Message boards :
Number crunching :
Special thanks for AK V8 application
Message board moderation
Previous · 1 · 2 · 3 · 4 · Next
Author | Message |
---|---|
MarkJ Send message Joined: 17 Feb 08 Posts: 1139 Credit: 80,854,192 RAC: 5 |
I too will add my voice to the chorus of Thank You's Thank you guys for all your hard work. Now how do we get Alex to look at optimizing some of the other BOINC projects like Einstein? :-) BOINC blog |
Spear Send message Joined: 15 Nov 01 Posts: 49 Credit: 6,365,604 RAC: 0 |
Use something like CPU-Z to check the older CPU's where it's less certain. And there are x64 version of the V8 app and I'm running them on Vista x64 and Xp64. |
derFunkenstein Send message Joined: 26 Apr 02 Posts: 25 Credit: 1,314,593 RAC: 0 |
yes, thanks to all for a great new worker - the SETI project will really benefit! |
zoom3+1=4 Send message Joined: 30 Nov 03 Posts: 65746 Credit: 55,293,173 RAC: 49 |
I too will add my voice to the chorus of Thank yous Last I looked unless Einstein is open source, He can't, It would be nice though. The T1 Trust, PRR T1 Class 4-4-4-4 #5550, 1 of America's First HST's |
Sniper Send message Joined: 9 Jul 99 Posts: 310 Credit: 2,831,142 RAC: 0 |
WoW!!! My Triple-Core Phenom is processing nearly TWICE as fast. (Will this make ET only half as far away?) :) |
zoom3+1=4 Send message Joined: 30 Nov 03 Posts: 65746 Credit: 55,293,173 RAC: 49 |
WoW!!! My Triple-Core Phenom is processing nearly TWICE as fast. (Will this make ET only half as far away?) :) I doubt It, Space is vast and ET could be anywhere in space. The T1 Trust, PRR T1 Class 4-4-4-4 #5550, 1 of America's First HST's |
AndyW Send message Joined: 23 Oct 02 Posts: 5862 Credit: 10,957,677 RAC: 18 |
|
John Clark Send message Joined: 29 Sep 99 Posts: 16515 Credit: 4,418,829 RAC: 0 |
Just wanted to post my "thanks" to all involved too. Tremendous improvement in crunch times seen here and no other apparent issues. I want to echo Andy's thanks to the port team and to Alex for the code. It's good to be back amongst friends and colleagues |
the silver surfer Send message Joined: 24 Feb 01 Posts: 131 Credit: 3,739,307 RAC: 0 |
A big " THANK YOU " to all the people involved in the porting !!! @ Joker @ Fred The new app works like a charm ! Unfortunately I`m only feeding my PENDING cache at the moment - Therefore my RAC is on the same level as usual, although I spend about 1/3 time less for each wu. But that`s no big deal, I`ll sit it out ! Happy crunchin` Kurt |
[KWSN]John Galt 007 Send message Joined: 9 Nov 99 Posts: 2444 Credit: 25,086,197 RAC: 0 |
A big " THANK YOU " to all the people involved in the porting !!! Pending??? I blew my cache down to .5 days before the switch to V8 and my pendings dropped to 32k. Right now I sit at 60k in pendings, and RAC has increased slightly. Clk2HlpSetiCty:::PayIt4ward |
Carlos Send message Joined: 9 Jun 99 Posts: 29833 Credit: 57,275,487 RAC: 157 |
THANK YOU. Just installed it last night. My cruch time were about 1:20 per wu. Now they are getting done in about 55 mins. Well done guys. |
Sniper Send message Joined: 9 Jul 99 Posts: 310 Credit: 2,831,142 RAC: 0 |
My crunch times for 74 point work units went from over 14k seconds, down to 6.6k seconds. Way to go with that app optimization!!!!!!!! |
Mr. Majestic Send message Joined: 26 Nov 07 Posts: 4752 Credit: 258,845 RAC: 0 |
|
Misfit Send message Joined: 21 Jun 01 Posts: 21804 Credit: 2,815,091 RAC: 0 |
|
archae86 Send message Joined: 31 Aug 99 Posts: 909 Credit: 1,582,816 RAC: 0 |
Fred W said: From the charts I published I was quite confident that the SSSE3x and SSE4.1 Apps were much faster than the SSE3. I have run through over 100 WU's using the SSE4.1 on my Q9450 since yesterday (all in the same AR band) and it looks like there are still a lot in the same AR band coming next, so I have just swapped to the SSSE3x App to do a direct comparison. I can provide additional support for your confidence from a small experiment on two Conroe-class systems. I tried downgrading my Q6600 and E6600 from the SSSE3x to the SSE3 ap versions overnight. By chance the Q6600 has caught a number of VLAR work units at exactly .008341 Angle Range (on the low end of VLAR). The advantage of the SSSE3x version is clear and substantial. SSE3 5 results, average 4798 CPU seconds SSSE3x 17 results, average 4094 CPU seconds The distribution of individuals is very clearly different, not even close to overlapping at this sample size. On the same host, I happen to have caught a single exact Angle Range in mid-AR also, .408612. Here the advantage of SSSE3x was much smaller: SSE3 5 results, average 3248 CPU seconds SSSE3x 11 results, average 3100 CPU seconds The distribution of individuals here is not nearly so impressive as at VLAR, with plenty of overlap, and room for doubt on the large-sample result. In case it matters, this Q6600 is running 25% SETI, with the rest being the current Beta Einstein Windows ap. My E6600 caught smaller comparisons of the same populations, with very similar results (large, unambiguous SSSE3x advantage at VLAR, much smaller, possibly questionable advantage a mid-AR, no results for SSE3 at VHAR on either host). As it also runs 25% SETI, all these ran with a current Windows Beta Einstein on the other core. I'm convinced enough of the SSSE3x advantage on these hosts to switch back, so won't be producing more data on this point. If I understood Jason's comment about stride reduction and memory conflict correctly (for the tweaks bearing the Xeon label), it is likely that were the Q6600 running four SETI's rather than one SETI and three Einsteins that the SSSE3x advantage would be yet larger. |
jason_gee Send message Joined: 24 Nov 06 Posts: 7489 Credit: 91,093,184 RAC: 0 |
caw.... thanks Grace :D *blushes* (Just an Old Fashioned Kindof Guy) "Living by the wisdom of computer science doesn't sound so bad after all. And unlike most advice, it's backed up by proofs." -- Algorithms to live by: The computer science of human decisions. |
[B^S] madmac Send message Joined: 9 Feb 04 Posts: 1175 Credit: 4,754,897 RAC: 0 |
|
kittyman Send message Joined: 9 Jul 00 Posts: 51468 Credit: 1,018,363,574 RAC: 1,004 |
I have installed it did I install the right one as it says Core 2 and I only have one cpu You have the right app........ And judging from the last result you returned, your crunch time has dropped almost in half......very nice. "Freedom is just Chaos, with better lighting." Alan Dean Foster |
jason_gee Send message Joined: 24 Nov 06 Posts: 7489 Credit: 91,093,184 RAC: 0 |
I have installed it did I install the right one as it says Core 2 and I only have one cpu Yep, to explain that peculiarity with a piece of nonsensical technical gibberish: The AMD/INTEL SSE3 build, actually uses "generic" core2 optimisation compiler flags ( /QxO ) and whatever the code runs on, it was still hand optimised by Alex for Core2 type machines ... So there you go, you're running a core2 optimised app on a p4 as it says ;D (It's the "generic" bit that limits it to SSE3 & let's it run on AMDs too) Jason "Living by the wisdom of computer science doesn't sound so bad after all. And unlike most advice, it's backed up by proofs." -- Algorithms to live by: The computer science of human decisions. |
KWSN Checklist Send message Joined: 4 Jan 08 Posts: 46 Credit: 3,282,541 RAC: 0 |
Thanks to all for making SETI@home a Greener project. IMHO, the best benefit. |
©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.