Don't know where it should go? Stick it here!

Message boards : Number crunching : Don't know where it should go? Stick it here!
Message board moderation

To post messages, you must log in.

Previous · 1 . . . 6 · 7 · 8 · 9 · 10 · 11 · 12 . . . 146 · Next

AuthorMessage
Profile petri33
Volunteer tester

Send message
Joined: 6 Jun 02
Posts: 1668
Credit: 623,086,772
RAC: 156
Finland
Message 1951788 - Posted: 24 Aug 2018, 12:27:55 UTC

 
State: All (18606) · In progress (1700) · Validation pending (9396) · Validation inconclusive (666) · Valid (6644) · Invalid (11) · Error (189) 


I know my machine is scaring fast and sometimes when I t est a debug build it creates some inconclusive results.
Should I be worried?
It is going down.
To overcome Heisenbergs:
"You can't always get what you want / but if you try sometimes you just might find / you get what you need." -- Rolling Stones
ID: 1951788 · Report as offensive     Reply Quote
Profile Zalster Special Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 27 May 99
Posts: 5517
Credit: 528,817,460
RAC: 242
United States
Message 1951815 - Posted: 24 Aug 2018, 15:35:12 UTC - in response to Message 1951788.  

Only if you start to hear this

https://youtu.be/0RNRVPIJxJk
ID: 1951815 · Report as offensive     Reply Quote
Profile Keith Myers Special Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 29 Apr 01
Posts: 13161
Credit: 1,160,866,277
RAC: 1,873
United States
Message 1951821 - Posted: 24 Aug 2018, 15:49:56 UTC - in response to Message 1951788.  

Seems to be about double the inconclusive rate for the 0.97 app from normal.
Seti@Home classic workunits:20,676 CPU time:74,226 hours

A proud member of the OFA (Old Farts Association)
ID: 1951821 · Report as offensive     Reply Quote
Profile Zalster Special Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 27 May 99
Posts: 5517
Credit: 528,817,460
RAC: 242
United States
Message 1951827 - Posted: 24 Aug 2018, 16:09:55 UTC - in response to Message 1951821.  

Seems to be about double the inconclusive rate for the 0.97 app from normal.


look at the number again and reference my link lol
ID: 1951827 · Report as offensive     Reply Quote
Profile Tom M
Volunteer tester

Send message
Joined: 28 Nov 02
Posts: 5124
Credit: 276,046,078
RAC: 462
Message 1951835 - Posted: 24 Aug 2018, 16:35:43 UTC

I wonder how well the ASUS 19 gpu Mining MB https://www.asus.com/us/Motherboards/B250-MINING-EXPERT/ would run on Seti.

In theory it "oughta" do very well. Has anyone tried it?

Tom
A proud member of the OFA (Old Farts Association).
ID: 1951835 · Report as offensive     Reply Quote
Profile Siran d'Vel'nahr
Volunteer tester
Avatar

Send message
Joined: 23 May 99
Posts: 7379
Credit: 44,181,323
RAC: 238
United States
Message 1951849 - Posted: 24 Aug 2018, 17:08:53 UTC - in response to Message 1951815.  

Only if you start to hear this

https://youtu.be/0RNRVPIJxJk

I really like listening to Motley Crue's songs, back in the 80's. I'm just glad I never went to a concert. They look dressed for Halloween. lol ;)

Siran
CAPT Siran d'Vel'nahr - L L & P _\\//
Winders 11 OS? "What a piece of junk!" - L. Skywalker
"Logic is the cement of our civilization with which we ascend from chaos using reason as our guide." - T'Plana-hath
ID: 1951849 · Report as offensive     Reply Quote
Phil Burden

Send message
Joined: 26 Oct 00
Posts: 264
Credit: 22,303,899
RAC: 0
United Kingdom
Message 1951853 - Posted: 24 Aug 2018, 17:33:20 UTC - in response to Message 1951835.  

I wonder how well the ASUS 19 gpu Mining MB https://www.asus.com/us/Motherboards/B250-MINING-EXPERT/ would run on Seti.

In theory it "oughta" do very well. Has anyone tried it?

Tom


It certainly isn't a cheap Mobo, Scan in the UK have for almost £200 All the slots bar 1 are PCIE-x1, rather than x16, isn't that golng to make a big difference?

P.
ID: 1951853 · Report as offensive     Reply Quote
Profile Keith Myers Special Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 29 Apr 01
Posts: 13161
Credit: 1,160,866,277
RAC: 1,873
United States
Message 1951855 - Posted: 24 Aug 2018, 17:39:17 UTC - in response to Message 1951853.  

It certainly isn't a cheap Mobo, Scan in the UK have for almost £200 All the slots bar 1 are PCIE-x1, rather than x16, isn't that golng to make a big difference?

Not for Seti, which has proven that a simple X1 slot has sufficient bandwidth to feed data to a gpu. Not the case with other projects like GPUGrid.net and Einstein. Not recommended to run anything but the highest bandwidth slots for those projects.
Seti@Home classic workunits:20,676 CPU time:74,226 hours

A proud member of the OFA (Old Farts Association)
ID: 1951855 · Report as offensive     Reply Quote
Profile petri33
Volunteer tester

Send message
Joined: 6 Jun 02
Posts: 1668
Credit: 623,086,772
RAC: 156
Finland
Message 1951883 - Posted: 24 Aug 2018, 21:31:03 UTC - in response to Message 1951821.  

Seems to be about double the inconclusive rate for the 0.97 app from normal.


There is nothing normal. Everything is inconclusive.
Like a superposition of zeroes and ones. Inconclusive ...
By the numbers now and then (666) it seems to be going D O W N. :)
To overcome Heisenbergs:
"You can't always get what you want / but if you try sometimes you just might find / you get what you need." -- Rolling Stones
ID: 1951883 · Report as offensive     Reply Quote
Profile Keith Myers Special Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 29 Apr 01
Posts: 13161
Credit: 1,160,866,277
RAC: 1,873
United States
Message 1951890 - Posted: 24 Aug 2018, 21:46:58 UTC - in response to Message 1951883.  

I have been noticing my inconclusives are the same as yours. Off by one pulse count compared to wingman. Don't know if it is the particular AR of those BLC06 tasks that the app has trouble with or not. Maybe the noise spectrum is different enough for the algorithm to have troubles counting.
Seti@Home classic workunits:20,676 CPU time:74,226 hours

A proud member of the OFA (Old Farts Association)
ID: 1951890 · Report as offensive     Reply Quote
Profile petri33
Volunteer tester

Send message
Joined: 6 Jun 02
Posts: 1668
Credit: 623,086,772
RAC: 156
Finland
Message 1951893 - Posted: 24 Aug 2018, 22:10:24 UTC - in response to Message 1951890.  

I have been noticing my inconclusives are the same as yours. Off by one pulse count compared to wingman. Don't know if it is the particular AR of those BLC06 tasks that the app has trouble with or not. Maybe the noise spectrum is different enough for the algorithm to have troubles counting.


Hi,
Would it be possible to grab some of those WUs to the edge of the hard drive and then later this weekend send them to me?
Especially if my inconclusives are from my latest test V0.97b1.

Petri
To overcome Heisenbergs:
"You can't always get what you want / but if you try sometimes you just might find / you get what you need." -- Rolling Stones
ID: 1951893 · Report as offensive     Reply Quote
Grant (SSSF)
Volunteer tester

Send message
Joined: 19 Aug 99
Posts: 13722
Credit: 208,696,464
RAC: 304
Australia
Message 1951895 - Posted: 24 Aug 2018, 22:15:04 UTC

As easy as it is to connect to a network these days, either wired or wireless, Thumb drives have been the best way of quickly transferring data between non networked devices. Luckily portable SSDs are now affordable, and with USB3x you can shift large amounts of data in a very short period of time. And as more RAM & NAND production comes online over the next 6-12months prices should continue to drop, as performance & capacity improve.
SanDisk's Extreme Portable 1 TB SSD reviewed.
Grant
Darwin NT
ID: 1951895 · Report as offensive     Reply Quote
Profile Keith Myers Special Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 29 Apr 01
Posts: 13161
Credit: 1,160,866,277
RAC: 1,873
United States
Message 1951897 - Posted: 24 Aug 2018, 22:16:40 UTC - in response to Message 1951893.  

I'll see if I can't get some preserved somehow. Either try and run Jeff's Windows app under Wine or fire up the Windows laptop.
Seti@Home classic workunits:20,676 CPU time:74,226 hours

A proud member of the OFA (Old Farts Association)
ID: 1951897 · Report as offensive     Reply Quote
Profile Keith Myers Special Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 29 Apr 01
Posts: 13161
Credit: 1,160,866,277
RAC: 1,873
United States
Message 1951904 - Posted: 24 Aug 2018, 22:32:58 UTC

I've been going through my list looking for good candidates. Even found one with TBar with the 0.97b1 app. The other thing I'm noticing is that A LOT of my inconclusive wingmen are running the older CUDA's, 23,42 and 50 and they are all having a particularly hard time returning ANY valid task on the BLC06 tasks. Almost all are early overflow 30 counts. I can probably discount any of the CUDA23 wingmen, but the 42 and 50 machines theoretically should work.

The cpu apps and SoG apps are the ones with the one pulse count differences.
Seti@Home classic workunits:20,676 CPU time:74,226 hours

A proud member of the OFA (Old Farts Association)
ID: 1951904 · Report as offensive     Reply Quote
Grant (SSSF)
Volunteer tester

Send message
Joined: 19 Aug 99
Posts: 13722
Credit: 208,696,464
RAC: 304
Australia
Message 1951909 - Posted: 24 Aug 2018, 23:01:30 UTC - in response to Message 1951904.  

I've been going through my list looking for good candidates. Even found one with TBar with the 0.97b1 app. The other thing I'm noticing is that A LOT of my inconclusive wingmen are running the older CUDA's, 23,42 and 50 and they are all having a particularly hard time returning ANY valid task on the BLC06 tasks. Almost all are early overflow 30 counts. I can probably discount any of the CUDA23 wingmen, but the 42 and 50 machines theoretically should work.

The cpu apps and SoG apps are the ones with the one pulse count differences.

What result does a CPU application give on those WUs?
Grant
Darwin NT
ID: 1951909 · Report as offensive     Reply Quote
Profile Keith Myers Special Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 29 Apr 01
Posts: 13161
Credit: 1,160,866,277
RAC: 1,873
United States
Message 1951910 - Posted: 24 Aug 2018, 23:11:16 UTC - in response to Message 1951909.  
Last modified: 24 Aug 2018, 23:13:35 UTC

Generally one pulse count more than 0.97.
https://setiathome.berkeley.edu/workunit.php?wuid=3107122320

Same with zi3v
https://setiathome.berkeley.edu/workunit.php?wuid=3108951161
Seti@Home classic workunits:20,676 CPU time:74,226 hours

A proud member of the OFA (Old Farts Association)
ID: 1951910 · Report as offensive     Reply Quote
Grant (SSSF)
Volunteer tester

Send message
Joined: 19 Aug 99
Posts: 13722
Credit: 208,696,464
RAC: 304
Australia
Message 1951915 - Posted: 24 Aug 2018, 23:27:29 UTC - in response to Message 1951910.  

Generally one pulse count more than 0.97.
https://setiathome.berkeley.edu/workunit.php?wuid=3107122320

So the GPU application needs to find/include that extra pulse.

All the various CPU applications agree with each other? So they are the reference everything else needs to match up with.

Same with zi3v
https://setiathome.berkeley.edu/workunit.php?wuid=3108951161

Be nice to have a CPU process the next resend.




It would be nice if the Scheduler could allocate any resends to a CPU.
Hmm. All new work to GPUs, all resends to CPUs- make that all inconclusive resends to CPUs. Time outs & out & out errors & invalids can still be done by GPU.
Grant
Darwin NT
ID: 1951915 · Report as offensive     Reply Quote
Profile Keith Myers Special Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 29 Apr 01
Posts: 13161
Credit: 1,160,866,277
RAC: 1,873
United States
Message 1951917 - Posted: 24 Aug 2018, 23:34:58 UTC - in response to Message 1951915.  

Generally one pulse count more than 0.97.
https://setiathome.berkeley.edu/workunit.php?wuid=3107122320

So the GPU application needs to find/include that extra pulse.

All the various CPU applications agree with each other? So they are the reference everything else needs to match up with.

Same with zi3v
https://setiathome.berkeley.edu/workunit.php?wuid=3108951161

Be nice to have a CPU process the next resend.




It would be nice if the Scheduler could allocate any resends to a CPU.
Hmm. All new work to GPUs, all resends to CPUs- make that all inconclusive resends to CPUs. Time outs & out & out errors & invalids can still be done by GPU.

Yes, that would be the best way to guarantee the validation process with a cpu being the final arbiter. Doubt that ever happens though with the general reluctance to change any of the existing server code.
Seti@Home classic workunits:20,676 CPU time:74,226 hours

A proud member of the OFA (Old Farts Association)
ID: 1951917 · Report as offensive     Reply Quote
Al Crowdfunding Project Donor*Special Project $75 donorSpecial Project $250 donor
Avatar

Send message
Joined: 3 Apr 99
Posts: 1682
Credit: 477,343,364
RAC: 482
United States
Message 1951927 - Posted: 25 Aug 2018, 0:12:57 UTC - in response to Message 1951853.  

I wonder how well the ASUS 19 gpu Mining MB https://www.asus.com/us/Motherboards/B250-MINING-EXPERT/ would run on Seti.

In theory it "oughta" do very well. Has anyone tried it?

Tom


It certainly isn't a cheap Mobo, Scan in the UK have for almost £200 All the slots bar 1 are PCIE-x1, rather than x16, isn't that golng to make a big difference?

P.
I've found that my system with the 6 cards in it, using the 1x USB-PCIe connectors is not an issue for me, the biggest drop dead issue that I ran into was the max supported # of GPU's. I think windows has a hard limit, but obviously with mining, some special OS allows them to blow past that limit. Not sure how it'd be done for BOINC though, as once I had hit that limit, I lost interest in trying to push beyond it, 6 GPU's was fine for me at that point in time. It'd be interesting if someone had the gumption and resources to get it, and see if they could find a way to make it work. Talk about trailblazing! ;-)

ID: 1951927 · Report as offensive     Reply Quote
Profile Brent Norman Crowdfunding Project Donor*Special Project $75 donorSpecial Project $250 donor
Volunteer tester

Send message
Joined: 1 Dec 99
Posts: 2786
Credit: 685,657,289
RAC: 835
Canada
Message 1951934 - Posted: 25 Aug 2018, 1:33:20 UTC - in response to Message 1951927.  

For kicks I priced it out on Newegg.
Board, 8600k/240mm cooler, 32G
18 various EVGA 1080Ti (can't get 18 the same, LOL), plus 1030
3 x EVGA 1600W
APC 5000 UPS

27 item(s): $27,797.79
Warranty & Services: $3,239.00
GST/HST: $1,553.59
PST: $387.53
Shipping: $34.97
Grand Total: $33,012.88 CAD

Add an SSD and 20 risers - $600
Plus 2 sheets of MDF, screws and glue, will that do it Al?

Ohh and the power bill !!!!!
ID: 1951934 · Report as offensive     Reply Quote
Previous · 1 . . . 6 · 7 · 8 · 9 · 10 · 11 · 12 . . . 146 · Next

Message boards : Number crunching : Don't know where it should go? Stick it here!


 
©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.