keep the ball in the fairway to reduce overall distance. $aving Tip #24... Play early enough that you feel guilty about buying expensive beer on the golf course. |
|
271)
Message boards :
Cafe SETI :
Rocky Horror Movie to be remade
(Message 797319)
Posted 13 Aug 2008 by JDWhale
Post: Ahhhh... fond memories of Saturday midnight viewings at the "Alabama" theater in the River Oaks district of Houston. Can't count the number of times I went. I hope Meat Loaf is recast... if not maybe Meat Loaf Jr. can play the part. |
|
272)
Message boards :
Number crunching :
AstroPulse errors - Reporting
(Message 797271)
Posted 13 Aug 2008 by JDWhale
Post: magpie2005 wrote:
Congratulations... You deserve a pat on the back for persevering, thanks for posting and allowing us to share your experience. I hope you don't mind that I point to your AP WU, wuid=309347223. The best thing about AP WUs seems to be that extended runtimes might fill in gaps of lean WU splitting and server outages. On the otherside... some might say that the lack of work is being caused by AP WU splitting and distribution. Comparing CPU times and credit claims of MB and AP WUs on your host, it looks like AP is underclaiming MB stock client by ~40%. As this is supposed to be addressed with the "adjustable" sliding multiplier within 30 days to narrow the gap, it seems that the early adopters are paying a premium to be the "guinea pigs". This is just the opposite of what I would expect; If the project administrators wanted to gain acceptance it seems they would be paying a premium in credits at first and allow the gap to narrow from the other direction. Folks might then be more accomodating to crunching AP during these "early" days; Especially considering the earlier distribution of "ghosts" and guaranteed missed deadlines due to issuing AP work to underqualified hosts. Just my opinion, JDWhale |
|
273)
Message boards :
Number crunching :
Saving $$ In a Tough Time
(Message 797102)
Posted 13 Aug 2008 by JDWhale
Post: Rather than paying $10 for a bucket of golf balls at the local driving range, I hit balls in open fields and shag them myself for free. Savings $50 / week. |
|
274)
Message boards :
Number crunching :
can posts disappear?
(Message 797084)
Posted 13 Aug 2008 by JDWhale
Post: So we come in and do alittle of this I could use some of that at my place ;-) |
|
275)
Message boards :
Number crunching :
The new #1 host.. AMD? :-D - EDIT: yes! :-)
(Message 796653)
Posted 12 Aug 2008 by JDWhale
Post: Seems like "operator error" to me, or maybe just lack of desire. |
|
276)
Message boards :
Number crunching :
AstroPulse errors - Reporting
(Message 796632)
Posted 12 Aug 2008 by JDWhale
Post: wuid=313778905 Then maybe the AP FAQ should be less cryptic. From the Astropulse FAQ... ========================================== What are the minimum requirements for my computer to run astropulse?
|
|
277)
Message boards :
Cafe SETI :
Shows you watched when we were younger.....
(Message 796559)
Posted 12 Aug 2008 by JDWhale
Post: I seem to be much younger than everyone else in here.... That was my fault... She looked "legal" to me ;-) Time to add...
|
|
278)
Message boards :
Number crunching :
AstroPulse errors - Reporting
(Message 796497)
Posted 11 Aug 2008 by JDWhale
Post: wuid=313778905 Another case of AP begin sent to "underspec" host... in this case 850 MHz PIII. This host received 2 AP WUs on 10 Aug (yesterday). Fortunately the host owner realizes the WUs will not complete before deadline and cancelled them. What happened to the "Minimum CPU: 1.6 GHz" requirement ? IMO, more work to the scheduler is still needed. |
|
279)
Message boards :
Number crunching :
Astropulse on slower machines
(Message 796074)
Posted 11 Aug 2008 by JDWhale
Post:
As long is it's 'fish' that you're feeding the kitties and not any whalemeat... I'd hate to have to 'tail slap' you....lol So if I read some prior posts correct... I can revert my app_info.XML file and not be worried about creating any more AP ghosts. Just for safety sake, I'll leave the file 'as is' and just cancel any AP work assigned. Except for the ones that have been reissued after validation... I'll wait to cancel them after a resolution is posted on that issue. I do hope someone is working on that particular issue... validating an AP WU, granted 0 credits, then issuing a redundant instance of the WU just to waste 30+ hours of CPU cycles. Afterthought: Also refrain from crunching AP WUs until the issuing of work to hosts that are guaranteed not to complete within deadline... like the 500-600 mhz PIII and Celerons. Please forgive me if this issue has already been fixed, I have read nothing addressing a resolution. |
|
280)
Message boards :
Number crunching :
Astropulse on slower machines
(Message 795952)
Posted 10 Aug 2008 by JDWhale
Post:
@Raistmer : Have you updated your app_info.xml to handle AP ? Maybe some of these AP WUs are your ghosts... wuid=309218975 wuid=310130812 wuid=309797058 wuid=309787810 wuid=309313971 wuid=309218975 [edit] On this one wuid=310489691 I see that your wingman is a 600 Mhz Celeron... maybe 800 CPU hours for him to finish. [/edit] |
|
281)
Message boards :
Cafe SETI :
Shows you watched when we were younger.....
(Message 795490)
Posted 10 Aug 2008 by JDWhale
Post: My favorite Martian. One of my all time faves... Broken spaceship in the garage... I think ET is already here... just invisible ;-) And don't forget... Mister Ed A horse is a horse, of course, of course... and no one can to talk to a horse, of course. That is, of course, unless the horse is the famous Mister Ed. @Uli : 'Get Smart' gets 2 thumbs up... Chaos vs. Control... sounds like my childhood ;-) |
|
282)
Message boards :
Cafe SETI :
Shows you watched when we were younger.....
(Message 795182)
Posted 9 Aug 2008 by JDWhale
Post: I Dream of Jeanie That Girl The Flying Nun |
|
283)
Message boards :
Number crunching :
Are you still using BOINC 4? Answer here
(Message 795174)
Posted 9 Aug 2008 by JDWhale
Post: wuid=311362368 No, I'm not still running BOINC 4... but my wingman on this AP WU is... 1) Will this host misreport credit as on MB WUs ? 2) This host seems to be a 500Mhz PIII... any guestimates on CPU time ? I'm thinking about keeping the WU on hold until near deadline. |
|
284)
Message boards :
Number crunching :
Snagged a couple of Astropulse WU's
(Message 794868)
Posted 9 Aug 2008 by JDWhale
Post: Say are these Astros added to our Seti@Home credits? There are NO credits for crunching Astropulse. So, are they added to S@H... It doesn't matter when they grant 0 credits? |
|
285)
Message boards :
Number crunching :
Snagged a couple of Astropulse WU's
(Message 794666)
Posted 8 Aug 2008 by JDWhale
Post:
My Q6600 @ 3.2Ghz takes ~34 hours on each of the first 2 crunched. Other Q6600 @ 3.0Ghz is taking ~36 hours on its first AP WU. |
|
286)
Message boards :
Number crunching :
Snagged a couple of Astropulse WU's
(Message 794634)
Posted 8 Aug 2008 by JDWhale
Post: My 1 AP unit is now at 15%, 24 hours down, up to 57h 48m to complete. So its doing about 7% every 12 hours. Sorry, I haven't even looked at the code yet, busy on another project. |
|
287)
Message boards :
Number crunching :
Snagged a couple of Astropulse WU's
(Message 794626)
Posted 8 Aug 2008 by JDWhale
Post: My 1 AP unit is now at 15%, 24 hours down, up to 57h 48m to complete. So its doing about 7% every 12 hours. Real world estimate : 160 Hours... 24 down... 136 to go :-( |
|
288)
Message boards :
Number crunching :
AstroPulse errors - Reporting
(Message 794559)
Posted 8 Aug 2008 by JDWhale
Post: wuid=310493783 34+ hours CPU time only to get: The extended attributes are inconsistent. (0xff) - exit code 255 (0xff) and Validate state Invalid I'm not a "happy camper"!!! |
|
289)
Message boards :
Cafe SETI :
The Last Person To Post Wins #67 - Clotted cream & Honey for POETs Day
(Message 794513)
Posted 8 Aug 2008 by JDWhale
Post: Is it too late to join the party? <3:11AM> Whoa.... Is that a whale on top??? |
|
290)
Message boards :
Number crunching :
Snagged a couple of Astropulse WU's
(Message 794511)
Posted 8 Aug 2008 by JDWhale
Post: I know there is an answer in here somewhere on these boards, however working 12 hours a day 7 days a week for now leaves little time to browse the boards, please forgive the question if so. I have downloaded several AP WU's on all my rigs both Intel and AMD, I see some are running now. I have not added to or This rig hostid=4256735 is no longer running "optimized" since a "detatch" at 7 Aug 2008 23:10:16 UTC. This host is now running stock client, DLing and running AP WUs without a problem. Cheers, JDWhale |
|
291)
Message boards :
Cafe SETI :
"Simple" Maths Problems II
(Message 794351)
Posted 7 Aug 2008 by JDWhale
Post: Question 24 (1 Point) : You are blindfolded before a table... Why is the table blindfolded ? And why do I get blindfolded first ? |
|
292)
Message boards :
Number crunching :
AstroPulse errors - Reporting
(Message 794211)
Posted 7 Aug 2008 by JDWhale
Post: wuid=309707897 945415709 4369968 7 Aug 2008 7:13:26 UTC 6 Sep 2008 7:13:26 UTC In progress --- New --- --- --- 939936711 3622100 3 Aug 2008 5:43:31 UTC 6 Aug 2008 23:32:53 UTC Over Success Done 186,108.40 719.06 0.00 945190258 4041963 7 Aug 2008 3:22:44 UTC 7 Aug 2008 3:24:20 UTC Over Client error Compute error 0.00 0.00 --- 939936710 4068471 3 Aug 2008 5:43:30 UTC 5 Aug 2008 23:50:53 UTC Over Success Done 155,793.30 719.06 0.00 2 hosts have already completed successfully... That is my instance on top... To crunch or not to crunch... That is the question. |
|
293)
Message boards :
Number crunching :
just thinking outloud..bear with me...
(Message 794024)
Posted 7 Aug 2008 by JDWhale
Post: thinking that Astropulse is gonna be the last nail in the coffin for anything less than a dual core... Not quite...remember that "lesser" hosts are not supposed to be fetching AP WUs, thus will continue same as present on MB work. |
|
294)
Message boards :
Number crunching :
AK_v8_win_SSE41.exe calling Microsoft?
(Message 793918)
Posted 6 Aug 2008 by JDWhale
Post: He is running a old Alpha client, my guess is this is the key line in the log.Engaging BOINC Windows Runtime Debugger I agree with arkayn... Seems that last bump in your OC (from 3690 to ~3970 Mhz) caused instability in your system and crashed the client, thus invoking the Windows Runtime Debugger, which tried talking to MicroSoft. To me, it is remarkable that on restart the WU completed and was granted credit. [edit]Not really since a valid "checkpoint" file was used on restart [/edit] |
|
295)
Message boards :
Number crunching :
AstroPulse WU Completed... RAC drops 66%
(Message 793777)
Posted 6 Aug 2008 by JDWhale
Post: wuid=310159610 Since my first wingman "errored out" after 40+ hours, I'm a bit leary about giving this WU the green light. I'll wait 'til second wingman reports before deciding to run or cancel. It's sad enough that AP WUs are underclaiming credit when completed, but a total waste of substancial CPU time is not going to be acceptable, IMO. BOINC On...on... JDWhale [edit] Oops... too late. This WU is already 50 CPU hours invested and 90% complete on the mighty E4500. About ~5.5 more CPU hours to completion. [/edit] |
|
296)
Message boards :
Number crunching :
AstroPulse WU Completed... RAC drops 66%
(Message 793616)
Posted 6 Aug 2008 by JDWhale
Post: So your Q6600 took about 34 hours to complete the unit. That Q6600 runs at 3.20Ghz.... My E4500 at 2.42Ghz is taking ~52 hours and should finish the first one ~1700 UTC today. There are 10 more AP WUs in that queue (~3 weeks running one core AP). This rig might compare more favorable to your Dell (similar clock) |
|
297)
Message boards :
Number crunching :
AstroPulse Ghost WUs !!!
(Message 793611)
Posted 6 Aug 2008 by JDWhale
Post: Saw a new message on the Frozen Penny, during a work request and subsequent download of new work, that I had not seen before..... 'fraid not, that means you were just assigned another ghost. |
|
298)
Message boards :
Number crunching :
AstroPulse WU Completed... RAC drops 66%
(Message 793602)
Posted 6 Aug 2008 by JDWhale
Post: wuid=310019506 122,440.10 CPU Seconds for 719.06 Credit on my Q6600 I figure if only crunching AP WUs, my RAC would drop from 6100 to 2040.... This ain't going to be pretty! BOINC on...on... JDWhale |
|
299)
Message boards :
Number crunching :
Seti Enhanced + AP is NOT working!
(Message 793531)
Posted 6 Aug 2008 by JDWhale
Post: [Incidentally I haven't tried to run Astropulse yet by making the app_info mods & downlaoding the files. I don't seem to get ghost AstroPulse WUs, though I suppose I could have missed some. ] @Jason: Funny... as soon as you posted, this wuid=311368279 shows in your task list. Guess that you jinxed yourself ;-) Cheers, JDWhale |
|
300)
Message boards :
Cafe SETI :
Hurricane Edouard/Dolly prayer
(Message 792737)
Posted 4 Aug 2008 by JDWhale
Post: One more time for Tropical Storm Edouard... I live in Sugar Land, TX, ~50 miles inland and smack dab center of Edouard' current forcasted path. Tropical storms can be devastating, in 2001, TS Allison hit Houston and dropped 37" rain over 5 days. It wasn't pretty! |
|
301)
Message boards :
Number crunching :
@ Astropulse Rumor - Reality
(Message 792707)
Posted 4 Aug 2008 by JDWhale
Post: ... researching OpenGL 2.0 "Fragment Shader" programming in this respect... not particular in Astropulse, but more general Seti@Home algorithms. My $10 GEForce 6200 doesn't do CUDA... ATI Radeon support w/minimal effort... Past OpenGL programming experience... Desire to learn GLSL and Seti provides interesting challenge... Cheers, JDWhale P.S. If someone wants to donate NVidia GTX280, I'd be happy to give CUDA a go ;-) |
|
302)
Message boards :
Cafe SETI :
Shows you watched when we were younger.....
(Message 792492)
Posted 4 Aug 2008 by JDWhale
Post: These lists are bring back fond memories... Here are a few more... Captain Kangaroo (Bunny Rabbit & Mr. Moose) Family Affair (Yep... Had a crush on 'Cissy') Courtship of Eddies Father The Ghost and Mrs. Muir Red Skelton Show (loved 'The Town Clown') Ed Sullivan Show (barely remember seeing 'The Beatles') Adventures of Superman (George Reeves reruns in the 60's) The Merv Griffin Show Jeopardy (Always losing to my older sister) Dennis the Menace Peanuts (seasonal specials w/Charlie Brown) The Dating Game I Love Lucy |
|
303)
Message boards :
Cafe SETI :
To the kitty people V
(Message 792433)
Posted 4 Aug 2008 by JDWhale
Post:
With Halloween approaching, figured I better try on my costume... see if it still fits ;-) |
|
304)
Message boards :
Cafe SETI :
To the kitty people V
(Message 792408)
Posted 4 Aug 2008 by JDWhale
Post: Mark that is a very cool picture! I agree... cool picture, but... I thought it was more of a "You better not f*** this up, too!", look ;-) |
|
305)
Message boards :
Number crunching :
AstroPulse Ghost WUs !!!
(Message 792159)
Posted 3 Aug 2008 by JDWhale
Post: Son-of-a-gun... the only rig I didn't update app_info.xml... my 1.6GHz single core laptop got a ghost AP WU. I will not run AP on my laptop... so I fear more ghosts will follow. Sounds like a solid solution for the laptop... better to cancel WUs upon issue than creating more 30 day ghosts. Can we say Ghost Busters ;-? |
|
306)
Message boards :
Number crunching :
AstroPulse Ghost WUs !!!
(Message 792098)
Posted 3 Aug 2008 by JDWhale
Post: I looked through the caches on all of my rigs and do not see any AP work there at present. If any has been issued to my rigs, it is not waiting in cache. @msattler : Here's one of your AP WUs... Have you changed your app_info.xml to understand how to crunch ? Son-of-a-gun... the only rig I didn't update app_info.xml... my 1.6GHz single core laptop got a ghost AP WU. I will not run AP on my laptop... so I fear more ghosts will follow. |
|
307)
Message boards :
Cafe SETI :
The MOON BAR
(Message 792055)
Posted 3 Aug 2008 by JDWhale
Post: So I meet the namesake of this fine establishment, tip of the ol' cowboy hat, Madame Moon. Just rode into town, thanks for the warm welcome... Sorry to offend any herbivores in the bar, but here in Texas 3 of 4 food groups are BBQ Ribs, BBQ Brisket, and BBQ Chicken... everything else is lumped together as "trimmings", the stuff you'd get from a salad bar (mostly pickles, onions and jalapenos). Around back, the smoker is always fired up and there is always somethin' good on the grill. Thanks for the Sangria... better make mine a double pour. I need something to stop this pounding in my head ;-) Cheers, JDWhale |
|
308)
Message boards :
Number crunching :
AstroPulse Ghost WUs !!!
(Message 791932)
Posted 3 Aug 2008 by JDWhale
Post: I think Mark actually got it right. Yes... The servers are sending out ghosts... and it's going to be a month 'til they are re-issued. Does anybody know how to edit my client_state.xml file to force a resend of the missing file? I am not afraid of the consequences. [edit] @WinterKnight : here's one [/edit] |
|
309)
Message boards :
Number crunching :
AstroPulse Ghost WUs !!!
(Message 791929)
Posted 3 Aug 2008 by JDWhale
Post: I think Mark actually got it right. Yes, I understand... Since I didn't update my app_info.xml file or install the AP client files, is it expected that I would be assigned ghost AP workunits? |
|
310)
Message boards :
Cafe SETI :
The MOON BAR
(Message 791916)
Posted 3 Aug 2008 by JDWhale
Post: So in celibration, I hosting a BBQ, enjoy. WOOHOO!!!!! Can I help... I'm pretty good cooking them babybacks... and an "ice-cold" Fosters Ale as accompaniment...   |
|
311)
Message boards :
Number crunching :
AstroPulse Ghost WUs !!!
(Message 791907)
Posted 3 Aug 2008 by JDWhale
Post: It should not download the work if you do not have the files in place for it though. Yes, I understood that statement the same way, my host has been running nothing but optimized AK_V8 with app_info.xml since mid March... this is Lovey, the first Windows host to be running AK_V8, the box I first deployed to, my first test host for the ported client. If I read the above statement a slightly different way... "If you are using an optimized application, then you need to replace or modify the app_info.xml file, you will get AP WUs". @Arkayn : Thanks, I've already modified my app_info.xml files, but the damage has been done with the first AP WU. @UL1 : I saw your earlier message about receiving one of the AP WUs on your MAC, I saw the same messages in my Win Vista log file. [edit] @msattler : Here's one of your AP WUs... Have you changed your app_info.xml to understand how to crunch ? [/edit] With 30 day deadlines... it's going to be a long wait to grant credit on the ghosts... fortunately my wingman on this WU is also running WIN AK_V8 and probably hasn't modified his app_info.xml either. |
|
312)
Message boards :
Number crunching :
AstroPulse Ghost WUs !!!
(Message 791896)
Posted 3 Aug 2008 by JDWhale
Post:
From the Astropulse FAQ... Can I choose to run only Astropulse? Can I choose not to run Astropulse? At present you must run both Astropulse and the original SETI@home. However, we plan to implement the ability to choose in the near future. ================ Lesson here... if you have the hardware, you will be issued AP work!!! Nevermind that you don't have the AP client. [edit] Only running optimized on all my rigs. If I wasn't running optimized, the AP client would have installed automatically. [/edit] |
|
313)
Message boards :
Number crunching :
AstroPulse Ghost WUs !!!
(Message 791891)
Posted 3 Aug 2008 by JDWhale
Post: Okay... I'll admit procrastination in updating my app_info.xml files to process the AP WUs... I was waiting for the official "An app_info.xml file will be made available here" section of the Astropulse FAQ to be created. Seems this didn't stop the project from issuing an AP WU to my host, which was immediately dismissed due to missing AP client, causing my issued WU to become a ghost. It will "timeout" on 02 Sep and then be reissued.... Sorry wingman. I doubt that I'm alone, and unless folks pay close attention to their message logs and "Computer Tasks" webpage, you won't even notice you own ghost WUs. Question is... How many others like mine are there ? Be warned folks, unless you want to be a Ghost Rider, do not postpone updating that app_info.xml file ASAP. BOINC on..on... JDWhale |
|
314)
Message boards :
Cafe SETI :
Shows you watched when we were younger.....
(Message 791805)
Posted 3 Aug 2008 by JDWhale
Post: Voyage to the Bottom of the Sea Flipper The Green Hornet Lost in Space |
|
315)
Message boards :
Number crunching :
Frozen Penny rebuild....and other upgrade notes......
(Message 791712)
Posted 3 Aug 2008 by JDWhale
Post:
If you bought it on a CreditCard... many have automatic insurance for 'stupid' (and also extent manufacturers warranty). |
|
316)
Message boards :
Number crunching :
Frozen Penny rebuild....and other upgrade notes......
(Message 791669)
Posted 2 Aug 2008 by JDWhale
Post:
Maybe you can return it ? |
|
317)
Message boards :
Number crunching :
CPU
(Message 791591)
Posted 2 Aug 2008 by JDWhale
Post: Any ideas about what could have happened or has the CPU burnt out or something equivalent. Any reason for the loss in performance. Whether the change in distribution is responsible I think it is unlikely. Looks like you're running 50% speed... the app reports... Processor Information: Model: Intel Core 2 Quad Q6600, 1.20 GHz Package: 4 Cores Good luck, JDWhale |
|
318)
Message boards :
Number crunching :
@ Astropulse Rumor - Reality
(Message 791226)
Posted 2 Aug 2008 by JDWhale
Post:
I've got early OpenGL background and am researching OpenGL 2.0 "Fragment Shader" programming in this respect... not particular in Astropulse, but more general Seti@Home algorithms. |
|
319)
Message boards :
Cafe SETI :
To the kitty people V
(Message 791207)
Posted 2 Aug 2008 by JDWhale
Post: 28" monitors, 100" televisions. What can I say ??? Went out for a ham sandwich... Came home with a TV... After all, it was the ex-wifes birthday ;-) |
|
320)
Message boards :
Cafe SETI :
To the kitty people V
(Message 791185)
Posted 2 Aug 2008 by JDWhale
Post: 28" monitor??? Yep... My 100" television was a tight fit ;-) Hope to commence season 3 of "MI-5" on DVD early next week. [edit] The kitties love to watch kibbles commercials when they visit ;-) [/edit] |
|
321)
Message boards :
Cafe SETI :
"Simple" Maths Problems II
(Message 791097)
Posted 1 Aug 2008 by JDWhale
Post:
Answer 23: ~182,849.470588235 |
|
322)
Message boards :
Cafe SETI :
SON OF WORD LINK
(Message 790096)
Posted 30 Jul 2008 by JDWhale
Post: Wild Cherry |
|
323)
Message boards :
Number crunching :
"Client detached"?
(Message 789926)
Posted 30 Jul 2008 by JDWhale
Post: Weird! Discussed last week in this thread Either manually cancel all your "detatched" WUs, or detatch and reattatch to the project. |
|
324)
Message boards :
Cafe SETI :
The Bullitt chase scene........
(Message 789719)
Posted 30 Jul 2008 by JDWhale
Post:
On Any Sunday 1971 documentary |
|
325)
Message boards :
Cafe SETI :
SON OF WORD LINK
(Message 788842)
Posted 28 Jul 2008 by JDWhale
Post: cock-eyed |
|
326)
Message boards :
Number crunching :
something odd is going on ...
(Message 788479)
Posted 28 Jul 2008 by JDWhale
Post: Leave it to Misfit Just what we need right now... more finger pointing ;-) |
|
327)
Message boards :
Number crunching :
One cause for partial WU request fulfillment.
(Message 788257)
Posted 27 Jul 2008 by JDWhale
Post: Thanks for the investigation & follow-up... What you suggest seems to explain the behavior I've seen, I did adjust the preference values you point to some time ago and have been fetching a bit more work per request. I'm parameter limited since the affected hosts run diskless with ramdisk, 2GB RAM each with a 1GB+ ramdisk. Currently running:
Use at most[b] [u] 1.00[/u][/b] Gigabytes disk space
Leave at least[b] [u] 0.00[/u][/b] Gigabytes disk space free
Use at most[b] [u]100.00[/u][/b] % of total disk space
Thanks for your help,JDWhale |
|
328)
Message boards :
Number crunching :
My Mac Pro seems to be processing W/U ok, but
(Message 788243)
Posted 27 Jul 2008 by JDWhale
Post: no credits are forthcoming. The last credit I received was on July 23. My account shows that all the credits are pending. It has never taken 4 days to get a return and I joined this hunt on May 18-1999. A quick look at your hostid=3732303&offset=20 shows credit is being granted... I understand some "stats" sights like boincstats are not updating due to changes in data exchange format a few days ago. You might also be suffering from Macs not being able to fetch new work when the project support folks are running new scheduler as discussed in this thread among other places. |
|
329)
Message boards :
Cafe SETI :
SON OF WORD LINK
(Message 787954)
Posted 27 Jul 2008 by JDWhale
Post: barbecue chicken |
|
330)
Message boards :
Number crunching :
New Credit Adjustment?
(Message 787789)
Posted 27 Jul 2008 by JDWhale
Post: ISn't that were the task duration correction factor comes in? The duration correction factor (DCF) is used to calculate estimated CPU times to run WUs and to calculate the amount of work to request from project servers. No effect on credit claim/granting. The difference you noticed between credit claims between your host and your wingman is due to Berkeley changing scheduler programs on the project servers. Last night (Berkeley time) they ran a different version/different parameters... then this morning they switched back to the old scheduler/parameters. WUs returned while running the "new" scheduler have lower credit claims. The new scheduler is having issues sending work to some Apple MAC computers, some other "side effects" have also been reported. |
|
331)
Message boards :
Number crunching :
New Credit Adjustment?
(Message 787774)
Posted 27 Jul 2008 by JDWhale
Post: Only ran one WU on the comp that was consistantly lower. This time my comp had the higher return as compared to my wingman. Think those CPU benchmarks have something to do with it. If you're like me and never shut down, or load BOINC at as soon as you boot while other system processes are loading, it can throw off the speed SETI sees your comp crunching at. I guess it can fluctuate. The "credit claim" depends on which version of the "scheduler" Berkeley is running at the time the WUs are returned. They have switched back and forth between schedulers a few times in the past couple days. |
|
332)
Message boards :
Number crunching :
Switching vista to xp
(Message 787669)
Posted 26 Jul 2008 by JDWhale
Post: I have a computer with a dual core that runs Vista. I just got myself a full copy of XP. I want to know how much faster Seti will run with XP. Do not expect Seti to run any faster using XP... a better choice for high performance would be simply switching to Windows 32 bit AK v8.0 SSE3 optimized client to double your RAC. Your PentiumD 2.8 Ghz RAC ~550 My PentiumD 2.8 Ghz RAC ~1350 Even if you decide to abandon Vista and install XP, I still recommend the optimized client above. Good Luck, JDWhale |
|
333)
Message boards :
Number crunching :
New Credit Adjustment?
(Message 787208)
Posted 26 Jul 2008 by JDWhale
Post: Just checked... discovered same behavior here... wuid=303990264 Maybe another "side effect" of the AstroPulse enabling changes? |
|
334)
Message boards :
Cafe SETI :
SON OF WORD LINK
(Message 787077)
Posted 25 Jul 2008 by JDWhale
Post: rubber duck |
|
335)
Message boards :
Cafe SETI :
SON OF WORD LINK
(Message 787019)
Posted 25 Jul 2008 by JDWhale
Post: horse shoes |
|
336)
Message boards :
Cafe SETI :
SON OF WORD LINK
(Message 787006)
Posted 25 Jul 2008 by JDWhale
Post: New Deal |
|
337)
Message boards :
Cafe SETI :
"Simple" Maths Problems II
(Message 786890)
Posted 25 Jul 2008 by JDWhale
Post:
140,000 |
|
338)
Message boards :
Number crunching :
Detatched WU Behavior Change ?
(Message 786887)
Posted 25 Jul 2008 by JDWhale
Post: For JD, unless you missed something in the logs, I'd have to say something different happened there. Yes... the "something different" was that Berkeley switched to new scheduler code on Monday, July 21, as Matt states here. Yesterday, July 24, the old behavior was restored for new detatches with the switch back to the old scheduler. What will the behavior be when Berkeley switches back to the repaired scheduler code next week? To accept detatched WUs or not to accept detatched WUs... That is the question. If the project sees fit to detatch WUs and is not going to accept any detatched WUs, then IMO, the project should cancel them on the affected host. Regards, JDWhale |
|
339)
Message boards :
Number crunching :
Detatched WU Behavior Change ?
(Message 786586)
Posted 24 Jul 2008 by JDWhale
Post: Hmmm... I think UL1's detatched WUs occurred before July 2... explaining prior behavior of allowing results reporting after detatch. I witnessed this with my prior auto-detatch on Jun 17 when I was out of town for a week... completed results kept reporting and getting credit when reported before WU validation by wingmen. Looking through recent WUs, I see others hosts detatched with "corrected" time reported in black rather than the deadline time in green. Yes detatched behavior has recently changed. [edit] Matt's post in Tech news puts timeframe on "new" scheduler Monday-Thursday AM. New detatched WUs are behaving as before (until they switch it again). [/edit] |
|
340)
Message boards :
Number crunching :
Detatched WU Behavior Change ?
(Message 786574)
Posted 24 Jul 2008 by JDWhale
Post:
Very well... Since the affected host DL'd another 360 WUs after being detatched by the project, I'll wait 'til 12:00 PDT to perform detatch/reattatch on my end. Many of the 360 WUs DL'd are shorties and are preempting the detatched WUs. Thanks for finding/explaining scheduler change, it sure would be nice if the scheduler cancelled the WUs on the host when being detatched though. Regards, JDWhale |
|
341)
Message boards :
Cafe SETI :
"Simple" Maths Problems II
(Message 786541)
Posted 24 Jul 2008 by JDWhale
Post: What's the thickness of the toothpick? I take it that this is a 2-dimensional problem and the toothpick cannot pass through the plane the lines are in. In 3-dimensions the probability is 100% that the toothpick crosses all lines as it passes though the plane ;-) |
|
342)
Message boards :
Number crunching :
Detatched WU Behavior Change ?
(Message 786480)
Posted 24 Jul 2008 by JDWhale
Post:
7/24/2008 1:52:26 PM||file projects/setiathome.berkeley.edu/04mr08af.32704.12342.13.8.239 not found . . 7/24/2008 1:53:26 PM|SETI@home|Started download of 04mr08af.32704.12342.13.8.239 7/24/2008 1:53:30 PM|SETI@home|Finished download of 04mr08af.32704.12342.13.8.239 wuid=304298285 Mine is the "detatched" task... The project is resending a detatched WU. This is a problem IMO. I didn't want to individually "cancel" 600+ WUs, so I tried deleting them from the project folder. Imagine my surprise when the project resent the WUs that I'd deleted. |
|
343)
Message boards :
Number crunching :
Detatched WU Behavior Change ?
(Message 786460)
Posted 24 Jul 2008 by JDWhale
Post: I'm not quite sure if it's a general problem, because amongst others I submitted these 'detached' WUs on 23rd: I see different behavior... Your "detatched" WUs still display deadline in green text... My recently detatched WUs display Time reported in black text. hostid=4369968&offset=1000 The main problem is that the WUs still exist on my host and were being crunched, then being ignored by project servers when completed or cancelled. Also, I see that resends are now enabled again... I just went into my project directory and deleted the detatched WU files, upon restarting BOINC all the detatched WUs were resent :-( Guess I'll have to manually cancel the ~600 WUs in the cache. |
|
344)
Message boards :
Number crunching :
Detatched WU Behavior Change ?
(Message 786419)
Posted 24 Jul 2008 by JDWhale
Post: Power failure early this morning has changed all work in queue' outcome to "Client detatched". I know this triggers a resend to new wingman. In the past (yes this happens occasionally) I continue to crunch the "detatched" WUs for about 24 hours and receive credit when completed and reported before the WU is validated by wingmen. This has changed. I am now not being credited or acknowledged in any way for completing the detatched WUs. I've cancelled some of the detatched WUs and when returned their outcome also remains "Client detatched" rather than showing "Client error" "Aborted by user" as in the past. Seems that the handling of these "Detatched" WUs has changed, possibly with the new scheduler code. Does anyone know a list of changes brought about with the "new" scheduler code? I guess I'll have to cancel all the work in the cache since the results are being ignored by the project. BOINC on, JDWhale |
|
345)
Message boards :
Cafe SETI :
SON OF WORD LINK
(Message 786100)
Posted 24 Jul 2008 by JDWhale
Post: prime number |
|
346)
Message boards :
Cafe SETI :
SON OF WORD LINK
(Message 786065)
Posted 24 Jul 2008 by JDWhale
Post: Meet Virginia |
|
347)
Message boards :
Cafe SETI :
SON OF WORD LINK
(Message 786064)
Posted 24 Jul 2008 by JDWhale
Post: Unison, Virginia |
|
348)
Message boards :
SETI@home Science :
Allen Telescope Array on PBS
(Message 786062)
Posted 24 Jul 2008 by JDWhale
Post: Allen Telescope Array featured on NovaNOW on PBS... In Houston it's on tonight at 8:00PM(CDT) Available online beginning 24 July at... ScienceNOW... The Search for ET |
|
349)
Message boards :
Cafe SETI :
"Simple" Maths Problems II
(Message 785989)
Posted 23 Jul 2008 by JDWhale
Post: If t=0 c=15, K=85 Well, since you retract your original answer, I'll step in with it... xfer rate of 85 degrees per minute... It would take much less than one minute to raise temp 10 degrees... ~ 0.1252 minutes |
|
350)
Message boards :
Cafe SETI :
SON OF WORD LINK
(Message 785866)
Posted 23 Jul 2008 by JDWhale
Post: elevator music |
|
351)
Message boards :
Technical News :
Mac Knifed (Jul 22 2008)
(Message 785862)
Posted 23 Jul 2008 by JDWhale
Post: UPDATE! "... Message from server: SETI@home Enhanced is not available for your type of computer." @Jeffrey : Just a quick peek at your results reveal that you are still running version 4.43 of BOINC. <core_client_version>4.43</core_client_version>You might consider updating to the current version. |
|
352)
Message boards :
Cafe SETI :
SON OF WORD LINK
(Message 785234)
Posted 22 Jul 2008 by JDWhale
Post: square dance |
|
353)
Message boards :
Cafe SETI :
SON OF WORD LINK
(Message 785118)
Posted 21 Jul 2008 by JDWhale
Post: flat beer |
|
354)
Message boards :
Cafe SETI :
Closed - Use the Last Two Letters to Make a New Word
(Message 785080)
Posted 21 Jul 2008 by JDWhale
Post: Icarus |
|
355)
Message boards :
Cafe SETI :
Closed - Use the Last Two Letters to Make a New Word
(Message 784921)
Posted 21 Jul 2008 by JDWhale
Post: lethargy |
|
356)
Message boards :
Cafe SETI :
New Moderator
(Message 784915)
Posted 21 Jul 2008 by JDWhale
Post: Prost!, Prosit! Any plans for the extra bucks? |
|
357)
Message boards :
Cafe SETI :
SON OF WORD LINK
(Message 784904)
Posted 21 Jul 2008 by JDWhale
Post: software engineer |
|
358)
Message boards :
Cafe SETI :
Closed - Use the Last Two Letters to Make a New Word
(Message 784902)
Posted 21 Jul 2008 by JDWhale
Post: geek (aren't we all) |
|
359)
Message boards :
Cafe SETI :
Closed - Use the Last Two Letters to Make a New Word
(Message 784873)
Posted 20 Jul 2008 by JDWhale
Post: incinerator |
|
360)
Message boards :
Cafe SETI :
Closed - Use the Last Two Letters to Make a New Word
(Message 784699)
Posted 20 Jul 2008 by JDWhale
Post: typewriter ( remember those ? ) |
|
361)
Message boards :
Cafe SETI :
SON OF WORD LINK
(Message 784698)
Posted 20 Jul 2008 by JDWhale
Post: Popper Stopper |
|
362)
Message boards :
Cafe SETI :
Closed - Use the Last Two Letters to Make a New Word
(Message 784559)
Posted 20 Jul 2008 by JDWhale
Post: undulate |
|
363)
Message boards :
Cafe SETI :
Closed - Use the Last Two Letters to Make a New Word
(Message 784446)
Posted 19 Jul 2008 by JDWhale
Post: LFTI |
|
364)
Message boards :
Cafe SETI :
SON OF WORD LINK
(Message 784207)
Posted 19 Jul 2008 by JDWhale
Post: Brand Grand <- QUADRUPLE SCORE -> |
|
365)
Message boards :
Cafe SETI :
Closed - Use the Last Two Letters to Make a New Word
(Message 784185)
Posted 19 Jul 2008 by JDWhale
Post: YMCA Remember the Village People? |
|
366)
Message boards :
Cafe SETI :
Closed - Use the Last Two Letters to Make a New Word
(Message 784184)
Posted 19 Jul 2008 by JDWhale
Post: pseudonym |
|
367)
Message boards :
Cafe SETI :
SON OF WORD LINK
(Message 784182)
Posted 19 Jul 2008 by JDWhale
Post: Grand Stand <-Score-> |
|
368)
Message boards :
Cafe SETI :
SON OF WORD LINK
(Message 784163)
Posted 19 Jul 2008 by JDWhale
Post: Hall Monitor |
|
369)
Message boards :
Cafe SETI :
SON OF WORD LINK
(Message 784155)
Posted 19 Jul 2008 by JDWhale
Post: Labor Union |
|
370)
Message boards :
Cafe SETI :
"Simple" Maths Problems II
(Message 784152)
Posted 19 Jul 2008 by JDWhale
Post: Sorry, JD: That means C is crossing without (the only, I assume) flashlight. Also assuming single-plank bridge with no hand-rails so big splash... C & D crossing simultaneously in opposite directions with D carrying the requisite flashlight conforms to all the rules and also lowers cumulative distance (6 vs. 8 bridge lengths). Rules only state that if two people cross at the same time they must go at slower persons pace and the flashlight is used, not that both people are going the same direction. Please, no assumptions. Cheers, JDWhale |
|
371)
Message boards :
Cafe SETI :
"Simple" Maths Problems II
(Message 784104)
Posted 18 Jul 2008 by JDWhale
Post: For that question, I will award JD & Jason a 1/2 Point.... Well Done both of you! A & D cross together (10 minutes) C & D cross together, opposite directions (2 minutes) B & D cross together (5 minutes) D carries flashlight on all crossings total = 17 minutes |
|
372)
Message boards :
Cafe SETI :
Closed - Use the Last Two Letters to Make a New Word
(Message 784059)
Posted 18 Jul 2008 by JDWhale
Post: RPG (my second programming language - IBM S/32) |
|
373)
Message boards :
Cafe SETI :
SON OF WORD LINK
(Message 783950)
Posted 18 Jul 2008 by JDWhale
Post: new beginning |
|
374)
Message boards :
Cafe SETI :
SON OF WORD LINK
(Message 783875)
Posted 18 Jul 2008 by JDWhale
Post: End Zone |
|
375)
Message boards :
Cafe SETI :
Closed - Use the Last Two Letters to Make a New Word
(Message 783872)
Posted 18 Jul 2008 by JDWhale
Post: onside |
|
376)
Message boards :
Cafe SETI :
Closed - Use the Last Two Letters to Make a New Word
(Message 783741)
Posted 18 Jul 2008 by JDWhale
Post: alchemy |
|
377)
Message boards :
Cafe SETI :
Closed - Use the Last Two Letters to Make a New Word
(Message 783693)
Posted 18 Jul 2008 by JDWhale
Post: escape |
|
378)
Message boards :
Cafe SETI :
Closed - Use the Last Two Letters to Make a New Word
(Message 783687)
Posted 18 Jul 2008 by JDWhale
Post: dye |
|
379)
Message boards :
Cafe SETI :
Closed - Use the Last Two Letters to Make a New Word
(Message 783658)
Posted 18 Jul 2008 by JDWhale
Post: rye ...as in whiskey... |
|
380)
Message boards :
Cafe SETI :
SON OF WORD LINK
(Message 783655)
Posted 18 Jul 2008 by JDWhale
Post: dark matter |
|
381)
Message boards :
Cafe SETI :
SON OF WORD LINK
(Message 783632)
Posted 18 Jul 2008 by JDWhale
Post: DNA sequence |
|
382)
Message boards :
Cafe SETI :
Closed - Use the Last Two Letters to Make a New Word
(Message 783609)
Posted 18 Jul 2008 by JDWhale
Post: Andromeda ...as in M31... |
|
383)
Message boards :
Cafe SETI :
Closed - Use the Last Two Letters to Make a New Word
(Message 783392)
Posted 17 Jul 2008 by JDWhale
Post: Włocławek |
|
384)
Message boards :
Cafe SETI :
"Simple" Maths Problems II
(Message 783144)
Posted 17 Jul 2008 by JDWhale
Post:
e^10.783844921 -+ 10E-5 |
|
385)
Message boards :
Cafe SETI :
Closed - Use the Last Two Letters to Make a New Word
(Message 783098)
Posted 16 Jul 2008 by JDWhale
Post: Resurrect Ctenophora |
|
386)
Message boards :
Cafe SETI :
Closed - Use the Last Two Letters to Make a New Word
(Message 782642)
Posted 14 Jul 2008 by JDWhale
Post: Dsongarian Basin in NW China |
|
387)
Message boards :
Cafe SETI :
Closed - Use the Last Two Letters to Make a New Word
(Message 782599)
Posted 14 Jul 2008 by JDWhale
Post: telecommunicate ( ! ) |
|
388)
Message boards :
Cafe SETI :
Closed - Use the Last Two Letters to Make a New Word
(Message 782582)
Posted 14 Jul 2008 by JDWhale
Post: lds slurpy |
|
389)
Message boards :
Number crunching :
OC and cooling issues (again)
(Message 782505)
Posted 14 Jul 2008 by JDWhale
Post: The Room is a balmy 84F. Depends on point of reference... In Texas summer 84F is balmy... my computer room runs 33C (91F)... w/Q6600 @ 3200 Mhz cores running 62-59C, quite a bit cooler that dcappello's Q9450. Though I haven't tried the Xigmatek S1283... I had cooling issues with my Kingwin RVT-12025 (similar design) until I finally got good contact that required removing the fan while mounting. Pushpin mount makes it difficult to get solid contact. Good luck, JDWhale |
|
390)
Message boards :
Cafe SETI :
Closed - Use the Last Two Letters to Make a New Word
(Message 782438)
Posted 14 Jul 2008 by JDWhale
Post: edify |
|
391)
Message boards :
Cafe SETI :
Closed - Use the Last Two Letters to Make a New Word
(Message 782215)
Posted 13 Jul 2008 by JDWhale
Post: tequila |
|
392)
Message boards :
Cafe SETI :
"Simple" Maths Problems II
(Message 782113)
Posted 13 Jul 2008 by JDWhale
Post: Though extraneous... I'm curious which two contestants were absent? |
|
393)
Message boards :
Cafe SETI :
Closed - Use the Last Two Letters to Make a New Word
(Message 782035)
Posted 13 Jul 2008 by JDWhale
Post: setbacks as in relapse ;-) |
|
394)
Message boards :
Cafe SETI :
Closed - Use the Last Two Letters to Make a New Word
(Message 781868)
Posted 12 Jul 2008 by JDWhale
Post: Falasafa Islamic philosophy ( ! ) |
|
395)
Message boards :
Cafe SETI :
Closed - Use the Last Two Letters to Make a New Word
(Message 781800)
Posted 12 Jul 2008 by JDWhale
Post: onomasticon ( ! ) |
|
396)
Message boards :
Cafe SETI :
Closed - Use the Last Two Letters to Make a New Word
(Message 781792)
Posted 12 Jul 2008 by JDWhale
Post: redware ( ! ) |
|
397)
Message boards :
Cafe SETI :
Closed - Use the Last Two Letters to Make a New Word
(Message 781746)
Posted 12 Jul 2008 by JDWhale
Post: slag |
|
398)
Message boards :
Cafe SETI :
"Simple" Maths Problems II
(Message 781431)
Posted 10 Jul 2008 by JDWhale
Post: Well Done to Jason Gee!... 1 Point... He takes the lead... 2216.53333... |
|
399)
Message boards :
Cafe SETI :
Closed - Use the Last Two Letters to Make a New Word
(Message 781406)
Posted 10 Jul 2008 by JDWhale
Post: Nnekwunwenyi |
|
400)
Message boards :
Cafe SETI :
Closed - Use the Last Two Letters to Make a New Word
(Message 781338)
Posted 10 Jul 2008 by JDWhale
Post: lei |
|
401)
Message boards :
Cafe SETI :
Closed - Use the Last Two Letters to Make a New Word
(Message 781314)
Posted 10 Jul 2008 by JDWhale
Post: gyroscope |
|
402)
Message boards :
Number crunching :
Declining RAC
(Message 781218)
Posted 9 Jul 2008 by JDWhale
Post:
I believe there is another contributing factor on the "red line" data... Wasn't there a switch from Q6600 to Q9450 CPU about the same time as the client changed from 2.4V to AK-V8 client? So while the curves might be "independently correct" they should not be used to gauge performance difference between 2.4V and AK-V8 clients on the "same" system. This is a wonderful chart for demonstrating the performance difference between 32-bit & 64-bit AK-V8 client... assuming same CPU speed and memory configuration is being used. Cheers, JDWhale |
|
403)
Message boards :
Cafe SETI :
Closed - Use the Last Two Letters to Make a New Word
(Message 781109)
Posted 9 Jul 2008 by JDWhale
Post: Hmongs (Asian ethnic group) |
|
404)
Message boards :
Number crunching :
AK V8 ported release ap. issues, install, questions etc.
(Message 780913)
Posted 9 Jul 2008 by JDWhale
Post: So after all the power failures, having to shut the machine down for one thing or another, reboots due to DVD burning etc., BOINC updated to my account this morning around 7 am, and when I arrived home at 5:30 pm, my machine was frozen. Counting the number of completed work units at 14, and my machine completes 4 per hour I'd say it was slightly over 2 hours that the system froze up. I was getting system freezes about a month ago on my Q6600... Yes system freezes, NOT BSOD. Swapped out my "high performance" Ballistix PC2-8500 with old faithfull Kingston ValueRam PC2-6400 and been stable even since. The Ballistix ran for about 2 months before starting to fail... this was the 2nd set of the product that failed for me and seems to be a recurring problem from reading reviews. |
|
405)
Message boards :
Number crunching :
Declining RAC
(Message 780896)
Posted 8 Jul 2008 by JDWhale
Post: Hi guys, Here's what I see... So counting the bins from the left beginning with bin#1... I see that during the month of May... when the new AK-V8 client was released everyone was happy and RAC was on the rise, AR distribution wasn't a concern. Working through June, the ratio of bin#4:bin#5 rose significantly... bin#4 yields more cr/hr, so RAC continued to rise in June. Enter July... now many more bin#5 than bin#4... bin#5 pays less cr/hr... thus we're seeing a fall from the June highs. RAC decline is to be expected with the current WUs being distributed. JMO... JDWhale [edit]@Fred - Thanks for gathering and sharing your data. [/edit] |
|
406)
Message boards :
Cafe SETI :
Closed - Use the Last Two Letters to Make a New Word
(Message 780496)
Posted 8 Jul 2008 by JDWhale
Post: incorrigible |
|
407)
Message boards :
Number crunching :
Progress counting backwards on some WU's
(Message 780304)
Posted 7 Jul 2008 by JDWhale
Post: I've been struggling with one machine for a while now and I haven't been able to pinpoint the problem. Maybe someone can point me in the right direction. First off.... Welcome to the message boards and congratulations on your recent 9 year anniversary of joining Seti@home. With a quick look through the available result history on your T7100 I can not find any WUs ending with any errors. I do see some variation of CPU times on "like" WUs (WUs with same Angle Range), but this sometimes happens depending on workload. Honestly, I've never noticed the Progress % decreasing but I admit that I do not pay attention to that number. I sometimes notice "To Completion" time increasing slightly, though. Since this your "work" computer, maybe you should enable the virus scanner and scheduled backups. Good luck, JDWhale |
|
408)
Message boards :
Cafe SETI :
Closed - Use the Last Two Letters to Make a New Word
(Message 780260)
Posted 7 Jul 2008 by JDWhale
Post: Icelandic (score) |
|
409)
Message boards :
Cafe SETI :
Closed - Use the Last Two Letters to Make a New Word
(Message 780060)
Posted 7 Jul 2008 by JDWhale
Post: iambize |
|
410)
Message boards :
Number crunching :
Thie guy has problems but doesn't seem to know it
(Message 780010)
Posted 6 Jul 2008 by JDWhale
Post: As to the science, if his results are bad they won't validate, so the science is safe. His wingmen will also get credit when the third work unit is returned. And the few good results do contribute to the project... so the project appreciates his participation. |
|
411)
Message boards :
Cafe SETI :
Closed - Use the Last Two Letters to Make a New Word
(Message 779990)
Posted 6 Jul 2008 by JDWhale
Post: erceiver < score > [edit] oops... sp. [/edit] |
|
412)
Message boards :
Cafe SETI :
Closed - Use the Last Two Letters to Make a New Word
(Message 779983)
Posted 6 Jul 2008 by JDWhale
Post: CSEM (Controlled Source ElectroMagnetic) |
|
413)
Message boards :
Cafe SETI :
Closed - Use the Last Two Letters to Make a New Word
(Message 779982)
Posted 6 Jul 2008 by JDWhale
Post: Olympics |
|
414)
Message boards :
Number crunching :
What a Frozen Penny looks like (56k warning)...
(Message 779980)
Posted 6 Jul 2008 by JDWhale
Post: I found that some concentrated tinkering with the timings on my PC2-8500 Corsair RAM, particularly the "Performance Level" parameter, helped get RAC up to its present 8264 (was 8500 a couple of days ago) on my Q9450. I think the 12M of L2 cache doesn't hurt, either. I think your "tinkering" is paying off... The frozen penny seems to have a distinct advantage on the VLAR (63.98CR) WUs where higher CPU freq performs better, but on the WUs requiring higher bandwith your "tuned" Q does very well at closing the gap. I wonder if this is a prime example of the P5K Deluxe advertised "Super Memspeed" at work ? |
|
415)
Message boards :
Cafe SETI :
Closed - Use the Last Two Letters to Make a New Word
(Message 779934)
Posted 6 Jul 2008 by JDWhale
Post: metropolis |
|
416)
Message boards :
Cafe SETI :
Closed - Use the Last Two Letters to Make a New Word
(Message 779899)
Posted 6 Jul 2008 by JDWhale
Post: Llano |
|
417)
Message boards :
Cafe SETI :
Closed - Use the Last Two Letters to Make a New Word
(Message 779806)
Posted 6 Jul 2008 by JDWhale
Post: yesterdays |
|
418)
Message boards :
Cafe SETI :
Closed - Use the Last Two Letters to Make a New Word
(Message 779571)
Posted 6 Jul 2008 by JDWhale
Post: Onondaga |
|
419)
Message boards :
Cafe SETI :
Closed - Use the Last Two Letters to Make a New Word
(Message 779488)
Posted 6 Jul 2008 by JDWhale
Post: Lying! G!spot |
|
420)
Message boards :
Number crunching :
seti@home networkstop
(Message 779415)
Posted 6 Jul 2008 by JDWhale
Post: ... My guess is that the host is new... I think you "nailed it". This is a new host, crunching it's first WU, and should request more work once the first WU is complete. |
|
421)
Message boards :
Cafe SETI :
Closed - Use the Last Two Letters to Make a New Word
(Message 779351)
Posted 5 Jul 2008 by JDWhale
Post: onryÃ…Â |
|
422)
Message boards :
Cafe SETI :
Closed - Use the Last Two Letters to Make a New Word
(Message 779264)
Posted 5 Jul 2008 by JDWhale
Post: ngu (Thai - ngu: snake) |
|
423)
Message boards :
Cafe SETI :
Closed - Use the Last Two Letters to Make a New Word
(Message 779237)
Posted 5 Jul 2008 by JDWhale
Post: rambutan <yummy> |
|
424)
Message boards :
Cafe SETI :
Closed - Use the Last Two Letters to Make a New Word
(Message 779232)
Posted 5 Jul 2008 by JDWhale
Post: xbox |
|
425)
Message boards :
Cafe SETI :
Closed - Use the Last Two Letters to Make a New Word
(Message 779225)
Posted 5 Jul 2008 by JDWhale
Post: Electrolux |
|
426)
Message boards :
Cafe SETI :
Closed - Use the Last Two Letters to Make a New Word
(Message 779069)
Posted 5 Jul 2008 by JDWhale
Post: Arecibo |
|
427)
Message boards :
Number crunching :
RAC and Alex V8 code
(Message 779067)
Posted 5 Jul 2008 by JDWhale
Post: Total volume of work still drifting down, but the Penny seems to have flattened out. With the current stream of WUs being dominated by 0.4 - 0.45 AR... I'm afraid Penny is right on target according to my estimates. Another 4 or 5 WUs returned compute errors over past 24 hours on that same rig. If that continues, or more 0.6 - 0.9 AR WUs are handed out, you can expect further RAC deterioration. I can also relate with Jason' comment...I sometimes forget to shutdown IE on my laptop only to discover a day or two later that IE has been a resource hog and not sharing any CPU time with S@H. Another behavior I recently noticed... I was having ISP/networking issues a while back (intermittent cable modem connection plays havoc with my switch) that for some reason was causing WU exit/restarts with messages in stdoutae.txt : 25-Jun-2008 03:51:25 [SETI@home] Task 16ap08ac.12776.15205.7.8.214_0 exited with a DLL initialization error. 25-Jun-2008 03:51:25 [SETI@home] If this happens repeatedly you may need to reboot your computer. 25-Jun-2008 03:51:25 [SETI@home] Restarting task 16ap08ac.12776.15205.7.8.214_0 using setiathome_enhanced version 528 25-Jun-2008 03:52:31 [SETI@home] Task 16ap08ac.12776.15205.7.8.214_0 exited with a DLL initialization error. 25-Jun-2008 03:52:31 [SETI@home] If this happens repeatedly you may need to reboot your computer. Well restarting BOINC and rebooting computer had no effect... but disabling network activity restored normal behavior. This same problem was occuring on multiple hosts. Good luck, JDWhale |
|
428)
Message boards :
Cafe SETI :
Closed - Use the Last Two Letters to Make a New Word
(Message 779056)
Posted 5 Jul 2008 by JDWhale
Post: Ayahuasca |
|
429)
Message boards :
Cafe SETI :
Closed - Use the Last Two Letters to Make a New Word
(Message 779026)
Posted 5 Jul 2008 by JDWhale
Post: encapsulate |
|
430)
Message boards :
Cafe SETI :
Smoking cessation......and digital cameras
(Message 778864)
Posted 4 Jul 2008 by JDWhale
Post: And a rare appearance of the kittyman and his brothers on Seti....that's me on the right in the gray t-shirt.... We know that's not you in the middle... he's burning one ;-) Nice pix! |
|
431)
Message boards :
Number crunching :
The sweetest little 4-core on the block
(Message 778821)
Posted 4 Jul 2008 by JDWhale
Post: It's not a question of "hiding" behind an older host. That is my first host_id on Boinc and, call me an old softie, but I am emotionally attached to it :P And I recognise the numerical ID easily when browsing host lists. This host started as a C2D E6400 on an Asus P5B D/L MoBo running XP Home (32 bit) with PC2-6400 (400MHz) RAM. It has morphed, via a Q6600 brain transplant, to a Q9450 on an Asus P5K D/L MoBo running Vista Home Premium (64 bit) with PC2-8500 (1066MHz) RAM. The only time that I have had a new host_id spawned was on the change of OS. Running down the newly-acquired (very small) cache on the new host_id and then transplanting the Boinc folder from the old installation successfully restored the familiar id (though it errored out on the WU's still in the old cache and downloaded new ones). I understand being sentimental about a hostid, but I think that each new host should sport a unique hostid... not share with prior hosts. Isn't that the point of having a host identifier? If for some reason the project wanted to identify all the work performed by a specific host, but that hostid has been reincarnated as the CPU and/or MB changed, the DB fetches based on hostid would not be accurate and would contain more rows than intended. I'm wondering what happens to the stats associated with a particular CPU type when you change from a C2D to Q6600 to Q9450... Are all the WUs and stats associated with the prior CPUs now polluting the pool of numbers for Q9450 hosts? For example...Does your host contribute to the Q6600 stats on this BoincStats host_cpu_stats page? I don't know... But IMO, it should since it crunched WUs as a Q6600... Don't you think? Just my opinions. The new entrant to the top 20 this morning that I expect to stay around is heinz and his skt (nice one heinz). Yep, this is one to watch... I quickly estimate _heinz host will reach RAC 9250-10,500 depending on WU mix. Cheers, JDWhale |
|
432)
Message boards :
Number crunching :
@ Astropulse Rumor - Reality
(Message 778391)
Posted 4 Jul 2008 by JDWhale
Post: Sorry didn't know "TRUE" referred to a HSF. A quick review shows my oldest Pending to have reported about 40 days ago. You may well be right in your estimate of when I swapped the processor (and the MoBo - the RAM upgrade came a little later) but comparing crunch times on a range of sample WU's with the other rigs inOkay... why hide behind an older host? Why not start from scratch and eliminate any confusion? My point is... here is another top 20 host... This host was just created in the past 24 hours. How can I believe the RAC that put the host in the top 20 when it was just created? So many ways to confuse the stats... Lets keep it simple and without a doubt. I can't even merge the same host with itself running the same OS... I find it hard to believe the posted hosts in the top 20. [My last OT post in this "AstroPulse" thread...] |
|
433)
Message boards :
Number crunching :
RAC and Alex V8 code
(Message 778329)
Posted 4 Jul 2008 by JDWhale
Post: And here is a graph of one of my hosts over the past month... Looks a bit like yours... though I think my deviations are a bit higher % yours... The deep dip on 26 June was due to network failure. 
|
|
434)
Message boards :
Number crunching :
Save Arecibo
(Message 778266)
Posted 3 Jul 2008 by JDWhale
Post: I'd like to volunteer a year service to Arecibo... Assuming accommodations are provided. |
|
435)
Message boards :
Number crunching :
@ Astropulse Rumor - Reality
(Message 778229)
Posted 3 Jul 2008 by JDWhale
Post: I reckon I've got the sweetest little 4-core on the block ATM. My Q9450 @ 3.56GHz on air with a TRUE has been on the first page of the top-hosts for the last few days. The only other non-Xtreme quaddie up there is an X3360 clocked at 4GHz by "Anonymous" so he's either got a very sweet chip or is chucking lots of Vcore at it and H2O cooling. "TRUE" what? Creation Date of 12 Jun 2007 22:33:49 UTC ? Hmmmm... IIRC, you started up the Q9450 ~65 days ago... so you might still be collecting credits from the prior CPU? I don't have a problem with "plugging in" a new CPU, but you really should do a reset and start from scratch when you do the swap. JMO. |
|
436)
Message boards :
Number crunching :
What a crappy case I have
(Message 778169)
Posted 3 Jul 2008 by JDWhale
Post:
Newegg has excellent photos/descriptions/reviews, do your research. If I can "look" at a case, then I open it up and examine the number of fans and locations, case air filters, hard drive cage and mounting methods, room to work. An example is when using the Heatpipe Direct Touch(HDT) HSF... Are the exhaust & intake fans in the correct location for the orientation of the HSF depending on the type of CPU being used(Dual or Quad)?... Yes it makes a difference. It's summertime in SE Texas and my Home Office maintains an ambient temp of 33C (CPU cores 62-56-56-58C). A case that provides proper/efficient airflow is important. |
|
437)
Message boards :
Number crunching :
RAC and Alex V8 code
(Message 778148)
Posted 3 Jul 2008 by JDWhale
Post: Thanks Looking through your history, I see a few compute errors returned in the past couple days... These are an effective way of decreasing RAC. Also you have been crunching a few VLAR WUs (63.98CR), which consume more CPU/CR than the mid AR WUs. Other than that, 10% swings are not uncommon as the WU distribution changes... at RAC 8700, 10% might bring you down to ~7900, though I think a range from 8000-9200 might be more typical for this host/client combo. [edit] update estimated RAC [/edit] |
|
438)
Message boards :
Cafe SETI :
Closed - Use the Last Two Letters to Make a New Word
(Message 775708)
Posted 30 Jun 2008 by JDWhale
Post: realm |
|
439)
Message boards :
Cafe SETI :
Closed - Use the Last Two Letters to Make a New Word
(Message 775706)
Posted 30 Jun 2008 by JDWhale
Post: reassure |
|
440)
Message boards :
Cafe SETI :
Closed - Use the Last Two Letters to Make a New Word
(Message 775140)
Posted 29 Jun 2008 by JDWhale
Post: otaku |
|
441)
Message boards :
Number crunching :
boinc sugestion
(Message 775135)
Posted 29 Jun 2008 by JDWhale
Post: the only way to get any usefull data about someones computer is to drill down to their results in hope that they are running the KWSN app to check their real cpu speed. Do not always believe the reported CPU speed... It does not take into account an adjusted CPU multiplier. (example: Q6600 running 400Mhz x 8 reports 3600Mhz rather than actual 3200Mhz). |
|
442)
Message boards :
Number crunching :
Bugged Chart on Scarecrow's Charts
(Message 775099)
Posted 29 Jun 2008 by JDWhale
Post: The "Results Waiting Validation" chart seems to be stuck with the last value charted from June 20th 20:10 UTC. I can't find a RWV, but the RFV (Results Waiting Validation) chart displays as a black box with a little red "X", like the image file is missing. 7 Days chart shows a large black box, other time frames display a much smaller black box. No grid lines, labels, title... nothing, zipola, nada. "Single View" graph displays just fine. Put XX and tune your goggles to Dos Equis. |
|
443)
Message boards :
Number crunching :
One P4. Zero Crunching.
(Message 775095)
Posted 29 Jun 2008 by JDWhale
Post:
Assuming no problems, perfect condition, running the AK-V8 SSE2 Windows client for 6 hours/day your RAC might reach ~80. I'm not sure what you could expect from available Linux clients. Good luck, JDWhale |
|
444)
Message boards :
Number crunching :
V8 app detects cpu speed wrong
(Message 774780)
Posted 28 Jun 2008 by JDWhale
Post:
Some folks view crunching as a competition... much like "drag racing". As such "what's under the hood" ends up being a "lying competition"... never reveal your "hole card". You will not be able to trust the information a user has supplied as being truthful or "up-to-date", too much "auto-probing" by the application might be construed as an invasion of privacy, IMO. Regards, JDWhale |
|
445)
Message boards :
Number crunching :
too durn hot i tell ya
(Message 773522)
Posted 25 Jun 2008 by JDWhale
Post: Well back in LA and it is Hot here...Had to remove my covers from my computer and it went from 63C to 48C. Guess I need a new case with some air flow... Know how it feels... just back from a week in Las Vegas where temps were running 105-110... Now back in Houston 95F feels cool ;-) To sweat or not to sweat... that is the question! |
|
446)
Message boards :
Number crunching :
Darn auto-detatch
(Message 773517)
Posted 25 Jun 2008 by JDWhale
Post:
Alright... I made it home and checked my host earlier described... It runs BoincPE with persistence (backs up every 6 hours). The host rebooted (possible power outage... Thunderstorms were in the neighborhood the night it rebooted) and restored BOINC from the last backup. I guess this caused the "forced detatch" as new WUs were DL and results reported between the last backup and the reboot. This causes double reporting and ghost WUs. Anyhow... it would still be nice if the project could/would cancel the detatched WUs. Thanks for all the feedback, JDWhale |
|
447)
Message boards :
Number crunching :
Darn auto-detatch
(Message 772981)
Posted 24 Jun 2008 by JDWhale
Post: The problem is that I still have the WUs and am crunching them... I do not want them "resent", I want them cancelled. They were resent to a new wingman 6 days ago. Now I am not getting credit when I report my completed results. I'll be home tomorrow, but nearly all the detatched WUs will be completed by then. It would be better if the project would cancel all my detatched WUs. I have no way to cancel these detatched WUs because I am on vacation for past week. JD |
|
448)
Message boards :
Number crunching :
Darn auto-detatch
(Message 772815)
Posted 24 Jun 2008 by JDWhale
Post: So I left home a week ago and this time left my cable modem connected so my hosts could report results daily. The day after I left, my host Lovey was detatched by the project making the entire queue (~300 WUs) redundant. Seems to me that if the project is going to detatch a host, the associated WUs should be cancelled on the host. For the past week Lovey has been crunching WUs and getting credit less and less each day. So much for "set it and forget it". |
|
449)
Message boards :
Number crunching :
Question for the Code Cutters
(Message 770864)
Posted 20 Jun 2008 by JDWhale
Post: What is involved in "optimizing" an app ? It's the use of SIMD (Single Instruction Multiple Data) instructions performing 4 additions or multiplications in the same time it takes a single addition or multiplication. It's harder to code, but can make programs much faster by using more of the CPU capabilities. HTH, JDWhale |
|
450)
Message boards :
Cafe SETI :
Old movies we should watch again.............
(Message 768569)
Posted 15 Jun 2008 by JDWhale
Post:
No, 'It All Came True' wasn't an available title, but I did watch 'Angels with Dirty Faces', 'The Petrified Forest' & 'Beat the Devil', among others. |
|
451)
Message boards :
Number crunching :
Crunching on every inch of hardware
(Message 768553)
Posted 15 Jun 2008 by JDWhale
Post: I have a GPU and an Xbox 360 that's not crunching anything and it irks me! Probably got a computer in my truck that could crunch something as well. What else can move those 1's and 0's around in a useful manner? I remember reading stories back in the days of SETI Classic of folks crunching on car stereos... really can't remember much in the way of details other than is was using a Linux variant. |
|
452)
Message boards :
Number crunching :
@ Astropulse Rumor - Reality
(Message 767965)
Posted 14 Jun 2008 by JDWhale
Post: I'm not quite clear on WU uploads & downloads, but a website can have more than one registered URL that points to that IP address. Understood. Both of these URLs seem to point to the exact same webpage. If one is "correct" for the project and the other is "wrong"... then the wrong one should point to and load the correct one to avoid confusion, IMO. But then I am only a programmer and am ignorant to details of webpage management... Maybe what I suggest makes no sense to project webmasters. When I went to attach to "SETI Beta" via BOINC Manager, I didn't see the project listed on on the popup selection list. So I performed a google search on seti+astropulse. The first returned result seemed the only reasonable result on the first page and after viewing, seemed especially reasonable. Only after attaching to it and DL ~150 hours of AP 4.33 work I view my "messages" where I was informed that I attached to the "wrong" URL and provided with the "correct" project URL. It's no problem... I'm just making comments about my "attaching to SETI Beta experience"... I'll detatch then reattach to the "correct" URL after my AP queue is empty. |
|
453)
Message boards :
Cafe SETI :
Gotta go.......
(Message 767815)
Posted 14 Jun 2008 by JDWhale
Post: Dang it Mark... I'm having trouble keeping up with your flip-flopping... You made it sound earlier like you were being forced by the kitties to "shutter down" for the weekend... Yet you're still pecking at the keyboard like you're part of a sewing circle... Glad you decided to stick around... now how about pouring me a drink? I'll catch the next round ;-) |
|
454)
Message boards :
Number crunching :
@ Astropulse Rumor - Reality
(Message 767812)
Posted 14 Jun 2008 by JDWhale
Post: Got there ok. have one wu says 165hrs.After I run my catche out of seti and rosetta, will see how it does. My first WUs were estimated at 105 hours, but are only taking ~32 hours to complete on Q6600 @ 3200Mhz. 2 more to crunch after current one finishes. Odd message in my log telling me that I attached to the "wrong" URL though... 6/13/2008 11:23:46 PM|SETI@home Beta Test|You used the wrong URL for this project 6/13/2008 11:23:46 PM|SETI@home Beta Test|The correct URL is http://setiweb.ssl.berkeley.edu/beta/ 6/13/2008 11:23:46 PM|SETI@home Beta Test|Using the wrong URL can cause problems in some cases. 6/13/2008 11:23:46 PM|SETI@home Beta Test|When convenient, detach this project, then reattach to http://setiweb.ssl.berkeley.edu/beta/ I never knew that you could attach to a "wrong" URL ??? |
|
455)
Message boards :
Number crunching :
Cooling is king........
(Message 767632)
Posted 13 Jun 2008 by JDWhale
Post:
Into the whiskey a bit early, aren't we? |
|
456)
Message boards :
Cafe SETI :
Smoking cessation......and digital cameras
(Message 767487)
Posted 13 Jun 2008 by JDWhale
Post: I kinda identify with 'Leaving Las Vegas'....... I love "Leaving Las Vegas"... (opening scene... shopping cart in a liquor store... I can relate) actually I like "going to Las Vegas" even more... heading there Tuesday for a long break ;-) Yup, I explained to a friend about 10 years ago that if he quit smoking (3 packs/day) he could easily afford that new 4x4 truck he wanted just with the fag savings. He quit and bought the truck... somethings become easier when you dangle a carrot. Been burning the midnight oil... Microsoft VisualStudio 2005 "trial" license expires today (needed for development with Intel compiler). I'll switch to a newer trial(90 days) version when I get back from Las Vegas. I'll be poking in later tonight to offer up some JD (JackDaniels) induced support. |
|
457)
Message boards :
Cafe SETI :
Smoking cessation......and digital cameras
(Message 767296)
Posted 13 Jun 2008 by JDWhale
Post:
I'll join Team Whisky with you for the weekend. You have my full support on that front ;-) I've got a shoulder that you can lean on, cry on, spill on. For DVD support... pop in a copy of "The Waterboy"... "You can DO it!" |
|
458)
Message boards :
Number crunching :
What's up with new embedded ""'s
(Message 766524)
Posted 12 Jun 2008 by JDWhale
Post:
Thanks... Yep, I've been kinda busy trying to figure out the AK V8 SSE client on AMD issues... got the blinders on :-) (Only 2 more days 'til Microsoft license expires, so I'm cramming). Heat doesn't bother me... I like going to Las Vegas in the summertime... Spent years in Thailand w/o AC and climatize easily... Especially when mangos and strawberries are both "in season" and the Margaritas are flowing freely. |
|
459)
Message boards :
Number crunching :
What's up with new embedded ""'s
(Message 766465)
Posted 11 Jun 2008 by JDWhale
Post: John Thanks... It figures that one day I don't check "Tech News", there's the answer :-) |
|
460)
Message boards :
Number crunching :
What's up with new embedded ""'s
(Message 766428)
Posted 11 Jun 2008 by JDWhale
Post: I see old posts have "" preceeding quotes on old posts. Is this a new "feature"? |
|
461)
Message boards :
Cafe SETI :
Old movies we should watch again.............
(Message 766065)
Posted 11 Jun 2008 by JDWhale
Post: I just watched Casablanca, Treasure of the Sierra Madre, and the Maltese Falcon. Welcome... I watched all the Bogart my county library had a few months ago...except one. Time for The African Queen with Katherin Hepburn and I can always watch Sabrina with Audrey Hepburn... Wonderful movies. |
|
462)
Message boards :
Number crunching :
@ Astropulse Rumor - Reality
(Message 765840)
Posted 10 Jun 2008 by JDWhale
Post:
Done. Thanks. |
|
463)
Message boards :
Number crunching :
@ Astropulse Rumor - Reality
(Message 765834)
Posted 10 Jun 2008 by JDWhale
Post: So... I though I'd checkout what all the hoopla was about and attatched to Beta last night... promptly DL 5 AP WUs. Crunched on one WU for almost 15 hours and reached 44% complete. Now BOINC Mgr says it's running the AP WU, but no more progress or CPU time is being logged. Task Manager only shows the 3 S@H WUs getting CPU time. Also my host has frozen/locked up 5 times in the past 15 hours. I've seen BSOD a few times in the past... maybe 3-4 times a month, but never total freeze/lockup. Coincidence??? I don't think so. I've suspended Beta project and rebooted and resumed S@H on all 4 processes. If stability resumes, I'll abort all Beta WUs in my cache. I could try setting the host to stock/factory clock and try again, but that is one consession I'm not willing to make.. Are there known pre-requisites... like BOINC 5.10.45, or running at stock clock that I'm unaware of? This was my first attempt at running Beta and I might be lacking in know-how. Regards, JDWhale |
|
464)
Message boards :
Cafe SETI :
Old movies we should watch again.............
(Message 765397)
Posted 9 Jun 2008 by JDWhale
Post: Blade Runner Good one... watched about a month ago... my son gave a HiDef version. Roman Holiday - Audrey Hepburn, Gregory Peck & a bearded Eddie Albert. |
|
465)
Message boards :
Cafe SETI :
Smoking cessation......and digital cameras
(Message 765392)
Posted 9 Jun 2008 by JDWhale
Post: It was the weasing at night... laying in bed... each inhale sounded like I was playing a harmonica... well not quite, but maybe you get the point. Just turned it off... no more cigs.... Every time I wanted one... I had sex instead :-P Good luck... |
|
466)
Message boards :
Number crunching :
Wish we could chose wingmen we would not accept?
(Message 765119)
Posted 9 Jun 2008 by JDWhale
Post:
Ehll would freeze over before Penny could load up with enough work. Another perspective on the whole issue that is taught at my former employer is that... "Diversity in the workplace is a positive." I think this mantra can be carried into the scheduling world. The last thing that the project needs is to always have the same sets of hosts paired together. From the validation perspective it is better to have the hosts "randomly" paired to limit introduction of possible patterns based on wingman selection. That sort of behavior could undermine the validity of results.... much like politics. |
|
467)
Message boards :
Number crunching :
Wish we could chose wingmen we would not accept?
(Message 765112)
Posted 9 Jun 2008 by JDWhale
Post: those are GREAT explanations, but it still leaves a question of refusing to work with certain people or at the least match the machine speeds and not have 1 wicked fast and one fossil slow working the same WU's. I think what you are asking for is too restrictive... What happens when 95% of the computers are very much slower than yours? You might not be able to DL work because all the WUs currently being assigned are already issued to slower computers. Would you rather not be issued any work than being issued work where your wingman is slower than you? Similarly, if 99% of the computers are faster than you, then you might not get work because you are too slow and no "compatible" wingmen are currently available. The schedulers are already complex enough and IMO restrict DLs too much when my hosts request work.... I have a host that continuously requests 7 days of work and are only granted 5 or less WUs per request. This host crunches 50 WUs per day and can only fetch 5 WUs per request! What you are asking might only allow a single WU to be DL per request... at that rate it might take 2 hours to DL the 100-200 WUs required to raise my cache to an acceptable level. That places too much burden on schedulers that are already overloaded. No thank you... Regards, JDWhale |
|
468)
Message boards :
Number crunching :
Problem with optimized client and Xeon Dual Proc. machine
(Message 765094)
Posted 9 Jun 2008 by JDWhale
Post: Looks like the Kitties had not copied anything since yesterday afternoon when they posted the link for the animated weather map. I was thinking the same thing... even looked like yesterdays weathermap :-( |
|
469)
Message boards :
Number crunching :
Ok, another wierd one after updating to optimized client.
(Message 765092)
Posted 9 Jun 2008 by JDWhale
Post: Exactly what I was looking for. Thanks for the advice, I will get the correct one, and use the other only on Core2 processors. One thought, might want to edit the thread, and instead of saying intel only, maybe Intel Core2 processors only, it might help others in my situation. Justa thought. Always happy to help... The SSSE3 client is also aimed at newer Xeon chips, not only Core2, thus we recommend users run CPU-Z to check processor capabilities before installing any particular client. Welcome to the "fast lane", JDWhale |
|
470)
Message boards :
Number crunching :
Ok, another wierd one after updating to optimized client.
(Message 765089)
Posted 9 Jun 2008 by JDWhale
Post: I upgraded my client on another one of my boxes to You refer to 2 different versions.... SSSE3x, then SSE3. If you are running a Pentium4 then you should try the SSE3 client. SSSE3x client is for Core2 and above. You should not try to run the SSSE3x client on that host. Try the SSE3 client. [edit : AMD does not have any SSSE3 capable hosts, thus the instructions that it is for Intel only... Also Pentium4 does not have SSSE3 capability.] |
|
471)
Message boards :
Number crunching :
Problem with optimized client and Xeon Dual Proc. machine
(Message 765061)
Posted 9 Jun 2008 by JDWhale
Post: I found a pretty recent link from a post by Gecko_R7, main links to Lunatics site (still down for me) and the other was to a site in Russia... That site check out ok, anyone know? Or any other options. It was Windows 32 bit AK v8.0 SSE2 (AMD/INTEL) client I am looking for, correct? Yes, the Russia link you refer to is good... If you still have problems, post again and I'll forward you a direct link via PM from my DL server. Good luck, JDWhale |
|
472)
Message boards :
Number crunching :
QX6700 user? Please help!
(Message 765022)
Posted 8 Jun 2008 by JDWhale
Post: Just my experience with the Q6600, the stock heatsink was horrible. I use an intel motherboard too, so no overclock, and I was getting very high temp readings. I just got a Zalman heatsink this weekend and it dropped my temps by 20c. And BTW-I tried the stock heatsink w/ silver thermal compound and it didn't help that much . . . It's pretty sad that a stock heatsink can't take full load w/ no overclock. I didn't have any heat issues with the stock HSF, even running a mild 5-10% OC on a Q6600 with ambient/room temps at 27C and inside the case temps of 34+C. Like you say... YMMV. The case was an Apevia X-Qpack Lan case... there wasn't room to mount anything other than stock HSF. That host would have been Wrongway until ~May 10... his performance is clearly documented. [Edit : Welcome to the forum Matt...] |
|
473)
Message boards :
Number crunching :
QX6700 user? Please help!
(Message 764961)
Posted 8 Jun 2008 by JDWhale
Post: Hi, I agree with MSattler that you probably have cooling problem. Check again your HeatSinkFan(HSF) mount on your CPU. Be sure to use "thermal" grease between CPU and HSF. Get help from friend or computer shop if needed. Use a program like CoreTemp to verify your CPU is not overheating. I do not know if you are OverClocking(OC), but since you use stock HSF you should not OC until you verify that cooling is not a problem. Good luck, JDWhale |
|
474)
Message boards :
Number crunching :
Which way will the credit go ?
(Message 764872)
Posted 8 Jun 2008 by JDWhale
Post: wuid=273489815 I like your "most pessimistic" optimism Joe...Thanks for your assessment. New wingman is an AMD Sempron running WinXP Pro and "stock" client... should find out in 3-4 days... |
|
475)
Message boards :
Number crunching :
sse ?
(Message 764840)
Posted 8 Jun 2008 by JDWhale
Post: Maybe JDWhale would/should like to chime in here? Benchmarking the SSE clients is a relatively slow and tedious task... (I'm not being critical, just stating a fact). Personally, I do not have any P3s or compatable SSE AMDs. There aren't many testers over at Lunatics running the trials and of those that are, results have been mixed with the AMD times being, as of yet, rather disappointing. I share frustration with users waiting for a release and Jason, Raistmer and others in trying to find an acceptable solution. The work involved in building multiple releases a few weeks apart is substancial, Jason's methodology in testing/benchmarking incremental changes is time consuming, but would be required in both cases (that is release of both interim and/or "final" builds). This is a *much* bigger chore than the SSE2 build-test-release cycle, caused IMO, to the different chip implementations/behaviors of SSE/MMX instructions. That all said... I've enlisted the help of a single tester to run trials of incremental changes that I've been making to the client (please do not offer to test, my support of a single tester is enough extra work already). With the current progress, we're seeing ~10 results returned per week and not a wide variety of ARs. As such evaluation of crunched WUs are slow and yet inconclusive, but a major performance improvement, if any, will show up. Bottom line... We've not abandoned the SSE client... it's just taking longer than we all want to be evaluated and released. Kind regards, JDWhale |
|
476)
Message boards :
Number crunching :
Which way will the credit go ?
(Message 764783)
Posted 8 Jun 2008 by JDWhale
Post: wuid=273489815 Stock client overflows with 31 pulse count. Credit .06 AK V8 runs to completion with 1 pulse count. Credit 75.60 As of yet, the possible deciding wingman hasn't been assigned. Will this WU be a total waste of my time with legitimate science and credit lost by being assigned to someone whom also overflows results? Yes... this situation is not uncommon overall and sometimes a penalty is assessed for not even commiting a foul... sometimes the "call" just isn't fair and sometimes it is :) Hopeful, JDWhale |
|
477)
Message boards :
Number crunching :
Linux port of Alex v8 code
(Message 764522)
Posted 8 Jun 2008 by JDWhale
Post:
I think I understand what you're asking... As far as I know, no such tool/benchmark exists. What I'm doing is to record completed WUs and perform my own eyeball tests of different clients on identical AR WUs. I'd love if the data collection were automated and could generate my own scatter plots. Happy crunching, JDWhale |
|
478)
Message boards :
Number crunching :
too durn hot i tell ya
(Message 764461)
Posted 7 Jun 2008 by JDWhale
Post: So far, so good.... <snip...snip>
Works fine with Vista32 Ultimate and IE7, too. Watch out Mark, the second wave is steamrolling straight at ya. -JDWhale |
|
479)
Message boards :
Number crunching :
Wish we could chose wingmen we would not accept?
(Message 764423)
Posted 7 Jun 2008 by JDWhale
Post:
Yes, I've noticed a few also, though they do not bother me as much as wingmen running old versions of BOINC and under reporting credit. If your wingman "errors out" quickly then there is no harm as the WU is resent to another host quickly. Just rest assured that your results are saved and the credit will come, eventually. Living in SE Texas, I too feel the pain of summertime heat and running the crunchers. Since they generate heat, they add to the AC costs... And my computer room stays at ~89F... the rest of my home is just a couple degrees cooler. In winter they add just the required amount of heat. BOINC on...on... JDWhale |
|
480)
Message boards :
Number crunching :
Linux port of Alex v8 code
(Message 764415)
Posted 7 Jun 2008 by JDWhale
Post: Hi everyone :) First off... Welcome to the S@H forums. Next... I'm sure that I speak for all others involved when I say that you're welcome. And last... Please be careful when comparing runtimes for WU's that do not have the exact same AR. It is not uncommon to see CPU times vary by 100-200 seconds with the same AR and when the AR changes, even slightly, the change in CPU times can vary even more. Happy crunching, JDWhale |
|
481)
Message boards :
Number crunching :
AK_v8_win_SSE41.exe
(Message 763953)
Posted 7 Jun 2008 by JDWhale
Post: Im not havuing any thermal problems, and I have a slight OC at 3.2Ghz. I have tested against all prime and otherwise tests and come up fine. My full load temps are around 55C tops. Tj Max is 95C. Hmmm.... Well if you want an earlier build, you'll have to switch to running an Apple. You are running the first release of the AK V8 for Windows. Looking through your errored results, all I've seen failed with memory errors when you were OC'd to 3200MHz like you say. My first suggestion is to reset your motherboard to default settings and to run your CPU at factory speed. If the program still fails, then post back. If all runs well, then I suspect that you do not have a stable system when OC'd. Have you also OC'd your memory? It might need a slight voltage adjustment, different branded memories have differing requirements when being OC'd. You mention another team member having the same problems, is your teammates system also OC'd? Give them the same advice to try "stock" speed. Please remember folks... this code was developed for Apple computers which are built to strict standards with a set of high quality components. Running this application on the variation of Windows PCs and their hardware introduces many possibilities for failure, especially when overclocking. Good luck, JDWhale |
|
482)
Message boards :
Number crunching :
Little help with CPU cooler please...
(Message 763754)
Posted 6 Jun 2008 by JDWhale
Post: [quote]My thanks go to JokerCPoC for the recommendation of the Kingwin heatsink, it definately made my choice easy. A pair of computer tech long nose clamping tweezers worked wonderfully for re-installing the fan on the heatsink after removing it to ease the installation of the heat-sink/CPU on the motherboard. [quote] Ditto... Thanks Joker for your tutorial & follow-ups on installing the Xigmatek/KingWin RVT-12025, releasing one side of the fan made it much easier to press the pushpin mounts... The hsf is now very securely mounted to the Mobo and keeping the Q6600 @ 3000MHz at 60-56C with 30C room temps. When I first mounted this HSF, I was having difficulty applying enough pressure at the corners to securely mount. With the fan released it is much easier. Cheers, JDWhale |
|
483)
Message boards :
Number crunching :
too durn hot i tell ya
(Message 763708)
Posted 6 Jun 2008 by JDWhale
Post: I hear you phud - 95+ here in Dallas TX - air has not shut off in days for my office - I have moved my heavy quads to other parts of the house (wifi cards) - just sold off a nice old mac mini since it can't crunch like the new boxes - looking forward to an update from apple so I can replace it. Not looking fwd Holy Cow... did a cold front just blow through Dallas... I see the temps fell off a cliff from 90F to 50F in just 1 hour, from 2-3PM. I sure hope it's heading toward Houston ;-) |
|
484)
Message boards :
Number crunching :
Where did he go?
(Message 763340)
Posted 5 Jun 2008 by JDWhale
Post:
If you're interested in following the optimized client development a bit closer... why not join Lunatics and become a tester. The forums are already setup and there is discussion of behind the curtain "Wizard of Oz" details. That aside... there is nothing stopping anyone from posting ideas here... Jason, Josef, Cruncher(I think) and others are frequent readers of this forum and will likely comment on interesting posts. It was discussions on this board that started me to look at the code and fetch the original AK V8 source. BOINC on..on... |
|
485)
Message boards :
Number crunching :
Where did he go?
(Message 762851)
Posted 4 Jun 2008 by JDWhale
Post: As never trying a 64bit Windows system I can't really comment on that.Ahh I see, you must've been wondering what I was on about then ;). Typically the 64 bit apps on Windows show a significant speed increase over 32 bit apps on the same hardware+64bitOS, so naturally as co-porter I am curious if this applies on who's machine, or there is some factor that may need to be addressed for future architectures. I did query about his not running release 64-bit SSSE3x and was told that he was running Vista 32-bit... That the WhalePort client was the fastest that he'd found for the hardware/OS configuration. Unfortunately, I didn't have a way to build a PGO client for SSE4.1, I'm finding that PGO with IPO applied yields another level of performance that I'd like to have seen on the SKT. Also coming into play, though with minimal effect is the SKT only having 4 memory slots vs. MacPro having 8 slots... but then every little bit helps. Regards, JDWhale |
|
486)
Message boards :
Number crunching :
AK V8 ported release ap. issues, install, questions etc.
(Message 762706)
Posted 4 Jun 2008 by JDWhale
Post: I'm running a Q9300 just a tad O/C'd vat 2.8 Gzh. You might want to keep some records of CPU times vs. AR and try both clients(SSSE3 & SSE4.1), then decide for yourself which is better/faster. With such variables coming into play as memory speed vs. CPU speed and latencies, each system is likely to have slightly different results. With the Q9300's "limited" L2 cache, this might be one chip that performs better without OCing unless the memory subsystem is finely tuned. Also you might find one client performing better at some ARs and vice-versa. Of course, you could just relax... knowing that whichever client you've chosen is much faster than the "stock" client. P.S. I went to high school with a John R. Bailey. You're not from northern California by chance? Class of '76 ? |
|
487)
Message boards :
Cafe SETI :
Something to grouse about - What's the price of road fuel where you are?
(Message 762493)
Posted 3 Jun 2008 by JDWhale
Post: They have safety meeting after safety meeting, training after training, and the PEOPLE still make mistakes. I highly doubt that any policy that any company ever makes or enforces, will EVER make anything 100% accident proof. They keep trying. I remember the safety meetings... Only reason anyone ever showed up was for the company paid beer afterwards. $3.69/gal 87 Octane in Sugar Land, TX |
|
488)
Message boards :
Number crunching :
Time Warner going metered for service
(Message 762295)
Posted 3 Jun 2008 by JDWhale
Post: If limits are imposed, then there must be a way to turn off unwanted content such as advertising. When I visit local weather sites and they automagically start up streaming ad videos that I do not want, why should I be forced to pay for the bandwidth they consume? Regulation on website content will become mandatory. This enforcement is not aimed at casual web surfing, but at those who abuse bandwith by DLing major porn from newsgroups and other multimedia content. Make the streamers pay, but do not penalize the rest of us. I'm part of a group of many whose service was switched between Comcast and TimeWarner last year as part of the Adelphia settlement. I'm saddened to now be a Comcast customer as my rates have increased and service has decreased. When my cable goes out it is typically a 5 day wait for a service call. Does this affect SETI usage? Only Hiamps with his BOINC Button abuse ;-) Regards, JDWhale |
|
489)
Message boards :
Number crunching :
37 hour WU?
(Message 762239)
Posted 2 Jun 2008 by JDWhale
Post: I right now have a WU with 37 hrs left on LaptopII. I see that I have the SETI WU, and three other WU's from other projects, one running, two "waiting to run". Your laptop reports being "Intel(R) Core(TM)2 CPU T7200 @ 2.00GHz"... I would expect much better performance if that information is correct. |
|
490)
Message boards :
Number crunching :
Downloads failing
(Message 762228)
Posted 2 Jun 2008 by JDWhale
Post: Hey JD, are you wireless by chance? I had 5 rigs all on wireless that had the same problem, that was the only common denominator. I upgraded to a "N" router because I think I was most likely maxed out on signal strength with my "G" router and all is well since. Also I just today increased my additional work buffer from .5 to 2.5 days and all was OK on downloads on a few rigs. Negative on these rigs... These are wired connections. I have a couple wireless, but they haven't DL any work today. Problem seems to have corrected. Downloads flowing smoothly again. Except I reached my daily quota a bit early thanks to some failed downloads. |
|
491)
Message boards :
Number crunching :
Downloads failing
(Message 762220)
Posted 2 Jun 2008 by JDWhale
Post: See Matt's latest daily post over in Technical News for a possible explanation (though even Matt is confused as to what's happening). Yes, I read his post... but no mention of download issues... maybe the project problem is bigger than just the assimulators... or DB entries for work are made before work files are created. IDK, but I found that I'm not alone. |
|
492)
Message boards :
Technical News :
Lunchtime Review (Jun 02 2008)
(Message 762213)
Posted 2 Jun 2008 by JDWhale
Post: Got a weird error, myself... got Ditto... About 50% of downloads are failing on my end... same message in log. Here's a result [Edit] Problem seems resolved. Downloads proceeded smoothly 'till daily quota was reached earlier than expected. I guess the failed DLs count against quota :-( [/edit] |
|
493)
Message boards :
Number crunching :
Downloads failing
(Message 762212)
Posted 2 Jun 2008 by JDWhale
Post: About 50% of my WUs are failing to download. Another Berkeley issue or am I the only one... 6/2/2008 4:48:02 PM|SETI@home|Started download of 07mr08ad.20495.12751.7.8.215 6/2/2008 4:48:02 PM|SETI@home|Started download of 06mr08ah.20398.77633.7.8.235 6/2/2008 4:48:03 PM|SETI@home|Giving up on download of 07mr08ad.20495.12751.7.8.215: file not found 6/2/2008 4:48:03 PM|SETI@home|Giving up on download of 06mr08ah.20398.77633.7.8.235: file not found 6/2/2008 4:48:03 PM|SETI@home|Started download of 07mr08ad.20495.12751.7.8.164 6/2/2008 4:48:03 PM|SETI@home|Started download of 06mr08ah.20398.77633.7.8.216 6/2/2008 4:48:05 PM|SETI@home|Giving up on download of 07mr08ad.20495.12751.7.8.164: file not found 6/2/2008 4:48:05 PM|SETI@home|Giving up on download of 06mr08ah.20398.77633.7.8.216: file not found 6/2/2008 4:48:05 PM|SETI@home|Started download of 07mr08ad.20495.12751.7.8.163 6/2/2008 4:48:05 PM|SETI@home|Started download of 07mr08ad.20495.12751.7.8.179 6/2/2008 4:48:07 PM|SETI@home|Giving up on download of 07mr08ad.20495.12751.7.8.163: file not found 6/2/2008 4:48:07 PM|SETI@home|Started download of 07mr08ad.20495.12751.7.8.116 Here's one of the returned results... <core_client_version>5.10.30</core_client_version> <![CDATA[ <message> WU download error: couldn't get input files: <file_xfer_error> <file_name>07mr08ad.20495.12751.7.8.164</file_name> <error_code>-224</error_code> <error_message>file not found</error_message> </file_xfer_error> </message> ]]> [edit] add url [/edit] |
|
494)
Message boards :
Number crunching :
How much does SETI like CPU cache?
(Message 762053)
Posted 2 Jun 2008 by JDWhale
Post: ...Another "out of the box/loop optimisation" could be: Maybe "pretty" fast... but still buggy. Your second "while" loop writes past the allocated buffer. "pfbuf" ends at pfbuf[pfbufsize-1] but you write 'til pfbuf[pfbufsize+blahinc-1]. Not good... you just stomped memory. Better luck next time... |
|
495)
Message boards :
Number crunching :
How to merge computers ?
(Message 761894)
Posted 2 Jun 2008 by JDWhale
Post: Not much help forthcoming... So I guess my next question is : What purpose does the "merge computers by name" serve if not to merge a host that has been given a new ComputerID by the project with the prior existing ComputerID ? I've seen folks merge computers that are not the same computer, why is there difficulty merging computers that are the same computer ? Maybe a moderator could move my post if it's in the wrong forum. Help, Answers, FAQ ? JDWhale |
|
496)
Message boards :
Number crunching :
Question about core 2 duo
(Message 760876)
Posted 30 May 2008 by JDWhale
Post: Just noticed this host as my wingman shouldn't he be running 4 cpu's instead of 2 Host The host you point to is a Core2 Duo running Darwin... 2 cores residing on a single CPU... reports as expected for this type processor. |
|
497)
Message boards :
Number crunching :
How much does SETI like CPU cache?
(Message 760855)
Posted 30 May 2008 by JDWhale
Post: The rule of thumb is that 10% of the code is responsible for 90% of execution time. SETI may not match that rule exactly, but that's the concept... How many places does the 90/10% rule come into play... In coding for sure, 90% of the code can be developed in 10% of the time... it's achieving the last 10% that breaks the back of many projects. From past experience in porting IrixGL applications to OpenGL we faced approx the same ratio... 90% of the port was completed in 10% of the time. Of course with OpenGL in it's infancy and not behaving according to the rules was a major challenge. Cheers, JDWhale A new app for Overclockers? Wouldn't that be something...Bet that would get a few peoples hair in a ruffle... I'm looking at different apps for differing ARs... maybe have 2-4 apps installed and switch between them, possibly via the app_info or similar mechanism, depending on the AR. [ducking for cover...] [Edit] That might not sound like I intended... The app is still capable of crunching all WUs, just different profiles (PGO) applied to optimize for different AR ranges, thus creating specialized apps for the work. [/edit] |
|
498)
Message boards :
Number crunching :
How to merge computers ?
(Message 760825)
Posted 30 May 2008 by JDWhale
Post: I've done it in the past without problem using the "merge computers by name". But today it isn't working. Earlier today my host Wrongway crashed BSOD. When BOINC/SETI restarted it started using another Computer ID, so the "new" Wrongway is physically the same computer running the exact same OS, only recovered from a system crash. The WUS (~500 of them) distributed to the "old" Wrongway are still being crunched and when reported are showing up under the "new" Wrongway. The project has determined that they are the same host, so why can't I merge them? When I try I get the message : Processing wrongway Can't merge host 4251730 into 4369968 - they're incompatible How can they be incompatible ?  Confused, JDWhale |
|
499)
Message boards :
Number crunching :
Predicting RAC?
(Message 760324)
Posted 29 May 2008 by JDWhale
Post: Are there any tables or other benchmarks for target RAC based on CPU type? Sorry, but in the PC world there are too many variables to base RAC upon CPU type alone. Here are a few that must be taken into account: CPU speed, Motherboard, Memory speed, Client application. There are Q6600s in RAC range of approx. 2000-7000 depending on variables. If you were to select a Mac, the "target" RAC is much easier to predict as the system configurations are more standardized. Regards, JDWhale |
|
500)
Message boards :
Number crunching :
SSE2
(Message 760108)
Posted 29 May 2008 by JDWhale
Post: I did wonder if you would be able to resist... ;)) It was the start of the Indy 500 on Sunday when the withdrawls started... Jason was trying some non-MMX "work arounds" on bits of SSE emulation and I started foaming at the mouth... I'll probably have to leave other work on that front to someone who has an affected AMD, I can't test AMD performance on my end :-( Cheers, JDWhale |
|
501)
Message boards :
Number crunching :
SSE2
(Message 759925)
Posted 28 May 2008 by JDWhale
Post: Speaking of which, I do have a quad processor (@ 500 Mhz) XEON server I could run an SSE only flavour test client on if you're still interested. I also have a dual processor PIII Tualatin system (1.4 Ghz) that's SSE only I could power up. With your eval license expiring, I wasn't sure if you were still looking. The 4-way system only takes roughly 26 hours per WU, so it'd complete maybe four per day for investigating... All in good fun... I slowed down 3 of my hosts by "burning in" the SSE code a couple weeks ago. No problems with validation, no crashes. My work on the SSE code has been turned over to Lunatics, clients have been built and early tests have been run. PIII performance was boosted by 5-20% over 2.4V client depending on AR. Performance issues of running the SSE client on AMD systems are hindering a general release. I know nothing about AMD, folks are still searching for the best solution. I've attained more "evaluation" licenses from Intel and am currently learning & experimenting with other compiler tools & optimizations. BOINC on...on... JDWhale |
|
502)
Message boards :
Number crunching :
SSE2
(Message 759734)
Posted 28 May 2008 by JDWhale
Post:
You seem to have discovered an "Easter Egg" I put in the SSE2 code... explodes when someone brags about passing a Whale ;-) Just kidding... my RAC might be higher if I'd abandon my coding adjustments and testing on a few hosts and just let them run "online" crunching 24/7. But I'm too busy trying to raise RAC for others. One side effect is that my own RAC isn't as high as it could be. Cheers, JDWhale |
|
503)
Message boards :
Number crunching :
found bug in latest v8 (I think !! or maybe BOINC)
(Message 758669)
Posted 26 May 2008 by JDWhale
Post: thanks guys ! but i still hate to lower my rac beacose of this game alone ! Well... if the game is really that compute intensive, you could tell BOINC to only run 3 WUs rather than 4 while you are playing the game. Then when you shut down the game just switch back to 4 WUs crunching. Might give you the best of both worlds... still crunching while you're playing without blocking. Goto BOINC Manager menu->Advanced->Preferences... Select "Processor Usage" tab then adjust "On multiprocessor systems, use at most XX processors" to be 1 less than the number cores you are running. HTH, JDWhale |
|
504)
Message boards :
Number crunching :
Q6600 OC core temps
(Message 758178)
Posted 25 May 2008 by JDWhale
Post: It is a bit warm in my room - 78F (25C) one of my fast Quads (6600) runs at: Howdy neighbor, I'm just outside Houston, ambient is 30C (86F)... Thurston (Q6600 @ 3.2GHz) w/Thermalright Ultra-120 Extreme 61-56-55-59 Wrongway (Q6600 @ 3.0GHz) w/Kingwin RVT 12025 58-52-55-56 I still need to do another couple cooling cycles for the AS5 to set properly. Cheers, JDWhale @Grant - Same here! |
|
505)
Message boards :
Number crunching :
Here's my story....it's sad but true.......
(Message 758161)
Posted 25 May 2008 by JDWhale
Post:
When the compiler license expired I swore off coding this weekend... But this Kentucky Straight Bourbon has almost got me to change my mind and fetch another evaluation license. We've got some puzzling behavior with the AK SSE client and I want to try a few more things... Unfortunately I lost my PIII in the divorce a couple years ago so code changes are not easy to evaluate. (Actually the divorce could have been much worse, the b!+(^ demanded one of the PCs... So I let her have the PIII ;-) |
|
506)
Message boards :
Number crunching :
Panic Mode On (7) Server Problems! Closed for Renovation
(Message 758155)
Posted 25 May 2008 by JDWhale
Post: THanks for clearing that up MArk, seems to be an intermittent problem. I think i'll just kick the router in a minute. Good advice... About 30 years ago I was having problems getting a "punch card" program to read through the IBM 2540 card reader/punch. The program was about 700 cards long and kept getting read errors. Out of frustration I hauled off and kicked the darn card reader! Broke a toe and walked with a limp for a year. The foolish things we do while still a teen.... Regards, JDWhale |
|
507)
Message boards :
Number crunching :
Here's my story....it's sad but true.......
(Message 758145)
Posted 25 May 2008 by JDWhale
Post:
I'm all in.... Too bad my compiler has expired or I could work on some new SSE4.1 code :-( I like drinking and coding ;-) |
|
508)
Message boards :
Number crunching :
Here's my story....it's sad but true.......
(Message 757663)
Posted 24 May 2008 by JDWhale
Post: My reasons for posting these musings are hidden, More likely he fried the CPU 'tween his ears.... Call it a Turkey Short. |
|
509)
Message boards :
Number crunching :
Who's Skull Prevails at Last
(Message 757466)
Posted 23 May 2008 by JDWhale
Post:
Ah.... OpenInventor.... Now you are talking my language... How I miss the camera, object, and lighting transforms... the actors... the followers... 3D interactive scenes... That is what I call FUN programming! Regards, JDWhale |
|
510)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 757430)
Posted 23 May 2008 by JDWhale
Post: Aww shucks. I may have to get round to loading my Vista 64 that's been sitting on the shelf for about 12 months and I've been avoiding like the plague ;( I've read on these boards that Vista cuts performance by as much as 10%, that XP is the better crunching OS. Of course, I've never held that view... There is no way that Thurston would gain another 10% by running XP. Also, caveat emptor when reading the reported clock speeds, we know they can be reported artificially low when you start adjusting the multiplier. |
|
511)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 757361)
Posted 23 May 2008 by JDWhale
Post: ======== S_S_E_____U_P_D_A_T_E ======== Well, my second set of Intel Evaluation licenses for compiler(ICC) and Performance Primitives(IPP) has now expired. With this event, I've had to stop further development on the SSE capable client. All is not lost... preliminary clients have been built, it seems from early benchmarks that PentiumIII hosts will benefit over KWSN_2.4V_SSE_MB. AMD based SSE hosts do not seem to show any benefit (possibly quite the opposite, showing performance loss). Since I do not have any SSE only hosts, I have relied on 3rd party, limited testing. In any case, I've posted my code changes over at Lunatics for possible inclusion to the source code tree. Also included in the updates is code that ports a previous Core2(SSSE3) only optimization to run on SSE and above. I've seen benefits on my SSE2 and SSE3 hosts, especially on VLAR (63.98CR) WUs gaining upto 10% over release AK SSE2 client on Gilligan. Whether this optimization will benefit SSE hosts is questionable... This optimization tickles parts of the SSE port that uses MMX code and requires an extra step to protect the FPU from corruption (translation: MMX switching is SLOW). Anyway... without a compiler, I'm done coding! Also, my evaluation installation of "Microsoft Visual Studio 2005 Professional Edition" expires in 21 days. I don't know if I can renew the evaluation license for that product. Yes, this is a required product for SETI client development on Windows Vista. Regards, JDWhale |
|
512)
Message boards :
Number crunching :
Who's Skull Prevails at Last
(Message 757071)
Posted 22 May 2008 by JDWhale
Post: Look what Who? dragged into the race, i sincerly wonder what that is? Let's not believe the date associated with a host, if so then Fred W started his Q9450 last June... We all know he acquired that chip less than a month ago. |
|
513)
Message boards :
Number crunching :
Curious about 2.4v versus AK V8 on E2200 Pentium Dual Core
(Message 756575)
Posted 21 May 2008 by JDWhale
Post: i have reinstalled 2.4V on both E2200's to see if the RAC's go back up. Hmmm... The errors seemed to have gone away when the OC dropped from 2850 to 2800MHz. I wonder if that helped any? |
|
514)
Message boards :
Number crunching :
SSE2, SSE3, SSSE3, etc
(Message 756202)
Posted 20 May 2008 by JDWhale
Post: John McLeod VII wrote: It was my impression that SSE used 64 bit registers and SSE2 used 128 bit registers. As CRMofTheCrop already said, SSE uses 128-bit registers. I think you are confusing SSE with MMX registers... Here is excerpt from Intel compiler documentation on intrinsics... Intel processors provide special register sets. The MMX instructions use eight 64-bit registers (mm0 to mm7) which are aliased on the floating-point stack registers. The Streaming SIMD Extensions use eight 128-bit registers (xmm0 to xmm7). Regards, JDWhale |
|
515)
Message boards :
Number crunching :
How to DL more work ?
(Message 755891)
Posted 20 May 2008 by JDWhale
Post:
Sorry, but no difference here with host Wrongway. This host completes ~4 WUs/hr, I try to fetch 200 WUs every other day. Today was requesting ~300 hours work and fetching 16 WUs/request rather than full 20 WUs/request. My hosts do not crunch any other project and I don't fiddle with parameters like DCF. Why doesn't the full allotment of 20 WUs/request get filled on this host? Regarding my host Lovey referenced at the beginning of this thread. This afternoon I brought her back to life after being "down" for 10 days. I do not expect to miss any deadlines from the ~300 WUs in her cache. So I still can't figure why the project rejected her request for more work 2 weeks ago with the lame excuse of "won't finish in time". She has been down due to a failed power supply for 10 days; Comes back online mounted on a lesser motherboard and still will not miss a deadline! I'll revisit this subject once Lovey' cache is depleted and she is requesting work again (in about a week). Regards, JDWhale |
|
516)
Message boards :
Number crunching :
The curious case of slower is faster
(Message 755074)
Posted 18 May 2008 by JDWhale
Post: LOL.......you can imagine how happy the kitties have been for the last few weeks since I have been running the new apps........and then playing a bit with RAM timings to try to tweak just a tad more out of it........ Just had a peek back through your reported WUs, and we see a significant increase in CPU times ~May 12 coinciding with the reverting back to 2.4GHz. Also of note are 2 WUs reporting with what I think are memory errors. I'd consider stressing the memory a bit and verifying it with memtest or similar before trying to press the OC back up. Regards, JDWhale |
|
517)
Message boards :
Number crunching :
AK V8 ported release ap. issues, install, questions etc.
(Message 754373)
Posted 17 May 2008 by JDWhale
Post: Hi, I know that I will not work on MMX client. A bit more work on SSE client today... nice improvements on Mid-AR WUs compared to KWSN_2.4V_SSE_MB client, though VHAR and VLAR WUs are bit slower :-( WU validation testing continues. That's all for today... time for bed. Regards, JDWhale |
|
518)
Message boards :
Number crunching :
SSE2
(Message 754271)
Posted 17 May 2008 by JDWhale
Post:
Hmmmm.... That box seems to have errored on at least 6 WUS in the past 3 days, most recent was returned within the past hour. I suspect memory errors and suggest a memtest is in order.... might also be time to reduce back to stock clock during it's golden years ;-) @Mark - I really don't mind if my PC feels the need to refill my glass/pitcher of margaritas... leaves me more time to get frustrated with the SSE code ;-) |
|
519)
Message boards :
Number crunching :
SSE2
(Message 754260)
Posted 17 May 2008 by JDWhale
Post:
I wish mine would suddenly want to clean the house... ;) [/quote] No, you really don't wanna go there.........rig on a rage is not a good thing......... [/quote] My computer loves to make strawberry-mango margaritas... We get along very well ;-) |
|
520)
Message boards :
Number crunching :
heat wave
(Message 754160)
Posted 16 May 2008 by JDWhale
Post: Temps are rising in the northwest.Our usual highs of around 56f are going to about 90f for the next few days, for the first time I'm running at default speeds.It's wierd having my temps in the 30s instead of the 60s.So back to work on my h2o project. I miss the year round average temps of 45-60F in the northwest (Eureka, CA). I remember one July 4th when we set a record high temp of 72F. That was a real "heat wave". |
|
521)
Message boards :
Number crunching :
SSE2
(Message 754156)
Posted 16 May 2008 by JDWhale
Post: Just had a windows error from the SSE2 version... Since each WU launches a new instance of the client program (AK_v8_win_SSE2.exe) when it starts, there seems to be something about the "new" WU that the program didn't find agreeable. After the Windows error, did a new WU startup ok or did you have kickstart BOINC to get things rolling again? Regards, JDWhale |
|
522)
Message boards :
Number crunching :
The curious case of slower is faster
(Message 754151)
Posted 16 May 2008 by JDWhale
Post: OK, you experts out there; riddle me this one then. Right. I've noticed cases on Gilligan where WUs with ARs of 0.409830 & 0.435152 repeatedly have CPU times of ~14,233 & ~13,362. Even though the WUs would show up in the same "bin", CPU times vary significantly with minor AR fluxuations. Since the WUs tend to come in groups of same ARs, the binned data can easily become biased with a particular ARs data. A tightly zoomed in scatter plot might be a beneficial display for analyzing data in question. Good luck with your memory OCing. Be careful, memory can overheat and cause problems without proper cooling. Happy BOINGing, JDWhale |
|
523)
Message boards :
Number crunching :
The curious case of slower is faster
(Message 753882)
Posted 16 May 2008 by JDWhale
Post: OK, you experts out there; riddle me this one then. Can you cite specific WUs with matching ARs at the different CPU clock that demonstrate this "odd" behavior? The few that I've been able to pair up have shown the higher CPU frequency have lower CPU times as I would expect. |
|
524)
Message boards :
Number crunching :
What is the cheapest all in one ASUS board...
(Message 753338)
Posted 14 May 2008 by JDWhale
Post: My host Wrongway(Q6600) was using ASUS P5B-MX until Saturday. No problems, never crashed.... at $70 (Newegg) can't be beat. |
|
525)
Message boards :
Number crunching :
Duo Core Core 2 Quad Core I have a headach, can someone just tell me what to buy?
(Message 752867)
Posted 14 May 2008 by JDWhale
Post:
To understand my SIG... You must watch the movie "Star Trek IV: The Voyage Home"... George and Gracy play a major part in the plot of the movie. Have you considered installing the optimized app on your hosts? You could boost your production by ~50% just by switching apps! And it's free !!! |
|
526)
Message boards :
Number crunching :
Duo Core Core 2 Quad Core I have a headach, can someone just tell me what to buy?
(Message 752852)
Posted 14 May 2008 by JDWhale
Post:
I think the MAC is supposed to turn you on ;-) Check PM for my SIG explained... Regards, JDWhale P.S. The mangosteens weren't cheap $6.99 / lb |
|
527)
Message boards :
Number crunching :
Duo Core Core 2 Quad Core I have a headach, can someone just tell me what to buy?
(Message 752837)
Posted 14 May 2008 by JDWhale
Post: I just want the best cruncher. [Edit] But if money is no object, buy a Mac Pro with dual X5482 CPUs, load it up with 8 sticks of matching memory, boot Windows XP-64 and run the latest optimized SSSE3X 64-bit client. [/edit] [Edit2] [OT] I was in Asian Supermarket today, mangosteens, durian, jackfruit... I miss Thailand :-( [/OT][/edit2] |
|
528)
Message boards :
Number crunching :
Duo Core Core 2 Quad Core I have a headach, can someone just tell me what to buy?
(Message 752834)
Posted 14 May 2008 by JDWhale
Post: I just want the best cruncher. Sawadee Khup, The best bang for the Baht remains the Intel Q6600 hooked to Asus or Gigabyte motherboard supporting Dual Channel DDR2 RAM, 2x1GB DDR2-1066 (PC8500) RAM kit, other components as you choose. Kind Regards, JDWhale |
|
529)
Message boards :
Number crunching :
The curious case of slower is faster
(Message 752728)
Posted 14 May 2008 by JDWhale
Post:
That shouldn't come as a complete surprise. Didn't you once report that you couldn't see any performance advantage when I raised Thurston clock from 3200 to 3375 GHz? Secondly, with the variation of ARs within each bin, does a zoomed in scatter plot reflect the same picture? Regards, JDWhale |
|
530)
Message boards :
Number crunching :
SSE2
(Message 752697)
Posted 13 May 2008 by JDWhale
Post:
I agree, but for testing the new clients I've pulled 2 reserves back into active duty. With outside temps already reaching 30C I'll have to retire both Ginger and MaryAnn very soon... I've identified a couple more SSE2 tweaks I want to try, since MaryAnn is my 2.66GHz Northwood, she will have to stay online a bit longer to run the time trials for evaluation. @Gas Giant - I'll switch Ginger (3.2Ghz Prescott) over to SSE2 clients... I'll do the release client first, but I also want to try two others for comparison. Maybe ~24 hours each, maybe longer if needed for similar ARs. Keep in mind that Jason sprinkles magic dust on his computer and recites ancient incantations while building the releases to give them a little extra power ;-) (So my personal builds would be slightly faster if Jason had build them.) [edit] Wait!!! I already ran those tests on Ginger, about 6 days ago. SSE3 is faster on Prescott. That was before I called MaryAnn back into service.[/edit] Best regards, JDWhale |
|
531)
Message boards :
Number crunching :
SSE2
(Message 752605)
Posted 13 May 2008 by JDWhale
Post: I just grabbed SSE2 for this one since it is still crunching its kinda hard for me to tell if the AK made any difference from it. I do hope that you meant the SSE2 client for this hostid=3762048. Your Q6600 would be better served running SSSE3x client. Kind regards, JDWhale |
|
532)
Message boards :
Number crunching :
SSE2
(Message 752084)
Posted 12 May 2008 by JDWhale
Post: Did I read somewhere that it took 4 days to clean off the margarita stains on that SSE3 emulation using SSE2 code? Those must have been the strawberry-mango margaritas, they sure are good, but strawberry stains can be difficult - Sorry :-( I've been more careful when working on the SSE2 emulation using SSE intrinsics and most of the work was done with the help of Southern Comfort & pineapple juice. Probably an easier cleanup when/if the time comes. All in good clean fun, JDWhale |
|
533)
Message boards :
Number crunching :
Requesting 0 seconds of work with 15 WUs in cache
(Message 751917)
Posted 12 May 2008 by JDWhale
Post: Hi, Do you have any "suspended" tasks... I don't know about BOINC 5.10.45, but the version I'm running, 5.10.30 will not request any work if any tasks are suspended. I've also heard that re-installing BOINC can "fix" this kind of problem. HTH, JDWhale |
|
534)
Message boards :
Number crunching :
Computer Error on WUs being reported
(Message 751287)
Posted 11 May 2008 by JDWhale
Post: It is heading to 2.00 am here, and sleep calls. Since these "high" temps are new today, since your blow out, my first thought is that you need to reseat your Turniq and go through the "hot and cool" cycles as required by AS5, assuming you are using AS5 of course. Even your idle temps of 60C with room temp of 25C is too hot, IMO... but then I don't run your particular chip, but I'd like to see idle temps no more than 50C for a 65nm chip. I'd suspect even lower for a 45nm chip. Good luck, JDWhale |
|
535)
Message boards :
Number crunching :
How to DL more work ?
(Message 751106)
Posted 10 May 2008 by JDWhale
Post: The earliest deadline in my queue is May 27 (more than 2 weeks away)... I can crunch other (shorties for example) WUs for a whole week without touching anything in my queue and still not miss a deadline. I could even turn Lovey off for 7-10 days and not miss a deadline. As luck would have, I get to test this statement. Last night, Wrongway lost his power supply. Today, I'm doing the CPU/MOBO shuffle. Wrongway and Lovey are going to swap Mobos and I'll leave Lovey down until I get a replacement PS (no hurry, I'm a budget shopper and that case requires a special "shallow" PS), or finish up modding an old Dell case to house Lovey with the proprietary Dell PS. I was intending to make this move for some time... These CPU and MOBOs have been mis-matched since March 9, when the WU from ehll burned up Thurstons Mobo and I last did the CPU/MOBO shuffle. Maybe SETI project knew somehow that I was going to suffer PS failure, and thus denied work with the "(won't finish in time)" message ;-) Damn! BOINC is good! Regards, JDWhale |
|
536)
Message boards :
Number crunching :
How to DL more work ?
(Message 750721)
Posted 10 May 2008 by JDWhale
Post: After reading the scheduler source code I think I understand what is happening... When a butterfly farts somewhere between me and Berkeley, one more result is disqualified from being sent to my hosts. Question : What time of day do butterflys fart the least? Regards, JDWhale |
|
537)
Message boards :
Number crunching :
How to DL more work ?
(Message 750660)
Posted 9 May 2008 by JDWhale
Post:
Sorry Gassy, but I disagree, I've seen it many times... WUs gets DL'd, WUs starts immediately with "high priority", putting those already running ahead in the queue to "waiting to run" status. The answers to my scheduling behavior issues probably lie in the source code, I guess I'll look there to try to understand why. @Mark - I know, You could tell me how, but then... I can figure out how, but I want to know "why?" BOINC..On...On, JDWhale |
|
538)
Message boards :
Number crunching :
How to DL more work ?
(Message 750611)
Posted 9 May 2008 by JDWhale
Post:
But I have plenty of room for shorties... The earliest deadline in my queue is May 27 (more than 2 weeks away)... I can crunch other (shorties for example) WUs for a whole week without touching anything in my queue and still not miss a deadline. I could even turn Lovey off for 7-10 days and not miss a deadline. Shorties DL'd would simply be crunched at "high priority" so not to miss deadline. At least that is the way they are usually handled when DL'd deep into the queue. Another host, Wrongway, is showing the same symptoms, only fetching 9-11 WUs/request. Wrongway started running the AK port ~1 week after Lovey started. Tomorrow I'll nudge Thurstons cache up a bit from 5 days up to 7 days worth of work (~700WUs). He was able to sustain that level and still DL 20 WUs/request a month ago when he was racing to catch the Skulltrail... I'll see if he can still do the same now, after 5+ weeks running the AK port. @Ned - My "connect every 0.0 days" indicates "always connected", IIRC, so there is no reason not to DL because of this. 'Til now I have left DCF under program control (currently DCF=0.125966], never changed its value manually. I'll wait a couple days before requesting new work, then lower the DCF just a bit and see if that helps to DL more WUs/request. Still confused, JDWhale |
|
539)
Message boards :
Number crunching :
AK V8 ported release ap. issues, install, questions etc.
(Message 750589)
Posted 9 May 2008 by JDWhale
Post:
The function called to determine CPU speed is ippGetCpuFreqMhz and is provided by Intel... Here is that functions description... The function ippGetCpuFreqMhz is declared in the ippcore.h file. This function estimates the processor operating frequency and returns its value in MHz as an integer stored in pMhz. Note that no exact value of frequency is guaranteed so that the estimated value can vary depending on the processor workload. Hope this helps, JDWhale |
|
540)
Message boards :
Number crunching :
How to DL more work ?
(Message 750547)
Posted 9 May 2008 by JDWhale
Post: Maybe a hint why Lovey can't fetch 20 WUs/request. Note the message about "won't finish in time" in below "messages" excerpt... The way I figure... 300 WUs @ 1.25 hours/WU =~ 375 hours 375 Hours / 2 Cores / 24 hours/day = 7.81 days My earliest deadline is May 27 (more than 16 days away) and latest deadline is Aug 22. So SETI scheduler/server is wrong in its assessment of what Lovey will complete. It is not a problem with my settings... rather a problem with Berkeley/BOINC servers. Since Lovey is the single host running the AK Windows ported code the longest (started on March 23)... I can only predict that other hosts will have similar issues 4-6 weeks after starting to run AK ported code. 5/9/2008 1:08:33 PM|SETI@home|Sending scheduler request: To fetch work. Requesting 379216 seconds of work, reporting 10 completed tasks 5/9/2008 1:08:35 PM||A new version of BOINC (5.10.45) is available for your computer 5/9/2008 1:08:35 PM||Visit http://boinc.berkeley.edu/download.php to get it. 5/9/2008 1:08:38 PM|SETI@home|Scheduler request succeeded: got 0 new tasks 5/9/2008 1:08:38 PM|SETI@home|Message from server: No work sent 5/9/2008 1:08:38 PM|SETI@home|Message from server: (won't finish in time) Computer on 99.6% of time, BOINC on 100.0% of that 5/9/2008 1:09:38 PM|SETI@home|Sending scheduler request: To fetch work. Requesting 379517 seconds of work, reporting 0 completed tasks 5/9/2008 1:09:43 PM|SETI@home|Scheduler request succeeded: got 6 new tasks 5/9/2008 1:09:45 PM|SETI@home|Started download of 03mr08ad.15214.71052.8.8.33 5/9/2008 1:09:45 PM|SETI@home|Started download of 03mr08ad.15214.71052.8.8.39 5/9/2008 1:09:49 PM|SETI@home|Finished download of 03mr08ad.15214.71052.8.8.33 5/9/2008 1:09:49 PM|SETI@home|Finished download of 03mr08ad.15214.71052.8.8.39 5/9/2008 1:09:49 PM|SETI@home|Started download of 03mr08ad.15214.71052.8.8.142 5/9/2008 1:09:49 PM|SETI@home|Started download of 03mr08ad.15214.71052.8.8.148 5/9/2008 1:09:53 PM|SETI@home|Finished download of 03mr08ad.15214.71052.8.8.142 5/9/2008 1:09:53 PM|SETI@home|Finished download of 03mr08ad.15214.71052.8.8.148 5/9/2008 1:09:53 PM|SETI@home|Started download of 03mr08ad.15214.71052.8.8.108 5/9/2008 1:09:53 PM|SETI@home|Started download of 12dc06aj.12812.2526.4.7.215 5/9/2008 1:09:57 PM|SETI@home|Finished download of 03mr08ad.15214.71052.8.8.108 5/9/2008 1:09:57 PM|SETI@home|Finished download of 12dc06aj.12812.2526.4.7.215 5/9/2008 1:09:58 PM|SETI@home|Sending scheduler request: To fetch work. Requesting 339883 seconds of work, reporting 0 completed tasks 5/9/2008 1:10:03 PM|SETI@home|Scheduler request succeeded: got 5 new tasks 5/9/2008 1:10:05 PM|SETI@home|Started download of 03mr08ad.15214.71052.8.8.171 5/9/2008 1:10:05 PM|SETI@home|Started download of 03mr08ad.15214.71052.8.8.174 Thoughts? Regards, JDWhale |
|
541)
Message boards :
Number crunching :
SSE2
(Message 750491)
Posted 9 May 2008 by JDWhale
Post: If you have time, can you post a status update on SSE2? ==== S S E 2____S T A T U S____U P D A T E ==== Well.... Gilligan has been crunching slightly different builds of the SSE2 code as applied to the Lunatics release the past 3 days. Next set of 4 results due to be uploaded in about an hour (~18:00 UTC). Unfortunately, not many results to view as they get purged ~24 hours after being reported. The fastest client so far seems to be SSE2x ("Xeon" optimization), the next cycle of WUs will be running with "O3" compiler optimized SSE2x client and will commence when the current cycle finishes. Each cycle takes ~18 hours to complete, thus the results are being reported about once daily. My code for SSE2 has been delivered to Lunatics and is scheduled to be applied to the code base and tested when time allows. Regards, JDWhale @Fred - I think Gilligan is getting ~5 WUs (VLAR & MidAR) per day with the AK optimized client. |
|
542)
Message boards :
Number crunching :
AK V8 ported release ap. issues, install, questions etc.
(Message 750232)
Posted 9 May 2008 by JDWhale
Post: Hi, after installing this AK_v8_win_SSSE3x on my quad q6600, my computing averages have decreased. Any thoughts on this? Looking at 2 results with similar Angle Range(AR) from before and after switching.... I see that your CPU time has gone from ~6800 to ~3800. Yes your CPU times have decreased, as expected by ~44% at this AR... So where is the problem? Regards, JDWhale |
|
543)
Message boards :
Number crunching :
How to DL more work ?
(Message 750100)
Posted 8 May 2008 by JDWhale
Post: 683k client_state.xml But you're probably not carrying 300 SETI WUs in your cache... Gilligan has ~25 WUs and state file is 63k Thurston has ~530 WUs and state file is 1135k I think the state files are consistant and reasonable for ~6 days cache. |
|
544)
Message boards :
Number crunching :
How to DL more work ?
(Message 750078)
Posted 8 May 2008 by JDWhale
Post: What remains a mystery is why the number of fetched WUs is changing... the prior Preference values were constant for at least the past 7 weeks... since before I went to Las Vegas 23 March. Nope, no log files... only files of significance are : 354k job_log_setiathome.berkeley.edu.txt 683k client_state.xml 683k client_state_prev.xml @John McLeod VII - It's Core2 Duo, Daily quota is 200 WUs, currently DL 79 WUs today. Add that is plenty for today... tomorrow I'll enable downloads to fetch another 60 or so. Hopefully I'll be able to fetch more than 3 per request. Thanks, JDWhale |
|
545)
Message boards :
Number crunching :
Alex V8 code ritched 800 Daily quota on Skulltrail.
(Message 750058)
Posted 8 May 2008 by JDWhale
Post: is it possible to increase it? Wrong.... "Tomorrow" for daily quotas is based on Berkeley time 07:00 UTC. Believe me... I've waited up 'til 02:00 Houston time to DL more work. |
|
546)
Message boards :
Number crunching :
How to DL more work ?
(Message 750056)
Posted 8 May 2008 by JDWhale
Post: They look okay to me... except for that "-1.0" for connected, but that's the same value Thurston shows, and he DLs 20 WUs/request. I agree... Yes... my other hosts fetch 20 WUs per request. <duration_correction_factor>0.128427</duration_correction_factor> Runtime estimates for ~300WUs in cache are 1:11:44 - 1:16:02 +2 reissued WUs have longer estimates of 1:49:13 & 2:42:05 But your mention of disk space might be instrumental in resolving my issue. This host runs "diskless", boots from SecureDigital USB flash memory and installs BoincPE to run from a 768MB Ramdisk (host has 2GB RAM). I just bumped the Boinc allotment from 250MB to 450MB, and proceeded to fetch 9 WUs/request. There is still 637MB free on the Ramdisk (Seti project directory is now 133MB in size, ~300 WUs, ~6.5 days cache) Values in Disk & Memory Usage of Preferences... Use at most 0.45 GB disk space Leave at least 0.00 GB disk space free Use at most 100.00 % of total disk space Write to disk 60 seconds Use at most 100.00 % of page file(swap space) What remains a mystery is why the number of fetched WUs is changing... the prior Preference values were constant for at least the past 7 weeks... since before I went to Las Vegas 23 March. I've now allowed space on Ramdisk for ~1500 WUs (450MB)... So I think there is plenty of room to fetch 20 WUs per request. Unless there is something I'm not taking into account. Boinc version is 5.10.30 - I can try to update, but it's not straight forward with BoincPE... I don't want to reburn the USB flash just yet. (Maybe soon, when I do the CPU/Mobo shuffle and need to rename hosts to match CPU.) Also, it always seems that as soon as I update BOINC, a new version gets released, making me instantly obsolete again. Maybe Berkeley is trying to ween BOINC version 5.10.30 hosts by cutting the WUs they are issued, If so, I'd rather see BOINC version 4.45 hosts weened ;-) Thanks, JDWhale |
|
547)
Message boards :
Number crunching :
How to DL more work ?
(Message 749998)
Posted 8 May 2008 by JDWhale
Post:
Clock hasn't changed, host runs nothing but S@H... ever.
Do you mean these? <time_stats>
<on_frac>0.995597</on_frac>
<connected_frac>-1.000000</connected_frac>
<active_frac>0.999484</active_frac>
<cpu_efficiency>0.991458</cpu_efficiency>
<last_update>1210271755.578125</last_update>
</time_stats>
They look okay to me... except for that "-1.0" for connected, but that's the same value Thurston shows, and he DLs 20 WUs/request. Thanks, JDWhale |
|
548)
Message boards :
Number crunching :
How to DL more work ?
(Message 749979)
Posted 8 May 2008 by JDWhale
Post: My host Lovey, is requesting 125 hours of work, but only fetching 3 WUs per request. A month ago Lovey was fetching 20 WUs per request, but that number has been decreasing... A couple days ago she'd fetch 5 per request, then 4, now it's 3 per request.... 5/8/2008 12:08:41 PM|SETI@home|Sending scheduler request: To fetch work. Requesting 464506 seconds of work, reporting 0 completed tasks 5/8/2008 12:08:46 PM|SETI@home|Scheduler request succeeded: got 3 new tasks 5/8/2008 12:08:48 PM|SETI@home|Started download of 04mr08aa.4854.18697.9.8.104 5/8/2008 12:08:48 PM|SETI@home|Started download of 03mr08aj.27365.21749.11.8.182 5/8/2008 12:08:53 PM|SETI@home|Finished download of 04mr08aa.4854.18697.9.8.104 5/8/2008 12:08:53 PM|SETI@home|Finished download of 03mr08aj.27365.21749.11.8.182 5/8/2008 12:08:53 PM|SETI@home|Started download of 04mr08aa.4854.18697.9.8.75 5/8/2008 12:08:57 PM|SETI@home|Finished download of 04mr08aa.4854.18697.9.8.75 5/8/2008 12:09:02 PM|SETI@home|Sending scheduler request: To fetch work. Requesting 450852 seconds of work, reporting 0 completed tasks 5/8/2008 12:09:07 PM|SETI@home|Scheduler request succeeded: got 3 new tasks 5/8/2008 12:09:09 PM|SETI@home|Started download of 04mr08aa.4854.18697.9.8.110 5/8/2008 12:09:09 PM|SETI@home|Started download of 04mr08aa.4854.18697.9.8.64 5/8/2008 12:09:13 PM|SETI@home|Finished download of 04mr08aa.4854.18697.9.8.110 5/8/2008 12:09:13 PM|SETI@home|Finished download of 04mr08aa.4854.18697.9.8.64 5/8/2008 12:09:13 PM|SETI@home|Started download of 04mr08aa.4854.18697.9.8.131 5/8/2008 12:09:17 PM|SETI@home|Finished download of 04mr08aa.4854.18697.9.8.131 5/8/2008 12:09:23 PM|SETI@home|Sending scheduler request: To fetch work. Requesting 437242 seconds of work, reporting 0 completed tasks 5/8/2008 12:09:28 PM|SETI@home|Scheduler request succeeded: got 3 new tasks 5/8/2008 12:09:30 PM|SETI@home|Started download of 03mr08aj.27365.22158.11.8.24 Any ideas ?? Thanks, JDWhale |
|
549)
Message boards :
Number crunching :
SSE2
(Message 748930)
Posted 6 May 2008 by JDWhale
Post:
You got that right.... I'm working with my ol' pal "Southern Comfort" right now to lay the SSE2 patch on top of the source Jason posted on Lunatics. I'll let Gilligan crunch a few before hollering back to Lunatics with the patched code... I believe Jason is enjoying a well deserved break at the moment, so I might have to stretch this out for a few days... ;-) Regards, JDWhale |
|
550)
Message boards :
Number crunching :
Corrupted reported crunch time
(Message 748423)
Posted 5 May 2008 by JDWhale
Post: The "No heartbeat ..." message must be round about the time that I was tinkering so that could have caused the resetting of the CPU time? Coincidence? I've seen similar behavior with the RC1 client on hosts that shared a network connection... When a host wasn't "plugged in" similar "no heartbeat" messages and invalid CPU times reported. Only seemd to occur on my 2 hosts running BoincPE. Haven't noticed any issues since release though. |
|
551)
Message boards :
Number crunching :
What happened?
(Message 748405)
Posted 5 May 2008 by JDWhale
Post:
You can probably keep running those detatched WUs for a day ot two if they're still in you cache... ought to keep them "pending" WUs under control for a bit. But then, if they're already gone... C'est La Vie... |
|
552)
Message boards :
Number crunching :
Top hosts
(Message 748360)
Posted 5 May 2008 by JDWhale
Post: Moving up quite fast due to the new apps. But who's moving down, multi cpu servers that have not installed new apps yet? Just wondering, I guess the list will settle down in a couple of weeks. The Windows 8-core hosts that were memory "bottle-necked" with prior clients will dominate the leaderboard in a few weeks... I can only imagine what a 16-core(or greater) server could do with the new AK Windows client. |
|
553)
Message boards :
Number crunching :
AK V8 Confusion. Help please?
(Message 748352)
Posted 5 May 2008 by JDWhale
Post: lol. I was just curios. I've always exited BOINC before installing my optimized apps. Do Not Try This At Home... FYI: I don't exit BOINC or suspend computation to change out a crunching client... but then I'm a professional. I could tell you how, but then I'd have to [you know what I'd have to do!] you. ;-) I don't like those restarted "hybrid" crunched WUs that occur when you shutdown BOINC to switch clients. Folks monitor my hosts, I want to keep the WUs purely crunched by a single client, so not to introduce "outliers". Regards, JDWhale @Robert W Johnston - I'd give you a direct link, but IDK if you want SSE4.1 or SSSE3 version... check the "sticky" thread 2nd from top of NC forum. [edit] Still don't know which client works best with the Q9300 chip... with the limited L2 cache memory bandwith becomes a much more important factor. [/edit] |
|
554)
Message boards :
Number crunching :
AK V8 Confusion. Help please?
(Message 748338)
Posted 5 May 2008 by JDWhale
Post: computation also before exiting, just in case...) Depends... Are you going to exit BOINC before installing ? Just by chance... Are you one of those guys who will change out an electrical switch without first flipping off the circuit breaker? Sometimes you get \\\\=~ZAPPED~=//. |
|
555)
Message boards :
Number crunching :
SSE2
(Message 748121)
Posted 4 May 2008 by JDWhale
Post:
I'm thinking of trying an SSE build in a few days, there are one or two SSE2 instructions that I've got to try to emulate with SSE code. I'll let you know if/when I get your client built. ==== S S E 2____S T A T U S____U P D A T E ==== First results are posted/validated. Though I consider myself unbiased, I'll leave the evaluation to others. I've still got a couple changes to try with the SSE2 client, but I'll limit testing thoses changes offline on Thurston for now. Note: I have rearranged Gilligans cache to flip-flop between WUs from the only 2 AR bins available. BOINCing AK SSE2, JDWhale |
|
556)
Message boards :
Number crunching :
SSE2
(Message 747659)
Posted 3 May 2008 by JDWhale
Post: ==== S S E 2____S T A T U S____U P D A T E ==== Acceptance tests finished and passed in ~6 hours. I've switched to running AK_WhalePort V0.2X (SSE2) on Gilligan. Expecting about 37% speedup on the midAR WUs out of the gate. Here's beginning of stderr for the WU just started... OS X optimized S@H Enhanced application by Alex Kan Version info: OS X SSE2 (Intel, P4-optimized v8-nographics) V5.13 by Alex Kan Windows Port: JDWhale V0.2X 1589 Mhz Work Unit Info: ............... Credit multiplier is : 2.85 WU true angle range is : 0.416690 Boinc...On...On, JDWhale |
|
557)
Message boards :
Number crunching :
Linux port of Alex v8 code
(Message 747656)
Posted 3 May 2008 by JDWhale
Post:
First 3 returned WUs show you running "stock" client... <stderr_txt> setiathome_enhanced 5.28 Revision: 26 g++ (GCC) 4.1.2 (Ubuntu 4.1.2-0ubuntu4) libboinc: BOINC 6.1.0 Work Unit Info: ............... WU true angle range is : 1.195146 Optimal function choices: . . . </stderr_txt> |
|
558)
Message boards :
Number crunching :
Linux port of Alex v8 code
(Message 747605)
Posted 3 May 2008 by JDWhale
Post:
@Crunch3r -- I guestimate that if the client scales linearly across all 8 cores you could hit RAC 20,000 with that SSSE3 client. I haven't looked, is there a better paying project ? |
|
559)
Message boards :
Number crunching :
SSE2
(Message 747601)
Posted 3 May 2008 by JDWhale
Post:
I didn't know you spoke Australian ;-) Is there no end to your talents? |
|
560)
Message boards :
Number crunching :
SSE2
(Message 747548)
Posted 3 May 2008 by JDWhale
Post: in my original post i did not mean any disrespect of the developers of the AK V8. None taken... I'd first looked at porting to SSE2 about a month ago... right after AK_WhalePort V0.2 was complete. The "intrinsic" code was confusing at first, I made a bad decision, chose a wrong path and got discouraged. As with most my porting efforts, I've found that the first attempt is usually not sucessful, but a great learning experience for the next attempt. Now with all the release testing over and my goal of getting Thurston into the "Top 20" achieved, I renewed my "evaluation licenses" from Intel and decided to take a second look at what is required to build AK SSE2 client. No longer intimidated by the SSE intrinsics, I see the solution is both simple and elegant. I'm confident my latest build will pass acceptance tests and I'll start crunching AK_WhalePort V0.2X (SSE2) on my P4M Northwood(1.6GHz) laptop shortly. Kind regards, JDWhale |
|
561)
Message boards :
Number crunching :
SSE2
(Message 747508)
Posted 3 May 2008 by JDWhale
Post: Pity you had to get teamed up with an anonymous 4.45 for that WU - he has at least one other in his pendings that has had to be re-issued on account of non-validation. I'm certain that not validating was my fault... I ran the acceptance tests on Thurston for the SSE2 client rather than on Gilligan before going "live" with that WU. After failing to validate, I then ran the acceptance tests on Gilligan(~4 hours) and sure enough some failed. Looks like Gilligan has a few VLAR, and MidARs (both sides of 0.4)... both posted and in his queue... I hope to snag a few VHARs currently being passed out to round out the collection. ==== S S E 2____S T A T U S____U P D A T E ==== I located/fixed what I think was the problem with yesterdays client (oversight on my part), and am now launching the 4hr acceptance tests on Gilligan with 2 new clients. Best regards, JDWhale |
|
562)
Message boards :
Cafe SETI :
I'm baaaaaaaaaaaack.........
(Message 747359)
Posted 3 May 2008 by JDWhale
Post: And I'll join ya, bud. I love the smell of SSE2 in the mornin'. Smells like... victory (with a little Southern Comfort mixed in). It's gonna be another late night on this end... (ehll, it's already late... 2am...) But I work better after downing a few...my laptop is soooo slow running the short bench WUs Watching the Les Paul docu on the side... good stuff. Makes me want to open my Chet Atkins archive. ...Back to the code... |
|
563)
Message boards :
Cafe SETI :
I'm baaaaaaaaaaaack.........
(Message 747342)
Posted 3 May 2008 by JDWhale
Post: Welcome back! Things just were not the same without the kitties. I'll drink to that! Welcome back, |
|
564)
Message boards :
Number crunching :
SSE2
(Message 747226)
Posted 3 May 2008 by JDWhale
Post: Hi, Okay... shaved almost 50% off the CPU time, but I'm discouraged that it didn't validate. The shortened WUs in the bench tool looked good enough for the "live" trial, but for now I've put KWSN R2.4 SSE2 back on the laptop while I work more on the AK SSE2 code. I'll post any developments to this thread. Regards, JDWhale |
|
565)
Message boards :
Number crunching :
Linux port of Alex v8 code
(Message 747164)
Posted 2 May 2008 by JDWhale
Post: VMWare appears very efficient, all CPU times and "clock" times seem spot on! Now I see why EMC spun off VMWare to it's own company... Good Stuff! |
|
566)
Message boards :
Number crunching :
Linux port of Alex v8 code
(Message 747149)
Posted 2 May 2008 by JDWhale
Post: [quote][quote] @Raistmer - Very good point. Wonder if Crunch3r would post some data from the log file. CPU times and clock time might differ some. |
|
567)
Message boards :
Number crunching :
Linux port of Alex v8 code
(Message 747139)
Posted 2 May 2008 by JDWhale
Post:
No doubt. But the compare is not complete without running 8 cores loaded native Linux. You might check out UL1's X5482 rig, though still slower on a per/core basis the clock is much closer to Crunch3rs rig. |
|
568)
Message boards :
Number crunching :
SSE2
(Message 747095)
Posted 2 May 2008 by JDWhale
Post: I'm running live with my first "cut" at SSE2 client on my P4M laptop Gilligan. It'll be a few hours before he reports, I'm predicting ~20% or better improvement on the mid AR WUs in his queue (currently 6.5 hours down to 5.0 hours). The code changes will require minimal effort to install on current ported code. Now it's just playing the "waiting game" to see if Gilligan can get a variety of WUs to see how other ARs perform. Regards, JDWhale |
|
569)
Message boards :
Number crunching :
"Task was reported too late to validate"
(Message 747086)
Posted 2 May 2008 by JDWhale
Post: I received this error message on these two work-units: Task ID: 826965581 and Task ID: 826962533. And you've got at least 4 more that are going to be granted 0.0 credits for the same reason. You could cancel any "detatched" WUs, you've got ~10 of them, and not feel guilty since they've already been assigned new wingmen in your place. |
|
570)
Message boards :
Number crunching :
SSE2
(Message 746981)
Posted 2 May 2008 by JDWhale
Post: LOL, so my "Swiss watch made from wet noodles" analogy isn't too far off then ... ;D LOL....I know my brain feels like "wet noodles" after sorting those MM_SHUFFLE MASKS. I think I'll implement macros for all SSE3 functions and restart with the original code.... we know the original works! My brain is soggy, I don't know anything right now! |
|
571)
Message boards :
Number crunching :
SSE2
(Message 746956)
Posted 2 May 2008 by JDWhale
Post:
I've had a go at the SSE2 last night with a pitcher of Strawberry-Mango Margaritas... Got it to build ~4am, but was crashing early in the WU... A quick debugging session after a couple hours sleep seem promising. Now I'm crashing at the end of my simple WU instead of at the beginning. anyway... point is: I've desimplified some of those SSE3 calls to SSE sequences... example: p1 = _mm_hadd_ps( f1_1, f1_2 ); becomes p1 = _mm_add_ps(_mm_shuffle_ps(f1_1, f1_2, 0x88), _mm_shuffle_ps(f1_1, f1_2, 0xdd)); and cd1 = _mm_addsub_ps(cd1, td1); becomes cd1 = _mm_shuffle_ps(d1=_mm_shuffle_ps(_mm_sub_ps(cd1,td1),_mm_add_ps(cd1,td1),0xd8), d1, 0xd8); Also, these macros work for a couple others... #define _mm_moveldup_ps(X) (_mm_shuffle_ps( (X) , (X) , 0xa0)) #define _mm_movehdup_ps(X) (_mm_shuffle_ps( (X) , (X) , 0xf5)) So..... if anyone wants to pick this up and have a go, the above might help you get started... Back to trying to get this thing to run on my P4M (Northwood) laptop. Regards, John |
|
572)
Message boards :
Number crunching :
One of my computers is down
(Message 746679)
Posted 1 May 2008 by JDWhale
Post: Come on guys... we are talking about a single WU here and risking screwing up ~5 WUs on his working system. One WU is no great loss when it comes to missing a deadline.... IMO, more trouble than it's worth. Just my $.02, JDWhale |
|
573)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 746642)
Posted 1 May 2008 by JDWhale
Post: Just some trivia for an old P4 Socket 478 3.2GHz oc'd to 3.6GHz running the KWSN 2.4 code here. I have now upgraded to the SSE3 AK V8 port and we'll see how it progresses...... FYI: My 3.2Ghz Prescott Ginger is approaching RAC 1000 after 3 weeks back online with the SSE3 AK port. Hmmmm.... Ginger results only show... Speed: 1 x 3200 MHz even though HT is enabled and is crunching 2 WU at a time... I see yours reports similar. So that must be "core count" and not "thread count" reported, a little misleading. Kind regards, JDWhale |
|
574)
Message boards :
Number crunching :
One of my computers is down
(Message 746634)
Posted 1 May 2008 by JDWhale
Post: and I think I may miss the deadline to send the result.is there a way I could get this sent to my current computer, so as to not miss the deadline? One of your hosts reported a result within 3 minutes of you posting this message, so I'll assume that you refer to your "other" host as being "down". I wouldn't worry too much about it since that host only has 1 WU currently "checked out" for crunching. BOINC projects are designed to handle missed deadlines, that WU will just get reissued to another host when your deadline passes without reporting results. Kind regards, JDWhale |
|
575)
Message boards :
Number crunching :
AK V8 ported release ap. issues, install, questions etc.
(Message 746575)
Posted 1 May 2008 by JDWhale
Post:
Your quota is 400 WU/day/host... It looks like you have your settings to "always connected" and cache set to ~3 additional days if even set at all. Of course I'm often wrong in my analysis. Your hosts will be able to crunch ~90 WUs/day each... So you will not meet your quota with your current settings... just steadily DL ~100 WU/day to keep ~300 WUs ready to crunch. BOINC On...On... JDWhale BTW... I've backed down Thurston OC to 3200 from 3375 MHz and watched my temps drop ~10C (59-52-55-57) room temp 28C. Also of note is that power draw has dropped from 256 to 200W, saving me $3.25/mo (25%) on electric bill for this host. |
|
576)
Message boards :
Number crunching :
V8 app problem
(Message 746563)
Posted 1 May 2008 by JDWhale
Post: http://setiathome.berkeley.edu/result.php?resultid=821256340 Could you post related messages from your logfile: "..\\BOINC\\stdoutdae.txt", for this hostid=4182637. You can search the log file for "13mr08ag.31579.2526.9.8.4_0" to find the interesting messages for this resultid=821998871. Just messages from the time that the WU started until the WU ended. This will confirm that the WU took more or less than the ~15 seconds as reported and show the "wallclock time" from start to end to time it was uploaded. Here is an example WU resultid=826453714 from one of my host that had similar reported CPU time: 01-May-2008 11:11:53 [SETI@home] Starting 23mr08af.18852.109002.15.8.0_1 01-May-2008 11:11:53 [SETI@home] Starting task 23mr08af.18852.109002.15.8.0_1 using setiathome_enhanced version 528 01-May-2008 11:12:09 [SETI@home] Computation for task 23mr08af.18852.109002.15.8.0_1 finished 01-May-2008 11:12:09 [SETI@home] Resuming task 23mr08ah.13667.20113.13.8.128_0 using setiathome_enhanced version 528 01-May-2008 11:12:13 [SETI@home] Started upload of 23mr08af.18852.109002.15.8.0_1_0 01-May-2008 11:12:13 [SETI@home] Starting 23mr08ah.13667.20113.13.8.166_0 01-May-2008 11:12:13 [SETI@home] Starting task 23mr08ah.13667.20113.13.8.166_0 using setiathome_enhanced version 528 01-May-2008 11:12:16 [SETI@home] Finished upload of 23mr08af.18852.109002.15.8.0_1_0 It might show other messages exposing the reason why it is reporting much less time than you observe, such as restarts or switching in/out of processing mode. Regards, JDWhale |
|
577)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 745332)
Posted 28 Apr 2008 by JDWhale
Post: Only 2.5 days before the Alex code port release! Nobody does it like Michael Buffer, "Let's Get Ready to Rumble". |
|
578)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 745244)
Posted 27 Apr 2008 by JDWhale
Post: A quick question: Do what you are comfortable with. Personally, I might switch clients 2-3 times a day on some hosts that are running client comparisons, and never even stop BOINC. But then, I am comfortable with the way I operate my hosts and would not encourage others, less comfortable, to do the same. Maybe, rather than running it all the way out, set it so that you'll only lose 0.5 - 1 days work if sonething goes awry when you switch over. Regards, JDWhale |
|
579)
Message boards :
Number crunching :
Budget Quad cruncher under $400.00
(Message 745193)
Posted 27 Apr 2008 by JDWhale
Post: At Mwave.com you can build your own quad cruncher for under 400.00 and at most stores I would bet. When I look at the QX9775 (SKT chip) at ~$1650 (US) and realize I can build 4 of your crunchers for the cost of a single SKT chip, I have to pinch myself and smile. (Granted I'll add ~$10 each for USB memory stick so I can run BoincPE) nice build hiamps... -JDWhale |
|
580)
Message boards :
Number crunching :
Recent price cuts on Intel CPUs
(Message 745089)
Posted 27 Apr 2008 by JDWhale
Post: It's nice that faster processors are getting cheaper. The bad thing is, that almost always, one has to buy a new motherboard and new memory to take advantage of (or even fit) the faster processors. True. But didn't Intel introduce the "Slot 1" to get away from the need for new Mobo to take advantage of newer processors ? Or was there another reason ? Granted that was a few generations ago, but seems the technology still only lasted for one generation... Just another marketing gimmick to get you to buy another motherboard. IMO. [edit] @David - I've worked with a few "Mad Scientists" over the years... Always a fun bunch when it comes time for a p!$$ up. [/edit] |
|
581)
Message boards :
Number crunching :
Recent price cuts on Intel CPUs
(Message 744969)
Posted 27 Apr 2008 by JDWhale
Post:
Like dcappello points out... If you live near a "MicroCenter", the Q6600 deal for $199(US) is a great price and has been available "instore" for the past 2 months, I bought one in early March. For those of us who are budget minded, like ME, a "no frills" system can be built by combining this chip with the GigaByte GA-P35-DS3L mobo ($100) and 2GB cheap DDR2-800 memory kit($50) and cheap graphics card for ~$550 total. Full time crunching on 4 cores at stock 2400MHz draws < 170 Watts. Same power consumption as my 3200Mhz Prescott, but easily 500% the performance. In my opinion, it's cost effective to retire the power hungry computers of yesteryear and replace up to 4 individual system with a single newer generation computer. In addition to less heat, lower power, simpler to manage, this solution is also quieter. Donate the old systems to your church or childrens day care for possible tax deduction. Regards, JDWhale |
|
582)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 744954)
Posted 27 Apr 2008 by JDWhale
Post: Voyager wrote: hey number 20 for a q6600! That's to cool...Very good JD... David wrote: JD - jeez, position 20 is great - I didnt think you would make it to page 1 :) For the true high-flyers... this is just base camp... But Thurston feels like he is On Top 'O the World. Thanks, JDWhale |
|
583)
Message boards :
Number crunching :
Gigabyte GA-P35-DS3 REV 2.1 ore 2.0
(Message 744736)
Posted 26 Apr 2008 by JDWhale
Post: Yes this is Thurston.... GA-P35-DS3L rev.2.0 Temps are a bit high... room temp is 27C Thurston - 243 Watts 27C KWSN_2.4V_SSSE3_MB  Regards, JDWhale |
|
584)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 744346)
Posted 26 Apr 2008 by JDWhale
Post: Observed.... RC1 client draws 13 Watts more than KWSN_2.4V client. CoreTemp, vDroop demonstrate indifference to optimized client version. [edit] 27C refers to room temperature... Thurston case is open... CPU & case fans not powered by Mobo connects (running full speed). [/edit] Thurston - 129 Watts 27C Idle  Observations: After 10 minutes with SETI client "suspended". Thurston - 256 Watts 27C AK_v8_win_SSSE3x_RC1  Observations: Kill-A-Watt readings observed as high as 263 Watts near 50% completion af all WUs. WUs still pretty much in-sync with all 4 starting within 5 minutes of each other. Draw normally staying within 253-258W range. Kill-A-Watt sampling begins 5 minutes after last RC1 client starts. Draw averaged over 3 hours. Thurston - 243 Watts 27C KWSN_2.4V_SSSE3_MB Observations: Watching the Kill-A-Watt, first 1/2 of WU was reading 246 Watts... last 1/2 draw was 238 Watts..... I didn't believe, so I closely monitored when the 2nd round of WU started up... I watched draw rise from 235 Watts at the tail end of all 4 WUs to 246 Watts when the 4 new WUs were crunching. Similarly when 2nd round of WUs reached 50% completion draw fell to 236 Watts. Power draw is averaged over 3 hours. Thurston will remain running RC1 client for next ~60 hours to compensate for running these tests in quest to reach RAC 7090... Regards, JDWhale |
|
585)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 744296)
Posted 26 Apr 2008 by JDWhale
Post:
Ummm...No!  Actual photo gathering stats... KWSN client had 3 hours, will give RC1 same time... so still ~1 hour before results are posted. |
|
586)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 744231)
Posted 25 Apr 2008 by JDWhale
Post:
Sorry buddy, I'll post Thurston (Q6600@3375) numbers but don't have a penny in the corral to test. I feel that the Q6600 firmly owns the "sweet spot", the most bang for the buck. That said... I've got a mobo and some quality memory available if someone wants to rush me a QX9770 Yorkfield chip to test ;-) (PM me for shipping address) All joking aside.... Tests are underway, expect results in a couple hours. (stage 2 beginning in ~45 minutes) I wish I knew how to reset the clock/KWh feature on the Kill-A-Watt so I wouldn't have to record values longhand. (I think I can record the numbers and still drink bourbon, I might not remember doing it tomorrow... but that's nothing new ;-) Trust me... I'm a professional!!! Ehll, I didn't even remember porting this AK code the next day ;-) @David - There might babble on this Friday night... ample supplies available ;-) Cheers, JDWhale |
|
587)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 744108)
Posted 25 Apr 2008 by JDWhale
Post: I seem to remember that while running the new app. Mark had to up his voltage,for the new app. ran hotter, or something like that.So will we have to tone down a bit? I guess with the increase of the app. this will not matter that much. In the interest of checking difference in core temps & vCore droop, I propose to divert Thurston to run R-2.4V SSSE3 client long enough to crunch 2WUs/core. (~4 hours) I'll follow up by running RC1 SSSE3X and post Core_Temp and CPU-Z screen caps to show temps & vCore idle and under load with each client... Anyn interest in this set of tests ? Might this information be beneficial? Regards, JDWhale @john deneer - Agreed. My room temp is 27C, case open, I'll snap the caps when the AC isn't blowing. Also I'll supply power through Kill-A Watt to get system draw. |
|
588)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 744054)
Posted 25 Apr 2008 by JDWhale
Post: FYI : I have been assured via PM from Francois (V) that the client errors that drained his cache on the SKT were in no way related to running AK_WhalePort V0.2x SSE4.1 client. In at least one way, the clearing of his cache was a good thing as he was already being granted 0.0 credit for crunching "detatched" WUs. wuid=251118044 Regards, JDWhale |
|
589)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 743991)
Posted 25 Apr 2008 by JDWhale
Post: I hope thats all it is... I use XP and the modified BOINC client and would hate to see the current cache crash here next week lol But I do not run SSE4.1 on my Q6600 whereas, V can run SSE4.1 on his SKT (QX9775) SSE4.1 is only for 45nm chips like penny. You can be safe and run SSSE3 on any Core2 and above. But to answer your question... for best performance you will need different build of the new client for QX9650 and QX6700 |
|
590)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 743930)
Posted 25 Apr 2008 by JDWhale
Post: - Might the developer group consider releasing these apps one day early, on Wednesday instead of Thursday, in case there are problems that create a massive load on servers requiring Matt’s attention? He is not usually “in the office†on Fridays. Again, this would be a remote possibility. . . I'm hoping that Jasons "May 1", translates to Berkeleys "Apr 30", IYKWIM |
|
591)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 743885)
Posted 25 Apr 2008 by JDWhale
Post: With Who?'s experience, is this suggesting there is an issue with the Crunch3r BOINC Manager (6.1.x) and the ported Alex code? I don't think so... Rather a Vista/BOINC issue... When I first noticed an issue with ~6 WUs exiting with errors, I didn't think much of it. Looking back at the exit code (0x40010004) and googling it, I see this in a BOINC FAQ: [FAQ: exit code 1073807364 (0x40010004) (on Windows Vista) Is this related to the subsequent canceling of all the WU's in the cache? I can't say, but I think the 2 events might be related. My question to the few who are following this thread... Please report if you've had similar experience so we can reasonably disqualify v8 ported code, AK_WhalePort, as a probable cause. Granted there are not very many VISTA users out there and the conditions for the error as reported in the FAQ is "Shutting down VISTA while BOINC is running". |
|
592)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 743853)
Posted 25 Apr 2008 by JDWhale
Post: crunch3r BOINC manager and the V8 port looked like they did get mad together ... all my workloads are gone in error. I am sorry, but maybe a blessing with all the detatched WUs. After the "detatched" WUs are reissued to new wingman and reported back, then you would get 0.0 credit. This "error" will cause you to start fresh, without detatched WUs in your cache. One of the first WUs to show error may have clue to failure: exit code (0x40010004). That error is diagnosed in this thread [Vista] Avoiding client error exit code 1073807364 (0x40010004). I don't know if you can relate, but might be caused by Vista shutting down too quickly for BOINC. Regards, John |
|
593)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 743677)
Posted 24 Apr 2008 by JDWhale
Post: [edit]message deleted[/edit] If you crashed... Mark had mentioned he needed to raise vCore a bit to account for the extra stress being put on his OC penryn with the new client. You probably have more information/resources on that requirement than I could help with. Check your BOINC log. About 4 days ago Thurston suffered from a "random project reset" where all WUs were deleted and detatched. More details on this were posted to this thread... like I said about 4 days ago. Since then I pointed a small fan directly on my memory and adjusted vDimm18 a bit. I don't have a lot of confidence in my BALLISTIX DDR2-1066 dimms. Looking good, John |
|
594)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 743649)
Posted 24 Apr 2008 by JDWhale
Post: [edit]message deleted[/edit] The Skulltrail has changed hostid and "detached"... New host is hostid=4262769 |
|
595)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 743641)
Posted 24 Apr 2008 by JDWhale
Post: With the release of the new app's about 1 week away, I have one question. I'll try one example... When wanting to find the MEAN of a vector(1D array), typically a programmer would write ~5 lines of code to loop over the all the elements to gather a sum, then divide the sum by the vector length. Simple enough and easy to read/understand, even for most non-programmers. Alex code might implement the same operation using maybe 100 lines of code, split the array into 4 or eight pipes, feed the "vector" part of the processor allowing 4 or 8 sums to take place in the same amount of CPU clicks as a single sum operation. Accumulate the sum from the separate pipes, then do the division. Not very intuitive, but making much more efficient use of the CPU and making the code much more complex to read and to write and to comprehend. Also requires coding some of the optimizations multiple times, once for each CPUs capabilities to gain maximum performance on each CPU type. This process uses what are called "intrinsic functions" that allow "low level" access to the CPU capabilities. This set of intrinsics change from CPU to CPU and are sometimes refered to as SSE, SSE2, SSE3, SSSE3, SSE4.1, MMX, ...etc. Explaining the need for more than one client, depending on which "instructions" your host(CPU) can understand. This was just the most simple of examples of how Alex speed up the SETI client. Regards, JDWhale [edit] The Skulltrail has changed hostid and "detached"... hostid=4262769&offset=1800 [/edit] |
|
596)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 743620)
Posted 24 Apr 2008 by JDWhale
Post: Starting this morning, one of my Skulltrail will be running John (JDWhale) version of the Alex V8 code, based on his request. I will use the other machine to try to catch up with my own code :) ... First results have posted hostid=4081480&offset=1620 These early results are hybrids, partially crunched with another client. "Purebreed" should be a bit faster ;-) @Martin - I only DL and install Intel ICC & IPP 32-bit evaluation copies. So the "real" clients will be even faster. In addition, 32-bit runs on both 32 & 64 bit OS. |
|
597)
Message boards :
Number crunching :
fairer Credit-system
(Message 743609)
Posted 24 Apr 2008 by JDWhale
Post: I equate credits to whale poop. Humpback whales crunch krill and make poop. Seti clients crunch WUs and make credits. End results are worth about the same in either case. [edit] After further thought, whale poop has more value... helps to make more krill. [/edit] |
|
598)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 743577)
Posted 24 Apr 2008 by JDWhale
Post: Starting this morning, one of my Skulltrail will be running John (JDWhale) version of the Alex V8 code, based on his request. I will use the other machine to try to catch up with my own code :) , that is a big task, because Alex/John code is amazingly fast. Thank you for allowing AK_WhalePort to run on your Skulltrail. The version I provided is V0.2X built using SSE4.1 IPP 32-bit. Since I do not have any hardware to test this version (45nm chip) I can only hope that this client version will perform well. I will be watching and waiting to see some results post. Note: AK_WhalePort is a "verbatim" windows port of Alex v-8 code... I only did what was necessary to get the source code to build & link substituting Intel IPP calls for Apple proprietary DSP functions. Again... Thank you Francois.. Best regards, JDWhale @mr.kjellen - Thanks... except for untold misfortune by the leaders I can not expect Thurston to rise any more... Since starting this quest/race the RAC to reach top 20 has risen by maybe 300-500. |
|
599)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 743515)
Posted 24 Apr 2008 by JDWhale
Post: I will pull the plug on Ginger when her cache empties in a couple days. Summer is approaching quickly here in SE Texas. Room temps are at 27C and Ginger stays in heat. Coupled with her energy draw puts me over my project budget of 1kW since Wrongway came online. She may be going down, but she is not going out. Is destined to my 21YO daughter come June. My focus is now pointed at "burning in" the release candidates so there are fewer (hopefully none), surprises upon release. I'm trying everything I know to break RC1... Used it as a coffee holder one morning... Had it fetch the post another...Manning the BBQ last night... Solid as a rock! We're talking "Top Shelf" stuff here. and have refrained from cancelling WUs where my wingam was running BOINC 4.45. I hate those... and sometimes you get a gaggle of WUs with the same wingman, who happens to go missing leaving you hold the pending bag for a couple months. Yes, modus operandi(MO) for Thurston avoid most "pending" pitfalls as credit is usually granted when the result is returned ("checked but no concensus" is the exception). As such I would expect RAC to stabilize 2 weeks after change is applied as long as he doesn't run out of work. That is one of the pitfalls of this MO. Without many pending, his RAC will drop like a rock when the work runs out... Thus a large cache is essential. Without another interruption, I expect Thurston to reach stable RAC in about 72-96 hours, target is still RAC 7090. G'Day, JDWhale |
|
600)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 743343)
Posted 24 Apr 2008 by JDWhale
Post: bump With the switch to RC1 SSE3 estimated RAC should be updated once again for Skipper, Lovey, and Wrongway; but I haven't plugged the data into my unofficial guestimator yet. Sorry still no updated estimates for RAC on these hosts. I've been busy trying get the holes on the data samples filled on Lovey to paint a more complete picture by our wonderful statisticians. With my running Lunatics RC1 on these hosts I agree that the value of AK_WhalePort displays are becoming of less interest. With the data painting a nearly complete fill(given the availability of variation in WUs), monitoring of Skipper, Ginger, Lovey and Wrongway should be terminated at your discretion. FYI : I will pull the plug on Ginger when her cache empties in a couple days. My focus is now pointed at "burning in" the release candidates so there are fewer (hopefully none), surprises upon release. A brief update on 2 of my hosts: Host = Wrongway: Q6600 @ 2520MHz  06Apr My estimated RAC = 4700 13Apr My revised estimated RAC = 5040 23Apr Newest revised estimate RAC = 5400 I realize that Wrongway is now running RC1 SSSE3X flavor, this fact has minimal (if any) impact on my evaluation. My earlier estimates on this host were intentionally conservative. This RAC 5400 number is the number you should expect if you are running a "stock clocked" Q6600 with DDR2-800 dual channel memory and the new client. Host = Thurston: Q6600 @ 3375MHz  06Apr My estimated RAC = ~7400 13Apr My revised estimated RAC = 7080 Recovery from the "Random Project Reset" 3 days past was fairly smooth, helped with the immediate DL of 80 reissued WUs that I used to my advantage to minimize "pending credit" build. Please understand... I am not using BOINC default schedueling with Thurston. I am employing techniques to minimize pending credits. I have only cancelled a couple WUs that were already validated by late wingmen and have refrained from cancelling WUs where my wingam was running BOINC 4.45. My employed methods are intended to reach "final stable" RAC in as little time as possible and are not intended to inflate RAC beyond what would be attained with more available time. YMMV Estimated RAC 7090 remains for Thurston. This should be your expected performance of a Q6600 @ ~3300MHz with DDR2-1066 dual channel memory. FYI: On the jukebox Frank by "Amy Winehouse"... How can anybody resist her sultry voice and honesty? IMO, totally deserving of her Grammy as "Best New Artist". Still searching for the "Nuclear Wessels", JDWhale |
|
601)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 743044)
Posted 23 Apr 2008 by JDWhale
Post: oupps ... my 2nd machine is running the merge of 2 versions of my code+ 2.4, I forgot to adjust the multiplier, please don t get mad, i just forgot.Will be fix in morning. [Edit] Post # 777 Wish I was playing the slots... [/edit] Maybe you refer to result like wuid=249775937. I understand, simple mistake. You only get "extra" credit if your wingman also "overclaim". Thank you for sharing... Your SKT are looking very good and climbing very fast. Congratulations to you! Regards, JDWhale |
|
602)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 742868)
Posted 22 Apr 2008 by JDWhale
Post: JD; Thanks for the offer... First a question back at you... Is your signup date really earlier than mine? I find it hard to trust anyone claiming to have joined the project before me. Just kidding.... With the pending release the new Windows clients only about a week away, It would be hard to gather a meaningful sample in the limited time. The folks monitoring hosts already have more on their plate than they can comfortably chew. Add to that the requirement of your being able to switch between OSX and WindowsXP/Vista on your MacPro as/when directed make this a difficult proposition. I hope you understand, JDWhale [Edit] Wait just a minute... I just spotted you at #30 spot and rising really, really fast. Does your Mac do Windows? Maybe I can sidetrack you to allow me to get even further up the ladder ;-) [/edit] |
|
603)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 742788)
Posted 22 Apr 2008 by JDWhale
Post: Thurston is positioned at #21 on the "Top Hosts" list as I write this. With some help from our fallen friend, #20 appears guaranteed within the next day or so. There is one host I'm in fear of jumping in ahead of me, UL1' Big Mac. Though currently sitting in #31, this host has the ability to leap several spots when fed the proper diet. I am begging The Curmudgeon to throw down the gauntlet if need be. Do not let Big Mac past. The quest continues, JDWhale |
|
604)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 742738)
Posted 22 Apr 2008 by JDWhale
Post: I know it's not related to your App JD, but on the Mac Pro, generally units of the same AR take within 200 seconds of one another to complete. If you look at any group of results for the Mac Pro you will find that finish times are very similar to one another. That is exactly the point I was trying to get across... In QS case, by switching to the Mac "xeon" client he is seeing a similar "narrowing" in the range of CPU times compared to when he was running the "core2" flavor in the same bins. Of course this observation is made with only a limited # of results reporting. The ranges could spread as more variety of ARs within the bins are crunched. A scatter plot, with points representing each result and vertical lines representing bin boundaries would bring more meaning to the range of CPU times per bin with these preliminary results. ( Just what our excellent chart people need to hear, more extraneous drivel from the peanut gallery ;-) [runs and hides before someone can toss an eraser at me] Cheers, JDWhale |
|
605)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 742724)
Posted 22 Apr 2008 by JDWhale
Post: Notes: Thanks for the new format, There is quite a bit more information and it took me a couple glances to really see the benefit. I like it! From this it can be seen that the mean values from the relatively few Xeon results so far would seem to indicate better performance than Core2, but that all the results for Xeon are between the max and min times for Core2 in any given AR band. This demonstrates the caveat that I have so often quoted. What I see, with the limited data so far, is that the Xeon client CPU times are more consistant (within a tighter range). Possible explanation for this is lesser influence by other process running on the host. I would expect the bars for the Xeon to extend downwards as more readings become available over the next few days, pulling the mean value down to demonstrate better performance. But only time will tell. Maybe, but if the CPU times remain so tightly grouped and less than the Core2 mean, doesn't that also indicate better performance?
Touche... |
|
606)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 742545)
Posted 22 Apr 2008 by JDWhale
Post: Just the facts! This reporter has just been informed of The Definitive Answer to the longstanding question of : Which client is faster on the same hardware: OSX or WinXP ? The pretty pictures will follow, but I'll give a brief account of what I have been made privy to... It is too close to call!!! It is a "photo finish" folks. DO NOT TEAR UP YOUR BETTING SLIPS!!! We are waiting for the photos to be developed. Personal thanks to :
for making this test possible. |
|
607)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 742379)
Posted 21 Apr 2008 by JDWhale
Post: It will be an interesting 4-6 weeks after the Alex code port is released to see how much the hardware can be pushed. It may also give an indication of any limitations due to the OS, within experimental error. Interesting few weeks indeed. I wouldn't be surprised to see a slight shift in which hardware configurations (speeds{cpu,fsb,memory}, fine tuning, L2 cache sizes) perform best. Client choice might become a bit confusing. I doubt I'll get in too much "hot water" with the Lunatics by leaking here that a great deal of analysis is being performed on benchmark results from the testers (self included as tester) in order to present as clear a choice as possible for which client provides best "overall" performance on your hosts. Different clients might be better at certain ARs than others and vice versa. Thus overall performance might vary depending on the WUs being served and chosen client. Keep on BOINCing, JDWhale |
|
608)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 742301)
Posted 21 Apr 2008 by JDWhale
Post:
I'm sorry... License restriction... you know... My only public invitation still stands, someone Who? can sidestep those restrictions and give us all a show of what is possible. @Gas Giant - Excuse me... Is that a Chupacabra as your avatar ? Or are roo just happy to see me ? May the BOINC be with you. |
|
609)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 742198)
Posted 21 Apr 2008 by JDWhale
Post: How about a weekly update with just minimal babble... With the switch to RC1 SSE3 estimated RAC should be updated once again for Skipper, Lovey, and Wrongway; but I haven't plugged the data into my unofficial guestimator yet. We should be seeing some updated plots/charts if/when the data fills in the AR bins. Host = Skipper: P4D-820 @ 2800MHz  06Apr My estimated RAC = 1280 13Apr My revised estimated RAC = 1342 Host = Lovey: E4500 @ 2420MHz  06Apr My estimated RAC = 2300 13Apr My revised estimated RAC = 2720 Host = Wrongway: Q6600 @ 2520MHz  06Apr My estimated RAC = 4700 13Apr My revised estimated RAC = 5040 Note here... As I'm writing, Wrongway is still running benchmarks for the Lunatics crew. After review of said benchmarks, I'll switch BOINC to run an RC1 client of my choosing (probably the one with the best performance in the mid-AR WUs we're seeing so many of. Warning : if there is an onslaught of VHAR or VLAR WU's I won't hesitate to switch to the best client for the data. Same holds true for Lovey. Host = Thurston: Q6600 @ 3375MHz  06Apr My estimated RAC = ~7400 13Apr My revised estimated RAC = 7080 Remember - Random unattended "Reset project" ~40 hours ago :-( At least it's going in the right direction! However tempting to switch... Thurston will remain on AK_WhalePort V0.2X until official Lunatics release... I want to see how accurate my estimated RAC of 7080 is. Host = Ginger: P4-540 (Prescott H/T) @ 3200MHz Sorry, Ginger is in the powder room and doesn't want her picture took. Trust me, her graph is climbing, but since she's been switching between KWSN R-2.4 (SSE3) and AK_WhalePort V0.2S (SSE3), and now running RC1 (SSE3), it's not a reflection of a single clients performance. Trust the charts/graphs posted by the experts. Regards, JDWhale |
|
610)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 741892)
Posted 20 Apr 2008 by JDWhale
Post:
The "Xeon" clients actually only refer to an additional AK optimization that at the time might have been intended to steer clear of Core2 with the limited L2 cache, etc. Intel has released so many variations of Core2, the optimization might now have a name other than "Xeon". Yes, The AK_WhalePort Xeon is/was faster than the Core2 build... After building SSE3 for Ginger & Skipper, I never thought to try SSE3 on Lovey. Until I ran the benchmarks on the new Release Candidates (RC) last night I didn't think that SSE3 would benefit the E4500 Core2 Duo. My numbers are mixed, thus I'll be running each of three new RCs on Lovey.. My guess is the smaller L2 cache might benefit at some ARs with SSE3. I thought others might benefit from my experiments when it comes to choosing the "best" client for some of the more economical Core2 Duos. As Murphy would have it... I was going to post my findings/results at Lunatics first thing this morning, but I've been exposed. (Jason might not let me "sample the cookie dough" next time :-( ) I'm considering a similar test for Wrongway, but you didn't hear me say that. Thurston is out of the question, he is staying 100% AK_WhalePort in his quest to the top 20 (then there might be a change). Cheers JDWhale |
|
611)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 741879)
Posted 20 Apr 2008 by JDWhale
Post: A message of comfort to those running SSE3 but who can't run SSSE3 ... Can't slip anything by my ol' buddy Fred W. [...mumble...grumble...] While running one of my convoluted benchmark ideas, I switched to running the latest SSE3 release candidate inside BOINC (single process only) on both Ginger and Skipper for a few hours as your detective work points out... I'd switched back to running AK_WhalePort V0.2S after the benchmarks run on both hosts as further detective work will confirm. Since being discovered, I'll now switch back to running the new SSE3 release candidate for a good portion of the day, this time on both processes live. This should help stabilize the numbers a bit, though I doubt there is much variation in WU ARs in the queue. I hadn't looked in detail at the "numbers" you've quoted. "I couldn't see the forest for the trees", so to speak.... Details...Details...Details. If you wouldn't mind, please keep an eye on Lovey also. Preliminary benchmarks on the E4500 chip are mixed, so I'll be running 3 different release candidates (SSE3, SSSE3, SSSE3X) here... your pictures will tell a better story than I can gather only by looking at the numbers. Happy Sunday/Monday as applies, JDWhale |
|
612)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 741464)
Posted 19 Apr 2008 by JDWhale
Post: And don't forget who? still has the fastest WINTEL machine around..... Welcome... Guy Fawkes was born April 13, was that the day you assumed the new name in his honor? Then I can only assume your "final version" will be on November 5, as a final tribute. Good luck, JDWhale |
|
613)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 741384)
Posted 19 Apr 2008 by JDWhale
Post: Thurston might have just caught a break... He DL'd ~70 resends of WU's that had past deadline or had problems DLing, etc. That works out to be ~18 hours worth, so I bumped them to the head of the queue. This might help offset starting with a fresh cache and slow the climb in "pending credits" a bit. The WU names all looked fine using "explorer", all looked fine in "client_state.xml". I'm pretty sure almost all were ~XXMar08YY, the new stuff. Step 300-500 hundred WU's into Thurstons cache and browse amongst the "detatched" WUs. They were all DLd over the past week 11-18 Apr. What causes a random unattended "Project reset"? Could be memory error, or maybe a "Vista" issue. I'm pretty sure the odd WU names were locally generated by a run awry BOINC. If others have had this issue, maybe it's a bug in the OS, If I'm an isolated case I might point to memory error. If it happens again, I'll be sure to run memtest before proceeding. Hmmmmm.... A closer look shows a definite pattern. The same WU names starting and ending at 1 minute intervals. Here is a longer snippet from the end of the event. It looks like Thurston went to sleep from 03:15 until I pressed the power switch at 04:06..... 19-Apr-2008 03:13:20 [SETI@home] Starting 5fe08ab.29541.273322.15.8.4_2 19-Apr-2008 03:13:20 [SETI@home] Starting task 5fe08ab.29541.273322.15.8.4_2 using setiathome_enhanced version 528 19-Apr-2008 03:13:20 [SETI@home] Starting ®mr08ab.27614.7025.6.8.205_0 19-Apr-2008 03:13:20 [SETI@home] Starting task ®mr08ab.27614.7025.6.8.205_0 using setiathome_enhanced version 528 19-Apr-2008 03:13:20 [SETI@home] Starting 'mr08ao.27675.6207.7.8.186_0 19-Apr-2008 03:13:20 [SETI@home] Starting task 'mr08ao.27675.6207.7.8.186_0 using setiathome_enhanced version 528 19-Apr-2008 03:13:20 [SETI@home] Starting ÿÿmr08ao.27675.6207.7.8.173_0 19-Apr-2008 03:13:20 [SETI@home] Starting task ÿÿmr08ao.27675.6207.7.8.173_0 using setiathome_enhanced version 528 19-Apr-2008 03:13:21 [SETI@home] Computation for task 5fe08ab.29541.273322.15.8.4_2 finished 19-Apr-2008 03:13:21 [SETI@home] Computation for task ®mr08ab.27614.7025.6.8.205_0 finished 19-Apr-2008 03:13:21 [SETI@home] Computation for task 'mr08ao.27675.6207.7.8.186_0 finished 19-Apr-2008 03:13:21 [SETI@home] Computation for task ÿÿmr08ao.27675.6207.7.8.173_0 finished 19-Apr-2008 03:14:20 [SETI@home] Starting 5fe08ab.29541.273322.15.8.4_2 19-Apr-2008 03:14:20 [SETI@home] Starting task 5fe08ab.29541.273322.15.8.4_2 using setiathome_enhanced version 528 19-Apr-2008 03:14:20 [SETI@home] Starting ®mr08ab.27614.7025.6.8.205_0 19-Apr-2008 03:14:20 [SETI@home] Starting task ®mr08ab.27614.7025.6.8.205_0 using setiathome_enhanced version 528 19-Apr-2008 03:14:20 [SETI@home] Starting 'mr08ao.27675.6207.7.8.186_0 19-Apr-2008 03:14:20 [SETI@home] Starting task 'mr08ao.27675.6207.7.8.186_0 using setiathome_enhanced version 528 19-Apr-2008 03:14:20 [SETI@home] Starting ÿÿmr08ao.27675.6207.7.8.173_0 19-Apr-2008 03:14:20 [SETI@home] Starting task ÿÿmr08ao.27675.6207.7.8.173_0 using setiathome_enhanced version 528 19-Apr-2008 03:14:21 [SETI@home] Computation for task 5fe08ab.29541.273322.15.8.4_2 finished 19-Apr-2008 03:14:21 [SETI@home] Computation for task ®mr08ab.27614.7025.6.8.205_0 finished 19-Apr-2008 03:14:21 [SETI@home] Computation for task 'mr08ao.27675.6207.7.8.186_0 finished 19-Apr-2008 03:14:21 [SETI@home] Computation for task ÿÿmr08ao.27675.6207.7.8.173_0 finished 19-Apr-2008 03:15:20 [SETI@home] Starting 5fe08ab.29541.273322.15.8.4_2 19-Apr-2008 04:06:38 [SETI@home] Starting task 5fe08ab.29541.273322.15.8.4_2 using setiathome_enhanced version 528 19-Apr-2008 04:06:38 [SETI@home] Starting ®mr08ab.27614.7025.6.8.205_0 19-Apr-2008 04:06:38 [SETI@home] Starting task ®mr08ab.27614.7025.6.8.205_0 using setiathome_enhanced version 528 19-Apr-2008 04:06:38 [SETI@home] Starting 'mr08ao.27675.6207.7.8.186_0 19-Apr-2008 04:06:41 [SETI@home] Starting task 'mr08ao.27675.6207.7.8.186_0 using setiathome_enhanced version 528 19-Apr-2008 04:06:41 [SETI@home] Starting ÿÿmr08ao.27675.6207.7.8.173_0 19-Apr-2008 04:06:41 [SETI@home] Starting task ÿÿmr08ao.27675.6207.7.8.173_0 using setiathome_enhanced version 528 19-Apr-2008 04:06:42 [SETI@home] Computation for task 5fe08ab.29541.273322.15.8.4_2 finished 19-Apr-2008 04:06:42 [SETI@home] Computation for task ®mr08ab.27614.7025.6.8.205_0 finished 19-Apr-2008 04:06:42 [SETI@home] Computation for task 'mr08ao.27675.6207.7.8.186_0 finished 19-Apr-2008 04:06:42 [SETI@home] Computation for task ÿÿmr08ao.27675.6207.7.8.173_0 finished 19-Apr-2008 04:06:42 [SETI@home] Starting 5fe08ab.29541.273322.15.8.4_2 19-Apr-2008 04:06:42 [SETI@home] Starting task 5fe08ab.29541.273322.15.8.4_2 using setiathome_enhanced version 528 19-Apr-2008 04:06:42 [SETI@home] Starting ®mr08ab.27614.7025.6.8.205_0 19-Apr-2008 04:06:42 [SETI@home] Starting task ®mr08ab.27614.7025.6.8.205_0 using setiathome_enhanced version 528 19-Apr-2008 04:06:42 [SETI@home] Starting 'mr08ao.27675.6207.7.8.186_0 19-Apr-2008 04:06:42 [SETI@home] Starting task 'mr08ao.27675.6207.7.8.186_0 using setiathome_enhanced version 528 19-Apr-2008 04:06:42 [SETI@home] Starting ÿÿmr08ao.27675.6207.7.8.173_0 19-Apr-2008 04:06:42 [SETI@home] Starting task ÿÿmr08ao.27675.6207.7.8.173_0 using setiathome_enhanced version 528 19-Apr-2008 04:06:44 [SETI@home] Computation for task 5fe08ab.29541.273322.15.8.4_2 finished 19-Apr-2008 04:06:44 [SETI@home] Computation for task ®mr08ab.27614.7025.6.8.205_0 finished 19-Apr-2008 04:06:44 [SETI@home] Computation for task 'mr08ao.27675.6207.7.8.186_0 finished 19-Apr-2008 04:06:44 [SETI@home] Computation for task ÿÿmr08ao.27675.6207.7.8.173_0 finished Any similar experiences welcome... Moderators feel free to move this & related posts to a new thread. Cheers, JDWhale |
|
614)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 741333)
Posted 19 Apr 2008 by JDWhale
Post: Thurston might have just caught a break... He DL'd ~70 resends of WU's that had past deadline or had problems DLing, etc. That works out to be ~18 hours worth, so I bumped them to the head of the queue. This might help offset starting with a fresh cache and slow the climb in "pending credits" a bit. @Mark - Odd that we'd both fail at the same time... Maybe E.T. is sending us a message that we can't ignore ;-) |
|
615)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 741322)
Posted 19 Apr 2008 by JDWhale
Post: Dagnabit - Stayed up 'til 1:30AM watching Dream for an Insomniac... get up at 4:00am to "flush-n-fill" the radiator, so to speak... Look in on Thurston and find him asleep... Some time between 1:30 - 4:00 BOINC hiccupped and drained the effin cache. Boinc was still running. Here's the log file and see a "Resetting project" shortly after I went to bed...:-( 19-Apr-2008 01:32:45 [SETI@home] Computation for task 28mr08ab.16850.6207.5.8.205_0 finished 19-Apr-2008 01:32:45 [SETI@home] Starting 27mr08ao.27675.6207.7.8.173_0 19-Apr-2008 01:32:45 [SETI@home] Starting task 27mr08ao.27675.6207.7.8.173_0 using setiathome_enhanced version 528 19-Apr-2008 01:32:47 [SETI@home] Started upload of 28mr08ab.16850.6207.5.8.205_0_0 19-Apr-2008 01:32:53 [SETI@home] Finished upload of 28mr08ab.16850.6207.5.8.205_0_0 19-Apr-2008 01:37:47 [SETI@home] Sending scheduler request: Requested by user. Requesting 0 seconds of work, reporting 4 completed tasks 19-Apr-2008 01:37:52 [SETI@home] Scheduler request succeeded: got 0 new tasks [color=red]19-Apr-2008 01:55:18 [SETI@home] Resetting project[/color] 19-Apr-2008 01:56:10 [SETI@home] Starting 5fe08ab.29541.273322.15.8.4_2 19-Apr-2008 01:56:10 [SETI@home] Starting task 5fe08ab.29541.273322.15.8.4_2 using setiathome_enhanced version 528 19-Apr-2008 01:56:10 [SETI@home] Starting ®mr08ab.27614.7025.6.8.205_0 19-Apr-2008 01:56:10 [SETI@home] Starting task ®mr08ab.27614.7025.6.8.205_0 using setiathome_enhanced version 528 19-Apr-2008 01:56:10 [SETI@home] Starting 'mr08ao.27675.6207.7.8.186_0 19-Apr-2008 01:56:10 [SETI@home] Starting task 'mr08ao.27675.6207.7.8.186_0 using setiathome_enhanced version 528 19-Apr-2008 01:56:10 [SETI@home] Starting ÿÿmr08ao.27675.6207.7.8.173_0 19-Apr-2008 01:56:10 [SETI@home] Starting task ÿÿmr08ao.27675.6207.7.8.173_0 using setiathome_enhanced version 528 . . . 19-Apr-2008 04:06:42 [SETI@home] Starting task ®mr08ab.27614.7025.6.8.205_0 using setiathome_enhanced version 528 19-Apr-2008 04:06:42 [SETI@home] Starting 'mr08ao.27675.6207.7.8.186_0 19-Apr-2008 04:06:42 [SETI@home] Starting task 'mr08ao.27675.6207.7.8.186_0 using setiathome_enhanced version 528 19-Apr-2008 04:06:42 [SETI@home] Starting ÿÿmr08ao.27675.6207.7.8.173_0 19-Apr-2008 04:06:42 [SETI@home] Starting task ÿÿmr08ao.27675.6207.7.8.173_0 using setiathome_enhanced version 528 19-Apr-2008 04:06:44 [SETI@home] Computation for task 5fe08ab.29541.273322.15.8.4_2 finished 19-Apr-2008 04:06:44 [SETI@home] Computation for task ®mr08ab.27614.7025.6.8.205_0 finished 19-Apr-2008 04:06:44 [SETI@home] Computation for task 'mr08ao.27675.6207.7.8.186_0 finished 19-Apr-2008 04:06:44 [SETI@home] Computation for task ÿÿmr08ao.27675.6207.7.8.173_0 finished 19-Apr-2008 04:08:34 [---] Starting BOINC client version 5.10.30 for windows_intelx86 19-Apr-2008 04:08:34 [---] log flags: task, file_xfer, sched_ops 19-Apr-2008 04:08:34 [---] Libraries: libcurl/7.17.1 OpenSSL/0.9.8e zlib/1.2.3 19-Apr-2008 04:08:34 [---] Data directory: C:\\BoincPE\\BOINC 19-Apr-2008 04:08:34 [SETI@home] Found app_info.xml; using anonymous platform 19-Apr-2008 04:08:34 [---] Processor: 4 GenuineIntel Intel(R) Core(TM)2 Quad CPU Q6600 @ 2.40GHz [x86 Family 6 Model 15 Stepping 11] 19-Apr-2008 04:08:34 [---] Processor features: fpu tsc pae nx sse sse2 pni mmx 19-Apr-2008 04:08:34 [---] OS: Microsoft Windows Vista: Professional Edition, (06.00.6000.00) 19-Apr-2008 04:08:34 [---] Memory: 2.00 GB physical, 4.21 GB virtual 19-Apr-2008 04:08:34 [---] Disk: 9.76 GB total, 9.29 GB free 19-Apr-2008 04:08:34 [---] Local time is UTC -5 hours 19-Apr-2008 04:08:34 [SETI@home] URL: http://setiathome.berkeley.edu/; Computer ID: 4087648; location: home; project prefs: default 19-Apr-2008 04:08:34 [---] General prefs: from SETI@home (last modified 09-Jan-2008 23:08:36) 19-Apr-2008 04:08:34 [---] Host location: home 19-Apr-2008 04:08:34 [---] General prefs: using separate prefs for home 19-Apr-2008 04:08:34 [---] Reading preferences override file 19-Apr-2008 04:08:34 [---] Preferences limit memory usage when active to 2045.88MB 19-Apr-2008 04:08:34 [---] Preferences limit memory usage when idle to 2045.88MB 19-Apr-2008 04:08:34 [---] Preferences limit disk usage to 4.88GB 19-Apr-2008 04:08:38 [---] Can't resolve hostname DimensionII in remote_hosts.cfg 19-Apr-2008 04:09:08 [SETI@home] Sending scheduler request: To fetch work. Requesting 3456003 seconds of work, reporting 0 completed tasks 19-Apr-2008 04:09:13 [SETI@home] Scheduler request succeeded: got 20 new tasks 19-Apr-2008 04:09:15 [SETI@home] Started download of 12mr08ab.21529.50559.15.8.167 19-Apr-2008 04:09:15 [SETI@home] Started download of 12mr08ah.19734.4571.10.8.45 19-Apr-2008 04:09:21 [SETI@home] Finished download of 12mr08ab.21529.50559.15.8.167 19-Apr-2008 04:09:21 [SETI@home] Started download of 12mr08ah.19734.4571.10.8.4 19-Apr-2008 04:09:22 [SETI@home] Starting 12mr08ab.21529.50559.15.8.167_1 I checked Thurston results and saw that he was still assigned all those WUs and detatched/reattached to project. Adjusted "DCF" to the value host Wrongway has and Thurston hit the ground running... I'll expect a bit of a hit to show up in the RAC graph for the next couple days. I hope this RMA'd BALLISTIX memory isn't showing signs of failure. @Joker - "But I haven't got a wiscence to shoot a fwickaseein wabbit." @The Gas Giant - I'm offering Who? a chance to run AK_WhalePort if he desires. I'd really like to see how it performs on the SKT and don't think I'd run into those pesky "licensing issues" if he runs the client. Back to bed for me... JDWhale |
|
616)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 741233)
Posted 19 Apr 2008 by JDWhale
Post: @user formerly known as "Who?" - If you're following this thread I wish you the best of luck with your code... (It is fun to have your "own" version, isn't it?) I'll try to nip this in the bud... I honestly do respect Who? and his effort and will stand for no Who? bashing! If I came across differently, I am truly sorry I gave everyone a glimpse at that message. If you are a coder, there is a bit of satisfaction in 'having your own code' to massage and tweak to your desire, experiment here, experiment there, what works, what doesn't. That is the only message I wanted to get across. Who's effort differs very much from the "direct", or "verbatim" port that I did with AK_WhalePort, rather it is my understanding that Who? chose a more difficult path by trying to bring Alex optimizations to the current KWSN (V-2.4) structure. A valiant effort and much more difficult task than what I tried to do. Please welcome Who? if he chooses to join us, if not then please let him be. I want no Who? bashing in this thread, only constructive observations. We all want to crunch WUs at "light speed", there is no telling where the next breakthrough will come from. Regards, JDWhale |
|
617)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 741191)
Posted 19 Apr 2008 by JDWhale
Post: Darn, 23 hrs and no comments on this thread - Nooooo, I need my hourly update on progress, or just rumours lol I've stopped posting Thurston' daily RAC graph since abandoning "The Race", So here's a teaser of what Thurston' been up to... Host = Thurston: Q6600 @ 3375MHz  Reminder that I'm employing Top Secret "conservative scheduling" methods so YMMV. Regards, JDWhale The rest of you please excuse me while I try to send a message a peer... @user formerly known as "Who?" - If you're following this thread I wish you the best of luck with your code... (It is fun to have your "own" version, isn't it?) |
|
618)
Message boards :
Number crunching :
Xigmatek HDT-S1283 or Kingwin RVT-12025 cpu cooler Tutorial for yer crunchers
(Message 741154)
Posted 19 Apr 2008 by JDWhale
Post: Thurston got jealous seeing PC1' picture on the internet and wanted his "Mug Shot" posted as well. But my Classic mug catches both :
|
|
619)
Message boards :
Number crunching :
Xigmatek HDT-S1283 or Kingwin RVT-12025 cpu cooler Tutorial for yer crunchers
(Message 741057)
Posted 18 Apr 2008 by JDWhale
Post: Thurston got jealous seeing PC1' picture on the internet and wanted his "Mug Shot" posted as well. As I said earlier, 3 hours downtime Wednesday so Thurston could relocate into this CoolerMaster case. The Thermalright Ultra 120 Extreme draws fresh air from the intake fan on top and blows to the exhaust fan on the rear. For extra cooling in the summer months, I'm thinking of putting a block of ice on top of the case above the intake fan ;-) Yep, he was crunching when this pic was snapped, it just looks like the fans are standing still. 
|
|
620)
Message boards :
Number crunching :
Xigmatek HDT-S1283 or Kingwin RVT-12025 cpu cooler Tutorial for yer crunchers
(Message 740928)
Posted 18 Apr 2008 by JDWhale
Post: Well so far the elevated motherboard is as Free of dust as is possible(23" above the floor). And I even have room now for Dominator type ram dimms now. Looking good... With that orientation, fan pointing down, I just don't see how you mount the fan after mounting the HS to the mobo, that rubber in the top left near the mobo that fixes the fan to HS seems impossible... And if you mount the fan to the HS before mounting the HSF assembly to the mobo, I don't see how to apply pressure (even with the "common" screwdriver) to the push pin locking mount in that same upper left corner (this also makes using backplate next to impossible, IMO). How do you do it ? [Afterthought : I think I see how... You pull the PS for extra room up there ;-) Not an option for me, because I prefer cases where the PS mounts at the bottom:-(] I might have to drive down from Las Vegas in about 8 weeks and have you show me how you do your magic on PC2 or PC3 when they get their new HSF. (Just kidding, 4 hours across the desert at the beginning of summer isn't the right conditions for a whale ;-) Cheers, JDWhale |
|
621)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 740670)
Posted 17 Apr 2008 by JDWhale
Post:
I'll just add that Thurston was running @ 3200Mhz until 7 Apr 2008 19:59:11 UTC when he was bumped to the present speed of 3375. There is no "Crunch3r" data for current settings and a great deal of the AK_WhalePort data might be from before the switchover as well. Regards, JDWhale |
|
622)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 740639)
Posted 17 Apr 2008 by JDWhale
Post: oh, uhm...I just cranked it up. LOL Correct, here is resultid=815986757 from my host Thurston (Q6600@3375Mhz), probably a closer compare to your OC E6600 than the other host I pointed to earlier. Look at the line "Version info:". Happy crunching... @John Galt 007 - I would share how I switch over without even stopping BOINC, thus no hybrid WUs, but it's not for the faint hearted and requires impeccable timing. BOINC on...on... |
|
623)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 740533)
Posted 17 Apr 2008 by JDWhale
Post: I believe that we have shown within error tolerance that the "different clients" are very close in performance. (Extensive offline benchmarks have been run that show the clients are within couple percents on the same host). This comparison between Marks HARV and my Thurston is going to show the difference between memory configurations Thurston is running dual channel DDR2-1125MHz, HARV is running single channel DDR2-760. Yes, this is exactly the hosts & clients I wanted to see. When I first hinted at this match up, I didn't know that Mark had hobbled his Q6600s memory and thought the matchup would be much more even. I think that I agree with your prediction, there will be even more significant disparity on VHAR WUs. Lesson to all being made at msattler expense. Optimize your memory. I think this is one place that the Macs have advantage over other PCs. The MACs are built to specification, whereas many of our "other" hosts are just thrown together without tuning the components. (Sorry Mark). Regards, JDWhale |
|
624)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 740527)
Posted 17 Apr 2008 by JDWhale
Post: I run the Xeon client exclusively because my FSB more matches that of a Xeon machine than a Core2Duo Mac (since they're laptop CPUs). Alex's Xeon port hasn't been updated since October 7, and Harpertown Mac Pros didn't come out until sometime after that. Low End Mac says Harpertowns came out January 8, 2008, so we're talking about a 3-month difference. From what I can see you are running the "Core 2" client, but of course I'm wrong quote often. I agree with you that it is the "#define compiler conditionals" that make the differences and alter the same source code to compile differently for each version. It sure is hard to read the first time through, if you know what I mean. OK... The reason I asked was when I look at the "version" line in stderr for your Mac E6600 I see this Version info: OS X SSSE3 (Intel, Core 2-optimized v8-nographics) V5.13 by Alex Kan This line only gets to stderr in the version of the Alex code I have when the "#define" for "USE_I386_XEON" is not enabled. When both "USE_I386_CORE2" and "USE_I386_XEON" are #defined, then you get the message Version info: OS X SSSE3 (Intel, Xeon-optimized v8-nographics) V5.13 by Alex Kan When both these directives we see the "xeon" message to std error and performance boost on all Core2 hosts. This is why I ask if you have tried the "xeon" client. Does what I'm saying make any sense. BTW - I haven't had the opportunity to try the code on a Harpertown yet, let alone a Mac Pro 8-core Harpertown. My first guess is that the Harpertown on OS-X will outperform on VHAR, but will underperform slightly on mid-angle ARs. Possible because the CPUs were released too late for Alex to optimize in time. Cheers, JDWhale |
|
625)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 740502)
Posted 17 Apr 2008 by JDWhale
Post: But at the end of the day if you guys think what you have is good enough ok. I don't see anyone saying you aren't helping out. You are entitled to your opinions. As anyone who has followed this thread knows, I have made my opinions known and do not accept any credit for the brilliant code assembled by Alex, I only answered a couple challenges by attempting to port his code to build/run under Windows. I am sorry that your opinions of me are so negative. That said, I'm curious if you have experimented and tried to run the "xeon" client for OS-X. The results in doing so with the Windows builds of the code have showna valuable performance boost on nearly all Core2 hosts tested so far. I don't even know if the "xeon" client will run on your Core2 host, thus I pose the question. It might be possible that the code that Alex built is so customized to the Mac that the crew at Lunatics could learn some more if you tried and posted your results. We are all in this project together, aren't we? Cheers, JDWhale |
|
626)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 740490)
Posted 17 Apr 2008 by JDWhale
Post: Excuse me, your host shows this :Intel(R) Core(TM)2 CPU 6600 @ 2.40GHz [x86 Family 6 Model 15 Stepping 6] You are correct, I apologize. I missed the original reference that your MAC was OC'd and thus not a close compare with Lovey, my E4500. I would hate to cause you a PITA by offering you to switch to Windows for a test. I know that if I had a Mac desktop I wouldn't want to "downgrade" either. Peace. BTW - I think we have the "Hardware out of the question" with the Mac Pro 8-core X5365 runs by UL1. Cheers, JDWhale |
|
627)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 740465)
Posted 17 Apr 2008 by JDWhale
Post: Your host compares fairly closely to my Lovey. Lovey is an E4500 chip running at 2420Mhz 2MB L2 cache at very nearly the same clock... Excuse me, your host shows this : Intel(R) Core(TM)2 CPU 6600 @ 2.40GHz [x86 Family 6 Model 15 Stepping 6] Where do you get 3.4GHz 4MB L2 ??? not from my E4500!
You must not understand, I think I have helped very many others, I do not see how your offer can help any further.
There are now 2 Skulltrail hosts in the top 20 hosts list. I offer my AK_WhalePort to run on them. 'nuf said, JDWhale |
|
628)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 740441)
Posted 17 Apr 2008 by JDWhale
Post: Something just occurred to me - since I've got a whole bunch of WU's completed in Leopard I could compare that to the XP client if you want. I know you've got a Harpertown Mac Pro doing SSE4.1 stuff, but if you really want to see if your SSSE3 or SSE3 clients are up to snuff, I think my machine would present a valid comparison. Your host compares fairly closely to my Lovey. Lovey is an E4500 chip running at 2420Mhz 2MB L2 cache at very nearly the same clock... The first comparison between OS-X and WindowsXP clients on the same MacPro has been presented just yesterday, a bit more crunching is in progress to try to fill in the gaps in ARs if possible. [Edit] The X5365 is not Harpertown and has been running the SSSE3 clients.[/edit] Regarding why Mark and I are competing with moderatly OCd Q6600 rigs with "different clients". I believe that we have shown within error tolerance that the "different clients" are very close in performance. (Extensive offline benchmarks have been run that show the clients are within couple percents on the same host). This comparison between Marks HARV and my Thurston is going to show the difference between memory configurations Thurston is running dual channel DDR2-1125MHz, HARV is running single channel DDR2-760. This is a test of how the clients perform on very similar, yet different hardware. The information should be of use to many with similar configurations. For my "nearly stock" Q6600 @ 2520MHz look at Wrongway as he approaches 5000 RAC. No I do not want any more hosts to test on, the picture is already "busy" enough. That being said, If anyone has a Skulltrail that they'd like to run AK_WhalePort trial on, please PM me. Please read that as Skulltrail Only invitation. Regards, JDWhale |
|
629)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 740247)
Posted 17 Apr 2008 by JDWhale
Post: Are there optimizations that have been introduced in the JDwhale & Lunatics versions that Alex could incorporate into an update to his v8 optimized app? In WhalePorting the code, I strived to maintain the original intent and only corrected what the compiler reported as errors. The one exception to this was substituting an Intel IPP function to perform correlation at one point in the code where the original code used a proprietary function to the same end. This change could be incorporated back to an OS-X build as IPP is already extensively used in the original code. Which is "faster", I do not know. Another similar substitution was made in one or two other places. I can not speak to what changes Lunatics have made. From what we've discovered and has been shown in the extensive amount of data collected and presented, your T8300 host might benefit slightly from running the Xeon optimized v-8 client rather than the Core2 client you are currently using. Try it, you might like it ;-) Regards, JDWhale |
|
630)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 740185)
Posted 17 Apr 2008 by JDWhale
Post: A fundamental question that spawned the work that gave rise to this thread was "Is the better performance of Macs purely down to Alex Kan's code or is OSX an inherently better operating system than Winows for crunching?" Thanks to UL1 for being our "guinea pig" and making his 8-core X5365 MacPro available for the trials. Also, thank you to Jason G and the Lunatics for providing us pre-release versions of your clients for comparison runs. And thanks to Fred W for diligently monitoring UL1s host and presenting the results. Though the picture of whether OS-X or Windows is going to be faster now seems unclear, of the results gathered from Windows clients, it is aparent that the Lunatics build is the "hands down" winner on this host. Good work guys, I couldn't have lost this race without you :-( Thanks to all, JDWhale |
|
631)
Message boards :
Number crunching :
Xigmatek HDT-S1283 or Kingwin RVT-12025 cpu cooler Tutorial for yer crunchers
(Message 740082)
Posted 16 Apr 2008 by JDWhale
Post: I also may convert these Xigmateks to use a mounting plate as the heatsinks are touchy when changing a fan(with the PC off), I was trying to add an air deflector that came with the xigmatek and two of the four plastic pins came up and yep the PC shut down instantly, then I found the heatsinks plastic pins were spread out too much and I had trouble mounting the heatsink until I was able to compress those two pins with some needle nose pliers, DejaVu... You know that I opted to put the KingWin equivalent (to Jokers Xigmatek) model HSF on Thurston a little over a week. I was bumping the OC ~6% and didn't want to pull the mobo to mount the backplate for a Thermatrite Ultra 120 Extreme I had available. To make a long story short, I should have pulled the mobo and used the Thermalrite. This morning I saw the core temps rising, suspended BOINC, then less than 1 minute later heard the pushpin mounts on the top of the KingWin HSF 'pop' loose. I still like the KingWin DHT design & coolong capacity but, like Joker is contemplating, think these models need the backplate for stability/endurance. Three hours later, Thurston has had a "whole body transplant" moved from the Antec P180 case to a CoolerMaster CM 690 case I picked up last week for $50 (clearance rack) when I was in Directron. I really like the flexability of this case, (has a cutout on the mounting tray behind the CPU mount for a fan to cool the mobo from behind), mesh top with twin mounts for fans uo to 120mm, side intake w/fan to feed the HSF with cool 'outside' air. FYI: Lovey moved into the vacated Antec case for now, so she is no longer "naked" ( though she does feel a bit like a "Hermit Crab". Thurston was down for 3 hours beachse after first attempt, I had to re-orient the Thermalrite HSF to point toward the back of the case (different from the HDT KingWin orientation), Lovey went down for 4 hours and is now with the "stock" HSF because of her minimal OC. "Don't worry, man, "She's cool". @Joker - Don't think twice... use a backplate with your Xigmatek HSFs, we've both had problems with the pushpin mounts. I don't know how to mount the fan after the HS is installed, but am sure that you will figure it out and let us know. Regards, JDWhale |
|
632)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 739844)
Posted 16 Apr 2008 by JDWhale
Post: And while I am giving the Penny a little rest, I thought I would do a test run than might interest a few more of you.... This host compares well with Thurston... HARV @ 3420MHz vs. Thurston @ 3375MHz (375x9)... about 1.3% difference in clock. Both are running the SSSE3, Xeon combination of respective ports. @Fred W. & archae86 This should show the benefit of the PGO work that Lunatics has done to gain a few percent over the "verbatim" AK_WhalePort w/o PGO. Hopefully enough data can/has be[en] collected over sufficient ARs this past week on Thurston for a meaningful one-on-one comparison with Marks Lunatics Pre-Release run on HARV. @Mark, what is HARV memory running at? Thurston is @ 1125MHz (375x3). Can you leave running long enough to collect enough data for comparison. I think there would be enough interest to warrant such a comparison. Like you say there are a lot more Q than penryn hosts ATM. Cheers, JDWhale |
|
633)
Message boards :
Number crunching :
I may be retiring a few PCs from my farm
(Message 739669)
Posted 16 Apr 2008 by JDWhale
Post:
Let me gaze into the crystal ball.... 6 Quads (I'll assume Q6600s and running stock 2.4GHz) running the soon to be released Lunatics AK port should combine for RAC ~31200... with draw of ~175W/host = ~1050 Watts.... Sweeeeet! RAC = 29.7/Watt Does that sound like a ballpark figure ? @Andy - Please do get back with the real numbers after the new client is released and RAC have stabilized somewhat. |
|
634)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 739129)
Posted 14 Apr 2008 by JDWhale
Post:
Thanks for remembering how similar outages last week hastened Thurston's withdrawl from the "Skulltrail Sprint". After a full nights rest on Thursday, Thurston hit the ground with newfound desire and wisdom to guide him. His cache still has enough WUs to last for a full 3-4 day outage if necessary. We know what can happen when whales are free to move undetected... Be afraid, be very afraid! BOINC... BOINC... BOINC, JDWhale |
|
635)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 738915)
Posted 14 Apr 2008 by JDWhale
Post:
Well, the above WU finished, I hit the "Update" button.... I am glad that both my wingmen are running R2.2 flavors and thus overclaimed, making me beneficiary. A closer look at the numbers (CR/Hr) for this WU shows the shift away from the VLAR WUs being in the sweet spot and, with the new AK-v8 Windows ports, now about the same as mid-range ARs of [0.4-0.5] flavor. Let the feast begin!!! BOINC onward thru the fog, JDWhale @David - This WU is on my E4500 running at 2420MHz, I think you can expect similar, maybe slightly lesser numbers on your E6300. |
|
636)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 738811)
Posted 14 Apr 2008 by JDWhale
Post: 600th post @KB7RZF - You can browse Gingers results, I'm switching her between KWSN R2.4-SSE3 and AK_WhalePort V0.2S daily to run same AR WUs on both clients for comparisons and analysis. While this is not going to compare exactly the same as the pending Lunatics release, it should give you a pretty good idea of what to expect. Regards, JDWhale |
|
637)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 738726)
Posted 14 Apr 2008 by JDWhale
Post: Time for a weekly update on all 4 hosts I have running AK_WhalePort v0.2. Yes, it's been two weeks since I posted these charts the first time. The new thumbnails are scaled 30x50% of the fullsize images compared to 30x30% on prior screen caps. Hope this is still not too confusing. Host = Skipper: P4D-820 @ 2800MHz   JDWhale on Apr 6 wrote: JDWhale on Mar 30 wrote:Lastly we see P4D-820 SSE3 chart, note it's very erratic and with it's just losing its 10 day cache will probably remain that way as Pending Credits will be rising. Thus, this chart bears no witness to the WhalePort effect as the others do. I'm really upset about blowing the cache away yesterday, I think this host could have shown the largest relative benefit, judging from what Fred W. has charted so far. Skippers RAC continues upward, though showing signs of leveling off. As one of only two of my hosts which had established RAC history, Skippers chart demonstrates different behavior than most others with their steady ramp upwards. The RAC performance vs. time exhibited here is more typical to what most hosts will exhibit. Who am I trying to fool, they are all going through he roof! My revised estimated RAC = 1342 This marks a 48% increase since switching to AK_WhalePort V0.2S (SSE3) Host = Lovey: E4500 @ 2420MHz    JDWhale on Apr 6 wrote: JDWhale on Mar 30 wrote:This is the mighty E4500 @ 2420Mhz left running WhalePort v0.1 while I was in Las Vegas Mar 23-27 and switched to v0.2 on Mar 28. This chip continues to amaze me. Purchased for $118(including an ECS mobo) the E4500 now coupled with the Gigabyte GA-P35-DS3L is a real performer as long as you do not allow 2 VHARs to run simultaneously. Here is a 113CR (AR=0.055) wuid=242481897, Lovey has been called in as tie breaker with a couple comparable hosts both running optimized clients... the results should prove interesting pitted against AK_Whaleport. Still about 17 WUs from being crunched, result should be posted at about 14 Apr 12:00 UDT... I'm excited, don't see WUs worth 113 credits very often. My revised estimated RAC = 2720 Host = Wrongway: Q6600 @ 2520MHz    JDWhale on Apr 6 wrote: JDWhale on Mar 30 wrote:This host is Q6600 @ 2520MHz switched from SSSE3 R2.4V to WhalePort v0.2 on Mar 28. Current graph shows some bouncing after Wrongways cache hit a low of ~20 WUs during the project servers issues last week. I've been working to rebuild his cache DLing in batches of 100WUs a couple times daily to mix it up a bit. Current cache ~475 WUs or ~7 days worth. My revised estimated RAC = 5040 Host = Thurston: Q6600 @ 3200MHz (for now)    Which brings us to Thurston, the star of my lineup. JDWhale on Mar 28 wrote: Thanks Crunch3r, do you mind if I hold off 'till my Q6600 makes it into the top 20 before I hand over the ported source? I figure it's got a chance to reach RAC ~6500 running at 3200MHz with the AK codes. JDWhale on Apr 6 wrote: JDWhale on Mar 30 wrote:This host is Q6600 @ 3200MHz I guess it didn't report results on Mar 27 explaining the deep V before switching to WhalePort v0.2 on Mar 28. If you've been following Thurstons plight, you know that he suffered a seizure after raising his CPU clock to 3375MHz and memory to 1200MHz. This resulted in Boinc crashing and wiping out his established cache of ~1100WUs. Server and DB problems made recovery of the cache especially when combined with the flood of VHAR WUs, Thurston had back to back days returning 200 WUs each. Since the ~1100 lost during the seizure and the inability to build ample cache to pursue Skulltrail in "high performance' fashion, I made the decision to detatch all WUs and start fresh after giving Thurston the night off. This released the ~1100 WUs to rereleased without them having to "timeout". Well Thurston is back online, cache is now ~580 or about 4-5 days. Will be adding to it in lots of 100 WUs 3-4 times daily until we reach 10 days reserve. Since the Skulltrail hunt, "The Race", is over, I've relaxed scheduling from "extreme" to "agressive" mode. My revised estimated RAC = 7080 Host = Ginger: P4-540 (Prescott H/T) @ 3200MHz I'm not posting a graph for Ginger because she is being switched between R-2.4 SSE3 and AK_WhalePort V0.2S (SSE3) clients as different ARs are downloaded in order to "fill the gaps" in the charts. Couple that her being offline for the past 3 weeks and you can imagine the graph, "steady drop for the past 3 weeks, up the past 4 days" I agree with Fred W. assessment of ~30% improvement in CPU times at mid-range ARs and 15-17% at VHARs. The way I do the switching does not create "hybrid" results, each result is processed entirely using the same client. (Until Murphy steps in and causes power outage or Ginger is restarted during transition from one client to the other.) ============================================== Earlier in this thread I offered up a teaser that I be posting predictions later. My first prediction was that Trevor Immelman will shoot "lights out" 66 and win "The Masters" by 5 strokes. Well I didn't post in time and I would have been wrong on both counts. Next up.... Mac Pros will continue to be the top RAC hosts on the project even after the Windows port of AK client comes public, there is room for one exception that involves overclocking. RAC 10,000 and above will only be seen by the same boxes that are currently capable. The gap will narrow, but the song remains the same. (There might be an exception to this prediction if the owners of the Windows based 8-core rigs can figure out how to feed their hosts without getting bit.) I could very well be wrong here also... time will tell. Sorry for being somewhat cryptic.... NOT!!! Regards, JDWhale |
|
638)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 738641)
Posted 13 Apr 2008 by JDWhale
Post: Just 1 quick question, will these apps also work on a P4 2.8GHz w/HT, I believe the highest set is SSE3, or are there plans to make an app that will work? I may have skipped over it if you already posted. Thanks While I can't speak for the Lunatics release, I brought Ginger (3.2GHz Prescott H/T) back online to evaluate AK_WhalePort (SSE3). Keep a watch out for pending performance report. Cheers, JDWhale |
|
639)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 738420)
Posted 13 Apr 2008 by JDWhale
Post: Whow! Stay tuned to this NC thread.... I will predict the future... (Of course we know how that's turned out from past experience.... ;-) @Curmudgeon - Watch your back... You've got a whale on your tail (for now...) ;-) BOINC... BOINC... BOINC... ON...ON... JDWhale |
|
640)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 738382)
Posted 13 Apr 2008 by JDWhale
Post: I just cut over to the x64 SSSE3 xeon ipp u8 pgo Ox r32 version.....don't know what the stderr output says yet, but that's what I am running for test. This is turning into Baskin Robbins... Think I'll have a triple scoop with BubbleGum, Chocolate Almond Fudge, and Strawberry Cheescake on top in a 64-bit waffle cone....LOL [edit] No, wait... make it Tiramisu on top [/edit] @Mark - Is this going to be the first "u8" live run? |
|
641)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 737958)
Posted 12 Apr 2008 by JDWhale
Post: [OT] Please folks, let's refrain from any and all Who? bashing in this thread. All are welcome as our goal is common, to increase project throughput. If any developers, Dr.Who? included, have contributions to make in reaching that goal I'm sure their efforts will be welcomed by all. We've stayed focused and well behaved, please let's keep it that way. Regrettably, in my overzealous comments while running my "Race to the Top", I'd made comments about tracking and hunting the Skulltrail. These comments were entirely non-personal in nature and were intended only as fodder of fantasy. [/OT] That out of the way... Congratulations to the Lunatics for a successful 64-bit client build and lowering the "limbo bar" once more. You guys rock!!! Regards, JDWhale @Martin - Thanks for the Sea Shepherd info. Real world heros out to make a real difference. Back when I was in high school, Country Joe & the Fish had a hit song with Save the Whales. (Maybe some of you remember Country Joe from Woodstock and "Give me an 'F'..." ;-) [edit] And what do you know, Country Joe lives in Berkeley!!! [/edit] |
|
642)
Message boards :
Number crunching :
Xigmatek HDT-S1283 or Kingwin RVT-12025 cpu cooler Tutorial for yer crunchers
(Message 737667)
Posted 12 Apr 2008 by JDWhale
Post: Am I the only person that seems to have problem with Speedfan. I've dumped it in the past because of inaccurate readings, but thought I would give it another go while reading his thread. But no it still saying impossible things. I'll be the first to take a swing.... Check/set the temp offsets are 15c, Configure->Advanced Chip? "INTEL CORE at $0 on ISA" Temperature x offset 15 be sure to "remember it" for each core Beyond that, the program is too confusing for me. Good luck, JDWhale |
|
643)
Message boards :
Number crunching :
Xigmatek HDT-S1283 or Kingwin RVT-12025 cpu cooler Tutorial for yer crunchers
(Message 737617)
Posted 12 Apr 2008 by JDWhale
Post:
Ditto here when I remounted Thurston's KINGWIN last night. (Same exact design, easier availability in the US) Only I used a small 6" chisel to press the locks under the fan. Experience allowed me to press down on diagonal mount points at the same time, locked in much better. Are we getting good at this, or what?
Same here fan underneath pointing skyward. The point should me stressed that with these "Direct Touch Heatpipe" heatsinks you are not allowed as much flexability with the mounting orientation, that the heatpipes must "cross the line" of cores as shown on the Arctic Silver mounting instructions. If you don't then the middle heatpipe alone does almost all the cooling. IMO. Installation on a Core2 Duo would require rotating the HSF 1/4 turn so that the fan blows to the front or rear of your case. But we're talking mounting on a Core2 Quad, so be sure the fan points either to the top or bottom of your case. I'm looking forward to a cool front blowing a bit later, so core temps should drop to upper 50s for most the night. Congratulations Joker... you're now as cool as a cucumber!!! |
|
644)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 737487)
Posted 11 Apr 2008 by JDWhale
Post:
Thurston is a Humpback Whale, watch Star Trek IV: The Voyage Home... San Francisco, transparent aluminium, save the planet [hint, hint]. Check out the wikipedia entry for Humphrey_the_whale, the true story lost humpback whale made famous with his adventures in and around San Francisco Bay as told in a childrens book. [edit]sp[/edit] |
|
645)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 737471)
Posted 11 Apr 2008 by JDWhale
Post: [OT] Last night, I got up in the middle of the night to p!$$, as usual I checked "the farm", only to find Thurston dark and quiet. Even though he had no WUs to crunch I left him powered up as always. So I hit the power switch and was tapping the "delete" key wanting to check BIOS settings before trying to boot him up. I was pleasantly surprised that he hadn't crashed at all, he just put himself to sleep because he was idle and I was only prompted for my password to proceed. First time for everything. I guess he really needed the "night off" ;-) [/OT] |
|
646)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 737431)
Posted 11 Apr 2008 by JDWhale
Post: Let me add something to this. I’ve noticed that you’ve been working in collaboration with the fine folks at lunatics so on a more serious note, for helping them in the kitchen to cook this new flavor of soup let me add a personal *Thank You*. You're welcome. Though I am not very active in the Lunatics, I did accept the invitation to join and was "fast tracked" and pointed in the direction of the developers forum there. (My first post was after a night of playing "PaiGow Poker" and drinking Glen Morangie 12 Yr Old Single Malt, just after creating AK_WhalePort v0.1 when I was first "enlightened" to the changes necessary that later became v0.2) Some discussion in the public forums, some PM exchanges on ideas, information sharing, and some benchmarking of their clients when I can be of help. Other than that I have kept my finger out of the pot. You know what they say about "too many cooks" ;-) In programming there is almost always more than one way to do a task, and style for code, once set, should be followed to maintain readability and maintainability. Though I haven't seen any code other than Alex' original tarball and my own port of that code, I am almost positive that the forthcoming Lunatics port will be both more readable and maintainable than either I have seen. Cheers, JDWhale [edit] @Joker - Let the hunt begin... Tho "The Race" is over, Thurston is back online and "doin' fine"... Temps have dropped after reseating the new HSF again last night (66-65-60-62). @John Neale - I've been a Trevor Immelman fan since his first Masters appearance and Gary Player is pure class. @Fred W - It's not polite to talk about the weight of a whale. [/edit] |
|
647)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 737414)
Posted 11 Apr 2008 by JDWhale
Post:
You are not alone, and the patience being exhibited by the posters to this thread is amazing. Not everyone is the business/profession of software development and sometimes it's difficult for "outsiders" to understand why public release of a software product takes so long. I might not have made it any easier with my public exploits and "over the top" claims. I'm comfortable that the public release is in good hands with the Lunatics crew and that they are dotting all the 'i's and crossing all the 't's and preparing clients for as many configurations as possible. All this takes time... lots of time. Couple this with the required testing on those configurations and the whole effort is volunteer by real people with real jobs who need to find some time for family and golf (remember it is "Masters" week ;-). I'm sure that most everyone knows all this, after all we are enthusiasts when in comes making our hosts, our personal computers, our computing farms perform to their best. ======================================================================= Brief history for anybody just joining the party/thread... As it was stated about 4 weeks ago in a thread far, far away that the task of porting Alex v-8 code was not easy, "if you think it is , DO IT!". Three days later, I had, and this thread was born. Thread readers became aware that an organized effort was already in progress, colaboration follows, Las Vegas break for R&R, research followed by enlightenment gets the Alex optimized GAUSSIAN code working using Intel IPP replacement function, more colaboration, testing, more optimizations (nicknamed "Xeon") discovered, more testing, more colaboration. Development has been at breakneck speed and my side of it is all posted in this thread as well as: "The Race", charts, plots, opinions, analysis, new clients for SSE3 hosts, more charts & plots and the personal graphs from my own hosts that have switched to running my "prototype port", AK_WhalePort. Pardon me for being self promoting, but if you haven't followed this thread from the beginning, you've missed a pretty good story in the making. All of us in the same boat, extremely favorable comments from some of the leading Mac Pro owners who have ruled the "Top Hosts" list for the past 6 months, no flames, just cooperation by the folks doing the work and words of support from those on the sidelines. As said, patience my friend, and enjoy being privy to seeing some of the tasks necessary doing development in preparation of a release and knowing the release will be the best it can be. Just my opinion, JDWhale |
|
648)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 737243)
Posted 11 Apr 2008 by JDWhale
Post: Do you Hyperthread? As mentioned a day or two ago, I've brought back the lovely Ginger (hostid=2166150) for an encore performance. Ginger is a 3.2 Ghz Prescott with 1.5GB DDR-400 in the form of a Gateway 508GE.. Ginger has been crunching since about 9 Apr 2008 13:00:00 UTC to set a very sparse baseline using: Optimizers: Ben Herndon, Josef Segur, Alex Kan, Simon Zadra
Version: Windows SSE3 32-bit based on seti V5.15 'Ni!'
Rev: (R-2.4|xP|FFT:IPP_SSE3|Ben-Joe)
CPUID: 'Intel P4 Pentium-IV (Prescott)'
cpus: 1 cores: 1 threads: 2 cache: L1=16K L2=1024K L3=0K
features: mmx sse sse2 sse3
speed: 3200 MHz -- read megs/sec: L1=4802, L2=4384, RAM=3303
A couple hours ago I introduced Ginger to AK_WhalePort V0.2S (For SSE3 capable hosts) and the results are trickling in. A few VHARs on short order with a limited assortment of ARs to follow until she runs out of work. I plan to keep Ginger up and crunching for about 1 week, switching between clients as/if varying ARs become available. Three VHARs have reported, CPU times range from 2400-3300 seconds for AK_WhalePort depending whether run concurrently or against mixed AR WUs. This compares to CPU times ranging from 2900-3800 seconds under similar conditions using the the baseline client. My initial view for VHAR WUs on this host is a 13-17% reduction in CPU times when running AK_WhalePort V0.2S vs. the competition. Anyway, the reason for running this test is to demonstrate possible benefit of Prescott hosts switching to the ported AK client when available. I'll leave further data presentation and analysis to others. Regards, JDWhale |
|
649)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 737158)
Posted 11 Apr 2008 by JDWhale
Post: WinterKnight wrote: Hope you have enough units to keep going through the present server problems, and realise that the RAC's for everyone will be taking a dip, not just yours. Thanks Andy. In order to free up the ~1100 abandoned WUs from Monday for redistribution, and in consideration of the project problems with download servers & DB, Thurston has detached from all prior WUs. The last results were returned just before 24:00 10 Apr UDT. Of course as soon as I hit the "Update" to return the last results, 40 "new" WUs tried to download, thus if you look at Thurstons results you will see that not "all" WUs have been detached. Host = Thurston: Q6600 @ 3375MHz (for now)  Final RAC = 5,549.47 The needs of the many outweigh the needs of the few [or the one]. -Spok Star Trek Boinc On, JDWhale |
|
650)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 736903)
Posted 10 Apr 2008 by JDWhale
Post: All, Well I've been privy to some early numbers and must apologize for my rant of hours past. Though Thurstons WU processing rate was quite high with the maelstrom of VHAR distribution over the past 48 hours, he was still very short of being able to crunch and return his download limit of 400 WU/day. In fact WinterKnight made the accurate assessment of reading Thurstons (and my) situation of being a "new" host trying to build a cache while being flooded with the VHARs. No Thurston is not a new host, just rebuilding his cache like a new host. There were not any of the said "errors", but there were 2 "user aborts" due to BOINC 4.45 client as wingman, and pretty fast turn around. @WinterKnight - I'm sorry for coming off brash. I'll try to keep my passion in check and the "cork in the bottle" after midnight, while awaiting the time that I can push the button to get more work. Humbly yours, JDWhale [edit] I just noticed that Wrongway, the other Q6600, has worked her cache to less than 12 hours, ~30 WUs, after dealing with the same VHAR attack, I hope we can get more work soon. [/edit] |
|
651)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 736822)
Posted 10 Apr 2008 by JDWhale
Post: Much has been speculated about NEZ over the years I can add some speculation... I wonder if NEZ has a backdoor to fetch WUs when us mortals can't? Makes me a bit more secure now about staying up with my ol' buddy JD (Old No. 7) last night to add a few WUs to Thurstons cache. I didn't want to be greedy, so I stopped at 100 WUs nightcap and hope to add another 100 or over breakfast & lunch. But then again.... |
|
652)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 736795)
Posted 10 Apr 2008 by JDWhale
Post: Keep crunching JDW Don't know NEZ, can you point to his host and I'll check. |
|
653)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 736790)
Posted 10 Apr 2008 by JDWhale
Post:
Howdy neighbor, Air is mighty thin up here, Can't breathe.... Bring oxygen!!! [edit] WinterKnight wrote: Edit] If it could get that many VHAR's the RAC would be ~6500 +/- 100 [/edit My ballpark estimate for yesterday is 7400 credits returned. Of course a "ballpark" is a big place with lots of room for error. BoincStats will be of little help in confirming with the project DB issues over the past ~40 hours. Can we wait on the keepers of the stats and hope the data is still available for fetching. FYI: I am often wrong, this might be one of those times. But then again... ;-) Also, it might be helpful to know that while many consider a day to be 00:00 UDT - 24:00 UDT, WU quotas seem based on Berkeley's clock (currently 07:00 UDT - 07:00 UDT) Thus I could DL some more work just past midnight Berkeley time. I don't know which timeframe should be used, but I prefer UDT day since I try to do an upload just before 24:00 UDT daily. Now back to bed for me. [/edit] |
|
654)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 736754)
Posted 10 Apr 2008 by JDWhale
Post: VHAR's over a few days only make up about 30% of all units issued, and even with the hopefully new further optimised apps the 63 and 50cr units are still going to take an hour or so to crunch. The chances that a host can crunch more than 100/cpu in any one day is almost nil, and for longer periods is worse odds than winning the lottery. Can we wait to see some quantitative numbers before make generalizations about the possibilities of the new clients ability to push the envelope. True for the "Average Joe" this does not become an issue, hopefully over the past few weeks I have shown that I am not not that "Average Joe" and tend to push the limits when it comes to computing. Until now, I have only used BOINCs scheduling methods, today I employed my "extreme scheduling methods" in order to maximize computing efficiency and have discovered that the project is not able to deliver ample workload to maintain maximum thoughput on my host with the new optimized client. Do I think I hit the "magic number" of 400 returned results within that 24 hour period? I answer NO, maybe 60-75% of that number, but I think I returned enough results that my quota should be raised for that period. Can we please wait to see some numbers? The point remains, IMO, if a host can return 200-300 valid results and it's cache has room, then the quota should be the "fixed limit" of 400 + the number of valid results. In a time that the project is in need of computing resources, why would you want to throttle back on the largest contributors? My CPU is "only" a Q6600 running with an agressive, but not extreme, 3375MHz OC there are much faster 4-core rigs whose owners are watering at the mouth to run the same client I am running, let alone the potentially faster client about to be released. IMO, this problem will not go away if ignored and needs to be addressed before the new optimized client is released. We are talking about enforcing limits on the enthusiasts of this project, that is not a good idea. Comments welcome, JDWhale P.S It's now 1:47AM in my time zone, only 13 more minutes until I can push the "Update" button and try to build Thurstons cache. I hope that I can grow his cache and am not stuck in the same position 24 hours from now. |
|
655)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 736684)
Posted 10 Apr 2008 by JDWhale
Post: JD it says you have 2038 work units in your cache ~1100 got abandoned ~55 hours ago when BOINC crashed and emptied the cache. It's documented in this thread. I've asked for help on how to detatch those 1100 WUs, none was offered. Thurston has ~460 WUs in his cache right now, the other ~500 have been returned and are in various states between waiting to be validated and waiting to be being purged. Extra stress on the project servers over the past 2 days DB issues are probably not helping. |
|
656)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 736674)
Posted 10 Apr 2008 by JDWhale
Post: Maybe we should all step aside and let the big dogs crunch..would it help if I lowered my catche.. I do have a bit of computing power..though it's just hardware. More apps. please.... Not at all, project server stats says it has 655,000 ready to send, just none to Thurston for another 5 hours. Must have been that one WU I cancelled today, my wingman was running Boinc 4.45 and underclaiming ;-) Regards, JDWhale |
|
657)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 736644)
Posted 10 Apr 2008 by JDWhale
Post: Host = Thurston: Q6600 @ 3375MHz (for now)  JDWhale wrote: My estimated RAC = ~7400 You can see why I didn't bother to post the daily update yesterday. I started to pass again today, as I'm fairly certain many who've been following see the same drop for today. On the brighter side, Thurston did manage to recover ~200 RAC from the low right after the DB server came back online. Thurston is facing a bit of another challenge right now. The project is refusing to serve us any new WUs for another 6-7 hours, says that he has reached his daily quota of 400 WUs..... Problem is that Thurston only has 475 WU's in his cache, those darn VHARs that Berkeley has been serving up are causing us problems. Thurston turns these around at a rate of ~4.5/hour per core (roughly 432 WUs/day) more than his "quota". We'll wait to see the numbers if one of our wonderful chartists can capture it. (I don't know if I'm making an impossible request, just asking ;-) Is there any to quantify the number of WUs returned in a 24 hour period? I'm curious if he returned more than his quota of WU's. How can you win a race when you are not supplied enough fuel? Quotas need to be revisited as the throughput of hosts jumps to another level as demonstrated with this host today! 'nuf said, JDWhale [edit] url and update cpu speed [/edit] |
|
658)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 736557)
Posted 9 Apr 2008 by JDWhale
Post: Surprising things can happen when the lights are off and no one is watching you. [edit] "Out of sight, out of mind", I've heard people say. [/edit] Cheers, |
|
659)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 736442)
Posted 9 Apr 2008 by JDWhale
Post: To satisfy my personal curiousity, partly fuelled by Skippers surprising performance on AK_WhalePort, I have decided to bring Ginger back for a brief trial of the ported clients. This is a P4 Hyperthread host with SSE3. Commencing 12:15 UDT 09 APR First up she will be running SSE3 R-2.4 Her stderr looks like this: [color=blue]Optimizers: Ben Herndon, Josef Segur, Alex Kan, Simon Zadra
Version: Windows SSE3 32-bit based on seti V5.15 'Ni!'
Rev: (R-2.4|xP|FFT:IPP_SSE3|Ben-Joe)
CPUID: 'Intel P4 Pentium-IV (Prescott)'
cpus: 1 cores: 1 threads: 2 cache: L1=16K L2=1024K L3=0K
features: mmx sse sse2 sse3
speed: 3200 MHz -- read megs/sec: L1=4349, L2=4040, RAM=3190
[/color]After establishing herself with a bit of history, I'll switch her to AK_WhalePort V0.2S (SSE3 client) for comparison. FYI : Ginger is a Gateway 508GE 3.2 GHz running stock except for 1 GB memory upgrade. Ginger_CPU-Z.jpg Ginger_CPU-Z_Mem.jpg Cheers, [edit] fix bbcode url @David - headers look fine here. [/edit] |
|
660)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 736163)
Posted 8 Apr 2008 by JDWhale
Post:
This WU is about the closest I can find to the one we shared. You should be glad Thurston was a no show and forfeited the match. You can be a hero to your kids instead of them having to carry you home after the massacre. (~44%) If our hosts should actually meet up soon, Thurston will not be so nice. |
|
661)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 736123)
Posted 8 Apr 2008 by JDWhale
Post: I'd think You'd want to hold off on that and look at the OCZ Vendetta 2, It has more deeper fin area and here's some info You've got to see(Flash required): Negative, the Kingwin RVT-12025 is available to me today, includes fan and for $32.47 total (+gas, but I need to visit a specialty supermarket near there anyway). Additionally, after installing I will lay Thurston on his side to allow gravity to help the heatpipes perform slightly better. Thanks for that excellent review!!! It confirms that we're both making very good decisions. You and I will have twin HSFs from different mothers. |
|
662)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 736087)
Posted 8 Apr 2008 by JDWhale
Post:
Thanks for reminding me, I'm putting the order in at Directron for the Revolution RVT-12025 now. $32.50 (tax included) and only 10 miles from where I live. I can pick up in the morning. I've got the Thermalright Ultra 120 Extreme here, and will use on the next system build, I just can't afford the extra downtime to pull Thurstons MoBo to install the backplate right now. I think the SKT master's been lurking and is onto the plan(sudden rise in RAC the past 4-5 days). The 92mm version works great, (remember Lovey was wearing it in the "PC Chick" centerfold last month), but we're hitting 85F these days and with the raised OC, it's just not enough for a Q6600, (sitting at 69-65-63-69 as I write). I don't like to run the AC until we get upwards of 95F. Ambient in the computer room is 26c right now and headed higher (~30c) tomorrow. Thanks, JDWhale |
|
663)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 736065)
Posted 8 Apr 2008 by JDWhale
Post: Here's CPU-Z caps of CPU and Memory tabs: Thanks, do the other numbers look Okay? [edit] @Joker -you're installing that Xigmatek HSF tomorrow aren't you? Could you report back with how well it works for you? Thurston needs help in the cooling department. That smaller Kingwin just isn't enough, also need a solution where I don't have to yank the mobo to install backplate. Need something quick and efficient. Thanks. [/edit] |
|
664)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 736029)
Posted 8 Apr 2008 by JDWhale
Post: Host = Thurston: Q6600 @ 3200MHz (for now)  JDWhale wrote: My estimated RAC = ~7400 From the looks of the Thurstons daily graph, Murphy must have paid us a visit today. Still do not know what happened, other than maybe the memory boost to DDR2-1200 was a little agressive. Thurston did not BSOD, one minute tearing through the first set set of WUs, three were going to be hybrids between his old clock and his new OC. The next minute BOINC had come tumbling down, system still up, just no BOINC. And no more cache of WUs. That quick and without warning. Backed off the memory a pinch by reducing multiplier from 3.2 to 3, bumped up vCore a tad, ran some more tests and went live. Here's CPU-Z caps of CPU and Memory tabs: Thurston_CPU-Z.jpg Thurston_CPU-Z_Mem.jpg Though some might consider todays loss of ground as just short of distasterous, I'll offer up a glimmer of hope: Thurston only lost ~2 hours of production, he has lost more than that in one day running benchmarks. His new diet, while not the leanest is tailored to get him back on into this race very, very quickly. I'm providing support and trying to keep him focused on the primary target. I think that within 24 hours we'll be able to see the recovery tactics in full stride. Just because you suffer a flat tire, doesn't mean you will lose the Tour d France, maybe no yellow shirt for that day, but far from losing the race. I've got to get back to work, afraid we're working into the wee hours tuning things up for a mad dash beginning tomorrow. Live long and BOINC even longer, JDWhale |
|
665)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 735935)
Posted 7 Apr 2008 by JDWhale
Post: The operation seems to have been a success in the end, but not without casualties. BOINC crashed and seems to have wiped out Thurstons cache of ~1100 WUs. They are nowhere to be found on the host, and still show as "Scarecrow" (out in the field, LOL). Detatch was not a selectable option. I am afraid that I have abandoned my wingman, gone AWOL. I am sorry. Thurston has gone back to live production and is gaining confidence with his newfound speed. Will post CPU-Z screen caps with nightly RAC graph. While this might be considered paramount to admitting defeat in the SKT hunt, do not count Thurston out just yet. Remember Skipper lost his cache in a different, but equally painful manner at the time of switchover to the AK_WhalePort V0.2S (SSE3) client and never missed a beat. FYI: I am willing to refund your pending credits, Please send a self addressed stamped envelope and $5 restocking fee and I will issue you an "IOU" for the abandoned credits. @David - Sorry but This WU was amongst the casualties. Please accept my sympathy for your loss. Yours in mourning, JDWhale Anyone have an idea how I can release my association with the now lost WU's |
|
666)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 735794)
Posted 7 Apr 2008 by JDWhale
Post:
Yep, just had a look and This WU is still about 360 WUs from what Thurston is crunching at the moment. You should see result validated by early Wednesday, my time not yours. (Just a stab at a target based on ~130 WU's in Fred Ws 0.400 - 0.50 : 78% bin) Though this is in some of the lower performing in Credit/hour ranges, at least it is reassuring that Thurstons getting through them nearly twice as fast as before. Due to Thurstons current diet (so to speak) I do not expect his RAC to rise much on the next 24 hours without a little bionic adjustment to compensate for the lack of carbs. "Better, Stronger, Faster" "We have the technology" "Thurston is that Whale" Better BOINCing with bionics, JDWhale (now scrubbing in and putting on surgical mask) @Fred - thanks for the updated charts and bin statistical data. |
|
667)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 735663)
Posted 7 Apr 2008 by JDWhale
Post: Anybody out there OCing the Gigabyte DG-P35-DS3L and Q6600 near or above 3300, feel free to PM me details. Well that's alright, after all, my boards are GA-P35-DS3L, I keep getting all the dang letters confused... A few pints of Fosters Bitter Ale during the BBQ this afternoon and a nightcap or two while composing makes it all well. We should be able to get you to 3200MHz without too much effort. Thurston on his prior MoBo of same make/model could do 3200 standing on his head and burping jellybeans and only consuming 172 Watts loaded to boot. That WU from ehll took out a host of champion potential :-( I haven't stressed the RMA'd replacement now serving Lovey (E4500). I'm looking back through the notes I made when Lovey was Thurston in drag. (Don't ask, it's all recorded in this thread.) That MoBo/CPU was OCCT STABLE at 9x360=3240MHz with loaded CoreTemp of 58-54-51-58. Remind me tomorrow and I'll send details if you're interrested. Ciao, JDWhale |
|
668)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 735657)
Posted 7 Apr 2008 by JDWhale
Post: [Here is a linear scale plot showing the one region with data from both aps. Thank you for sharing, I should have alerted a few, like you, of Thurstons hostid earlier, but felt that the other hosts provided enough exposure to validate my experiment. Coupled with my switching Thurston to v0.2 within 24 hours of returning from Las Vegas, I don't know how much more data could have been collected. I'm thankful that Wrongway had 5 days of crunching that were captured and is still running without any hardware variation, this makes him a much better example for documenting the switchover on Q6600 hosts. That tight grouping is beautiful, it demonstrates that regardless the WUs crunching the CPU times are predictable. I'm thinking that this will lead to more consistant host RAC with less dramatic variations, especially when coupled with the much more efficient crunching of WUs with AR {0.4<=XXX<=1.25}. If Thurston can fetch some more VHAR WUs, I'd like to try running him so that he crunches 3 VHARs and one other WU to try to gain insight of the "mixed WU behaviors" on multi core hosts with this new optimized client/application. The whisky is starting to kick in, Does any of that make sense? Cheers, JDWhale |
|
669)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 735649)
Posted 7 Apr 2008 by JDWhale
Post: Time for a weekly update on all 4 hosts I have running AK_WhalePort v0.2. Yes, it's been a full week since I posted these charts the first time, Sorry, but I've grew the image sizes just a bit because of the large change in vertical range. The new thumbnails are scaled 30x50% of the fullsize images compared to 30x30% on the original screen caps. Hope this is not too confusing. Host = Skipper: P4D-820 @ 2800MHz JDWhale on Mar 30 wrote: Lastly we see P4D-820 SSE3 chart, note it's very erratic and with it's just losing its 10 day cache will probably remain that way as Pending Credits will be rising. Thus, this chart bears no witness to the WhalePort effect as the others do. I'm really upset about blowing the cache away yesterday, I think this host could have shown the largest relative benefit, judging from what Fred W. has charted so far. Last week Skipper was last in the line, but this week he is first in the lineup. Well Skipper surprised me, by not even seeming to notice that he didn't have a built cache. Looks like SSE3 dual cores will also benefit immensely from a switch to ported AK code when available. Makes me wish that I didn't decommission my 3200MHz Prescott two weeks ago, I might have to bring back Ginger for an encore engagement of one week to checkout her hyperthreading with the SSE3 version of the build. My estimated RAC = 1280 Host = Lovey: E4500 @ 2420MHz JDWhale on Mar 30 wrote: This is the mighty E4500 @ 2420Mhz left running WhalePort v0.1 while I was in Las Vegas Mar 23-27 and switched to v0.2 on Mar 28. And as such, does not exhibit the substantial ramping up in RAC when put on v0.2, she had already been ramping. Little Lovey is the only host to run v0.1 for a significant amount of time and as such I would expect her to "top out" ~4-5 days before the other hosts. My estimated RAC = 2300 Host = Wrongway: Q6600 @ 2520MHz JDWhale on Mar 30 wrote: This host is Q6600 @ 2520MHz switched from SSSE3 R2.4V to WhalePort v0.2 on Mar 28. A solid performer with a very modest OC limited by the Asus P5B-MX MoBo, yada-yada-yada... My estimated RAC = 4700 Host = Thurston: Q6600 @ 3200MHz (for now) Which brings us to Thurston, the star of my lineup. JDWhale on Mar 28 wrote: Thanks Crunch3r, do you mind if I hold off 'till my Q6600 makes it into the top 20 before I hand over the ported source? I figure it's got a chance to reach RAC ~6500 running at 3200MHz with the AK codes. JDWhale on Mar 30 wrote: This host is Q6600 @ 3200MHz I guess it didn't report results on Mar 27 explaining the deep V before switching to WhalePort v0.2 on Mar 28. Well Thurston is unhidden, and I haven't had to try "extremely creative scheduling" yet, though I left the gate open last night and quite a few very sweet WUs got away before I could crunch them... or something like that. Actually, what happened was that I accidentally allowed his cache to fill to 10 days. I've been keeping his cache ~6 days, only downloading enough WUs to run for 24 hours at a time. This allows me to get a sampling from the WUs being offered up every day rather than overloading on just one days offering. Make sense in some sort of twisted way? I hope so. My estimated RAC = ~7400 Whatever it takes to catch Mr. Skulltrail....(no taunting intended) Congratulations Fred W., You must have read my mind ;-) This effort means that Thurstons bionics need some tweaking, I'll start my boosting the CPU to ~3300MHz and re-evaluate after some WUs post. Anybody out there OCing the Gigabyte DG-P35-DS3L and Q6600 near or above 3300, feel free to PM me details. Remember he is now "Better, Stronger, Faster than before". 'nuf said, JDWhale |
|
670)
Message boards :
Number crunching :
Q6600 Vs Q9450 - Is It Worth The Extra $$$'s
(Message 735648)
Posted 7 Apr 2008 by JDWhale
Post: Thinking of building a new machine and I'm wondering whether to go for a 9450 or just get a 6600 and OC the hell out of it. Don't know your prices down under, but here I can get the Q6600 for $200 (US) can't beat it, IMO. |
|
671)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 735607)
Posted 6 Apr 2008 by JDWhale
Post:
Like I said, "I'm not aware".... I was careful to keep Thurston undercover as best I could, but when you are growing 3 inches per day it's not too long before you stand out in a crowd. I hope you got in early enough to collect some 2.4V samples for comparison. I'll tell you when I decide uptick Thurstons OC for that last minute pedal to the metal, push to the end, win by a nose, photo finish. (IFF needed ;-) My guess that UTC time with the results buffer flushed will be enough information for you to make necessary DataVac adjustment, or should I tweak the version in stderr a bit just for certainty? -JDWhale =========================================================================== Oops seems I've over ran my original post..... Here's my quick thoughts why my adjusting Thurstons OC is acceptable.... Thurston started running AK_WhalePort v0.2X just one week ago yesterday (I think. It's been a busy week, but the app is dated 3/30, so I think 8 days is fairly close). To accurately settle on RAC can take several weeks, depending on wingmen delays, WU diet, extraneous computer usage, etc. My fiddling with Thurstons clock in the last couple days of this 16 day race is just my way to help him reach what might actually his true RAC. Who knows? (What he knows, we don't know. Exactly!). I'll try to remember to post a CPU-Z screen cap. Up for discussion and time for those last minute guesses on just what Thurstons RAC will be come April 15. I'll post my prediction for all 4 of my hosts running AK_WhalePort v0.2 when I post daily the RAC chart in about 2 hours. Current guesses are JokerCPoC 13,000 David 6,740 Pilot 5,880 msattler It already hurts........ The Curmudgeon 6,000 +/- 250 UL1 7,000 Fred W Wh*? Interesting.... Seems I've glues two posts into one some how.... Did you know that "50 thou a year will buy a lotta beer"? "I'm a peeping-Tom techie with X-ray eyes", JDWhale |
|
672)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 735558)
Posted 6 Apr 2008 by JDWhale
Post: It may be easier to port Alex's code to Linux, or now that you've done the really hard work, port your version to Linux. I haven't slept for two days trying to get my head around it. I don't like to be beaten! I know the feeling, I couldn't stop 'til it ran back on that first attempt. Some things still weren't right, Jason PM'd me and told me where the client was hanging up on mid-angle WUs. I've got the fix to help you get over that hurdle when you're ready. (good for shaving 20 minutes, Thurstons runtimes on those specific WUs and documented in Fred W barcharts & data earlier posted in this thread). It's in the bits brought to life with "USE_VECTORIZED_GAUSSIAN" #define. Cheers, JDWhale [edit] I take that back, I'm not aware of any data gathering on Thurston, rather the CPU time difference was on another host[/edit] |
|
673)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 735553)
Posted 6 Apr 2008 by JDWhale
Post: We did a furthest from any ocean check back 20 years ago when I was working in NW China. ...Where were you? I've been in Urumqi and Turpan, and traveled the road between a couple of years ago. Turpan is just about 100 miles from their version of our National Test Site. Assuming you refer to Lop Nor, that is the a** hole I speak of. |
|
674)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 735526)
Posted 6 Apr 2008 by JDWhale
Post:
[OT] We did a furthest from any ocean check back 20 years ago when I was working in NW China. If there is an a** hole on this planet, I think we were within viewing distance ;-) At about 1500 miles to the Pacific, 2000 miles to the Indian, we didn't even care to think about escaping through the USSR to the Arctic, but it was equally far, escape was something we dreamed about, but was highly impractical. Please, this is no reflection on the Peoples Republic of China, I truely enjoyed several years in the country and loved our host countries co-workers and the residents of the compounds nearby town & village. So much so that every day after work, I'd be on my push bike for the 11 km ride into town for iced down Tsing Tao beer and lamb shish-ka-bobs. Many a "pub-crawl" was staged amongst us ex-pats through the town and we ran the hash weekly. My arse still shivers at the thought of sitting on the block of ice. [/OT] BOINC On..Onnn... JDWhale Andy Lee Robinson wrote:
@Andy - IIRC, Intel supplies it's compiler free of charge on Linux for "not-for-profit" use, Don't know if IPP is available though. Client distribution should not have the same restriction as I'm faced with on Windows. |
|
675)
Message boards :
Number crunching :
Welcome to the 9 Year Club
(Message 735406)
Posted 6 Apr 2008 by JDWhale
Post: WooHoo. I am now officially 9 years crunching. BooYah!!! True, my AK_WhalePort client will never be released. The scene depicted with my "do a little jig" comment is from the cover of a childrens book titled "Whale Port" by Mark Foster and illustrated by Gerald Foster. Thus the "Welcome to Whale Port" reference, not AK_WhalePort. I'm sorry for the confusion. |
|
676)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 735367)
Posted 6 Apr 2008 by JDWhale
Post:
BoincStats has a wealth of information, though sometimes slightly out of sync for the detailed views, IMO. Thurston is the host running Vista.  Live BoincStats JDWhale Hosts page  Thurston's Detailed BoincStats page |
|
677)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 735353)
Posted 6 Apr 2008 by JDWhale
Post: Thanks, guys. Okay, I've seen that a couple times before, am running now across my 4 hosts to compare with my pending predictions. Andy can only quantify the past, but we're looking to the future... Timbuk3 - The Futures So Bright (I Gotta Wear Shades) Kind regards, JDWhale |
|
678)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 735340)
Posted 6 Apr 2008 by JDWhale
Post: UL1 wrote:
Fred W wrote:
Excuse me, what is/ where can I find more information about andyk and his "app". |
|
679)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 735251)
Posted 6 Apr 2008 by JDWhale
Post:  Host = Thurston: Q6600Walter Winchell used to open his broadcast with: "Hello Mr. and Mrs. America and all the ships at sea." I'm sorry, but that phrase holds a bit of nostalgia for me... though born in 1958, it was probably a bit before my time. So was Humphrey Bogart, and he is one of my all-time favorite actors. Too bad he had to keep up the "macho smoking image" on screen. I digress. Thurston fell a bit short of my predictions for today. Yep, he ran ~2 hours of "uncredited" benchmarks last night, and equally as much today plus a little "AK_WhalePort v0.3" work on the side. I make no excuses, but as stated much earlier in this thread, Thurston is not a dedicated S@H cruncher. He is my "desktop" system and does much more than crunch WUs. So here it is, midnight in my time-zone and Thurston sits at #31 in the "Top Hosts" list, just 50 RAC below my prediction of 24+ hours ago. Put your seat belts on folks, in less than 24 hours I will update my prediction for Thurstons RAC to be reached in ~10 days or less. The number is no longer 6500! Guesses welcome... BOINC 'til it hurts, JDWhale |
|
680)
Message boards :
Number crunching :
Farm auction has closed...and the winner is.....................
(Message 735240)
Posted 6 Apr 2008 by JDWhale
Post: zeitgeistmovie.com wrote:
Somebody lied. |
|
681)
Message boards :
Number crunching :
Welcome to the 9 Year Club
(Message 735234)
Posted 6 Apr 2008 by JDWhale
Post: WooHoo. I am now officially 9 years crunching. BooYah!!! Think I'll have a beer with some friends and do a little jig...  Welcome to Whale Port. |
|
682)
Message boards :
Number crunching :
Farm auction has closed...and the winner is.....................
(Message 735220)
Posted 6 Apr 2008 by JDWhale
Post: The 455 Buick......... Yea, a old buddy of mine used to have the GS Stage1, only mod from factory was the NOX, kept a really low profile. Replaced the GS markings with Skylark, a real sleeper. Made a ton of money street racing, had the knack for the hustle. What a ride!!! |
|
683)
Message boards :
Number crunching :
Xigmatek HDT-S1283 or Kingwin RVT-12025 cpu cooler Tutorial for yer crunchers
(Message 734886)
Posted 5 Apr 2008 by JDWhale
Post: Mornin all Youse Frog Legs! ;) Thanks for the "heads-up" and your tutorial. The KingWin Revolution (similar design, less capacity) keeping Thurston cool came with the fan already mounted to the HSF. If I decide to upgrade to the Xigmatek HDT-S1283 as summer nears and room temp rises to ~30C, can you do this chore for me? IIRC you're planning for a couple more of these HSF if this one is effective. Regards, JDWhale |
|
684)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 734511)
Posted 5 Apr 2008 by JDWhale
Post:  Host = Thurston: Q6600Thurston is back on track after yesterdays sidestep, Currently #35 in the Top Hosts list. ~8 hours earlier than predicted, YeeHaw! Tomorrows targeted position is #30 and RAC 5400. Benchmarks are scheduled on one core this evening with estimated runtime of 2 hours. Yesterday I cancelled 2 WUs where my wingmen was running Boinc 4.45 and were under reporting. Boinc On..On..., JDWhale |
|
685)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 734318)
Posted 4 Apr 2008 by JDWhale
Post: Latest update from the Frozen Penny (data collected at 17:55:21 UTC today): I can't form words.... gotta pick my jaw up off the floor!!!!! Mighty, mighty, mighty nice!!!! I just did a quickscan through Pennys results and didn't see anything in that AR bin with CPU time ~1250, though I did see a couple that exited with error that requested credit with similar CPU times. Of course I didn't look at all 1400 WUs either :-( |
|
686)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 734277)
Posted 4 Apr 2008 by JDWhale
Post:
Ya, it was good alright... good for me! Believe me when I say, "You don't want WhalePort V0.3". I'm building especially for Mark to run trials on Frozen Penny over the weekend. Should help Thurston catch up just a bit, if you know what I mean ;-) Codename: "Silent Hairball". Looks like it's running just fine.... but under cover its regurgitating the same WUs over and over again in disguise. I wonder if the kitties will recognize the symptoms of Frozen Penny coughing up a hairball? |
|
687)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 734250)
Posted 4 Apr 2008 by JDWhale
Post: And the bios starts thinking that it knows all...........LOL. ROFLMAO david wrote:
Remember back before your trip to Sydney to get the Thermalrite? We exchanged a few messages about memory...I went with BALLISTIX DDR2-1066 about a month ago. First set got fried along with the GA-P35-DS3L. (More details back about 5 days ago in this thread.) Yessir, it was the WU from ehll that took them both out and dern near killed ol' Thurston. But his cerebral cortex survived. "Gentlemen, we can rebuild him. We have the technology", they said. "We have the capability to build the world's first bionic whale. Thurston will be that whale. Better than he was before. Better, stronger, faster." The rest is history being documented in this thread ;-) Sorry, I'm still hungover from quenching my thurst[on] during the all night coding session and 2 hours sleep, tend to ramble on and fantasize a bit when I'm hungover and delirious. Anyway, the memory is running "stock" with VDIMM set at 2.2V... Mind you with this memory "stock" is overclocked, so they say. Works out with perfect 3.0 multiplier at this bus speed. Cheers, JDWhale |
|
688)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 734200)
Posted 4 Apr 2008 by JDWhale
Post: Speaking of speed, what speed is Thurston running at? Here's some CPU-Z captures... Thurston_CPU-Z.jpg Thurston_CPU-Z_Mem.jpg HTH [edit] Sorry, I can't show you the matter/anti-matter parameters, you'll have to figure those out for yourself... Be careful, the wrong mixture and poof!, you cease to exist. [/edit] |
|
689)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 734175)
Posted 4 Apr 2008 by JDWhale
Post:
Yep, Orca sounds nice, and they might be prettier to look at, but I am of the more intelligent Humpback species of whales. Hunted to the brink of extinction, now on the comeback trail. In seriousness, AK_WhalePort will not see public release, but is seeing service as a 'testbench' for some of my new ideas. AK_WhalePort is also the vehicle I'm employing to slap the kitties silly in about a weeks time and takedown the Skulltrail. Cheers, JDWhale Guess I need a better picture of my tail as an avatar, or maybe one of me breeching and waving for the camera. Sheesh... Orca couldn't find krill in a bathtub. |
|
690)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 734087)
Posted 4 Apr 2008 by JDWhale
Post: Looks like the ante for a place has been raised by another 200 CB of RAC. Thurston still at 44 with the RAC at 5000+. No fear here Curmudgeon, Thurston "playing possum" is in the past... this time tomorrow? Thurston has a late night/early morning reservation for spot at #35 ! watch your back!. |
|
691)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 733995)
Posted 4 Apr 2008 by JDWhale
Post:  Host = Thurston: Q6600Tough day it seems for Thurston, while most of the WUs crunched from deep in the cache validated, his diet in the past last 24 hours included a large side order of VHARs which are still awaiting Berkeley to validate. Thurston appears to be gaining a bit on his wingmen, probably need to try to grow the cache a little closer to what the Frozen Penny has, I'd hate to be waiting for wingmen to post as this race nears the finish line.  Host = Skipper: P4D-820Thought an update on P4D-820 SSE3 progress was due. ~10% gain in RAC since starting on AK_WhalePort v0.2 SSE3 5 days ago, even with losing his 10 day cache at the transition point. Live long and BOINC often, JDWhale [edit] Make text links point to host rather than chart. [/edit] |
|
692)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 733902)
Posted 3 Apr 2008 by JDWhale
Post: msattler wrote:
I know the DB crash is Marks fault, I blame Misfit, but we all know it was Frozen Penny crunching at warp speed that did it... Some of the anti-matter I ported must have escaped at -200F. AK_WhalePort SSE4.1-Xeon build has been suspended due to lack of insulation on the anti-matter chamber. LOL.... Put on your seatbelt folks, the next 24 hours is going to make LeMans look like a go-cart race.... YeeeeeeHaaawwwwwww... Y'all know that I'm a Texan by choice, can you tell? @Curmudgeon - Better save Thurston a seat... @Mark - please, only chicken for the kitties tonight. BOINC On..On... JDWhale |
|
693)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 733772)
Posted 3 Apr 2008 by JDWhale
Post:
I think the good Dr. is taking the night off, seems he's been on call too much lately ;-) Let me know how the kitties like the AK_Whale_Tail... It might take some getting used to...LOL |
|
694)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 733768)
Posted 3 Apr 2008 by JDWhale
Post: Congratulations on returning to number 48, even if the competition is ~500 CS higher. Number 38 and above awaits your attendance. Thanks for the invite, I'm on my way.... Scotty: "I'm giving it all she's got, Captain!" Thurstons RAC gain ~220 on 02 April.  Host = Thurston: Q6600 03 Apr 00:37 UDTCheers, JDWhale |
|
695)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 733670)
Posted 2 Apr 2008 by JDWhale
Post:
Hate to admit it, but tonight is going to filled with tension and might explain my "possum like" behavior.... A bit over 3 weeks ago, on March 8, THURSTON reached #48 on this same list, I've posted his local RAC chart and will do so again after todays close, Thurstons RAC on that "ill-fated" night was 4369. Then along comes the WU from ehll, forcing the BSOD and subsequent thrashing of a MoBo that just won't POST. That fiery WU not only took out Thurston MoBo, but also fried his memory. It was the same memory that was in the mighty E4500 host (Lovey) when this thread started and would cause Lovey to lock up every 7-8 hours. The same memory that crunched the first WU with the original AK_WhalePort that started all this hoopla. A couple days ago I put the RMAd replacement memory back into Thurston, Crucial BALLISTIX DDR2-1066, bumped the voltage up to 2.2V, bumped AK_WhalePort from V0.2C (Core2) to V0.2X (Xeon). We come full circle. Though this time around Thurston is running at a much more modest 3200Mhz rather than the 3600MHz and His memory has slowed from DDR2-1200MHz, He is making his run with AK_WhalePort and not skipping one single beat. But my spine is tingling as I observe with a sense of deja vu. (Or like one of my favourite all-time baseball players has said, "It's DejaVu all over again". Yogi Berra) So in the past 3 weeks the RAC for the same position (#48) on the "Top Hosts" list has changed by ~500. And here we sit, on the same rung of the ladder, only a little bit higher in the air. Again we come full circle. Yep, tonight I'm a bit scared, thus I'm just a whale in waiting, "playing possum" so to say. Hoping that tonight we break from the full circle and break out instead. BOINC onward though the fog, JDWhale [edit] I just realized that the performance boost being seen with v0.2X might be related to the memory speed going from DDR2-800 to DDR2-1066. After todays close at 24:00 UDT, I'll switch back to v0.2C and recheck the numbers. You can check the results yourself, the AK_WhalePort version is displayed in stderr. [/edit] [edit2] On the other hand hosts Wrongway(Whale2) & Lovey(Whale1) have not been shut down in more than 2 weeks and shows similar benefits as Thurston when running V0.2X over v0.2C After all, I will not be converting Thurston to v0.2C at 24:00 UDT and will continue running v0.2X. Addin that I ran v0.2C last night for a few WUs and the results agree with the charts, V0.2X is ~8% faster than V0.2C in the mid-angle WUs. [/edit2] |
|
696)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 733579)
Posted 2 Apr 2008 by JDWhale
Post:
Shhhhhhh.... I'm playing possum ;-) |
|
697)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 733569)
Posted 2 Apr 2008 by JDWhale
Post:
What are you talking about? That is GREAT NEWS!!! If Frozen Penny is stuck in the mid-angle rut, then Thurston has a much better chance of overtaking her. Of course this only true if the kitties can and will stick to the Chicken diet ;-) SSSE3 only at that please, SSE4 is for p****'s! -JDWhale @Adrian - I officially decree from this day forward the name of the application I am running to be "AK_WhalePort"! Hmmmm let's not tell Green Peace, they might invade Alaska looking for the whales. |
|
698)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 733513)
Posted 2 Apr 2008 by JDWhale
Post: i have a e8400 .. need another tester ?? So while Penny is "testing" the new port live, gaining ground on Thurston, I've scaled Thurston back to 2.4V trying to fill in the VHAR hole in the performance charts for comparison (while I've got the VHAR WUs available.) Dang nab it, I might never catch you at this rate :-( But who?s Skulltrail might be within reach even if you aren't... I'd rather bag an 8-core OCd X9775 than a 4-core X9650 anyway. A bit more impressive when mounted and hung on the wall, if you know what I mean :-) All in good fun, my friend, and all in good time.... |
|
699)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 733478)
Posted 2 Apr 2008 by JDWhale
Post:
Aren't they made by FreeScale ? That's about all I know and only found that out a couple days ago researching function calls. It's that the vector processing unit on the Power Macs that gave them the video editing advantage over Intel based systems? I think that we used to have similar hardware hooked to both out inhouse systems and mainframes... Used to call them "array processors"... Some the size of phone booths made by a company named Floating Point Systems (FPS) out of Beaverton, Oregon. Department of Defense got real worried when we started exporting them to China back in the mid-1980s.... Sorry, I'm rambling... must be time for bed. Ciao, JDWhale |
|
700)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 733468)
Posted 2 Apr 2008 by JDWhale
Post: WinterKnight wrote: Why not call it "Gee! Whale soup from a Kan" Sorry Andy, but I am a Humpback whale. I hope that all of us have seen Star Trek IV: The Voyage Home and know what an important part the Humpbacks play in our future. One of the better StarTrek movies, IMO. ("Where do you keep the nuclear wessels?") LOL... You better not go putting me in a can ;-) Jason and the crew at Lunatics have put much more work into this project than I have. Ultimately they have the responsibility and naming rights. While it is true that we collaborated, Jason has not seen my ported code, nor have I seen his. I am just glad to have helped a bit and to be a part of the project. My WhalePort code is not in a state that would meet coding standards required for a release. I make no excuses and have no desire to make it so. Most my past experience has been in the R&D environment where standards are lax and we always have passed our code to those than can make it meet standards, release it, and support it. Rapid prototyping, proof of concepts, hotfixes, and such is a totally different type of coding than a lasting, maintainable project like this must have. I hate neat code, because I don't want to write it. You know how is with some folks, nobody else can read their handwriting. That is the way I code, it's not very readable. I have my personal version that I can tweak and try new things with. I am content away from the spotlight.... But not until I can Catch a FREEZING PENNY!!! Andy Lee Robinson wrote:
Intel makes their compiler and some other tools available for free on Linux & Mac as long as it is not for profit. There should not be the hurdles/restrictions to port to Linux. Though I'm not sure about finding an IDE like MicroSoft Visual Studio that's available for Linux. Others might comment. To the Moon, JDWhale |
|
701)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 733413)
Posted 2 Apr 2008 by JDWhale
Post:
Showoff !!! But, do you know what I just saw..... Thurston is the TOP Q6600 on the project The top hosts offset=40 still has him at #2 just behind the leader, but if you look at Thurstons current RAC he has passed the leader. Nevermind that MadCat might not have reported yet today. Sorry MadCat, your turn is coming soon enough. Thurstons RAC gain ~186 on 01 April.  Host = Thurston: Q6600 02 Apr 01:00 UDTLike the late great Jackie Gleason used to say... "BANG!, ZOOM!, right to the Moon!" -JDWhale [edit] fix url [/edit] |
|
702)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 733376)
Posted 1 Apr 2008 by JDWhale
Post: Before someone accuses me of being a CPCW, I admit it. I've just cancelled a WU where my wingman was running BOINC 4.25... Here it is wuid=240743754 I've been running Benchmarks on one of Thurstons cores about 8 hours today with about another hour to go before I can get back to crunching 4 cores. Please excuse me if I don't want to accept "under credit" for one WU while trying to chase down a host running at nearly half again Thurstons clock. Add in that he is running Alex Ported in short bursts. Sure, Mark will tell you that he is running benchmarks too, that he is suffering the same as I. But "Frozen Penny" runs the benchmarks much faster than "Thurston" can and gets back to crunching before him. @Mark - Thurston's got Penny in his sites Buddy, she can run[WUs really, really fast], but she can't hide!!!! LOL..... BOINC On..On... JDWhale [Edit] Thank you to all... We're all in the same boat, wanting to meet the project crunching demands one way or another. I'm just glad I could contribute in some way and can't wait to see the first host with RAC 20,000 as a result. It might not be long and there is good chance it will belong to... you know Who?, We will see Skulltrail at it's finest if he opts to run Alex code. No problem, all are welcome as this is truly a group project we are attached to. JMHO [/edit] |
|
703)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 733197)
Posted 1 Apr 2008 by JDWhale
Post: WU : 26ja07ab.16641.18068.9.7.202 WU : 28dc06ag.14133.23794.16.7.218 WU true angle range is : 2.387831 CPU time : 725.108248 sec CPU type : GenuineIntel Intel(R) Core(TM)2 Quad CPU Q6600 @ 2.40GHz [x86 Family 6 Model 15 Stepping 11] Number of CPUs : 4 Operating System : Microsoft Windows Vista Professional Edition, (06.00.6000.00) [edit] wuid=244705636 [/edit] |
|
704)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 733186)
Posted 1 Apr 2008 by JDWhale
Post: Though Fred can provide more widespread analysis and total assessment... A quick peak at the 0.39 AR 72.23 credit WUs that we're both choking on right now shows me that Thurston ~16% faster than your Penny at this AR. Of course you're running at 4000Mhz and Thurston is only at 3200Mhz, so the comparison is not at all fair ;-) Mind you that neither Mark nor I are running Chicken+ (sorry if this is not the preferred moniker). Mark might run it for short bursts for benchmarking. I'm running WhalePort v0.2 and am holding nothing back and disclosing everything I've learned and done with the crew working on the Chicken+ port. As said before, I'm restricted by Intel from sharing WhalePort, but I do not want my short-term effort to be lost. I hate to reitterate that point, please understand my position. By "already optimized", I assume you're referring to the earlier Lunatics KWSN, Cruncher built 2.4V version. While it's true, that code was/is "optimized" and demonstrates a remarkable performance over the "stock" app. The source code in "Alex Kan src-v8.tar" is a total rewrite of several functions that highly leverage the SIMD instruction sets on more than one level (SSE2 SSE3 SSSE3), addin differenly coded algorithms based on processor characteristics and you can see that the code is complex and lends itself to many possible permutations. Give those folks doing all the work at Chicken+ the credit they deserve, That is where your "ported" app will likely come from. I'm just a big mouthed fish that got lucky with some "trial development tools". IMO...At first & second look, some functions bear absolutely no resembelance to each other. The methods of optimization are beyond my comprehension, I have no working knowledge of SIMD instructions and often find myself laboring over just a few lines of code trying to understand the method. This isn't the stuff you're likely to learn at school without at least a half dozen prerequisites, more likely only after already in the workplace a while and paying Intel dearly for a customized course/workshop. Just my opinion, YMMV. I don't pretend to be a coding expert and fall far short on Windows code (my working knowledge is in low level 2D Graphics, 3D-Graphics(OpenGL) and performance computing on the SGI Shared Memory MultiProcessor super computers of yesteryear. Like I've stated before, "Kudos to the folks that understand & develop this stuff". I'm just a porter that prefers hitting golf balls. I think I'll head to the practice range now. Cheers, John |
|
705)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 733167)
Posted 1 Apr 2008 by JDWhale
Post:
Though Fred can provide more widespread analysis and total assessment... A quick peak at the 0.39 AR 72.23 credit WUs that we're both choking on right now shows me that Thurston ~16% faster than your Penny at this AR. Of course you're running at 4000Mhz and Thurston is only at 3200Mhz, so the comparison is not at all fair ;-) Whereas Marks "Frozen" Penny running at 4580MHz with 2.4V SSE4 shows virtually the same CPU times as Thurston running AK WhalePort v0.2 at the same AR. BOINC On..On... JDWhale |
|
706)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 733151)
Posted 1 Apr 2008 by JDWhale
Post:
I'm counting on you running 2.4V as the only way Thurston could even come close to catching Penny. Remember that Thurston is only a 3.2GHz Q6600 chasing your 4.58GHz X9650 Extreme... You've got almost 1.5 times the clock rate and 1.5 times the L2 cache. I estimate that my WhalePort of AK code is ~1.6 times the speed of the 2.4V code... This might prove an interresting challenge ;-) I estimate that if Penny were running the ported code full time that she just might reach RAC of 10,000. I wouldn't stand a chance in ehll of once the "Chicken Plus" port comes public... So a little friendly "exhibition match" to demonstrate ported potential for us "little Q6600 guys" in the mean time.... I might be substituting in other "minor tweaks" to help gain advantage if I can find some, so watch your back, Penny ;-) Cheers, JDWhale |
|
707)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 733067)
Posted 1 Apr 2008 by JDWhale
Post:  Host = Thurston: Q6600 01 Apr 02:22 UDTUpdate on Thurstons RAC 6500 Challenge shows that RAC climbed ~230 yesterday and it seems the slope trended higher. I haven't even performed any creative scheduling yet, well not very creative anyway ;-) He was even crunching on 3 cores for part of the day while running some benchmarks. So I wonder what's up...? Any guesses on what day he'll reach 6500 ? Will he even reach 6500 ? Any predictions on where he'll top out relative to Marks "Frozen Penny" ? Maybe BoincStats can help with the predictions, I think I'll go have a look. [edit] Boincstats says he did 6,432 on 31 Mar... two weeks like that and 6500 is in reach. [/edit] Cheers, JDWhale @OzzFan - Back @ you! Seen Iron Man yet? |
|
708)
Message boards :
Number crunching :
AMD host burned by WU - Wasted 80 CPU hours
(Message 732951)
Posted 31 Mar 2008 by JDWhale
Post: No more problems with that WU... Crunched and verified! Thanks all. |
|
709)
Message boards :
Number crunching :
AMD host burned by WU - Wasted 80 CPU hours
(Message 732917)
Posted 31 Mar 2008 by JDWhale
Post: wuid=237465654 Just came across this WU in one of the queues... Too bad the host that claimed 1374.19 credits errored out. Wasted 80 CPU hours for nothing :-( Just coincidence that both failures were AMD's ? |
|
710)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 732834)
Posted 31 Mar 2008 by JDWhale
Post: The 4 simultaneous VHAR WUs on Whale2 completed and posted. I'm happy to report only ~10% performance hit from running them one at a time on that particular host. Much less impact than running 2 simultaneous on Whale1 where the performance drop was closer to 70%. I am leaning to the smaller L2 cache per core being the culprit since Whale1 is actually running faster memory on a newer[better?] chipset and only feeding 2 cores. Whale1: Gigabyte GA-P35-DS3L & E4500 @ 2420MHz & DDR2-800 Whale2: Asus P5B-MX & Q6600 @ 2520MHz & DDR2-800 (though MoBo limited DDR2-667)I know the MoBo & CPUs are mismatched, catastrophic events of 8 March, remember? "WinterKnight" wrote:
In your case as with archae86 the issue might be more bandwith related as you're trying to feed 4 cores vs. 2 cores. I'm thinking that this "memory bottleneck" will likely disappear with the enhanced use of SIMD instructions offered up by Alex. Of course many H/W variables also come into play as we're seeing with Whale1 and it's limited L2 cache cause it's own issues. We are really seeing that all clock speeds are not created equal, even within the Allendale/Conroe chips. Of course, I might not have a clue since I'm really more of a coding hack than a hardware expert. What can I say, "Opinions are cheap" ;-) "David" wrote:
'Tis a bit funny that you raise that point. My original intent was to download the BOINC code and investigate implementing a different scheduling mechanism to raise "overall" performance, as well as adding some features to the Boinc Manager UI to make it easier to control groups of tasks. I never got around to anything more than building the libraries/executable when the challenge to port Alex-v8 code to Windows was made. I do hope to get back to the original intent some day soon, though since I'm only interrested in running S@H, any changes I make to Boinc scheduling code will likely have limited practical value away from S@H, thus will probably remain private :-( We'll see, I haven't had a chance to "look" into the Boinc code yet, beyond building it that is. Cheers, JDWhale |
|
711)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 732803)
Posted 31 Mar 2008 by JDWhale
Post: archae86 wrote: The updated Whale1 and Whale2 graphs now include a useful amount of results near Angle Range 2.35. They also have a very clear different in scatter, with the E4500 results much more varied in CPU time at a given Angle Range than the Q6600. Thanks archae86, for collecting the data and generating the scatter plots. I've got a comment to make about the variation in CPU times being witnessed, especially pertaining to VHAR WUs on Whale1. I am scheduling these VHAR WUs to run one at a time vs. WUs of other ARs, there was one instance when I purposely ran two of these WUs concurrently, this can be seen when the CPU times came in near 1800sec vs. the 1000-1250sec when run one at a time. We know from past experience that the E4500 chip with it's 1MB L2 cache per core causes a severe performance penalty when these WUs run concurrently. Since these WUs typically come in bunches and often run concurrently at "high priority" due to the shortened deadlines, the data in the graph might be a bit misleading from what your "typical" user can expect. I'm controling Whale2 similarly, but have not allowed more than one to run at a time yet. That is going to change right now. I'm kicking off 4 VHAR WUs to run at the same time, so when the next datas are presented we should see some data points at the "worst case" location for the Q6600 Whale2. I do not expect as much variation as Whale1, but we should be able to see some variation from the 1050sec seen in the current plot. I'm sorry if by controlling what WUs are free to crunch at any given time is not entirely up to the BOINC scheduler, but I feel that I can get better throughput by taking some control to prevent known bottlenecks and raise overall performance. So, am I manipulating the data? Maybe yes, but I'm at least telling you about it, and why. Finally, here is the new graph for the Smithfield host, for which I certainly concur that this version of the port is showing a very nice speedup. <img snipped> To help fill in the blanks, I've just set Whale3 back to running KWSN_2.4_SSE3-Intel-P4_MB while there is some variation of ARs in the queue. This should help to paint the overall picture a bit better. I won't let it go for very long, just long enough to get some more data points for comparison. Regards, JDWhale |
|
712)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 732742)
Posted 30 Mar 2008 by JDWhale
Post:
Wrongway and Lovey swapped CPUs back on that ill-fated March 9 CPU shuffle, we'll call that the morning after the shipwreck. The USB sticks they boot from have the names embedded and are matched to the MoBo, not the CPU. Yes, Lovey was the "naked" PC that there have been pictures posted on the internet. Her MoBo was donated to Thurston when his failed, while her processor is mounted to Wrongway. and the new Lovey was born from the spare parts bin and has since gone thru a number of operations, but she is still naked and likely to become a quad soon (Frys has the Q6600 for $190 in some parts of the country.) Other hosts have included Ginger, MaryAnn, Stubby, and Thurston was once up a time called Professor when I was trying out Ubuntu-64bit. You do remember Stubby, don't you? |
|
713)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 732728)
Posted 30 Mar 2008 by JDWhale
Post: So, I'm running WhalePort v0.2 on 4 hosts now, 2 x q6600, 1 x e4500, 1 x p4D-820 I like the charts that Fred W. & plots that archae86 have been posting so much I thought I'd give it a personal touch and tell you a little story of what's in the works for the next 2 weeks. All you with the bigger and faster quads & octos will have your day soon enough, I hope that you don't mind if I get the jump on you and share my experience. One common factor affects my Core2 hosts depicted below. On Mar 08 I suffered MoBo failure. Probably pushing the box a bit too hard. Q6600 running at ~3600MHz (405x9) with DDR-1066 memory pushed to ~1200MHz. Then along the "firebreathing" WU from ehll and took out the MoBo (that's my story and I'm sticking to it, Pilot is my corroborator, check the NC board about that time, you'll see ;-) Next morning was the MoBo/CPU shuffle, thus the behavior changes evident in the charts at that time. Hosts changed names, WU's were ghosted and detatched, clock rates changed, etc...I won't bore you with all the details, but the charts bear witness to the changes in host configurations. Host = Lovey: E4500This is the mighty E4500 @ 2420Mhz left running WhalePort v0.1 while I was in Las Vegas Mar 23-27 and switched to v0.2 on Mar 28 Host = Wrongway: Q6600This host is Q6600 @ 2520MHz switched from SSSE3 R2.4V to WhalePort v0.2 on Mar 28 Host = Thurston: Q6600This host is Q6600 @ 3200MHz I guess it didn't report results on Mar 27 explaining the deep V before switching to WhalePort v0.2 on Mar 28. This is the host that I have hopes for reaching RAC ~6500 before the next 2 weeks, be warned that I might employ "creative schedueling" if necessary. I will be unhiding hosts as it approaches being the top Q6600 in the "Top Hosts" list in the next couple days, though I'm sure that you can find it in the "Top Hosts" list if you look hard enough. It is also the host that I do development, testing, benchmarking, and browsing on and am using now to compose this message, so I don't feel too guilty about employing that "creative schedueling" to level the playing field a bit. No, I will not be cancelling results for this run to the top, just some results might stray a bit from the default processing order. It's out in the open, I'm telling what I'm going to do, you be the judge. Host = Skipper: P4D-820Lastly we see P4D-820 SSE3 chart, note it's very erratic and with it's just losing its 10 day cache will probably remain that way as Pending Credits will be rising. Thus, this chart bears no witness to the WhalePort effect as the others do. I'm really upset about blowing the cache away yesterday, I think this host could have shown the largest relative benefit, judging from what Fred W. has charted so far. Anyone guess the theme of my LAN/Workgroup ? Kind regards, JDWhale |
|
714)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 732617)
Posted 30 Mar 2008 by JDWhale
Post: What about 'Alex Kan's Chicken Soup'? Yep, 4 sets of 4 bytes worth at a time... (UB128bit) |
|
715)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 732588)
Posted 30 Mar 2008 by JDWhale
Post: I'd have to suggest that it is Alex Kan's v8.0 for Windows ... Though I'm guilty of possible first coining the "WhaleApp" term, I'd prefer phrased as "WhalePort" if asked ;-D |
|
716)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 732486)
Posted 30 Mar 2008 by JDWhale
Post:
Maybe... "Speed Limits" could/should be imposed. In the sake of "fairness" for all... NOT!!! ( though it might save more than a few MoBos from going up in smoke ;-) ============================= I don't know if it's been mentioned yet, but IIRC there is another optimization level, in at least one part of the code, for XEON processors, above and beyond the SSSE3/CORE2 stuff I've focused upon to date. Because I do not have the hardware to test on, and due to license restrictions, can not redistribute, I do not know what other performance benefits Alex has coded for the "high end" chips. In this regard, it's possible that the Mac Pro still has an Ace in the hole with regard to the WhalePort. Regards, JDWhale |
|
717)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 732473)
Posted 30 Mar 2008 by JDWhale
Post: If someone[you know W..?] could loan me a fully loaded Skulltrail system (for a year or two), I'd love to benchmark it with the WhalePort of Alex' V8 code. Hint, Hint... [edit]Maybe just lift the restrictions on my "Evaluation Licenses", or grant a "Not-For-Profit" license, for ICC & IPP so that I can distribute the app.[/edit] |
|
718)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 732442)
Posted 30 Mar 2008 by JDWhale
Post:
Glad to hear that I'm not the only one losing sleep... That whole week before holiday, from the day this thread started, I was getting by on ~2 hours sleep a night. I kept telling myself that I was in "training" for the Vegas scene ;-) Truth be told, I had to go to Las Vegas just to get some rest. Try telling that one to the folks you're gambling with at 3:00AM. I had to resort to "catnaps" in the afternoons just to keep alert at the tables. While driving up into Death Valley, I pulled onto the shoulder just to fit in a 45 minute nap when I felt myself nodding at the wheel. Don't know if you pulled an all-nighter last night or not, but at least I got a couple hours sleep while you were running benchmarks early this morning. Keep with it Mark, you know the saying... "You can rest when you're dead" ;-) BOINC Onward, through the fog! |
|
719)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 732358)
Posted 30 Mar 2008 by JDWhale
Post: @JDWhale, Thanks for sharing details on the IPP correlation issue, works a charm. You're quite welcome, I'd never chased down that issue if you hadn't pointed out the problem with the early app hanging in that spot. This is a collaboration, not a competition ;-) Regards, JDWhale |
|
720)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 732289)
Posted 30 Mar 2008 by JDWhale
Post: Good Job guy's........... I too, was curious about that, but am happy to report the same temps as when running SSSE3 V2.4V on Q6600. Q6600 running at 3200MHz (9x356)... CoreTemp reports 63-57-57-61 with room temp at 26C. HSF is KingWIN Revolution RVT-9225 in Antec P190 case. Note HSF is rotated 90 degrees so that the fan blows toward the top of the case... this aligns the "Heatpipes" perpendicular to the bead of AS5 and cools much better than when fan is pointed toward the back of the case. Not a problem with the Antec P180 which has exhaust fans both on top and on rear. HTH and YMMV, JDWhale |
|
721)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 732256)
Posted 30 Mar 2008 by JDWhale
Post: Very nice presentation Fred. I would not expect any performance improvement on VHAR WUs, I'm understanding that there is no Gaussian search performed on those WUs. The main (only?) difference between v0.1 and v0.2 was to turn on the part of Alex code that implements vectorized gaussian detection (I don't mean to sound smart, I'm just reading the comments/descriptions in the code. For all I know that part of the code walks your dog for you ;-). The vectorized gaussian code seems to knock ~20 minutes off the CPU times for Intel Core2 hosts running at ~2.4GHz over the non-vectorized code regardless of AR. Thus we see the 1200 second step between v0.1 and v0.2 for the ARs presenting so far. It's been documented that I ran out of time before leaving on holiday to debug this part of the code, thus the intermediate (v0.1) version. It's interresting to see how improvements in different parts of the code impact run times. ================================================================= On another note, my Smithfield has posted enough results to be included for those that will benefit from SSE3 capabilities... hostid=2938512&offset=60 is showing similar performance gains as already presented for the SSSE3 capable chips. This host is a Dell XPS400 running a P4D at 2.8GHz. (~30% faster than KWSN_2.4_SSE3-Intel-P4_MB) This is my HTPC hooked up to a HiDef front projector and Dolby digital 5.1 receiver. It rarely gets used, but I'm heading in there right now... My son just brought over a digital remastered HiDef version of "Blade Runner" !!! (ie: "Do Androids Dream of Electric Sheep?") BOINC On...On... |
|
722)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 732050)
Posted 29 Mar 2008 by JDWhale
Post: Stay tuned for some news, kitties and kittyettes........ I've been locked in the dungeon since returning from Las Vegas 40 hours ago... got plenty of coffee, but haven't been out to restock the beer ;-( I'll trade you ;-) |
|
723)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 732007)
Posted 29 Mar 2008 by JDWhale
Post:
First few results have posted/validated SSE3 with ~23% improvement over KWSN_2.4_SSE3-Intel-P4_MB on my Smithfield PD-820 2.8GHz Dell XPS400. Sadly, while installing the program last night, I botched the app_info.xml file and the entire cache was wiped out in an instant. Followed by hard times getting work to feed the host over the past 12 hours :-( Will post a link once more WU's post. Regards, John |
|
724)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 731798)
Posted 29 Mar 2008 by JDWhale
Post: Beautiful!!! I've got to get me a "Frozen Penny"... LOL I noticed on the Whale plots, the 20 minute savings per WU between versions 0.1 and 0.2 best depicted my these mid-range WU's that fill my current cache. BOINC ON!!!! |
|
725)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 731651)
Posted 29 Mar 2008 by JDWhale
Post: I'm not trying to be a "task master" here, carrying a whip. I also simply do not have the skillset to dive in and help, not without taking some serious time to just get up to speed. I'm just trying to be an advocate for the bulk of the user base...so that it helps the project more than just targeting "newer" systems would... The question that comes to mind is "What percentage of that bulk run optimized applications?" If the percentage is not high then there is little impact by providing an optimized app, further diminished if the performance gains are not great. Sorry if that sounds cruel, I do not have that intent, just being a realist. I tried to build for SSE2, but the errors are great in number and the code looks like it has already been modified once at many of those locations. It might be fairly simple to make it work, but I regret that I have no desire learning the routines/intrinsics. Coupled with the fact that my Intel Evaluation Licenses expire in a little over 2 weeks, there are other things I'd rather pursue. Possibly running VTune to identify hotspots/bottlenecks or substituting more IPP routines to increase readability/maintainability as long as performance does not suffer. |
|
726)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 731583)
Posted 29 Mar 2008 by JDWhale
Post:
Until now, I only focused building for Core2, I am building now for SSE3 and will try on my P4D. [Edit] First few comparisons test positive... 25-35% speedup vs. KWSN_2.4_SSE3-Intel-P4_MB I'll go live on the P4D with the ported AK app [/edit] |
|
727)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 731543)
Posted 28 Mar 2008 by JDWhale
Post: I would like to thank JD for the encouraging Port of AK's code, now under local test. Further, I would like to thank Crunch3r for his encouragement to JD and offer to help resolve the legal bits should JD want to move towards a public release. Pure speculation is ~8900 if performance scales linear. If my runs with KnaBench are indicative, I'm seeing just under 40% decrease in runtimes overall with some ARs closer to 60% decrease. I'm guestimating RAC should rise by factor ~1.6. for my Q6600 against SSSE3 R2.4V. Am I optomistic? Yes, very!!! I've still got a couple tweaks to try, but will be running them on a yet undisclosed host, my desktop Q6600. It's only got 16 days to make it's move. Running Vista, so it's not as fast as those XP boxes. LOL... Of course, your performance will depend on whether Crunch3r throttles back the code or not ;-) JK Regards, JDWhale [Edit] I could not have gotten this code where it is today without Jason pointing out where my first attempt was hanging. That hint persuaded me to pursue getting VECTORIZED_GAUSSIAN part of the code working and to find the get around to the Mac corr() function with Intel "ipps" calls, Also it was a discussion on Lunatics forum that disclosed a bug an another part of the code. Jason had run into the same brick walls that I hit and his pointers are much appreciated.[/edit] |
|
728)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 731506)
Posted 28 Mar 2008 by JDWhale
Post: ... Thanks Crunch3r, do you mind if I hold off 'till my Q6600 makes it into the top 20 before I hand over the ported source? I figure it's got a chance to reach RAC ~6500 running at 3200MHz with the AK codes. @Blu Dude - I think I tried with VS2005 Express, IIRC that product wasn't supported with the Intel compiler. Anyway the main issue is with Intel products ICC & IPP... I still have a bit more than 2 weeks before evaluation period expires. |
|
729)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 731481)
Posted 28 Mar 2008 by JDWhale
Post: I'm putting the JDWhale v0.2 app on the mighty E4500 host at 17:20 UDT 28 Mar 2008, if anyone is interrested in monitoring/comparing Yes, still running at 2418/2420MHz, I relocated the chip from the original ECS MoBo to Gigabyte DS-P35-DS3L about a week ago and matched clock rate as best I could. Should you reach the stage of wanting test hosts, I can offer a Core 2 Duo and a Core 2 Quad (Q6600) under WinXP for which I have a pretty good existing history of CPU timings for a broad set of Angle Ranges runningRev: (R-2.4|xT|FFT:IPP_SSE3|Ben-Joe) I also just started running v0.2 on this hostid=4251730 q6600 @ 2520MHz... Current result offset is at 560 with ~300 results running SSSE3 r-2.4V returned in the past 13 hours. This box also was left running the past 5 days while I was away with network activity suspended if you're interrested. A few v0.1 results were crunched in the past 12 hours and will continue running v0.2 for the time being. I'll try to force update daily at 23:00 UDT. As for outside testing, I fear that since I've built the application with EVALUATION LICENSES from both Intel(ICC & IPP) and MicroSoft(VS2005PRO) I'm restricted from sharing the application with others. :-( Kind regards, JDWhale |
|
730)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 731407)
Posted 28 Mar 2008 by JDWhale
Post: Debugging session this morning seems to have been a success, plus a bit of research in how Mac Altivec correlation/convolution likes its parameters delivered. I think that I've successfully activated the VECTORIZED_GAUSSIAN portion of AKs optimized v-8 code using Intels ippsConv() in place of the Macs routine. Am seeing a bit of performance boost when compared to SSSE3 r-2.4V application in KnaBench and better performance than v0.1 that's been running the past 5 days. I'm putting the JDWhale v0.2 app on the mighty E4500 host at 17:20 UDT 28 Mar 2008, if anyone is interrested in monitoring/comparing hostid=4251855&offset=260 is where the newest results will start showing up. If first few results post/validate without issue, I'll let it run for a while for comparisons. Boinc On...On.... |
|
731)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 731290)
Posted 28 Mar 2008 by JDWhale
Post: So, I made it home and was happy to find my E4500 host still crunching away on Alex-v8 ported to Windows app... I checked a couple results against my wingmen and didn't find any mismatches. I've uploaded the 137 results completed during my absence. Though I didn't check them all, I found a couple that didn't validate, but in both cases it is my wingman that is suspect having many other results that didn't validate, causing reissue. hostid=4251855 E4500 Core2Duo at 2420MHz A couple interresting WUs.. wuid=240008090 wingman is OC Q6600 running SSSE3 r-2.4V wuid=240008235 wingman is Q6700 stock clock running SSSE3 r-2.4 wuid=239939252 wingman is Mac Xeon 5150 running AK OS X SSSE3 optimized I'll start debugging the problem with the USE_VECTORIZED_GAUSSIAN blocks of code after a few hours sleep. Regards, JDWhale |
|
732)
Message boards :
Number crunching :
How do I run a machine without a HDD?
(Message 731009)
Posted 27 Mar 2008 by JDWhale
Post: What must I do in order to make a computer run the operating system and BOINC from the system RAM? Which operating system is best suited for such a configuration? And how much RAM is recommended? I use BoincPE... Boots minimal XP installation from USB memory stick, auto launches BOINC, persists data to network share. You can probably get by with as low as 512MB RAM on dual core, 1GB for quad core if you keep your WU cache relatively small. Personally I run 2GB Dual Channel (memory is cheap these days) because I've setup to have 768MB RAM disk. I'd recommend 512MB ramdisk if you run with 1GB RAM. Caveats: It ain't easy to setup, but the info is on the web if you are persistent. Project detatched WUs are commonplace, esp after power failures and can require maintainance by cancelling the detatched WUs else you get 0.0 credits granted for "Received too late to validate" if both wingmen have already reported (Shouldn't be much of a problem if you keep your cache small, ~1 day). WindowsXP licensing might become an issue. FYI: Be sure to shutdown "A43 file manager" after use... there seems to be a memory leak that can be fatal. BOINC On..On... |
|
733)
Message boards :
Number crunching :
Your SETI farm power consumption?
(Message 730716)
Posted 26 Mar 2008 by JDWhale
Post: On the other hand, in a more humid place such as Louisiana, Florida, or Ohio, you won't find them at all. I live in SE Texas (Houston area), 95F summer temps are commonplace, often with relative humidity > 80%. I hate running A/C in the flat, so have gone the opposite of the swamp cooler by running a dehumidifier and an area fan or two to circulate the air. 85F at 40% humidity is comfortable to me with a slight breeze. I do have the A/C to provide a bit of relief when the ambient in the flat gets near 90F. I must admit that I climatize very easily, possibly due to my time in SE Asia without A/C or years spent in NW China without proper heat. Neither the warm or cold bother me much, I enjoy both for different reasons and dress accordingly. I hate paying the electric company and enjoy my monthly 1000KWh ($90) bills. Also coming in from the outside where the temps and humidity are high, it still feels much cooler with the lowered humidity and a slight breeze. Cheers, John |
|
734)
Message boards :
Number crunching :
CPU Performance Question
(Message 730635)
Posted 26 Mar 2008 by JDWhale
Post: I have two machines. One has a QX6700 quad core 2.66 chip with 4GB RAM and Windows XP SP2. The other has 2 Xeon 5355 2.66 chips (8 cores) with 4GB RAM and Windows XP SP2. Each machine has the same client version and the same enhanced app on it. I have posted a result from each machine below of similar sized tasks. Why do my cores in the QX6700 run the task so much faster than the cores in the dual Xeon 5355 machine? Any ideas? Your case is normal for 8-core host running SSSE3 2.4V app and might be a memory/cache contention issue. Problem should ease with optimized application transition to V8 based intrinsics code. Keep your eyes and ears on this forum, a solution might be near. |
|
735)
Message boards :
Number crunching :
How much RAC will, or, can I expect...
(Message 730630)
Posted 26 Mar 2008 by JDWhale
Post: ...out of my laptop (DELL XPS M1530): Depending upon the application(stock v. optimized) and actual crunching CPU percentage... somewhere between 0-600. |
|
736)
Message boards :
Number crunching :
Your SETI farm power consumption?
(Message 730461)
Posted 26 Mar 2008 by JDWhale
Post: I thought I'd keep this thread alive for a bit longer(Or until the OP decides differently), As I just got My bill for the month and I somehow saved $7.09 on the Electric Bill($120.86-7.09=113.77) and I only used 865KWhs to boot. Next month I should see a rise in the electric bill as I have the swamp cooler powered up and set on the low setting as It's getting into the high 70's to low 80's outside(It's a warm spring here). I tapered off on a P4 single core and a Prescott and powered them off before I left home. That made some room in the monthly KWh to throw another Q6600 together... I've already got the MoBo, DDR2 RAM, graphics card and HSF, so if MicroCenter still has the CPU for $200 and I don't go totally bust here in Vegas, I'll try to get the chip soon after my return home. Only issue will be keeping the chips cool enough as ambient will rise to ~85f in a few months. I sure hope those "Direct Touch Heatpipe" HSF's are up for the challenge. @Joker - Made the drive over to Baker and up toward Death Valley today, beautiful. Desert flowers are in bloom lots of yellow & purple... temps topped out at 80F. Simply a wonderful 220 mile loop back through Pahrump. Much better drive now than in July or August. If I'd known that "The Mad Greeks" Gyro Emporium was in Baker I'd have skipped breakfast so I could give it a try. I love Gyros and can't remember the last time I eaten one. Well that's all the break I get, gotta head back to the PaiGow Poker table for another 4 hour run. BOINC On..On... |
|
737)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 729776)
Posted 24 Mar 2008 by JDWhale
Post: So, I had a torrid debugging session lasting a few hours before hopping on the Vegas plane Sunday. Discovered an issue and worked in a quick fix to turn off VECTORIZED_GAUSSIAN, fixed up function GetPeak() to work with different data passed in f_weight array. Maybe too much detail for many, maybe not enough for some.... Will do more debugging work to resolve(hopefully) VECTORIZED_GAUSSIAN algorithm when I get home. All test units pass muster when tested against SSSE3 2.4V app. I had time to squeeze in 2 WUs and posted results before rushing out the door. This is one. Offset is at 360 to help you find the other. Note the bug describing the version of the code. I was trying to get an SSE3 version built for my laptop when I realised that my laptop can only do SSE2. No new WhaleApp for the laptop for now.. But I did leave the E4500 based host, pointed to in WU above, crunching the new app while I'm away. Assuming the 2 already completed WUs validate, I'll upload the ~5 days work upon my return home. Hopefully the electricity doesn't fail and there are a wide range of ARs in the queue. Results for the 2 crunched WUs appear ~20% faster. Regards, JDWhale |
|
738)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 729286)
Posted 23 Mar 2008 by JDWhale
Post: Pilot wrote: Can't really say if the High temp readings were the cause or result of the failed MB. I prefer to think that Berkeley let some "Fire Breathing" WUs get out. LOL! |
|
739)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 729192)
Posted 22 Mar 2008 by JDWhale
Post: Congratulations Jason... Looking good my friend!!! @Mark - Your Seti City Kitty is now on Starbucks payroll ? |
|
740)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 729180)
Posted 22 Mar 2008 by JDWhale
Post: Nice, nice. I wish I could claim credit, but I've not provided anyone with what I've got so far. Currently I'm only crunching some of the low and high angle ranges, whereas these are mid-angle. There appears to be more than one port being tested. This is a good thing! [edit]I've got a feeling that Mark is testing some penryn optimized (ssse4) code, but I could be and often am wrong.[/edit] I'm leaving for Las Vegas Sunday, still loading up the farm with work for while I'm gone. JDWhale |
|
741)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 729104)
Posted 22 Mar 2008 by JDWhale
Post:
Thanks for offering to save me a sip... quite gentlemanly of you. If serious, drink of choice at home is a fine MACALLAN Single Malt, the older the better and just one sip is enough ;-) Seriously though, We've all got Alex Kan (AK) and others to thank, I just worked persistantly for one day to get the src-v8 tarball to build in the Windows environment. Nothing magic about it. Something is still amiss, and I can't quite put my finger on it. Now I'm having to try to understand portions of the source, and I'm far from being a mathematician. It is much easier to "port" existing code than it is to develop it firsthand, or understand it. Hat's off to the guys & gals that understand this stuff. If anyone is aware of bugs or gotchas in the code pointed to above, please point me to a fix and I'll try to make this happen. I think that I've resolved the convolution/correlation issue in gaussFit.cpp. Someone correct me if I'm wrong, the operator is symmetric so correlation becomes convolution , either operation will work. The only issue is to allocate enough memory for the output and to reposition to exclude taper zones. (if using ippsConv()) thanks again for the support... The original E4500 host chip has moved from the ECS MoBO. Yep the UPS guy delivered the replacement for the failed Gigabyte GA-P35-DS3L. I tried to keep the clock the same @2420MHz with 2x1GB DDR2-800 Patriot. Switch was made at 21 Mar 23:00 UTC. For those keeping close tabs, you might notice a shift in the numbers at that time. Possibly due to the higher memory speed or different chipset. @Pilot - I think I got some of those "HOT" WUs you referred to in another thread and took this MoBo down. Timeframe matches ~Mar 9... Nevermind that the Q6600 was running 3600MHz on air with the memory screaming at 1200MHz. CoreTemp was good 61-56C on PRIME and passed STABLE ...then along comes a HOT wu... and like Pat Travers sings "Out go the Lights". Cheers all, JDWhale |
|
742)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 728773)
Posted 21 Mar 2008 by JDWhale
Post:
[Edit:timed out while composing] Just DL'd a bunch of VHAR WU's that might not make deadline after I shutdown the outside connection. So I'm going to continue them either 1,2,3 or 4 at a time to see how much impact it has on CPU times I think that we saw the best-worst case already 31% increase in CPU time (860 vs. 1130 seconds) for 1 and 4 processes. Just interrested where 2 & 3 simultaneous VHAR processes fall in the mix. Should show up as discreet, clearly distinguishable, gathers for same AR vs. CPU time on plots. I'll crunch them for a short while, then switch back to KWSN_2.4V_SSSE3_MB to run comparisons... For clarification, this host always has 4 cores crunching, I'm only varying the number of VHAR WUs to run simultaneously. On systems with more memory bandwith, this might not be as much an issue. [/edit] |
|
743)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 728755)
Posted 21 Mar 2008 by JDWhale
Post: as long as you don't inhale, i'm sure you could get away with it as well. When playing cards, I like to get the waitress to bring me a snifter of Grand Marnier so I can swirl it and smell it, maybe put a drop on the tongue with the stir stick... Ummm lovely. One's got to keep a level head when at the tables in Las Vegas. But at home, ahhh that's a different story altogether ;-) We digress! Alright, I was preparing the farm to run unattended while I'm away when I came across this host of mine that the project had decided to detatch a couple days ago. (Haven't looked through the logs, don't care too. This did however yield an opportunity. Looking at the results I see that I've been getting 0.0 credits granted due the now famous "Task was reported too late to validate". So I decided to cancel most the "detatched WUs" and turn loose the WhalePort on those few VLAR & VHAR that have a chance to report "early enough" to get validated. There were maybe 10 detatched VLARs, and 6-8 VHAR's that were not "detatched", but that I turned loose anyway, currently running are 4 VHAR WU's simultaneously... In the past, this created the worst case scenario with horrible performance swings from only running only 1 or 2 at a time. 3 others ran with mix of other AR WU's. An example VLAR WU. We'll see how it goes... FYI Host is:
|
|
744)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 728510)
Posted 21 Mar 2008 by JDWhale
Post: Welcome to the party... looking at a function named "analyze_pot" right now..... Hmmmm Is that Legal ? "you know that you're over the hill.... When your mind makes a promise that your body can't fill." -Lowell George RIP |
|
745)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 728485)
Posted 21 Mar 2008 by JDWhale
Post: Mark, does your "Seti City Cat" keep changing colors, or are the frozen margaritas catching up with me??? It's only 10PM here in Texas and I've still got a good 4 hours debugging before my eyes go cross. Please tell me it's not the tequila :-( -John P.S. Little Feat - Waiting for Columbus playing on the jukebox tonight :-) |
|
746)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 728457)
Posted 21 Mar 2008 by JDWhale
Post: I kind of took this quick port on in response to a couple challenges by folks here on the MB that made it sound hard... So I thought I'd see what I could do in a day. Definitely, not you Brian.... Sorry if I implied so. I will not spend money on development software and persued this attempt with "evaluation" products from Intel & Microsoft. Just give me a JAVA IDE for what I need ;-) Beyond that, I do not have the appropriate hardware to test with, and I really didn't appreciate the "do it yourself then" attitude from some folks, as I wasn't the one who has been teasing people with supposed wonders of great code for months, if not years now (not sure how long it's been)... Ditto... Please see PM. Regards, JDWhale |
|
747)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 728428)
Posted 21 Mar 2008 by JDWhale
Post: Here's an interresting WU <snip> @Fred - The dice don't stand a chance ;-) I thought my quad teammate was/is running stock, that is why I brought up the comparison... I've got some compatable WU's, both VHAR & VLAR, coming up on my quads, but don't want to suspend a bunch of tasks. So I don't know if I'll get them crunched before I leave. But if I finish getting this code sorted... I might be able to turn it loose on one before I leave. Okay, so my 10% OC'd E4500 @ 2418 might not be able to keep up with an OC'd quad, the point that I failed to make is the performance we're seeing from a chip with only 1MB vs. 2MB L2 cache/core on the Q6600. Running virtually clock for clock, the Conroe didn't run away from the Allendale. Rather the preliminary findings are that the "Alex-v8" port to WinXP is ~17% faster per core than v2.4 for this AR on a lesser machine. But then it just might be the new memory sticks I put in that explains the performance we're seeing ;-) -JDWhale @Mark - We'll see, I kind of took this quick port on in response to a couple challenges by folks here on the MB that made it sound hard... So I thought I'd see what I could do in a day. I really don't enjoy coding all that much any more. This was the first code I've messed with in nearly 2 years... other than the BlackJack "basic strategy" tester I wrote in Java a couple months back to test my play (the "Simpsons" yell at me when I make a mistake). Personal record is 500+ hands without making a mistake, but as time drags on the mistakes become much more frequent. Cheers, |
|
748)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 728344)
Posted 20 Mar 2008 by JDWhale
Post: Here's an interresting WU where I was teamed with a Q6600 running optimized 2.4V... So if he isn't OC'd we are running about the same clock as my E4500 at 2418MHz though I've only got 2MB shared L2 vs. quads 2x4 shared L2 cache. I replaced the mismatched memory with a 2x1GB DDR2-800 Mhz set from Fry's ($44) yesterday. So I've been crunching both processors for the past 24 hours or so, CPU times are stable as I've only been crunching the 63.98 credit WUs. To bad the MOBO tops out at DDR2-667MHz... maybe a little more with the 10% OC. I'm moving that CPU to another MOBO so will probably get a new host ID later tonight. Still working to resolve the mid-angle issues though I'm happy with the low and high angles for now. I've run out acceptable WU's to crunch for now so am switching back to 2.4V since I'm heading to Las Vegas for some R&R. I'll Check back in about a week. Ciao, JDWhale |
|
749)
Message boards :
Number crunching :
Welcome to the 9 Year Club
(Message 727736)
Posted 18 Mar 2008 by JDWhale
Post:
I was already participating in "distributed.net" in mid-March, 1999, when I first hear about "SETI@Home". In the news/web blurb it was stated that the project would be available for DL on April 5, 1999. On the evening of April 5th, I visited the website and registered... Thus my entry at #16 on the class page for 6 Apr 1999, the date/times on that page are UTC. If I had internet access at work back then, my "join" date would be Apr 5,1999, The first day that the project came public! I'm pretty sure that I DL'd the original "screen saver" to my trusty Pentium 60 at that time, but I know that my registration date is true... Tue Apr 6 02:36:53 1999. I didn't have any insider connections, I'm just a "geezer" in Texas. I explained to my then 17 year old son that I was making a lifetime commitment to supply computer proccessing power to this project and have been crunching ever since. Boinc On..On... |
|
750)
Message boards :
Number crunching :
Welcome to the 9 Year Club
(Message 727519)
Posted 18 Mar 2008 by JDWhale
Post: I thought the first day to register was April 6, 1999... I know that I waited in anticipation, How did some of you get in "days" ahead of me? |
|
751)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 727461)
Posted 17 Mar 2008 by JDWhale
Post: John, well done and congratulations on putting your money where you mouth is. Thanks, They ran, but won't validate, I'll try again and not send the WU's in till the numbers match Next time I won't promote "real time". I've got a bunch of detatched WUs that are being validated by wingmen and will end up "too late" to be accepted... Lots to practice on. LOL I wonder if the runtimes are pretty close, though? Later, John |
|
752)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 727432)
Posted 17 Mar 2008 by JDWhale
Post: Okay... we're back on... I found the error of my ways in gaussFit.cpp. The substituion for the convolution call was incorrect and causing buffer overrun to the output buffer. [Edit] Thanks for the tip, you know who you are![/edit] That's what I get for pulling a replacement function out of thin air, not knowing the exact behavior of the original, and not carefully reading the user guide for the IPP replacement. 12 hours and a pint of Vodka can have that effect. I digress... Unfortunately, I also changed some compile options and don't have any VHARs left for comparison. Prior crunched WU's can not be used for comparison as the host was left unattended and WU's were stalling, thus favoring the WU left running without contention. I've launched a direct comparison of WhaleApp vs. 2.4V...on WU's from same splitter block (I hope that term is correct)... 2.4V wuid=236103546 wuid=236103433 vs. WhaleApp wuid=236103598 wuid=236103652 The WhaleApp WUs haven't reported yet, just kicked them off simultaneously, will perform update as soon as they finish, I predict before 23:00 UDT. Similarly the 2.4V WUs above kicked off simultaneously and virtually at the same time. (never mind the crud at the beginning list, it just shows that I tried to crunch them with the earlier WhaleApp, but they both stalled less than a minute in the run. -------------------------- I know that this is not the best host, but it is representative of what is currently available for folks on a budget. Intel Core2Duo E4500 @ 2418 MHz on ECS 945GCT-M with mismatched DDR2-667 Running WindowsXP diskless via BoincPE BOINC On..On... |
|
753)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 727347)
Posted 17 Mar 2008 by JDWhale
Post: Hi all, Thanks for all the positive feedback... but all is not well with the WHALEapp just yet. Some WU's enter what seems to be "wait state", mostly within the first minute of execution. Not what you want for benchmarking or daily driver :-( But those VHAR WUs sure haul a**! I'll try changing some compiler options as suggested and try again. I really don't want to debug this, too many years on the job have burned me out. I'll answer the obvious question before someone asks... Yes those constant 913 second run times are true for the 17 credit VHAR WUs. Only thing was I set BOINC to run on only "1" processor on this E4500 Core2 Duo. No memory or cache contention, the whole 2MB L2 mostly for the single process. Just what "High Priority" WUs deserve! I had to shutdown the system to replace that flakey Ballistix memory, the only sticks I had laying around were unmatched 2 x 1GB DDR2-667. This system actually runs better single channel than with the unmatched sticks, but because I've configured the RAM drive (remember BOINC_PE) to use 768MB, that didn't leave a whole lot of memory for the OS & processes, so I went with the unmatched sticks for now. Anyway, I'm putting the little C2Duo back on KWSN_2.4V_SSSE3_MB.exe for the time being. Cheers to all, John |
|
754)
Message boards :
Number crunching :
Windows port of Alex v8 code
(Message 727273)
Posted 17 Mar 2008 by JDWhale
Post: After 13 hours nonstop porting, the first WU processed validates...resultid=783586431 Processor isn't very impressive E4500 running @ 2418 mHz on a crippled ECS 945GCT-M MoBo with disfunctional Crucial Ballistix memory that fails about every 6-7 hours. OS is Windows XP running diskless with BoincPE. Build is using : MS Visual Studio 2005 Professional Intel C++ 10.1.020 Intel IPP 5.3.2.073 Here is the compile commandline: /c /O3 /Og /Ob2 /Oi /Ot /Oy /GT /GA /I "C:\\Program Files\\Intel\\IPP\\5.3.2.073\\ia32\\tools\\staticlib" /I "C:\\Program Files\\Intel\\IPP\\5.3.2.073\\ia32\\include" /I "C:\\Program Files\\Intel\\Compiler\\C++\\10.1.020\\IA32\\include" /I "../../../boinc/win_build" /I "../../../boinc" /I "../../../boinc/api" /I "../../../boinc/api/win" /I "../../../boinc/client/win" /I "../../../boinc/" /I "../../../boinc/lib" /I "../../image_libs" /I "../../jpeglib" /I "../../db" /I "../../glut" /I "../../" /I "../" /D "USE_IPP" /D "USE_SSSE3" /D "USE_I386_OPTIMIZATIONS" /D "USE_I386_CORE2" /D "__INTEL_COMPILER" /D "WIN32" /D "_MT" /D "NDEBUG" /D "_WINDOWS" /D "CLIENT" /D "_CONSOLE" /D "NBOINC_APP_GRAPHICS" /D "_MBCS" /D "_VC80_UPGRADE=0x0710" /GF /FD /EHsc /MT /Zp16 /GS- /Gy /GR /Yc"..\\StdAfx.h" /Fp".\\Release/seti_boinc.pch" /Fo".\\Release/" /W3 /nologo /Zi /Gd /TP /FI "win-config.h" /fp:fast /Qprec-div- /Qprec-sqrt- /Qfp-speculationfast /QxO I'm not very familiar developing on Windows with these tools (first try, I just installed VS2005PRO & Intel Compiler yesterday). If anyone has suggestions for better options, please let me know and I'll try. I'll let it run while I get a few hours sleep and check some other AR's, Now it's 3:35am, time to get some shut eye. These WU's have already been detatched by the project, so nothing lost by trying to crunch them with first attempt at ported code. Regards, JDWhale |
|
755)
Message boards :
Number crunching :
Welcome to the 9 Year Club
(Message 727240)
Posted 17 Mar 2008 by JDWhale
Post: 21 days and counting..... |
|
756)
Message boards :
Cafe SETI :
How old are you?
(Message 726394)
Posted 15 Mar 2008 by JDWhale
Post: I was born the same year as Madonna, Michael Jackson, & Prince. I am a few months younger than all 3 as my birthday is not until December. Hi Keith, I'll guess that you are 4+ months shy of 21 Blue Moons and can join uli & I in celebrating 21 Blue Moons on 31 Dec 2009. You will turn 50 in December, 2008. I will be waiting for you. |
|
757)
Message boards :
Cafe SETI :
Thought of the day.
(Message 726335)
Posted 15 Mar 2008 by JDWhale
Post: Well, like they say in your century, I don't even have your telephone number. [edit][/edit] |
|
758)
Message boards :
Number crunching :
PC Power Consumption -- Word Problem
(Message 726298)
Posted 15 Mar 2008 by JDWhale
Post: And one last point. Since you all are telling me power is so cheap, I think I'll bring my old dual AMD CPU machine out of the closet as a dedicated cruncher. I'm sure it has a little life left in it. That thing at IDLE uses more electricity than my main PC at full load. [edit] You've got nearly 8.5% savings just with the new power supply... That's great! But don't bring back the "dead" just because of what others call "cheap electricity"...[/edit] I've got a q6600 running diskless with onboard-graphics ASUS P5B-MX that is currently plugged into a Kill-A-Watt... Draw is 143W or $10 per month (a tad bit cheaper with my provider ($.0903/kWh), than yours).. [edit2] This host generates 5 times production (RAC) when compared to my Prescott 3.2GHz that draws 160W. Newer, higher efficiency processors do much more computation for less cost! Combine with high efficiency power supplies for maximum impact! Compute Green!!! [/edit2] BOINC Green Please! |
|
759)
Message boards :
Cafe SETI :
Thought of the day.
(Message 726255)
Posted 14 Mar 2008 by JDWhale
Post: You can tune A Paino but you can't Tuna A fish Ummm...... What's a Paino ? |
|
760)
Message boards :
Number crunching :
PC Power Consumption -- Word Problem
(Message 726248)
Posted 14 Mar 2008 by JDWhale
Post: My head is hurting trying to figure this out. I want to see if you guys get the same answer I did. So far, you calculated savings in kWh.... 12.24 kWh per month savings. You forgot to apply cost per kWh... 12.24kWh/M * $0.09738/kWh = $1.19/Month Here is my calculation factoring 365 days / year with 17W difference.... 17W * 24 Hour/Day * 365 Day/Year / (12 Month/Year * 1000 W/kW) = 12.410 kWh/Month Savings... 12.41 * $0.09738/kWh = $1.21 / Month Savings HTH... |
|
761)
Message boards :
Cafe SETI :
How old are you?
(Message 725916)
Posted 14 Mar 2008 by JDWhale
Post: JD , Thanks for the invitation, but I've lived in Texas for years. Closest I get to Mendocino is when I can pick up some of North Coast Brewing Co.s heady No.38 Stout. Train depicted on the label was called The Skunk and used to make the trip between Willits (town of my birth) and Ft. Bragg. Plus I just love the company logo... If you know what I mean ;-) BOINC On On... |
|
762)
Message boards :
Number crunching :
Any Ideas on Whats wrong with computer?
(Message 725906)
Posted 14 Mar 2008 by JDWhale
Post: This computer has never been overclocked and unfortunately I am not there right now to do anything with it. It has been restarted twice and still is erroring out. I will have to have it shut down for awhile I guess. It seems fine for the programs my Mother runs on it but has some serious problem... Just happened across the host of a wingman of mine with the same symptoms as yours... Makes me wonder how much of a hit the project might be taking, how many similar hosts exist... Set-n-forget types might let hosts run in a state like this for a long time, eh? He is not running an optimized client, isn't there a way to issue reset from the project? |
|
763)
Message boards :
Cafe SETI :
How old are you?
(Message 725561)
Posted 13 Mar 2008 by JDWhale
Post: "Ian SB" wrote:
 ~~~We have a WINNER~~~ +++++++++++++++++++++++++++++++++++++++++++++++++++ @Ian - I guessed yours, now you've guessed mine. Now how deserving is that! (You didn't make an early morning trip to Cambridge to look up David, did you?) We both share having to wait more than 2 years for our first Blue Moon, 1955 & 1961, yours and mine respectively. Let this be a reminder that they don't happen often and that each one in passing should be celebrated. Your Moon Pie is in the mail...;-) Enjoy! +++++++++++++++++++++++++++++++++++++++++++++++++++ As Ian has deduced through diligent research and an overwhelming desire to taste one of America's most nostalgic treats, I'll turn 50 Years in October, 2008. The Blue Moon Calculator was my informational source and is a very good read for anyone interrested. What's in the future:
|
|
764)
Message boards :
Cafe SETI :
How old are you?
(Message 725387)
Posted 13 Mar 2008 by JDWhale
Post: "Ian SB" wrote: I'm getting desperate for this pie! Wazza matter Ian? Can't get your hands on Moon Pie in England? I sense that maybe you are grasping at straws here. Might you pay a visit to a gent near Cambridge named David. He'll have all knowledge to help your guess. Or you could just sit back and relax with one of these fine books...  or  while dreaming about your first ever Moon Pie. NOT!!! @DR C.E.T.I. - Sorry about my "insulted" remark... all in good fun, my friend. "uli" wrote: Next Blue Moon 12/31/2009 "uli" wrote:
You must have found the mystical Blue Moon Calendar. Question is "If I just turned 50, then how many blue moons old am I?" 12/31/2009 I will come of age as you predicted... 21 at last. ( you will too, won't you? I think that we have the same Blue Moon Age ;-) What a night it will be... End of the month, ...year, ...decade, and to top it off..... a "21 Blue Moon Salute"! GO for it... Take the final step... You've almost got it! 
|
|
765)
Message boards :
Cafe SETI :
How old are you?
(Message 724950)
Posted 12 Mar 2008 by JDWhale
Post: "Dr. C.E.T.I." wrote:
Is that the year I was born or my age? either way I'm insulted, my friend... Bring on your best bowler, let me hit for the century... (or at least my age)   Sorry, No Moon Pie for you! Perhaps you'd like some  instead?
|
|
766)
Message boards :
Cafe SETI :
How old are you?
(Message 724920)
Posted 12 Mar 2008 by JDWhale
Post: JD, born somewhere around May 1957. uli, nice try... I was born up in mendocino and raised in humboldt county, ca. up where the "fog" wasn't always fog, and the redwood trees grow to 325' or more when in search of the "worlds tallest tree". US-101, can you dig it? Lots of hippies back in my day. While you are the closest so far, I'm afraid you're still in the fog. Sorry, no Moon Pie for you today. For those in search of the only clue to my age, in this thread you must dig deep.   I got a "speeding" ticket in front of this giant redwood log "Souvenir Shop" in the summer 1976. Intersection of Broadway & Wabash... |
|
767)
Message boards :
Cafe SETI :
How old are you?
(Message 724853)
Posted 12 Mar 2008 by JDWhale
Post: "Ian SB" wrote: Now, as for yourself - I believe I witness less blue moons than you, so I'll hazard another guess and say you were 52 in December? I do request you refrain from Pink Floyd's "WYWH" for one day because... You may not "Have a Cigar". In other words, "Sorry, but no Moon Pie for you today". Thought to ponder : How often does February not have a Full Moon? 
|
|
768)
Message boards :
Number crunching :
Getting 0 credit while other 2 wingmen do.
(Message 724814)
Posted 11 Mar 2008 by JDWhale
Post:
Does anybody know if the "feature" to cancel redundant results has recently been disabled? Would be beneficial for those of us with detached results and our wingmen... Hate to waste processor cycles. |
|
769)
Message boards :
Cafe SETI :
How old are you?
(Message 724309)
Posted 10 Mar 2008 by JDWhale
Post: This July my age will be two greater than the year of my birth (minus 1900 of course). Hmmm... so right now, you are 1 year older than the year of your birth, You must be 54, born in July, 1953 ? [edit] Just barely between 14 and 54 ;-)[/edit] Which would make you 22 Blue Moons ;-) |
|
770)
Message boards :
Number crunching :
Short Deadline WU for March causing 85C to 90 C CPU Temps
(Message 724248)
Posted 10 Mar 2008 by JDWhale
Post:
Might be similar to UL1 issue on another thread? Third host seems to be another case of "Client detached", looking back at that hostid=4246483&offset=320 on 6-7 March yields a run of "Third hosts", I think that XEON had detached/reattached by project. But he is crunching so fast with a small cache it's hard to catch him. |
|
771)
Message boards :
Number crunching :
Short Deadline WU for March causing 85C to 90 C CPU Temps
(Message 724244)
Posted 10 Mar 2008 by JDWhale
Post: I have been seeing some short deadline WU on my system that are resulting in both very high CPU temps as well as resulting in many cases in Computation errors. I am running the Chicken Apps. for Core 2 and even when I backed down to stock speed for my Q6600, the high temps still occured with these WU. Temps have soared to as high as 122C in one case before the system did a BSOD. Most often it will spike for a few seconds at 85 to 90C, before erroring out and moving on to the next WU. Might have been one of these that caused my MOBO to fail late Saturday :-( RMA should bring a replacement in a few days. Also a few more BSOD's on another system. Guess I'll suspend any still in the caches 'til conclusive. |
|
772)
Message boards :
Number crunching :
optimized app and quad core
(Message 724206)
Posted 10 Mar 2008 by JDWhale
Post:
Thanks for clarifying that.... Rather it seems that the "new" host in these tests running the SSSE3 2.4V app reflect my prior "sweet and sour"(*) conclusions when compared to the prior host. (*) Not to be confused with #6 at nearby Chinese takeout, "Sweet and Sour Chicken";-). |
|
773)
Message boards :
Number crunching :
optimized app and quad core
(Message 724195)
Posted 10 Mar 2008 by JDWhale
Post:
Thanks, Fred W, both for these tests and for all the presentations you make to help us understand some of the stats and bring them into focus. So, if I'm reading this chart and data correctly, it looks like marginal improvement running Crunch3r's old 2.4 ssse3 build rather than the 2.4V ssse3 build. Assuming the app refered to as "new" was the 2.4V app. The newer app seems to make the "sweet" WUs slightly less sweet, but also makes the "sour" WUs slightly less sour. "Sweet" refers to the higher credit/CPU second while "sour" refers to lower credit/CPU sec WUs. I realize that you refered to Angle Range(AR) vs. CPU Time in your presentation, please excuse me for expressing my conclusion based on "current" requested credit running newer BOINC versions crunching those ARs. Boinc OnOn... |
|
774)
Message boards :
Number crunching :
Just some figures with the new heatsinks :)
(Message 724176)
Posted 10 Mar 2008 by JDWhale
Post:
Then you must have the GA-P35-DS3 Rev 2, and not the GA-P35-DS3L Rev 2.0 shown in the picture ;-) G-Day... |
|
775)
Message boards :
Number crunching :
Just some figures with the new heatsinks :)
(Message 724126)
Posted 10 Mar 2008 by JDWhale
Post:
That's odd...This board is stamped "Rev 2.0"... I've got a couple of these boards, both with the same NB HS and port configurations... Hmmm ??? |
|
776)
Message boards :
Cafe SETI :
How old are you?
(Message 724062)
Posted 10 Mar 2008 by JDWhale
Post:
You're either 48, or else you will be 48 before July 4th |
|
777)
Message boards :
Number crunching :
"Task was reported too late to validate" ?
(Message 724047)
Posted 10 Mar 2008 by JDWhale
Post:
I've similar issue (project generated detatch events) on a host that runs BoincPE with persistence. I've had a few power outages recently that causes some WU's to crunch and report twice (WUs that have already reported before the power drop are recrunched when power and the network backup, still containing the uncrunched WUs, is restored to the host). Seems that the project doesn't like this and will sometimes detatch (and reattach?) the host without making the WU's redundant. I know that I've never pressed the detatched button, rather the project seems to do it. Maybe we could call these zombies since they seem risen from the dead. A different issue presents when a host is issued WU's and drops power before the next backup...Those WUs become ghosts (I think), the recently sent WU's show up on the project task list for that host, but the host doesn't have the WU's or any record of them, they are not detatched, and wingmen are left hanging til expiration. This wasn't a problem back when we had resends:-( What I can do for these ghosts, if I notice, is drain the cache, then detatch/reattatch. This should cause new results to be generated and sent out before deadline. Is there such thing as being redundant before having quorum? |
|
778)
Message boards :
Number crunching :
Your SETI farm power consumption?
(Message 723506)
Posted 8 Mar 2008 by JDWhale
Post: So the House Finch might be It, maybe. They look nearly identical too when seen side by side. Finches are pretty small (wingspan < 10"), the bird in the pic is much too big. Any way you can measure it? In any case, Back to topic, I'm hoping My next electric bill won't go above $176 as It's warming up around here, We actually had a day that was in the mid 70's(Ducks for cover). Wide range of temps here in SE Texas 30-80F like a yo-yo lately. Electricity bill might go up just a bit this month to $100 as I bring the new Q-Ball online and dial her in while letting the P4 Prescott drain her WU cache. |
|
779)
Message boards :
Number crunching :
Your SETI farm power consumption?
(Message 722867)
Posted 7 Mar 2008 by JDWhale
Post: ... As the War against the sparrows is on going as I've escalated the war with upturned Tacks and wood screws glued in place in the pics You'll have to look really close here and there. Okay, I saw the sparrows, Mustang, & treadmill... I'm confused with the plumbing pics though. And what I really liked was the girl falling out of her bikini on the web hosting site. You're right, I might need to go back and look really, really close... |
|
780)
Message boards :
Number crunching :
Your SETI farm power consumption?
(Message 722760)
Posted 6 Mar 2008 by JDWhale
Post: I have just set up my first quad core and was suprised to find that it actually uses less power than an old PIV machine! Andy, I didn't know that you were expecting. Congratulations on the new addition! My electricity bill just went up $3.65/month with the new OC boost :-( Come summertime I'll need to back off to 3200MHz and 170Watts :-( Guess it's time to donate another P4 to charity to cover expenses :-) |
|
781)
Message boards :
Number crunching :
attn Bad to the bone
(Message 722665)
Posted 6 Mar 2008 by JDWhale
Post: JD, the trouble is he's overclocked it, so it doesn't matter, if it was stock config, then it would be something to remark about. His overclock isn't as extreme as Who?s SKT near 4GHz with 8 cores, yet this guys RAC is very close with his 4 cores at a lower speed. Us with Wintel boxes can't run Alex code, but we are free to OC to try to level the playing field ;-) Besides, you can buy the DELL XPS 720 H2C factory overclocked and liquid cooled. When I go to spend my credits, the merchant doesn't ask if they were "OC" earned. LOL, All is fair in BOINC land. All validated results benefit the project, that is why I am here. My newfound philosophy is BCC: "Build it, Clock it, Crunch it". Live Long and BOINC |
|
782)
Message boards :
Number crunching :
attn Bad to the bone
(Message 722626)
Posted 6 Mar 2008 by JDWhale
Post: I'm sorry but he should be got rid of from the project. If you look at the most recent 600 task he's got over 400 of them have been aborted/cancelled by user. And funnily enough there all short units. Hi y'all, I wonder if it's too late to abuse the update button and try to snag some of them shorties ;-) @SATAN - Please don't go, just sit back and let The Osmonds help you reset your perspective with their 1971 break out hit One Bad Apple. Okay, so in our case it's three BTTB apples, you get the point. G'day from Texas, John Edit: Y'all better watch your back. This little 4 core Vista box hostid=3954050 has broken into the top 20, and core-for-core is the top box on the project with RAC = 1375/core running the Crunch3r Vista build. I wonder if maybe Crunch3r put in a back door to unlock the apple tweaks? :o Explanation anyone? |
|
783)
Message boards :
Number crunching :
optimized app and quad core
(Message 722488)
Posted 6 Mar 2008 by JDWhale
Post: Excuse me, but while on the Core2 Quad subject... On this particular processor, Kentsfield, can anyone attest that 2.4 ssse3 flavor is indeed faster and therefore a tastier treat than 2.4 sse3 running on XP 32bit? There might be differences in performance between the Duo's and Quads because of the extra bandwidth required to feed 4 cores on the same size bus. Also, I realize that with the WU Credit Multiplier pending, Crunch3r's build of 2.4V is poised to become the de facto standard, but for right now XP users have a choice between 2.4 or 2.4V. I am partial to 2.4 because of the cache and memory speed printed in the results. While maybe not terribly accurate, I feel it can be used as an indicator of BIOS configuration issues and is therefore somewhat useful. Sorry for that semi-extraneous setup, now I'll get to my second question: Have any of you an opinion on which among the flavors of 2.4 and 2.4V is faster overall on XP 32bit? Please include which of the Core2 chips you are using when responding so that we can reference with Intel tables regarding L2 sizes, frequency, etc. TIA |
|
784)
Message boards :
Number crunching :
Free VNC solutions...
(Message 722304)
Posted 5 Mar 2008 by JDWhale
Post: Voyager wrote:
Daniel wrote:
JDWhale wrote:
I use... TightVNC as viewer running on Vista RealVNC as viewer and server running on XP UltraVNC as server running on BoincPE All for free and working in harmony together. Server is not installed on my single Vista box. Since it is the only system with monitor, keyboard, and mouse, I use it to perform server maintainence when necessary. For example, when it's time to install a new version of BOINC just bring up VNC connection to a server, stop BOINC service, update BOINC, restart Boinc service, disconnect. So you install VNC server on your "remote" systems. Launch VNC viewer (client) on any system to access any of your remote systems. I'm most familiar with RealVNC, been using it for more than 8 years. (Used to control remote Linux servers running Slackware distribution crunching S@H Classic from my Windows95 host.) Though to be honest, once set up, I almost never had to revisit. It is highly configurable to meet the needs of even the most demanding power user (commercial products are available if needed). You might want to keep display complexity to a minimum on servers and set server properties to remove wallpapers and background patterns and to disable "user interface effects", in order to enhance performance. These tweaks hold extra value with a wireless server/viewer connection. One server property setting that I've found very useful is to allow clipboard updates to/from clients. Cutting and pasting across systems... how great is that? Good luck, John |
|
785)
Message boards :
Number crunching :
I got a new mouse........
(Message 722302)
Posted 5 Mar 2008 by JDWhale
Post: Voyager wrote:
Daniel wrote:
I think this topic needs a thread to call home... |
|
786)
Message boards :
Number crunching :
Just some figures with the new heatsinks :)
(Message 721962)
Posted 4 Mar 2008 by JDWhale
Post: I am having exceptional performance out of this HSF assy using AS5 and KINGWIN Revolution RVT-9225 92mm H.D.T. CPU Cooler on a Q6600 running slightly OC. 23c to 25c at idle and 34c to 38C crunching SETI 24/7 at room temps of around 70F.It appears the heat pipe direct touch design is working for me well and it is quiet. http://www.newegg.com/product/product.aspx?item=N82E16835124017. I am pleased anyway /salute @Dennis - Thanks for the heads-up on this remarkable HSF. Temps are comparable to what I get with the Thermalrite on a similar rig running @ 3200MHz. While the KINGWIN fan, running fullspeed at 2800RPM, is quite a bit louder that the Scythe 120MM case fan I've attached to Mr Thermalrite, it does move enough air to sail a schooner.  The KINGWIN RVT-9225 is keeping the quad core nice and cool while crunching...  Note the adjustable baffle in the KINGWIN directing air back across the DDR2 1066 sticks. That additional airflow will help when we've settled on the CPU overclock and bump these sticks back up to 2x600MHz 5-5-5-15. No fear here. The North Bridge HS gets plenty hot on the Gigabyte GA-P35-DS3L motherboard, the extra airflow off the KINGWIN helps keep it cool. Now it's time to shop for cheap case to make this brand new cruncher feel at home. Boinc OnOn... |
|
787)
Message boards :
Number crunching :
Your SETI farm power consumption?
(Message 721475)
Posted 3 Mar 2008 by JDWhale
Post: Your electricity is MUCH cheaper than ours (~ half the price). Where I live, it pays to shop around, the provider that I terminated service on 12 Jan was charging 50% more than my new provider ($0.0903 vs. $0.144) same electricity, it all feeds the same grid, just much cheaper. You can also spend more if you want to feed the "green" thing by choosing a provider that only generates with renewable resources (hydro, solar, wind, etc). Live Long and BOINC |
|
788)
Message boards :
Number crunching :
Your SETI farm power consumption?
(Message 721421)
Posted 3 Mar 2008 by JDWhale
Post: Yep :( I've capped my computer load at 1KW. With my energy provider (Electric Company) currently charging $.0903/KWH, monthly crunching estimate is $66 with or 66% of my total consumption. Costs can vary from month-to-month with changes in raw fuel costs. I chose not to "lock-in" a fixed rate which was slightly more during off peak months, when A/C use is at a minimum. Here in Texas, electricity can be a cutthroat business with offers coming in the mail weekly to switch energy providers. Some companys are publicly traded, some have recently been taked private by venture capitalists. Competition tends to keep them rather honest IMO. Where I live, I have choice of ~8 different providers. No change in lines or meters, owners/operators of actual equipment remains the same regardless of your choice in provider. Fees for equipment/services (~$6/Month), are included your monthly consolidated bill from your provider.. @Andy - Thanks for starting this thread I've just purchased the "Kill-a-Watt" meter mentioned earlier in the thread, and able to monitor my computer load and see the fluxuations as I'm tweaking boxes. Recent Intel Quad box draws 170W when crunching away, about the same as my 3.2 GHz Prescott box, but is RAC is nearly 600% of Mr Prescott, and generates less heat. Here in Houston, Texas, less heat is a very good thing :-) The P4 Prescott is now destined to become my daughters "new" computer, making room in the energy budget for another Q6600 build. She turns 21 years come June, and is thrilled to be getting Daddys old computer. Cheers, John |
|
789)
Message boards :
Number crunching :
Pending Credit
(Message 721338)
Posted 3 Mar 2008 by JDWhale
Post: Pending credit: 25,310.93 |
|
790)
Message boards :
Number crunching :
Just some figures with the new heatsinks :)
(Message 721149)
Posted 3 Mar 2008 by JDWhale
Post:
I'll have to make that trip up Death Valley in 3 weeks, so heat shouldn't be a factor. In June I'll be bringing my daughter to celebrate her 21st birthday. I could still find time to make the 1 hour drive down to Baker though, if you change your mind. I have access to Frys, Directron, and MicroCenter nearby my home. Just PM me. Check you later, |
|
791)
Message boards :
Number crunching :
Which do I want. Windows XP or VISTA.? Why?
(Message 721131)
Posted 3 Mar 2008 by JDWhale
Post: All is well JD and thank you for your concern. I had a power supply go out after I left work and it was down a few days on the XP rig and except for a few power outages due to weather and reboots for updates both run 24/7. Again, TY JD /salute @David You present good evidence. I'll stop pleading the "inconclusive" argument. 20% difference in RAC is very significant, and to me, simply amazing ;-) |
|
792)
Message boards :
Number crunching :
Which do I want. Windows XP or VISTA.? Why?
(Message 721121)
Posted 3 Mar 2008 by JDWhale
Post: P.S. interstitial housekeeping - great phrase, I had to look that one up ;-) ROFL... Edit: Does anyone know of a way to fetch both granted and pending credits for a specific host? How about the % of time BOINC client is running 98.0976 % While BOINC running, % of time work is allowed 99.4641 % Average CPU efficiency 0.954279data from the user host page, might that shed some light. Also, I've looked at the daily updates for the past 60 days on BoincStats fairly closely. The total credit bar charts seem to undulate when overlayed, indicating, IMO, that there will be reversals in the lead position. @Dennis - Sorry if I invaded you privacy by accessing your stats. Respond quickly, and I'll edit out the link. Go Team Air Force!!! Regards, John |
|
793)
Message boards :
Number crunching :
Which do I want. Windows XP or VISTA.? Why?
(Message 721100)
Posted 3 Mar 2008 by JDWhale
Post: I agree completely that RAC is a bad indicator, and the best possible test is actual crunch time. I agree that this is an interresting experiment, but in and of itself might prove little. If more contributors were to conduct similar experiments, the combined results would be more conclusive. If one of these hosts cannot obtain work and its queue has run dry, the results are skewed without any evidence that a problem occured. BOINC OnOn... P.S. interstitial housekeeping - great phrase, I had to look that one up ;-) Edit: Does anyone know of a way to fetch both granted and pending credits for a specific host? |
|
794)
Message boards :
Number crunching :
Which do I want. Windows XP or VISTA.? Why?
(Message 721086)
Posted 2 Mar 2008 by JDWhale
Post: I set up 2 identical rigs except for HD size,and Vista uses a 4 gig Flash drive for ready boost advantages, both still in use, for this very question. AMD 6400's. XP is averaging a 150 to 250+ higher RAC. /salute Just a quick look at some CPU times of WUs with same credit, 63.98 for example, on both boxes, I see no advantage either way. RAC is not a good judge for comparison as it is too easily influenced by factors that are out of direct control. Like pending credits, download mix of WU's. Rather a better comparison is actual CPU times from both hosts for similar WU's. IMO. |
|
795)
Message boards :
Number crunching :
Just some figures with the new heatsinks :)
(Message 721085)
Posted 2 Mar 2008 by JDWhale
Post:
LOL... My darn hands get in the way mounting the Thermalright on the P35-DS3L boards, I think about the same size as the Turniq. I'm heading over to Directon, ~15 minute drive, tomorrow morning to pick up a Kingwin Revolution H.D.T. to tryout on my new Q-Ball. Copper heatpipes make direct contact with CPU, no aluminum to inhibit heat transfer. You might be thinking, "What can a $30 HSF do? It's too cheap". I figure what the heck, for $30 it deserves a shot. Especially so, after reading a few reviews. I need a good HSF when the ambient temp reaches 85F. I'll ask the folks in the storefront when they'll be stocking the Scythe Orochi. Actually I'm planning a morning trip up 127 to Death Valley Junction, so Baker was on my route. Heat doesn't scare me, I live in Houston and don't use AC, just a dehumidifier and a fan. Take care, |
|
796)
Message boards :
Number crunching :
I got a new mouse........
(Message 721017)
Posted 2 Mar 2008 by JDWhale
Post:
Is that a hotplate way up there on top? Dude, you can use that to cook omlets on ;-) |
|
797)
Message boards :
Number crunching :
Which do I want. Windows XP or VISTA.? Why?
(Message 721013)
Posted 2 Mar 2008 by JDWhale
Post: Getting a new pc. which system will work best for crunching? I've got Vista Ultimate on my quad core, the only box I keep a monitor attached to. I love it. I don't run many apps besides BOINC and some Visual Studio C++, and of course IE7.0 & Media Player. Performance is great with 2GB RAM and a $10 GeForce 6200 LE. I'll only revert to using XP if tortured:-( Sorry guys, Edit: But on the otherhand, my dedicated crunchers run XP with either very small hard drives or bootup & launch BOINC from USB with no hard drive. |
|
798)
Message boards :
Number crunching :
Mac Pro CPU time vs. Angle Range plots ?
(Message 721002)
Posted 2 Mar 2008 by JDWhale
Post: (not least of which was redundancy - after 37 years with the same company! - so I had to sort out ongoing gainful employment).F. Sorry to hear that news Fred, best of luck to you. Nice work on the crossplot. My first impression is to route all the VLAR WUs over to Mark, err Core2Extremes ;-) Edit: Insert foot in mouth :-( Edit2: Remove foot from mouth :o Edit3: Repair edits :-) |
|
799)
Message boards :
Number crunching :
I got a new mouse........
(Message 720958)
Posted 2 Mar 2008 by JDWhale
Post: <snip image> And to think, we used to need a room full of similar hardware just to keep 1 MB of memory on the IBM 360/65 cool. Many a mouse fell prey to our lookout kitties over the years in that room. Edit: Just joking with the guys down at the mom&pop auto parts store, I took a ThermalRight Ultra 120 Extreme in a couple weeks ago, asked them if they could order me one for my Hundai... LOL |
|
800)
Message boards :
Number crunching :
Mac Pro CPU time vs. Angle Range plots ?
(Message 720952)
Posted 2 Mar 2008 by JDWhale
Post: Very good then. All good things take time ;-) |
|
801)
Message boards :
Number crunching :
Mac Pro CPU time vs. Angle Range plots ?
(Message 720920)
Posted 2 Mar 2008 by JDWhale
Post: Just wondering if any of our great chartists have gathered/plotted any data, using Fred W.s data vac, from one (or more) of the leaders in the current Mac Attack? Possibly plot against msattler penny rig or some other reference. I think the results might prove interresting. If there is low correlation bewteen the datas, might this make selection of "WU Credit multiplier" more difficult? Or will only data collected from "stock" applications be given consideration? Kind regards, John |
|
802)
Message boards :
Number crunching :
Just some figures with the new heatsinks :)
(Message 720773)
Posted 2 Mar 2008 by JDWhale
Post: dennis wrote: I am having exceptional performance out of this HSF assy using AS5 and KINGWIN Revolution RVT-9225 92mm H.D.T. CPU Cooler on a Q6600 running slightly OC. 23c to 25c at idle and 34c to 38C crunching SETI 24/7 at room temps of around 70F.It appears the heat pipe direct touch design is working for me well and it is quiet. http://www.newegg.com/product/product.aspx?item=N82E16835124017. I am pleased anyway /salute Thanks for the heads-up, the price is right, and fan is included :-) I'll pick one up on Monday . Might even keep the cruncher cool with room at 85F. |
|
803)
Message boards :
Number crunching :
Just some figures with the new heatsinks :)
(Message 720771)
Posted 2 Mar 2008 by JDWhale
Post:
Thanks for the info, I'll remember to check that when mounting the next Ultra-120 on the Q6600 I'm burning in now (haven't found a case for it yet). Rosewill HSF is keeping at 61-60-59-57C crunching 4 cores with CPU at 3060MHz (340x9). FYI: CoreTemp v0.96.1 on my loose mount are @ 55-54-49-48C Ambient: 21C in Antec P180 Q6600 on Gigabyte GA-P35-DS3L OC at 3200MHz(400x8) AS5 I'll consider lapping come summertime when ambient is 30C if needed. G'day, John |
|
804)
Message boards :
Number crunching :
I got a new mouse........
(Message 720710)
Posted 2 Mar 2008 by JDWhale
Post: Another VNC user here. All of my machines can be accessed from the laptop and because it is across the local LAN I can get full screen, high res images which make the experience comparable to being logged on to the machines locally. Not experience any problems so far and have named each of my machines logically which makes remembering their IP numbers easy too! The major plus is that it is much cheaper than a KVM and less wiring too! I'm wondering why you must remember IP addresses, I'm using DHCP and refer to my hosts by name within my "Island" workgroup : Gilligan, Skipper, Thurston, Lovey, etc. I agree cheaper & more freedom than a KVM switch. |
|
805)
Message boards :
Number crunching :
Just some figures with the new heatsinks :)
(Message 720691)
Posted 2 Mar 2008 by JDWhale
Post: Tell Me about It, I've seen this on all 3 of My PCs, I'm looking at replacing an Old heatsink with a New one soon in May(a Tuniq Tower as the current one is too hard to reinstall and to maintain due to Dust and Cat Fur), I'd love to get one of the New Scythe Orochi heatsinks for use On PC2 and It's Hot B3 Q6600 cpu, But so far in the USA It's not for sale as of yet as far as I can tell, Heck I can't even Google a review of that Scythe heatsink. :D I might get the Scythe since It's so big and now I know what Scythe wants for It $78.80 before shipping and hopefully the shipping from Scythe USA won't be very much. @Joker If you shop around at Directron and find something you want, let me know. I'll be in Las Vegas 23-27 March, and again 18-21 Jun. I'll meet you half way. I Live near Directron in Houston and can pick up without shipping. Have you looked at the Thermalright Ultra 120 Extreme, the HS David is posting about here? $53.99 at Directron 120mm fan not included, they also have a wide variety of Scythe products. FYI: Houston has 8.25% sales tax. PM me if interested. Disclaimer: I have absolutely no affiliation with Directron, It's just a store in my neighborhood and I am available to help out a fellow cruncher. @David Do those heat sinks mount firmly enough? I found there to be a little slippage, but then I didn't remove my MB and just used the Rosewill backplate that was already in place. I'm happy enough with the cooling, just the HS isn't as secure as I'd like. Regards, John |
|
806)
Message boards :
Number crunching :
I got a new mouse........
(Message 720677)
Posted 2 Mar 2008 by JDWhale
Post: Oh yeah......I thought about trying a wireless.....but I was worried about compatiblity issues....... Ever considered to installing/using a Virtual Network Console (VNC) app rather than going the KVM route? All comunication via network interface on your LAN, fewer wires... LOL...I wired up enough already...:-) Just launch VNC viewer, scroll through your list of hosts... and a window pops up with remote desktop fully under your control. Like magic :-0 yessirree, I swear. Course for bios tweaks you'll still have to plug in, but after your hosts are stable, easy as pie. It allows you to spread the hosts around, include the HTPC in the other room, the PC in the bathroom, plus the one in the garage... There is no stopping you now!!! Edit: Looks like trouble upstream... uploads are failing :-( Edit2: 5 min later...Nevermind, just a temporary hiccup :-) Edit3: You do have a PC in the bathroom, don't you? ;-) |
|
807)
Message boards :
Number crunching :
I got a new mouse........
(Message 720671)
Posted 2 Mar 2008 by JDWhale
Post:
I just had to have a look at the LX3 after all the hoopla, that looks like my kind of mousy... a mousy with a tail...wireless is OK in a pinch, but for maximum clickability I require a wire! Know whatta mean? |
|
808)
Message boards :
Number crunching :
I got a new mouse........
(Message 720669)
Posted 2 Mar 2008 by JDWhale
Post: Well...... So what did you do with the old dead mousy? Reason I ask is that my cornsnake, Hermione, just loves thawed out mice...:-) 'Course there might be fight with the kitties, don't you think? |
|
809)
Message boards :
Number crunching :
Wondering about pick of the draw..
(Message 720537)
Posted 1 Mar 2008 by JDWhale
Post: <snip> I'm frequently on trips of duration up to a fortnight. It's not uncommon for me to make this (or similar) settings a few days before departure to load up on WUs to crunch unattended. I won't keep power to my cable modem, cable TV box, water heater and such while on leave for extended periods. But I refuse to shut down the S@H crunchers, lightning storms are an exception. I would be hurt if my assigned WUs missed deadline. I don't believe in "jiggering" the system either, rather just using the system as best I can.
Ah,then by Hogwarts special courier it is. ;-) As Team Hash House Harriers might say, "BOINC OnOn...". |
|
810)
Message boards :
Number crunching :
Francois, Do You Need Help With Bail?
(Message 720512)
Posted 1 Mar 2008 by JDWhale
Post: It's a bit sad, really... Who?ever came up with the speeding ticket based on CineBench & 3DMark makes us laugh at first impression; ironically, IMO, consumers who purchase Skulltrail should be taxed (fined) in the same manner purchasers of Gas-guzzlers are taxed, without 8500 lb exception. In much the same way most gas-guzzling behemoths tend to perform poorly in high performance racing, SKT isn't much of a S@H cruncher, either. 'nuf said... Edit: fix above url |
|
811)
Message boards :
Number crunching :
Francois, Do You Need Help With Bail?
(Message 720493)
Posted 1 Mar 2008 by JDWhale
Post: Carlos wrote: ...Do I need to say more? I completely agree with you. I'm in the process of replacing my older P4s with $500-600 Q6600 rigs. Hope to get ~4000RAC each with minor OCs on air. The newer rigs draw less than 170Watts, about the same as the P4's, but with 4-8 times RAC per box. that Translates to 4-8 times RAC per Watt!!! Plus, the church loves receiving the P4s as donations :-) I cap my SETI power consumption @ 1000W ($66/month) so it looks like I'm limited to 5 Q6600 crunchers with monitor, etc. Anyone know the power draw of the SKT fully loaded, 4 FB-DIMMS, with ATI HD 2900 or dual Nvidia 8800 GTX running SLI (only 2 cards each). I bet the lights in the room dim when the SKT comes to life. LOL Forget the initial cost, I can't afford to operate the darn thing! BOINC on, John |
|
812)
Message boards :
Number crunching :
Wondering about pick of the draw..
(Message 720471)
Posted 1 Mar 2008 by JDWhale
Post: I'll be glad to swap my 17 credit WUs for any 63/75/84s you've got on hand ;-) I suggest you try this on that rig that hates the shorties (VHAR WUs)...
|
|
813)
Message boards :
Cafe SETI :
Knight Rider
(Message 720028)
Posted 29 Feb 2008 by JDWhale
Post:
Did you get the custom KITT paint job? I bet that would be a very expensive option ;-) |
|
814)
Message boards :
Number crunching :
V8 Optimized App for Windows and Linux
(Message 720011)
Posted 29 Feb 2008 by JDWhale
Post: ... However from what i've seen so far, results are interesting. I think the tests don't mean much unless we attach sizes to the operations. Here are some runs with 8198KB data size. Also, different parameters (sizes, operation, etc) can be assigned to each thread to reduce number of screen caps. I'm editing the first images to reduce forum thread size. Core2 Quad Q6600 3.2GHz w/ DDR2 800MHz @ 1000MHz   Edit: @Martin - Thanks for the tip ! |
|
815)
Message boards :
Number crunching :
V8 Optimized App for Windows and Linux
(Message 719990)
Posted 29 Feb 2008 by JDWhale
Post: More of the same. I wonder if there is any difference between versions 1.0 & 1.1 :o |
|
816)
Message boards :
Number crunching :
V8 Optimized App for Windows and Linux
(Message 719966)
Posted 29 Feb 2008 by JDWhale
Post: Crunch3r wrote:
Core2 Quad Q6600 3.2GHz w/ DDR2 800MHz @ 1000MHz Read on 4 cores...  and reading/writing simultaniously...  MB: Gigabyte P35-DS3L MEM: 2x1 GB DDR2 DIMM 800MHz @ 1000MHz Sorry, I can't shut her down to reset memory OC right now :( |
|
817)
Message boards :
Cafe SETI :
Knight Rider
(Message 719836)
Posted 29 Feb 2008 by JDWhale
Post: The new voice is too human for my taste. Also, Using a Mustang instead of a Pontiac. I partially agree with your dissent of using the Mustang. But Pontiac doesn't make a mainstream muscle car anymore, and the Chevy Camaro is already associated with Bumble-Bee from the "Transformers" movie. Personally I would like to have seen the brand spanking new for 2008 Dodge Challenger cast as KITT. My guess is that Ford felt threatened by the upcoming competition and paid a premium product placement fee. |
|
818)
Message boards :
Number crunching :
Performance of MAC running Windows app via VMWare
(Message 719688)
Posted 29 Feb 2008 by JDWhale
Post:
A very good read, thanks. You will notice that I only quoted a small section of the report. The part that scares the bejesus out of me. Do I read this correctly as your memory subsystem alone draws 85 Watts. Whoa, I've heard that the FB DDR2 DIMMS were power hungry, but this is ridiculus!!! So, I'm guessing that if you were to populate all 8 slots with matching memory, the memory system alone might draw ~120Watts. No processors, no fans, no video card, just the chipset/memory! Please someone tell me that I'm reading this wrong :( My Q6600@3.2GHz and 2GB DDR2-800@1000MHz with RAC ~4000 draws 170W when crunching on 4 cores according to the Kill-A-Watt. (definitely a no frills system) The memory bandwith benchmarks actually look better than I expected, I think that you're near the top of the class in that department, I didn't see latency numbers, but from this FB-DIMM Memory paper I'd assume ~50ns for the 8 ranks on 4 channels that you are running (those are dual rank FB-DIMMs, aren't they). I'm terminating the Intel D5000X dual quad core rampage of recent. In my opinion that is not a high performance/cost effective solution for S@H with the current Windows applications. You might be able to pay for a quad core box like this hostid=3954050 just with the savings on electricity :0 Boinc on... |
|
819)
Message boards :
Number crunching :
Francois, Do You Need Help With Bail?
(Message 719338)
Posted 28 Feb 2008 by JDWhale
Post: I just added a 2nd Skulltrail to my stack :) Nice! Just out of curiosity and I understand if you can't until after the product is actually released or maybe because of some other nondisclosure reason, could you post images of your BIOS settings? Regards, John |
|
820)
Message boards :
Cafe SETI :
Knight Rider
(Message 719306)
Posted 28 Feb 2008 by JDWhale
Post: I think KITT and Mike Traceur will meet up with Danny McCoy and the gang from the show "Las Vegas", did anyone notice that it was at the "Montecito Casino" that Mike was playing poker in episode 1? Moreover, I predict that Mike & Danny know each other from the Iraq war. Future episode prediction... Synopsis: Something cataclysmic happens to KITT and he goes berzerk, Danny McCoy comes to the rescue in his ultra cool classic yellow Camaro convertable... and Mike Cannon toubleshoots KITT and does an emergency BIOS reset (similar to brain surgery) to return KITT to normalcy. That is, if you call KITT normal ;-) Did episode 2 already air? Please don't tell me that I hit the nail on the head with my prediction! STOP the PRESS...I just read that "Las Vegas" has been cancelled :( But what an episode that could have been. |
|
821)
Message boards :
Number crunching :
GPU crunching on G8x
(Message 719284)
Posted 27 Feb 2008 by JDWhale
Post: Why not spend $650 for a Tesla C870 GPU Computing Processo (128 cores). And not much more expensive if I can them get for $200 ea. I am also thinking that with dual quad core XEONs, 8 WUs can crunch simultaneously. With 4 cores on CPUs, maybe with affinity set with 2 per CPU, I'm thinking that memory contention won't be an issue with those restrictive FB-DIMMS. Am I right here, that the goal is to have the WU loaded entirely on the GPU so that GPU crunching will not require much bus bandwith? Is there still the restriction that each GPU process will require a cpu core? Sorry, if I am asking stupid questions, but it has been a while since I've caught up on this project development thread. Does anyone care to speculate on the potential RAC such a rig might reach, I imagine 4 cpu cores might be ~1000RAC each == 4000RAC. + ???RAC on the quad 8800 GTs. I do not have a clue what might be possible on them. I know that 6-8 years ago I witnessed GPU crunching, Seismic 3D Prestack Velocity Panels for analysis, on the SGI Onyx Reality. Very impressive at the time. I am excited with the performance potential as this project developes. Keep up the good work, "Hats off" (translates to "respect") to all involved. Kind regards, John |
|
822)
Message boards :
Number crunching :
V8 Optimized App for Windows and Linux
(Message 719262)
Posted 27 Feb 2008 by JDWhale
Post: LOL... Descended from indeed. Though I've never been there, Daddy-O was born & raised near Bunceton, MO. :-D I guess it shows :) |
|
823)
Message boards :
Cafe SETI :
Knight Rider
(Message 719259)
Posted 27 Feb 2008 by JDWhale
Post: What is everybodies thought on the use of a different voice for kit? Personally, I like Val Kilmers voice. Seems not quite as "robotic" as William Daniels was. Maybe a reflection on technological advances in the 25 years of elapsed time. Also, the new KITT seems to be better programmed for emotional response. |
|
824)
Message boards :
Number crunching :
V8 Optimized App for Windows and Linux
(Message 719211)
Posted 27 Feb 2008 by JDWhale
Post: Crunch3r wrote: Finaly :) Thanks Crunch3r, for your persistence. After I've had the bruise on my forehead a while, I might learn to open the door before trying to exit the room :) I haven't had time to DL the V8 code you pointed me toward yet, but I'll assume that part of the hand massaging in that code reduces main memory reads and does more with the data each time it's cache. Working smart not hard is the better path. Thanks for being persistent with me, now it's time for me to go deep into my cave and hibernate. Ciao, John |
|
825)
Message boards :
Number crunching :
V8 Optimized App for Windows and Linux
(Message 719176)
Posted 27 Feb 2008 by JDWhale
Post: I'm curious if anybody following this thread either has or has access to the Dell Precision T7400 and is willing or able to run some benchmarks... SiSoftware SANDRA (Lite version is free for personal use), or maybe Everest Home Edition(Sorry I can't find link to that specific product today, maybe not as reliable, but easier to run from a USB device without doing an install) read/write/latency tests. Maybe Dell is a similar partner with Intel and can get their hands on the "good stuff". Cheers, John |
|
826)
Message boards :
Number crunching :
V8 Optimized App for Windows and Linux
(Message 719166)
Posted 27 Feb 2008 by JDWhale
Post: Alinator wrote: LOL... Thanks for that report, Alinator. I saw this Mac Pro wiki earlier today showing the MacPro use of Extensible Firmware Interface (EFI). Available when booting OSX & linux, but not available when booting WinXP using Bootcamp. Apple's Boot Camp provides backwards-BIOS-compatibility, and dual and triple boot configurations can be achieved.[24] These operating systems are easily installable on Intel x86 based Apple computers:
|
|
827)
Message boards :
Cafe SETI :
Knight Rider
(Message 719140)
Posted 27 Feb 2008 by JDWhale
Post: Cool. I used to love the old Knight Rider series. Hope they have got the new series right. Look forward to that making it's (slow) way over here :) Thanks for the info. Not to give too much away, but just enough to keep you drooling. Knight Industries Three Thousand(KITT) comes in the form of a Limited Edition Ford Shelby GT 500 KR. A beautiful car, though goes a bit away from my "green" roots. Maybe not if you only drive it on special occasions like Saturdays up to the public park for a Tender Loving Hand Wax Job, or to go watch Friday night drag races. ;-) |
|
828)
Message boards :
Number crunching :
V8 Optimized App for Windows and Linux
(Message 719128)
Posted 27 Feb 2008 by JDWhale
Post: JDWhale wrote:Can someone running OSX on the MAC Pro please post some memory bandwith and latency test results? @UL1 - Thanks, I wish I knew. Not being familiar at all with MAC, I don't even know if such test programs exist (but I think they must). So, I am appealing to those knowledgable about benchmarking MAC under OS X to share memory bandwith & latency test programs. It seems that all the benchmarks focus on video editing, encoding, (etc.), which as resolutions and video quality rise, can put more demand on memory performance when using an application that runs across all 8 cores. I'm still looking, John |
|
829)
Message boards :
Number crunching :
attn Bad to the bone
(Message 719119)
Posted 27 Feb 2008 by JDWhale
Post: I do believe that this thread has potential of going viral without the video. I think I'll stop chasing the intel 5xxx chipset bandwith issues under Windows, put my blinders on and focus tuning my new Q6600 on WinXP not to have any memory bottlenecks. Maybe I should not have become active in this forum and just stayed a lurker :( regrettably yours, John |
|
830)
Message boards :
Number crunching :
attn Bad to the bone
(Message 719102)
Posted 27 Feb 2008 by JDWhale
Post: sorry JDWhale hope you were not offended being called a geezer :-) it wasn't meant in any derogatory manner.. No offense taken, I enjoy the jab ;-) (@Adrian see PM I sent) Sorry if I broke protocol by using a specific systems RAC and specs to help make my point about how the top MACs are performing against the other equivalent hardware. The MACs are truely the top performers. I call them as I see them when it comes to high performance computing, that was my job for many years. I am very new to using this type forum and still have much to learn. I am very sorry if I offended anyone. Kind regards, John |
|
831)
Message boards :
Number crunching :
attn Bad to the bone
(Message 719093)
Posted 27 Feb 2008 by JDWhale
Post: hello mate That geezer would be me, and I never referenced anyone by name, only pointed to a system with unusually high RAC for its specs. Cherry picking doesn't bother me, I simply pointed out that systems operated in that fashion should not be used for comparisons, except against other cherry pickers, of course. The "other" system that referred to does not seem to be operating in a "cherry picking" mode, rather steadily crunching all WU's issued. If my Q6600 only processed WUs with AR ~0.126878 it could attain a RAC ~6500, but I am not that picky :) G'Day [edit] @Adrian - I Tip my hat to you, great performing rig! [/edit] |
|
832)
Message boards :
Cafe SETI :
Knight Rider
(Message 719084)
Posted 27 Feb 2008 by JDWhale
Post: Hey guys, is this reference the Knight Rider film? Not a movie (US translation for "film" :), rather Knight Rider(2008) is TV series that broadcast it's debut episode a week ago past Sunday. Next generation of Knight Rider(1982) a popular TV show from the '80s. OC'd US muscle cars on steroids:) Woohoo!!! [edit] FYI: 28 September 1982, was 3 months over 11 blue moons ago :) [/edit] |
|
833)
Message boards :
Number crunching :
V8 Optimized App for Windows and Linux
(Message 719072)
Posted 27 Feb 2008 by JDWhale
Post:
In that review, it is stated that, [the] "V8 delivers considerable torque to the rear wheels thanks to 16MB of total L2 cache, dual 1333MHz front-side buses, a staggering 21GB/s of memory bandwidth, and eight Core-microarchitecture CPU cores running at 3GHz each". Yet in the shootout when they pitted quad-core Xeons against Opterons, the V8 choked on the Memory bandwith tests, yielding a paltry 4GB/sec for the dual XEON 5355 system. Seems that benchmark was running two threads, running on the same chip, thus utilizing only one of the available FSBusses. Even by doubling the that figure to compensate, bandwith might have reached 8GB/sec, still less 37% efficiency!!! Does anyone know of similar benchmarks on the MAC Pro running OSX with all 8 slots filled with matching FB-DIMMS? Only when I can see those OSX benchmarks will I believe that this performance discrepancy is totally code related. I apologize if this an old subject that has been discussed before. With me being rather new to these boards if someone could point me to any prior discussion I'd be much appreciative. TIA. Can anyone come up with a 4 core MAC Pro to compete against this 4 core Vista box currently in 17th place on the Top Hosts list running Crunch3r's Vista build? Show me one with higher RAC and I might then believe that it is the code. This one doesn't count because he is a cherry picker! @Crunch3r - I've only just began to look at the stock S@H code, Do you know how I can get it to build using MS Visual C++ 2005 Express Edition? (Where is the .prj file?) I would like some time gaining familiarity and experimenting before I even think about porting code. |
|
834)
Message boards :
Number crunching :
V8 Optimized App for Windows and Linux
(Message 719023)
Posted 27 Feb 2008 by JDWhale
Post:
Unless the optimized MAC code employs a technique to lower the memory bandwith required, then I do not believe the performance gains are all code related. I have yet to see a 5000x based system running XP or Vista that can beat the memory performance of a tuned dual channel DDR2 system. Though I have no basis for this assumption, I think the MAC Pro running OSX has broken through the 10GB/sec memory bandwith to feed those 8 cores more data than XP or Vista. But then again I am frequently wrong. Can someone running OSX on the MAC Pro please post some memory bandwith and latency test results? |
|
835)
Message boards :
Number crunching :
Performance of MAC running Windows app via VMWare
(Message 719009)
Posted 27 Feb 2008 by JDWhale
Post: I’ve got two 5000x systems and can't find any other explanation then Alex's code is better. I’ve poured all over my work units making comparisons to other computers and can't figure it out other then the Mac has a 1.5 times faster app on the V8. I wish it was just a driver issue. When I look at results from your E5320 rig, something just doesn't seem to add up. The processing times seem too high with that 2.6GHz OC... Compare these 2 results of nearly similar AR's one from your E5320 rig, one from my E4500 rig Core2duo with similar OC frequencies. My rig crunched the WU in half the time. I think there is something other than code slowing you down. This difference can not be explained by Alex code, can it? My e4500 is still ramping up, only been online a couple weeks. I expect RAC ~2000 in another 2-3 weeks. If you have 4 times the processing power, then you should be able to attain nearly 4 times the RAC. I just lost a whole day behind a power failure, and all my WU's were "cancelled by server" when it came back online. I wish I could get my hands on a 5000x system and enough memory kits to run some benchmarks and do some tuning. I still think that the problem is with the memory subsystem not having the bandwith. Can you confirm that you are running 8 memory slots, fully populated with matching sticks, in dual channel interleaved mode. Have you done any benchmarking to get synthetic performance numbers? Good luck, |
|
836)
Message boards :
Number crunching :
Why am I a volunteer tester?
(Message 718610)
Posted 26 Feb 2008 by JDWhale
Post: ... Thanks, Josef. Just as I thought, "Total Credit 0", never did any work for the project. Guess it was just my good looks that earned me "Volunteer tester" status. On the other hand, it might have my entertaining of the troops (picking up the bar tab) that earned me a pass ;-) |
|
837)
Message boards :
Number crunching :
Where to buy? - Any suggestions?
(Message 718589)
Posted 26 Feb 2008 by JDWhale
Post: MicroCenter is blowing out the Q6600 right now for $199. FYI: In-store only. Just picked me up another one ;-) You might want to call ahead. The store in Houston ran out after 2 hours Friday past. I walked in this morning when they opened at 10:00AM and they had one on the shelf, SLACR in a retail box. 30 day exchange only, no refund policy on CPU's. Good luck, John |
|
838)
Message boards :
Number crunching :
Why am I a volunteer tester?
(Message 718580)
Posted 26 Feb 2008 by JDWhale
Post: Would running any beta or alpha project, like QCN alpha or malariacontrol.net beta also make you show up as a volunteer tester ? Thanks for being my wingman on this Blu Dude. Taking a closer look at BoincStats, I did find out that I attached my laptop to SETI Beta some time ago (Probably partying it up in Las Vegas when I ran out of SETI work, explains why I can't remember:-) Can't find where I ever did any work for the project though. The project probably didn't have any work to distribute when I tried :-( Maybe just attaching and trying to download work is enough to get the "Volunteer tester" tag. Seems like the tag is cheap enough, almost like getting something for nothing. Does anybody know how I can find out if I ever completed any SETI Beta WU's? |
|
839)
Message boards :
Number crunching :
Credit?
(Message 718565)
Posted 26 Feb 2008 by JDWhale
Post: I'm saving mine up for a free toaster! I think NEZ is saving up for the Hyperdrive. Or was it a Warp Drive? Can't remember, too many reboots. Pricetag: Infinite credits! And he's almost got enough ;-) |
|
840)
Message boards :
Number crunching :
Validity of benchmarks
(Message 718508)
Posted 26 Feb 2008 by JDWhale
Post: Just looking at the review & benchmarks of this Crucial Ballistix Tracer memory kit. I especialy like the You Tube video on page 3 where the memory is being exercised by the SiSoftware Sandra Memory Read Test. Reminds me of the Northern Lights. Look at the bandwith and latency graphs when overclocked to 1200MHz compared to not OC'd. Check the 3% benefit on Super_PI to 8M digits when the memory is OC'd to 1200MHz And best of all, this 2 X 1GB kit can be had for less than $100 right now! I can't wait to team this kit with a new Q6600 running at 3.2GHz (8*400MHz). I know that when I overclock it to 1200MHz I can expect about 7.5% increase in memory bandwith and 6.5% better latency. Is there any doubt that this memory will boost performance with S@H on my OC'd Q6600? |
|
841)
Message boards :
Number crunching :
GPU crunching on G8x
(Message 718487)
Posted 25 Feb 2008 by JDWhale
Post: Not that I will drop the $$$ for the MOBO, but just a possibility question. Suppose I pick up the MOBO from the new Intel Skulltrail, populate with a something affordable like dual XEON E5410's with the 12MB L2, and 4 of the GeForce 8800 GTS 320MB. Do you see where I'm going here? 1 X D5400XS Motherboard ~ $ 500 estimate - pricing not yet available
2 X Intel XEON E5410 CPU ~ 600 Actual $275 ea. right now at NewEgg
4 X EVGA GeForce 8800GTS 320MB~ 800 Actual $190 ea. with MIR now at NewEgg
1 X Antec TP Quattro TPQ-1000 ~ 200
Case,memory, harddrive, etc ~ 500
------
Total $2600Crunching on 4 GPU's and on 4 CPU cores.Possible? Issues? |
|
842)
Message boards :
Number crunching :
Validity of benchmarks
(Message 718453)
Posted 25 Feb 2008 by JDWhale
Post: The problem with so many things with BOINC and SETI is that they are estimates, and the estimates are used to predict future events. Ummm.. are you sure you're in the correct thread? |
|
843)
Message boards :
Number crunching :
Validity of benchmarks
(Message 718452)
Posted 25 Feb 2008 by JDWhale
Post: Well, since I was cut off from being "off topic" in another thread, here's my response: Ahh a breath of fresh air and a fresh topic. Thank you.
The only application that I am interrested in attaining maximum performance is crunching S@H. I don't run web servers & I don't play 3D first person shooter games. Heck some my machines do not even have hard drives in them, they boot from USB, they were built to crunch. Motherboard, CPU, Memory, USB stick, cheap video card if none onboard, case optional. No hard drives, DVD drives or expensive video cards that run up the system price and electricity bill. My running benchmarks is a key step in the iterative process of maximizing the system performance by tweaking the system at it's lowest levels of assembling hardware and BIOS performance settings. I would include running a S@H benchmark if a comprehensive one could be made available, but I fear that with the wide assortment of hardware and assortment of Work Units(WU) being split and combination of WUs being executed at any given time on dual- or quad- or octa-core systems, such a suite of S@H benchmarks might be too comprehensive to ever finish. I do my own S@H benchmarking by using preemptive WU schedueling and comparing CPU times for comparable WUs executed with WUs of all types, but that is a different topic. I call that "schedueling WU's" rather than benchmarking. So what I do to tweak performance of WU crunching is to run and analyse synthetic benchmarks that measure memory bandwith and latency. This part is pretty straight forward, maximize bandwith, minimize latency, do not let one suffer too much because of the other. Try to minimize Super_PI run times, not only on a single core, but on all available cores. Lower is always better on single core, but a significant performance as the number of cores running Super_PI increase tells me that I am memory bandwith restricted, the same can be evidenced by running 4 cores of VHAR WUs simultaneously my thoughts on that have been posted on other threads. The problem with running S@H as a benchmark is that it takes too long to complete the tasks when I am doing the hardware tweaking. Maybe an hour or two to crunch the WU's, getting the right mix of ARs downloaded, and synchronization is a real b!^&#. But what I can do rather quickly is to fire up Everest to check memory bandwith and latency, run Super_PI to check some number crunching combined with memory bandwith, and run Prime95 to check stability and core temperatures. This testing can be performed quickly. Prime95 is not really a necessary stress test since my overclocking is not extreme but I do run it after Super_PI has reached it's limits. So I then record the results and go onto another tweak iteration within 5 or 10 minutes. I move that the "controlled test" isn't done properly. This is why many companies (including AnandTech.com) have started developing their own test suites to get a more accurate picture of the expected role. AnandTech has done a great job in developing their own server tests which paint a much more realistic picture as to what the hardware can do. Doesn't apply... this is a SETI Number Crunching forum, not a generalized server performance forum. Folks that run S@H on servers built for other purposes and actually do just let SETI run on otherwise idle CPU cycles probably do not make maximizing S@H performance a priority.
Not unicorns, but true racing machines just like Thoroughbreds are among horses, bred to finish first.
True, run the benchmarks across a wide range of hardware and pick the sweetspots that overlap. I don't think that you will need to get so specific as to the chipset on motherboards, processor family is probably good enough. In the case of S@H, weight the results toward the hardware distribution contributing to the project, and maybe toward the newer technologies a bit. Most of the major project performance boosts will be with the newer hardware as it proliferates due to lowered costs and reduced power consumption that make them cheaper to own and operate.
Feedback is good, but please do not dismiss the benefit of using benchmarks as a tool to measure performance. Yes focusing totally on a single benchmark is wrong, but when a single benchmarks results clearly show a problem, then one has to wonder why and see if adjustments can be made to resolve the issue. IMO. Like OZZY says "I'm going through changes" and so is computing technology. Rock on OzzFan, John |
|
844)
Message boards :
Cafe SETI :
How old are you?
(Message 718410)
Posted 25 Feb 2008 by JDWhale
Post: 3 months short of 21 blue moons Sorry, no Moon Pie for you today :-( But you can try again tomorrow :-) |
|
845)
Message boards :
Number crunching :
Where to buy? - Any suggestions?
(Message 718409)
Posted 25 Feb 2008 by JDWhale
Post: MicroCenter is blowing out the Q6600 right now for $199. FYI: In-store only. Just picked me up another one ;-) |
|
846)
Message boards :
Number crunching :
Specs for Core2 Extreme QX9775
(Message 718407)
Posted 25 Feb 2008 by JDWhale
Post: What I'm hoping to get out of this thread is probably a little different, because I'm looking at the programming side. Some idea of the dynamics at work as far as threads & memory bandwidth requirement is nice. What I gather I'm seeing so far is 8 cores all operating on different data, cramming the Memory &/or FSB until they choke. [I would argue this phenomenon would still occur on desktop DDR2 etc ...and MACs too ] Yes, yes I've noticed on my E4500, running Single Channel because I didn't have a matched set of sticks at the time, that setting in BOINC user preference to only use 1 processor, that I could run 2 WUs faster by wall clock time than running those same 2 WUs simultaneously. That factoid hastened my purchase of a matched set of sticks and tuning bus frequency for data throughput. Though I've just started looking at the seti_client code (fetched from berkeley last night) what I'm seeing in the first file I've looked at to interesting and some experiments that I want to try. Can ths project by built using MS Visual C++ 2005 Express Edition like BOINC can? Also maybe I should look to Lunatics site and start with the optimized code, It may turn out that 8 cores working cooperatively on a single task [or perhaps one per socket] in finer grained parallel fashion may be a more efficient use of the bus ( Reduce Bus throughput requirement by 8(or 4), cleaning up the cache demands at all levels and hopefully bringing the software back to being CPU bound (which we can overclock ... :P) More yes, yes, yes!!! With the release of multicore systems, I pretty sure that efficiency can be obtained by using finer granularity than just launching more processes. Use more threads in the same process! Have you looked at the data transfer rates between cores? On my Q6600 running at 3.3GHz I'm getting over 300GB/sec intercore transfer rates on 32kB-64kB block sizes!!! This reminds me of programming the SGI Power Challenges & Origin servers. Huge amounts of shared memory(16-128GB) with 16-64 CPUs all working on 1 common task rather than matching the number of tasks to the number of CPUs. Just a thought...Maybe check if the WU's processing could be set up streaming across threads, or just break up the WU in to chunks assigned to respective threads. Of course one of the prime benefits to all this is that performance would not suffer if running on single core system. And I guess that you could still honor the BOINC request of max number of processers and just use that many threads. If your quad core is running hot, just cut the number of threads to 3 ;-) Please slap me if I'm sounding stupid. This discussion might be better off in a new thread and bring in some new programming blood. Edit: [ot] Just got back from MicroCenter with a new Q6600 chip for $199!! WooHoo!! [/ot] |
|
847)
Message boards :
Number crunching :
Specs for Core2 Extreme QX9775
(Message 718399)
Posted 25 Feb 2008 by JDWhale
Post: Then I proffer: what good is maximum potential if it can never be achieved? Because they give you a performance target. Ok OzzFan, Just because the benchmark is "Synthetic", does not mean that they post the theoretical limits, rather I've never seen a benchmark that came close to posting theoretical maximums. Those Synthetic Benchmarks truely are achieving said performance under a controlled test! Obviously you feel that this type of testing is worthless, but it does give some of us programmers a target to aim for. And to me, that information is not worthless. I want my programs to be able to achieve that performance when necessary. Okay we disagree on the usefullness of the numbers and I see the charts & graphs as something more than "pretty pictures" just to look at, rather a valuable tool to use when assigning block sizes for memory access in this case. It's okay to disagree, even if, in your opinion I'm wrong. I will not discuss the validity of benchmarks any more in this thread, their validity is off topic, IMO. Start another thread, please. |
|
848)
Message boards :
Number crunching :
Specs for Core2 Extreme QX9775
(Message 718362)
Posted 25 Feb 2008 by JDWhale
Post: I just don't trust synthetic benchmarks anymore. They really don't indicate real world performance (but I'll admit they're fun to look at!). Maybe not, but they can tell you performance under ideal circumstances. Of course, actual (real world) performance will be different for everybody. The "synthetic benchmarks" really are moving data at the rates indicated, and do provide basis for comparison. Isn't that what benchmarks are supposed to do? Take, for instance, the EPA Mileage Estimates used to rate the fuel efficiency of automobiles here in the USA. Very few people, if any, truely do achieve those City/Highway numbers, though I do get pretty darn close. Driving habits and conditions vary for each driver, but the numbers do still provide a basis for comparison, their purpose. |
|
849)
Message boards :
Cafe SETI :
How old are you?
(Message 718352)
Posted 25 Feb 2008 by JDWhale
Post: 3 months short of 21 blue moons Sorry, no Moon Pie for you today :-( |
|
850)
Message boards :
Number crunching :
Specs for Core2 Extreme QX9775
(Message 718291)
Posted 25 Feb 2008 by JDWhale
Post: I did an 'enforced test' of this when I got my octocore (Dell Precision 490 workstation, OEM version of Intel 5000 board, supplied with 2 x 1GB FB DIMMs @ 667). It felt really slow as a SETI cruncher - speeded up by about 25% when I put in an extra two DIMMs, as you can see here. Thanks for refreshing our memories with that thread, now imagine feeding 8 cores like Skulltrail is doing, and I think you see performance stepping back to similar to before your memory upgrade!
Looking back at the BIOS settings photos in Tomshardware review, it seems that with the X5400DS (Skulltrail) board with it's 4 memory slots can be configured as
|
|
851)
Message boards :
Cafe SETI :
How old are you?
(Message 718172)
Posted 25 Feb 2008 by JDWhale
Post:
I remember curling up to take naps between the memory banks, or was it the massive floating point unit, attached to the IBM 360 Model 44 that I cut my FORTRAN teeth on (that was the warmest spot in the computer room). I, too miss all those blinking lights on the front panel. Now if I could only remember the toggle sequence to clear core. Must have toggled one time too many and cleared my own core. LOL |
|
852)
Message boards :
Cafe SETI :
How old are you?
(Message 718156)
Posted 25 Feb 2008 by JDWhale
Post: 3 months short of 21 blue moons A Moon Pie for the first correct answer ;-) |
|
853)
Message boards :
Number crunching :
Specs for Core2 Extreme QX9775
(Message 718142)
Posted 25 Feb 2008 by JDWhale
Post: OzzFan wrote: No, I can't attribute it to bandwidth issues entirely. Before I ran Vista on my Xeon machine, I was running XP SP2 and would consistently get RACs of 2,400 or so. Since switching over to Vista, my RAC has dropped to what it is now (roughly 14-1500). Granted, I'm also running CPDN on this machine too, but its too large of a drop to be considered for bandwidth issues alone. Unfortunately, I like Vista more than XP so SETI will just have to suffer the consequences. Sorry to hear about possible Vista performance hit, I'm running Vista Ultimate on my Q6600 (shows up as Vista Professional) and am very pleased with performance. And I agree with the whole Vista Experience, it pains me to have to use XP anymore. Crunch On, OzzFan. It's time for my "Wheels of Confusion" fix ;-) |
|
854)
Message boards :
Number crunching :
Where to buy? - Any suggestions?
(Message 718135)
Posted 25 Feb 2008 by JDWhale
Post: Directron has some specialty gear at great prices, picked up Thermalright 120 Ultra Extreme there the other day. I've been buying from them for years, plus it helps that they are in the neighborhood. Generally though, prices are better at NewEgg and I've found NewEggs online chat with customer service representative very handy for instantly resolving any billing issues that may arise. Delivery is always fast with their 3-business days UPS service and shipping is often free:-) Be sure to subscribe to their newsletter for promo pricing on a wide variety of items weekly. I've also picked up some items at Frys.com. Sometimes I think that they misprice items by applying rebate deductions directly to their merchandise and then still offer & grant the rebate. Got a couple of video cards from them in the past 3 weeks, free after mail-in rebate. Granted, it's not the latest hi-performance video cards, but the GeForce 6200 more than meets my needs. it has a heatsink and not a fan, and is a power miser, so is both quiet and is cheap to operate. They also had the Antec P180 case for $70 after $30 MIR, though that was too expensive to ship, so I picked that up at the local Frys store, same price. That giant Thermalright HS from Directron fits perfect on a NewEgg Q6600 in the Antec case from Frys. Maybe I got carried away with details, but the point is that it pays to shop around and high performance computing need not be too expensive. Sometimes the best prices can be found in your own backyard. Good luck, John P.S. I'm still keeping my eye on you in my rearview mirror, Bob in Fl |
|
855)
Message boards :
Number crunching :
Specs for Core2 Extreme QX9775
(Message 718031)
Posted 24 Feb 2008 by JDWhale
Post:
I would tend to believe those numbers OzzFan, I just had a peek at some of your results. Granted you are running stock application, but the variations in runtimes for your VHAR WUs is maybe 30%, I think the differences can be attributed to the varying WU mix and with it, the varying bandwith requirements to feed the cores. It's been my experience that you need in excess of 8GB/sec to get similar runtimes on VHARs feeding 4 cores. You might try setting your local preferences to use at most 3 processors antdsee if your CPU times stabilize. If they do, then it confirms a bandwith issue. Sandra just might be your friend. |
|
856)
Message boards :
Number crunching :
Specs for Core2 Extreme QX9775
(Message 717975)
Posted 24 Feb 2008 by JDWhale
Post: my DDR2-800 running dual channel @ 960 gets ~ 7.5GBytes/sec too [On SiSoft Sandra Lite]. Checking the List of device bandwidths wiki, the theoretical is somewhere between: Right, I was using the Oz-imposed theoretical max of 8 times bus frequency* multiplier, not taking into account dual channel. So according to the "Synthetic Benchmark" results and Wiki theoreticals, we're at ~ 47% theoretical that's a bit more believable. That puts us in the ballpark of what OzzFan was saying of actual being ~theoretical/2.0 Thanks, John |
|
857)
Message boards :
Number crunching :
Specs for Core2 Extreme QX9775
(Message 717891)
Posted 24 Feb 2008 by JDWhale
Post:
Should I believe that my DDR2-800 PC-6400, OC to 1000Mhz, PC-8000 yields theoretical bandwith of 8.0GB/s and benchmarks at 7.55GB/sec with SiSoftware Sandra benchmarking tool. Is 5.6% under theoretical maximum believable? I guess what I'm asking is if you are familiar with the tool, and if so can I believe the computed benchmarks? Go OZZY! |
|
858)
Message boards :
Number crunching :
Specs for Core2 Extreme QX9775
(Message 717854)
Posted 24 Feb 2008 by JDWhale
Post:
Did you take into account that it is "1600MHz dual independent Front Side Busses" ? Lets look at Dells just released Precision T7400 specs, another Intel D5xxx MOBO solution running the X5482 Xeon like some of the leading MAC PROs. Memory Bandwidth Up to 32.0 GB/s of theoretical memory bandwidth for 667MHz memory in quad channel mode i.e. 4 DIMMs Up to 38.4 GB/s of theoretical memory bandwidth for 800MHz memory in quad channel mode i.e. 4 DIMMs They also claim data transfer rates of 12.8GB/s or about 33% of theoretical max. Come on Ozz, my DDR2-800MHz gets 6-7 GB/s running Dual Channel mode, this D5400XS should at least get better than that in "dual channel interleaved" mode. Shouldn't it? Again I say, "Something just ain't right about the Skulltrail!" |
|
859)
Message boards :
Number crunching :
Specs for Core2 Extreme QX9775
(Message 717842)
Posted 24 Feb 2008 by JDWhale
Post: A couple of possibly related clues (just observations really): Go back to my quote from the Intel released QX7775 spec that started this thread... The 1600 MHz Front Side Bus (FSB) is a quadpumped bus running from a 400 MHz system clock making 12.80 GBytes per second data transfer rates possible. The 12800 GB/s is 12.8 GBytes. In specs "B" is for bytes, "b" is for bits, IIRC, it's been many years since I used my 56Kbps modem (Kbps==Kilobits/sec);-) -John |
|
860)
Message boards :
Number crunching :
Specs for Core2 Extreme QX9775
(Message 717811)
Posted 24 Feb 2008 by JDWhale
Post: jason gee wrote:
That was a nice story about helping your sister, Jason, we should all be so lucky to tweak MAC Pro hardware to fit the application, hands on, rather than trying troubleshoot from the sidelines. Sounds like fun! Those Wolfdales are great chips, I love both high performance and saving energy, if only the auto makers could deliver on that front. Now to the meat... Warning: The following are my expressed opinions and are not those of my former employer or any other persons, living or dead. I take full responsibility for expressing my own ignorance and am not being forced to make these statements. It's my understanding, both from lurking in this forum for many moons, and from my hands on experience with these Multi Beam WU's, that there are different resource needs depending on the WU's being crunched and the hardware a user employs. I fear that high bandwidth on multiple cores most directly affects CPU times when crunching multiple Very High Angle Range(VHAR) WU's. Crunching 2 of these WUs simultaneously can hinder the performance of a Core2 Duo by as much as 50% or more, if your memory subsystem isn't configured optimally. Crunching these VHAR WU's demand an almost constant "Read from memory". They just can't get the data fast enough, if you throw a faster processor, you better tweak the memory some more, too. These VHAR WUs also have the greatest credit reward potential. I won't get into any more details about that in fear of cherry picking WUs, (Yes I've tried it, purely for testing purposes;-), and feel that highly efficient scheduling of WUs has more potential RAC benefit than cherry picking). Crunching other types of WUs have other requirements and the memory subsystem is not stressed as much, faster CPU's are just what the Dr. ordered for these. But consider having to run these "other", non-VHAR WUs in an environment where the bus is already saturated trying to the VHARs that are running. Now you've slowed down the non-VHAR WUs because of the memory bottleneck. (I've been solving these types of behavior in my workplace for years, remove bottlenecks/increase performance, repeat.) I don't profess to be an expert, these are just my opinions gathered from my observations and experimentation. [ot] Please excuse my [apparent] obsession with trying to milk the most of every CPU cycle, but after all, it's hardware that I paid for and I'm the one paying the electricity bill. The obsession might also come from years of trying to get just 1-2% performance increases, at a time, in high performance computing environments. Knocking just those 1-2% might mean the difference between bonus and penalty on very large scale pre-stack depth migration jobs that would require an much as 6 months to complete running 64 CPUs in shared memory. I worked as a programmer in seismic data processing and 3D visualization/interpretation. Yes, I might be dating myself, I'm sure that with todays compute clusters the runtimes are down to just a month or two(probably mostly waiting on the analyst). Witnessing SGI demonstrate the migration computations on the graphics hardware was very impressive, but it was just a demo at the time. I can't wait till I can try crunching SETI on a GPU. Of course, then I'll have to buy a real video card :-( [/ot] I'm just trying to get a grip on what this D5400XS/QX9775 system called Skulltrail, just now coming to market, might be capable of with the optimum configuration. There seem to be so many ways to configure the memory and the reviews aren't helping because each one seems to be using a different memory configuration, the bandwith just doesn't seem to be there. I'm convinced Intel has crippled this board by not putting enough memory slots to feed the 8 cores from the required FB_DIMMs, and fully utilize the chipsets dual branch, dual channel; dual channel interleaved; quad channel, call it whatever you want, memory subsystem! Looking back at the page that Alinator referenced, Frontside Bus Properties
Bandwith 12800 MB/s
Memory Bus Properties
Bandwith 20640 MB/swhere is all the bandwith going? This system only reads at 3.8GB/sec with the configuration that this reviewer chose. Do the FB-DIMMs only allows this board to read from memory at 30% efficiency? Is that the performance penalty for using FB-DIMMs? Isn't there anything that can be done with the configuration options to tweak this thing to respectability?Like I've stated before, my core2 Duo E4500 at 2528Mhz on an ASUS P5B-MX runs the VHAR WUs faster than the Skulltrail, granted I process only 2 at a time, but I fear that I'm nearly saturating the bus (Read=6GB/sec, Latency=74.1ns according to Everest Home Edition) with just 2 cores if running 2 VHAR WUs concurrently. Just imagine the performance penalty trying to feed 8 higher power cores that are running 58% faster with less bandwith. I build my own systems, so this box (Skulltrail) is not for me. I'm just trying to get my head around all the numbers. Remember that it was my career 5 days a week to get the most that the hardware had to offer and to increase efficency to collect the bonus rather than paying a penalty. Sorry I rattled on for so long, It's time to go knock about some golf balls. G'Day, John |
|
861)
Message boards :
Number crunching :
Why am I a volunteer tester?
(Message 717698)
Posted 24 Feb 2008 by JDWhale
Post:
I doubt that I ever ran BOINC alpha, as I'd never heard of BOINC 'till a couple months before SETI Classic shut down and I switched over. I guess that it's possible that I attached to SETI Beta once upon a time when SETI had no work. Can't really recall. Thanks, John |
|
862)
Message boards :
Number crunching :
Specs for Core2 Extreme QX9775
(Message 717692)
Posted 24 Feb 2008 by JDWhale
Post: Hmmmm... Thanks for straightening me out Alinator. Now I'm totally confused, I've seen that page from Toms before before... Clear as day "Quad Channel". Can only using 1 FBDIMM per channel explain the 3.8GB/sec on the read benchmark on this page ? Granted 7.1GB/sec for writes is stellar, but come on 3.8GB/sec reads will not even feed a Core2 Duo properly crunching SETI (my e4500 chip on ASUS P5B-MX reads at 5.7GB/sec (according to Everest) and still can't do 2 VHAR's without penalty! Something is still amiss on this board. Maybe it would perform better with Dual Channel, 2 slots each. LOL The reviews just do not add up, only constant theme is that this board is bandwith constrained. [edit] Maybe the Anandtech wasn't too screwed up after all, under photo of the memory on the page you linked it states Dual-channel memory mode is supported as well.and in the photo of the bios shows "Branch mode" options (Single channel). [/edit] |
|
863)
Message boards :
Number crunching :
Why am I a volunteer tester?
(Message 717593)
Posted 24 Feb 2008 by JDWhale
Post:
I have never run SETI Beta. [edit] at least not consciously ;-)[/edit] Is there another way I could have become a tester? |
|
864)
Message boards :
Number crunching :
Specs for Core2 Extreme QX9775
(Message 717588)
Posted 24 Feb 2008 by JDWhale
Post: Not so sure that you guys have read the stories correctly. It appears that the Skulltrail does have 4 channel memory just like the MacPro. As I read the story the complaint is the need for 4 sticks to take advantage of that. Most of the overclock tests were only done with 2 sticks, thus the memory bottleneck. I disagree, you better read the review again. The testbed states they have 4x1GB sticks of 800Mhz FB-DIMM occupying the 4 slots, the BIOS display shows each of 2 channels with 2GB each. The other channels displayed in the BIOS as "Not Installed" because they are not on the motherboard. All 4 slots are occupied in the MOBO photos. Memory performance would have been increased (nearly doubled) if they would have had one slot per channel rather than 2 slots per channel(i think), or 8 memory slots like the similar DELL or MAC PRO systems. [edit] Other boards based on this chipset might have 4 channels, but I fear that the D5400XS, the board teamed with dual QX9775 CPU's to make Skulltrail, does not! [/edit] |
|
865)
Message boards :
Number crunching :
Performance of MAC running Windows app via VMWare
(Message 717538)
Posted 23 Feb 2008 by JDWhale
Post: JD your right, the ability of Windows to only work in Dual channel is the real problem. Surely Windows can use Quad Channel, Dell sells Windows based T7400 workstation using the e54xx & x54xx chips with Quad Channel (12.8GB/sec). It might be just a matter of finding and installing the drivers. |
|
866)
Message boards :
Number crunching :
Specs for Core2 Extreme QX9775
(Message 717535)
Posted 23 Feb 2008 by JDWhale
Post: I'm not sure if the X5400 series board is able to use both, it's just only set for DDR2 at the minute. I think a firmware/bios update from intel will allow it to take DDR3. I guess it would have added to the confusion if the D5400XS used different memory (DDR3 -vs- FB-DIMM) than the other D5000 boards. So instead, our friends at Intel crippled the board by only using Dual Channel FB-DIMMS. And in all the hyperbole (hyper bull) you know Who? was spreading, this little factoid was conveniently quashed. 'nuff said... |
|
867)
Message boards :
Number crunching :
Specs for Core2 Extreme QX9775
(Message 717504)
Posted 23 Feb 2008 by JDWhale
Post: Sad because the Mac Pro have exactly the same board. I think it's pretty poor of Intel not to give access to Quad Channel memory on there "BEST" system, yet us mac users get it on standard chipped systems. If they just would have went with DDR3-1600 the bandwith would be closer to 12GB/sec Dual Channel rather than the paltry 7GB/sec that it's getting now. When you are spending that kind of cash for MOBO and CPUS, what is another $300 for 2GB of DDR3 RAM? I like in the review where they say that it's more of a "game developer's" system than a "gamers system". LOL |
|
868)
Message boards :
Number crunching :
Specs for Core2 Extreme QX9775
(Message 717473)
Posted 23 Feb 2008 by JDWhale
Post: The picture is coming together for me now... After reading a very good review of SkullTrail at anandtech.com, the dual QX9775 crunching mystery has been solved. The main bottleneck is the D5400XS motherboard only having Dual Channel FB-DIMM memory system -vs- MAC PRO uses Quad Channel. Half the bandwith of the MAC PRO. The box is doomed to perform less than DDR2 Dual Channel systems because of increased latency of the FB-DIMMs. Bad news when high bandwith is required. Final words on page 16 sums it up: "Skulltrail is closer to perfection than Intel's V8 platform ever was, but it's still not quite there. Let us use desktop memory and Intel will have a winner on its hands." I can now sleep at nights. |
|
869)
Message boards :
Number crunching :
Performance of MAC running Windows app via VMWare
(Message 717425)
Posted 23 Feb 2008 by JDWhale
Post: JD, if you run XP through Bootcamp on the Mac, it is running natively, so it won't perform any better than the same configuration PC box. Unless Apple have gone through the drivers etc and cut everything out that wasn't needed. I'm just looking over the results being posted by UL1's X5365 box now [OT] short night after too much bourbon, my head hurts[/OT], what is baffling me is the poorer than I'd expect performance on the shorties. Looking at this wuid=226606715 where his box is paired with a Q6600 that is seems to be running at stock 2400Mhz. The X5365 has 50% longer CPU time!!! I fear that the MAC is being memory choked, just like the Skulltrail(SKT). I don't think that WindowsXP has the proper drivers to use the quad channel memory system to its potential. I can't help but wonder if this isn't the same problem that is holding SKT back, both in crunching and in SiSandra results memory bandwith benchmark. Something is wrong with the Intel 5xxx chipset running Windows!!! |
|
870)
Message boards :
Number crunching :
Performance of MAC running Windows app via VMWare
(Message 717330)
Posted 23 Feb 2008 by JDWhale
Post: Started a new experiment yesterday: I've installed Win XP Pro x64 on my 3.0 OctaCore and Seti is running now with high priority... Performance looks dismal... about 2 times CPU time for the shorties... 800 sec on OS X vs 1500+ secon WinXP... I roughly guestimate you will top out ~6500 RAC on that XP box... Possible still enough RAC to make you the top Windows box with 8 cores maybe in the same ballpark as SkullTrail. Is this due to IO bottleneck? I don't think so, the other WU's also underperform OS X. Why won't Windows perform on the same hardware? It might be something in the Windows Core I/O that doesn't allow the performance. I'd be interested in the results of "Everest Benchmarks - Read" or SiSoft Sandra "Memory Bandwith" with similar benchmarks on MAC OSX, if available. Why can't the same hardware perform similar CPU times on the two different OS platforms? Does the difference lie in the OS? Does Windows limit performance somehow? If so, and WinXP can not even come close to competing on identical hardware, then it only demonstrates that Windows hinders performance. Maybe Windows is lacking the proper drivers to allow the hardware to perform near potential. I've heard rumors that newer drivers must be "slipstreamed" to the XP distribution to raise performance. But I've not followed up since I do not have the hardware. That must be a chore. Maybe the differnece lies with the more efficient MAC "Alex Kan" code. Call me ignorant but with 20+ years as a coding professional, I need to see similar benchmarks before I can dismiss this discrepancy to application code! I don't know... but someone at Intel and Microsoft does, and it's probably a closely guarded secret. UL1, thank you for trying my proposed experiment... I would understand if you abandon this as a failed attempt whenever you see fit. Dang it... I'm curious enough that I want to find room in my budget to purchase the hardware just to try to explain the differences. But honestly, I can not justify the expense. Cheers, John |
|
871)
Message boards :
Number crunching :
Crunching order
(Message 717065)
Posted 22 Feb 2008 by JDWhale
Post:
While I'm not even a honcho, I figure that since "pending work units" require resources amounting to disk space on the project servers, they can not be deleted until completed and validated. The project wants to clear out the earliest work units sent when possible. Thus your computer runs the earliest work units that you have been issued whenever possible. |
|
872)
Message boards :
Number crunching :
Crunching order
(Message 717059)
Posted 22 Feb 2008 by JDWhale
Post: I have heard explanations in the past for this question, but I'm just not satisfied with the answers that I have gotten, so I'm asking it again. Are you familiar with FIFO (First In-First Out). The first WU that is downloaded to your computer is the first one processed. The exception to this rule is when WU's in your queue might miss their deadline, then they scheduled to run at High-Priority, superceding the FIFO rule. Happy Crunching [edit] I think this the same as what Ned just said. [/edit] |
|
873)
Message boards :
Number crunching :
Specs for Core2 Extreme QX9775
(Message 717031)
Posted 22 Feb 2008 by JDWhale
Post:
Sorry, answer is no to both questions. Looking at VHAR results for the SKT, where bandwith is really put to the test with 8 cores we see a severe performance penalty when put up against the leading MAC Pros, with CPU times nearly double! Sorry, I can't find WU's crunched by both systems right now, I had a links to at least one a couple days ago, the the WU results have already been deleted. [edit] CPU times of 1.5 times, not nearly double as stated above. [/edit] |
|
874)
Message boards :
Number crunching :
Specs for Core2 Extreme QX9775
(Message 717003)
Posted 22 Feb 2008 by JDWhale
Post:
www.hothardware.com shows SiSandra results SiSandra results. Check the "Memory Bandwidth" test. or checkout TomsHardware review. Both discuss the lack of bandwith hindering chipset performance. |
|
875)
Message boards :
Number crunching :
Specs for Core2 Extreme QX9775
(Message 716977)
Posted 22 Feb 2008 by JDWhale
Post: http://download.intel.com/design/processor/datashts/31912801.pdf "The 1600 MHz Front Side Bus (FSB) is a quad-pumped bus running from a 400 MHz system clock making 12.80 GBytes per second data transfer rates possible." Q: When will we see these data rates? |
|
876)
Message boards :
Number crunching :
Francois, Do You Need Help With Bail?
(Message 716946)
Posted 22 Feb 2008 by JDWhale
Post: So, I know that a lot of journalist went emotional on the FB-DIMMS, proving how little they know about hardware, to design a memory controler able to handles more than 4 threads, it is a huge efforts, that required a lot of smart minds on the problem, in 2008, intel will be releasing something new about this, I will start explaining more soon, but basically, FB-DIMM todays are the top of art of 8 cores memory controler, it even beats a machine that AMD has with 16 Cores, and beats world records every day! So what you are saying is that the current SKT memory subsystem is not yet able to fully utilize performance potential of FB-DIMM. Does this explain the 7GB/sec ceiling on memory reads? Maybe we can expect this figure to nearly double, that the true and all powerful SKT is only a few (2-10) months away? |
|
877)
Message boards :
Number crunching :
Francois, Do You Need Help With Bail?
(Message 716808)
Posted 22 Feb 2008 by JDWhale
Post: Nothing personal! this is just DATA from the TOP 20 list of SETI ... Is the SKT host that you've been running here at maximum memory performance, or have you been holding back, or is the motherboard that you have presented to us one of those manufacture defects. Why does your SKT have bandwith problems that will not allow it to achieve greater than 7GB/sec? That is not enough bandwith to feed 8 cores. Yes, the processors are fast and you can overclock all you want, but the problem is that you have not demonstrated that you can feed the beast and the processors are being starved. My DDR2 PC6400 gets 8 GB/sec and even that is not enough to fully feed my single little old Q6600. How can it be enough bandwith to feed 8 cores that are running faster? Please help me to understand, I beg you. Is it possible that you forgot to flip the switch and open up the second memory channel. |
|
878)
Message boards :
Number crunching :
Memory Speed?
(Message 716760)
Posted 22 Feb 2008 by JDWhale
Post:
Seems like we're thinking alike, that's the same heatsink I used. In case you haven't seen it Anandtech review was quite good. I used the same Scythe fan they used in the review, nice and quiet and plenty of airflow. Be sure your case have 7" clearance from top of motherboard (fits perfect in my Antec P180). I'm not crazy about the wire brackets to hold the fan in place. Though they do seem to work, I'd feel more secure with screws or a plastic clip. The HS that I bought had the modified mounting bracket, so that resolves the issue in the review about need to "bend the Socket 775 adapter". And you probably already know that you'll have to pull the motherboard to mount the backplate. Good luck |
|
879)
Message boards :
Number crunching :
Memory Speed?
(Message 716691)
Posted 22 Feb 2008 by JDWhale
Post: It does support 1066, but the price its a little out of reach at the moment. The PC gets used for other things, including Video editing, so thats why I was thinking of the 4GB. With 2GB it is ok for editing, but the extra 2GB might help a little with speeds when video editing etc. Sorry, I'm not familiar with video editing and it's requirements. I guess the first thing I'd checkout would be if your system was running low on memory and starting to hit the swap file when you are running your program. If not, then adding more memory probably will not help much, but replacing your memory with faster memory can sometimes make a huge difference. Hardware prices vary so much regionally also. On my current DS3L board, I'm running Kingston ValueRam KVR800D2N5K2/2G (2x1GB sticks). The motherboard allows OCing the memory to 1000Mhz at 5-5-5-15 timings. And the memory is cheap! CPU runs at 8x400Mhz for a 3200Mhz OC with a good aftermarket HSF. I'm assuming that when doing your editing that you're shutting down BOINC to free up it's memory, granted that it's only about 200MB, but every little bit helps. Good luck, |
|
880)
Message boards :
Number crunching :
Memory Speed?
(Message 716633)
Posted 22 Feb 2008 by JDWhale
Post: I am heading down to Sydney early next week and while I am there I was considering upgrading the RAM on one of the Quad Cores. I am running a Q6600 on a Gigabyte P35-DS3 motherboard If you're doing the upgrade to help S@H crunching, and it's not needed for other reasons, then 2GB total is plenty. Also, that board supports 1066Mhz (PC8500), doesn't it? I'm trying the Crucial Ballistix 2GB (2 x 1GB) 240-Pin DDR2 SDRAM DDR2 1066 (PC2 8500) Dual Channel Kit Desktop Memory in my next build with the P35-DS3L next week. Reports say it'll do 3-3-3-8 1T @ 800Mhz, but I can't confirm yet. Good luck |
|
881)
Message boards :
Number crunching :
V8 Optimized App for Windows and Linux
(Message 716520)
Posted 21 Feb 2008 by JDWhale
Post: It'll just be the machine getting back at me for calling it Ethel. I'll look around tomorrow and see what I can find. Ethel, huh? Wasn't that the term once used for high octane gasoline? Fitting name for a MAC PRO! LOL |
|
882)
Message boards :
Number crunching :
V8 Optimized App for Windows and Linux
(Message 716506)
Posted 21 Feb 2008 by JDWhale
Post: SATAN wrote: JD, wont happen. Hmmm... UL1 had it working for QMC trials, granted not with the newer MAC though. |
|
883)
Message boards :
Number crunching :
What Upgrade?
(Message 716490)
Posted 21 Feb 2008 by JDWhale
Post:
Where I live, MicroCenter is blowing out the Q6600's right now for $199, on special. Very easy 3200(8x400mhz) OC on a Gigabyte D3SL($90). At that price, it is a no-brainer for a relatively low cost crunching machine. |
|
884)
Message boards :
Number crunching :
V8 Optimized App for Windows and Linux
(Message 716488)
Posted 21 Feb 2008 by JDWhale
Post:
Me thinks that you will have the fastest XP box around. Hopefully this will be after you get all 8 banks of memory populated ;-) Can't wait to see this! |
|
885)
Message boards :
Number crunching :
Performance of MAC running Windows app via VMWare
(Message 715979)
Posted 20 Feb 2008 by JDWhale
Post: UL1 wrote:
Thanks for the heads up. It looks like Bootcamp XP is the way to go for my test, eliminating the VMWare overhead. This should allow for as close to apples-to-apples comparison of optimized applications bewteen MAC OSX and MAC WindowsXP. I'll look around and see if I can find a fully blown MAC PRO (Dual X5482's and 8x2GB) to run the tests. Come to think of it, I'd better get 2 systems so I can run the test simultaneously ;-) It's not going to happen!!! At least not in my BOINCing room. My guess is the MAC OSX box will get to about ~13000 RAC given enough time, and the MAC WindowsXP box would definitely be the fastest Windows cruncher on S@H and be in the same ballpark. This is a shootout that I can only dream about. Happy Crunching, John |
|
886)
Message boards :
Number crunching :
Performance of MAC running Windows app via VMWare
(Message 715699)
Posted 20 Feb 2008 by JDWhale
Post:
I agree. But the question is: Will the MAC PRO running WindowsXP on OSX be able to crunch faster than similar hardware running WindowsXP native? And similarly: Can the MAC optimized application run faster on WindowsXP running OSX than the Windows optimized application native? |
|
887)
Message boards :
Number crunching :
Performance of MAC running Windows app via VMWare
(Message 715692)
Posted 20 Feb 2008 by JDWhale
Post: Don't know how that would go, but I've been thinking about trying it the other way 'round. Running Mac OSX on my Windows PC through VMWare (OSX86Poroject Howto), [ With a newer CPU runnin XP 32 bit I get smaller than I exepcted CPU overhead running other VMs for Linux(Ubuntu) and Vista ] Yes, that is the the other way I'd be interrested in. Running the optimized MAC application on that configuration (OSX on Windows). If the assessments that the MAC application is so much faster than the optimized Windows app, would the increased performance be enough to offset the VMWare overhead? Please keep us posted if you decide to give it a go. |
|
888)
Message boards :
Number crunching :
Performance of MAC running Windows app via VMWare
(Message 715687)
Posted 20 Feb 2008 by JDWhale
Post: I had a MacPro running Windows XP Pro via Bootcamp and crunching QMC. When I stopped this experiment after two weeks it had reached an RAC of about 7.000 and was still rising... Just looking at the QMC Top Hosts, RAC of 7,000 would put you firmly in 2nd place behind 16 cores of Intel X7350. Am I reading the numbers right? I am assuming you were running 8 cores of either X5482 or X5365. Either way, I would think that RAC of 875/core is quite impressive. Of course that just might be my ignorance showing as I know absolutely nothing about QMC. |
|
889)
Message boards :
Number crunching :
Performance of MAC running Windows app via VMWare
(Message 715483)
Posted 20 Feb 2008 by JDWhale
Post: Just a thought, I don't know if it's even possible and I realize that there would be some overhead. What kind of performance could I expect from the optimized application running in WindowsXP via VMWare? Wouldn't it be a hoot if the MAC could run Windows version of Seti@Home faster that a similar configured native Windows box. Of course I'd want to run the test the other way as well and install Darwin on my native Windows box to see how the optimized MAC application performed there. I first used VMWare, at home, about 10 years ago and as recently as 5 years ago at work, the performance was painful to run Windows 2000 on my Dell Precision workstation running RedHat native. Is the VMWare performance any better these days? If I had the cash to throw down on a MAC, I'd be running the test myself, if possible! -John |
|
890)
Message boards :
Number crunching :
Win a Skulltrail at Hothardware.com
(Message 715446)
Posted 20 Feb 2008 by JDWhale
Post:
Back in the early days, maybe before what is now called "SETI Classic", US Robotics sponsored a sweepstakes giving away a 56K modem each day for 56 days. I entered every day... and finally won on the 55th day. You can't win if you don't play. This box might not be able to keep with the latest MAC Pros, but I'd like to go for a ride anyway. |
|
891)
Message boards :
Number crunching :
The cache & downloading WUs?
(Message 715264)
Posted 19 Feb 2008 by JDWhale
Post:
So many ways to control the cache, so many different behaviors. I've gone with setting "local preferences" to: Connect about every 0.0 days Additional work buffer 5.0 days Setting connect interval to 5 days causes download havoc with the flood of "shorties" coming down the pipe due to the short deadlines. YMMV |
|
892)
Message boards :
Number crunching :
STD_ERR return
(Message 714875)
Posted 18 Feb 2008 by JDWhale
Post: Do you mean a new version of BOINC or a new version of SETI@home. You have the current version of BONC 5.10.30 It's time to update to the newest Seti@Home application. Good luck |
|
893)
Message boards :
Number crunching :
Long computation times
(Message 714861)
Posted 18 Feb 2008 by JDWhale
Post: http://setiathome.berkeley.edu/workunit.php?wuid=219150807 You are right, something is affecting your performance. Check the benchmark scores for your computer. Measured floating point speed 1505.29 million ops/sec
Measured integer speed 279.49 million ops/secThese numbers should be much, much higher, especially the integer speed, is only about 5% of what you should be seeing. I noticed that at least 2 of the low performance results were "restarted" during execution. Have you made any changes to your Computing preferences/Processor usage recently, possible allowing it to enter "low power" state by setting: Suspend work if no mouse/keyboard activity in last xx minutes Check that you have this value set to 0 for the profile that you are using. Good luck, John |
|
894)
Message boards :
Number crunching :
Rash of Validate errors
(Message 714660)
Posted 18 Feb 2008 by JDWhale
Post: For the past several upload attempts, I'm typically seeing this: Hi Walt, On Feb 16, I noticed a string of those same errors on my laptop that started about 18:00 UTC on Feb 14. And now that I look at the results list for that system, I see a bunch with "Validate error" from Feb 15-16. I recall noticing a "runaway" workunit, one that had exceeded it's estimated runtime and was still less than 50% complete back on Feb 13 or 14 and had cancelled it. Coincidence? Maybe. Anyway, the problems went away after restarting BOINC and I'm now getting credit for my completed WU's once again. Yeah:-) Good luck, John |
|
895)
Message boards :
Number crunching :
single or dual channel
(Message 714636)
Posted 18 Feb 2008 by JDWhale
Post: However much RAM you install, try to go with dual channel wherever possible. I agree with OzzFan.. Go with dual channel when possible for maximum speed transfering data between CPU and RAM. Take particular care in the the requirements of dual channel with "matched" memory sticks, memory performance can actually decrease otherwise. Looking at results 749134830 and 749056752 from your Q6700, I can see where your system might be somewhat memory bandwith limited. Both results requested the same credit and have similar flops, yet CPU time varies by ~35%. The slower of these WU's was probably choking trying to read from memory. This was just a quick look at your results, I wouldn't be surprised if some variations were ~50%. This type of behavior can/will occur when 2 or more WU's of "Very High true_angle_range"(VHAR) are run concurrently. Your crunching performance will improve on this type of WU's by tweaking your memory performance. Good luck, John |
|
896)
Message boards :
Number crunching :
Client claiming negative credit?
(Message 714481)
Posted 17 Feb 2008 by JDWhale
Post:
I have a bit of an issue with the "unused cycles" argument when it pertains to newer "speedstep" processors. If I wasn't running S@H, then my CPU would throttle back to 6x multiplier and power comsumption would decrease considerably, rather S@H keeps the CPU with the pedal to the metal and causes the processor to create more cycles so it can run. I'm a dedicated cruncher, so throttling back isn't something that I'd want to do, but S@H uses more than "unused cycles" these days. Hmmm.. Like Bob Dylan once wrote "For the times they are a-changin" and so is computing technology. -John [Edit] BTW... Black Sabbath Volume4 was probably my favorite album back in high school. GO Ozzy!!! |
|
897)
Message boards :
Number crunching :
Motherboard for q6600
(Message 714360)
Posted 17 Feb 2008 by JDWhale
Post: I have been asked for advice,but I thought it would be better to open it up to the forum. I've got my Q6600 on a... Gigabyte GA-P35-DS3L [$90+SH] with easy 3200 OC [8x400MHz] Kingston ValueRAM KVR800D2N5K2/2G PC2-6400 CL5 runs at 1000MHz [2x500MHz] Good luck, John |
|
898)
Message boards :
Number crunching :
Client claiming negative credit?
(Message 714354)
Posted 17 Feb 2008 by JDWhale
Post: Maybe slightly off topic, but I'll give it a try anyway... The last time I checked, credit is useless. That is, until the credit can be used for something. Granted, bragging rights have some value if you need an ego boost, but that is about it! I started running the Seti@Home project for the science back on the first day that the download was made public and have crunched 24/7 since on a variety of hardware. I've retired the old Pentium 60, PII's, PIII's, and SGI Power Challenges, and am in the process of retiring the P4's in favor of energy efficient Core2's. Rather than focusing on credits granted and older BOINC versions, wouldn't it be more appropriate to limit hosts to those that are more energy efficient. Try to raise the bar on the science performed per KWh. BOINC is going GREEN, I fear NOT!!! When you consider the energy consumed by those inefficient hosts whose sole reason for still being powered up is for BOINCing, it's quite wasteful. Please don't get me wrong, Pentium 60's and PII's can and still do valuable science for the project. Just at too a high of a price. Let the BOINC projects take some responsibility for the BOINC footprint on the global environment. Regards, John |
|
899)
Message boards :
Number crunching :
Client claiming negative credit?
(Message 714123)
Posted 17 Feb 2008 by JDWhale
Post: There's a pair of settings which can be put into the project configuration to warn users that a change in minimum allowed core client is coming. One setting says what the minimum will be, and any hosts which are running a lower version will get a message in each Scheduler reply saying: Mark... It my understanding that there will not be any "strange message popping up" on anybodys computer, rather a subtle message in the log, viewable under the "Messages" tab of the Boinc Manager, and probably unnoticed by the most of those receiving said message. (Joe please correct me if I am wrong). Joe... What happens when the elapsed time passes and the "minimum version" is enforced? Are WU's blocked from being sent to clients not meeting the minimum BOINC version criteria? Thus not "wasting" cycles on a host working on a result that would ignored upon completion anyway. It is my opinion that this is the most elegant and non-intrusive solution. My main concern is that of impact to the projects data throughput. Can we quantify the amount of work that would be lost by those exiled (for lack of a better term) hosts? Regards, John |
|
900)
Message boards :
Number crunching :
Client claiming negative credit?
(Message 713366)
Posted 16 Feb 2008 by JDWhale
Post: Bottom line... As long as the science is valid, the credit granted is irrelevent! And the door shuts. |
|
901)
Message boards :
Number crunching :
Client claiming negative credit?
(Message 713352)
Posted 16 Feb 2008 by JDWhale
Post:
From your earlier post... In other words, you get what you ask for, unless it is higher than what is currently correct, and if you claim a lower amount than what is currently correct, that is what you get, and the wingman is not punished. So if I'm reading your earlier post in this thread correctly, If you ask for 0.0, then you get 0.0, even though the "currently correct" credit might be 50.14 and your result is used to validate your wingman who gets the 50.14 because that is what he claimed. What happens in the case that user one claims 25, wingman claims 102, and the currently correct credit is 63.98. Both original users validate without issue, but both have in correct credit claims? Did I miss something? |
|
902)
Message boards :
Number crunching :
Client claiming negative credit?
(Message 713328)
Posted 16 Feb 2008 by JDWhale
Post:
I have issue with the proposition of granting differing credits... seems a bit dishonest, like taking advantage of a cashiers mistake when they give you too much change. If there are really that few of the 4.xx BOINCers taking part, just eliminate them to appease the masses. |
|
903)
Message boards :
Number crunching :
Client claiming negative credit?
(Message 713305)
Posted 16 Feb 2008 by JDWhale
Post:
I've a feeling than in addition to those who set-it/forget-it, there are those mom & pop types that are only running SETI because junior installed it while at home on spring break, or something similar. Years pass and never an update to BOINC, though SETI updates the crunching application automatically as new versions become available. What if SETI were to install a "do nothing" application, like a detatch from project, to those volunteers systems that haven't a minimal version of BOINC installed. Something along these lines would attack the problem at the source, WU's never get DL'd to those out of date systems so there isn't an issue with credit handling after processing. Seems to me that since the mechanism is already in place to update the application as needed, the framework would already be in place to implement this approach. Of course, we'll be losing production capacity in a time that the project is in need of more crunching power, not less. I wonder what the impact to the project would be. Now I'll run and duck for cover;-) |
|
904)
Message boards :
Number crunching :
Your SETI farm power consumption?
(Message 712969)
Posted 15 Feb 2008 by JDWhale
Post: Honey, I had to upgrade to the Core2 Quad to help save the enviromment. And with it drawing 150W less than our current system, will pay for itself in electricity savings alone in just over a year. Yeah, like that is going to work.... |
|
905)
Message boards :
Number crunching :
Best way to wind down a machine?
(Message 712807)
Posted 15 Feb 2008 by JDWhale
Post: One of the machines I am running here has got a rumbling hard drive. I don't think it is in immediate danger of failing, provided I don't power it down again, but I want to run the queue down to virtually no reserve so when (if) it does fail I don't lose too much work for the project and cause grief to whoever the unlucky wingman is. Select "No new tasks" button from "Projects" tab of Boinc Manager. Do an "Update" when all tasks have finished. |
|
906)
Message boards :
Number crunching :
Client claiming negative credit?
(Message 712792)
Posted 15 Feb 2008 by JDWhale
Post:
I don't get it... How can adjusting the clock cause the CPU time to go negative? Maybe it's just ET showing off. |
|
907)
Message boards :
Number crunching :
Client claiming negative credit?
(Message 712755)
Posted 15 Feb 2008 by JDWhale
Post:
Hmmm... So in this case, the offender would be granted -38 credits. That could create a whole new race, see who can get the most negative RAC. Could make things more interesting. LOL |
|
908)
Message boards :
Number crunching :
Sirius Club
(Message 712635)
Posted 14 Feb 2008 by JDWhale
Post: So I guess that I'm about just about 1/4 light-year beyond Sirius at this time. So that's Bob I see in my rearview ;-) |
|
909)
Message boards :
Number crunching :
Your SETI farm power consumption?
(Message 712546)
Posted 14 Feb 2008 by JDWhale
Post: This just leaves me with this thought, "Why the h... does my machines use so much juice"? Well, back to the drawing board :) Hey Chris, Going back to the link that Jason provided..fairly comprehensive list of graphics cards power ratings There's a link on that page for "eXtreme PSU Calculator". A tool where you put all the specifics of your configuration and it tells you your total draw. That might help a bit in determining what's sucking up all the "juice". |
|
910)
Message boards :
Number crunching :
Did everyone go on Vacation?
(Message 712339)
Posted 14 Feb 2008 by JDWhale
Post: I did manage to launch a Q6600 @ 2.6GHz(I'd tried as high as 2.86GHz[not very stable or as quick as 2.6GHz-ram setting problem, overheating? don't know really], Wish I'd bought a more modern Intel motherboard instead of the AW9D-MAX as someone somewhere said to get a better NB heatsink and My problems would be over, wrong. I'm looking at getting a DDR2 compatible P35(Asus or Gigabyte) or an Asus X38(Also DDR2) of some sort for as little cost as I can manage as a replacement. Dang Joker, someone had a Q6600/GA-P35-DS3L combo for sale last week at a great price ($350). Instant 3.2Ghz OC w/DDR2 PC6400 OC @500Mhz. El Cheapo screamer!!! Can I say "Back up the truck"??? |
|
911)
Message boards :
Number crunching :
Your SETI farm power consumption?
(Message 712286)
Posted 14 Feb 2008 by JDWhale
Post: I was also confused about graphics cards power consumption/ heat myself too, so I went searching and found this: Great link!!! National retailer F**'s is giving away the EVGA e-GeForce 6200 LE (after $50 Mail-in-Rebate) through 14 Feb (hope that wasn't a commercial advertisement ;-) 18W max!!! Just the card I need for my next OC'd Q6600!!! I'm just angry that I need a video card at all when I don't even have a monitor attached! |
|
912)
Message boards :
Number crunching :
Questions about crunch times
(Message 712124)
Posted 13 Feb 2008 by JDWhale
Post: Can someone please tell me how "anonymous" the #1 person on the top computers lists has the same exact computer as I do and seems to run the same v8 version of an optimized client and it takes him 1000 seconds or less for each work unit and it takes my computer on average 2700 seconds. That is a massive difference for the same setup. Hi Eric, Looking at history of your computer and #1 person computer, I see that you are crunching comparable WU's faster that #1. #1 history your history Just keep your system up and running steady and your RAC will continue to grow as pending WU's come to fruition. Good luck, |
|
913)
Message boards :
Number crunching :
c2 vs quad
(Message 711729)
Posted 12 Feb 2008 by JDWhale
Post: hyper pi one core=15.110 four cores=23.777, quite a difference what does it mean? I'm only using one stick of ram right now I think that you are memory bandwith limited and should evaluate again after replacing the 2nd matched memory stick to your system. That's assuming that your MB is Dual Channel capable and the memory sticks must be matched or performance could suffer even more. Even then I'd try to limit running to no more than 2 VHAR WU's concurrently. Seti seems to run much better with a mix of AR's. Good luck, |
|
914)
Message boards :
Number crunching :
How much memory?
(Message 711682)
Posted 12 Feb 2008 by JDWhale
Post: Another thought I had was to share a power supply between 2 crunchers with onboard video. Minimal cost, heat, and maximum throughput. I agree, lots of variables. Just wondering how much power is actually needed with a mATX board with onboard graphics & NIC, E4500 CPU w/HSF, 512MB, onboard NIC & USB. No O/Cing. No keyboard, mouse, hard drive or case fans needed. Hard to imagine drawing more than 200W. Like I said, just a thought. |
|
915)
Message boards :
Number crunching :
How much memory?
(Message 711661)
Posted 12 Feb 2008 by JDWhale
Post: I'd suggest no more than 256MB per processor core, though 1GB sticks are the current sweet spot in pricing. And performance will increase with 2 sticks running dual channel. A couple 256MB sticks on BoincPE would be plenty with the reduced minimal Windows memory footprint. I can't speak to the licensing legality as I only have one box running BoincPE from a legitimate XP license. Another thought I had was to share a power supply between 2 crunchers with onboard video. Minimal cost, heat, and maximum throughput. Good luck, |
|
916)
Message boards :
Number crunching :
How much memory?
(Message 711633)
Posted 12 Feb 2008 by JDWhale
Post: I'd suggest no more than 256MB per processor core, though 1GB sticks are the current sweet spot in pricing. And performance will increase with 2 sticks running dual channel. For additional savings on a truly dedicated cruncher, both initial cost and recurring electricity consumption, you might consider BoincPE running from a cheap USB stick (no hard drive to purchase or spin). Though setting it up is not for the faint at heart. Good luck, |
|
917)
Message boards :
Number crunching :
c2 vs quad
(Message 711574)
Posted 12 Feb 2008 by JDWhale
Post: Possibly suffering from memory bandwith shortage... as these VHAR WUs tend to preempt other WUs, it's not uncommon for 4 to run at a time and saturate the FSB. I don't see any evidence of other AR WUs in your history suffering such a performance issue, there are even a couple of instances of VHAR WUs that seemed to have crunched in a mix with other ARs and performed admirably. You might try running Hyper_PI and compare performance of single core vs 4 cores, 20% increase in runtimes are acceptable, 100% increase is not! I had similar issues with an economy ECS motherboard and an E4500 chip, the dual channel performance was very poor with only about 10% boost over single channel performance. Good luck, |
|
918)
Message boards :
Number crunching :
Happy Old Year Graphs
(Message 696898)
Posted 3 Jan 2008 by JDWhale
Post: Well spank my a$$ and call me Sally. I love your pizza and beer inspired graphs Scarecrow. Maybe you or someone else can provide insight as why the 38 year spike in Transitioner Backlog cleared up within a day or two. Cheers, |
|
919)
Message boards :
Technical News :
Hooray? (Aug 08 2007)
(Message 616701)
Posted 9 Aug 2007 by JDWhale
Post: Or "Jim Beam" 'll do in a pinch !!! > Great news bring on the beam (multi beam that is) |
time
Uli. I'm feeling quite chuffed with myself actually. The Team Stats page took me about a month to build; in between my travels, guests, family celebrations and hang-overs! 
Previous 20
©2024 University of California
SETI@home and Astropulse are funded by grants from the National Science Foundation, NASA, and donations from SETI@home volunteers. AstroPulse is funded in part by the NSF through grant AST-0307956.