New AstroPulse for GPU ( ATi & NV) released (r1316)

Message boards : Number crunching : New AstroPulse for GPU ( ATi & NV) released (r1316)
Message board moderation

To post messages, you must log in.

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

AuthorMessage
Profile [AF>EDLS] Polynesia
Volunteer tester
Avatar

Send message
Joined: 1 Apr 09
Posts: 54
Credit: 5,361,172
RAC: 0
France
Message 1263789 - Posted: 23 Jul 2012, 17:42:43 UTC
Last modified: 23 Jul 2012, 17:55:20 UTC

I tell you here when downloading my 2 units GPU is finally completed ...


Alliance Francophone
ID: 1263789 · Report as offensive
Profile [AF>EDLS] Polynesia
Volunteer tester
Avatar

Send message
Joined: 1 Apr 09
Posts: 54
Credit: 5,361,172
RAC: 0
France
Message 1264024 - Posted: 24 Jul 2012, 10:05:31 UTC

About 1 hour 20 for the operating time on my gtx 470


Alliance Francophone
ID: 1264024 · Report as offensive
Profile HAL9000
Volunteer tester
Avatar

Send message
Joined: 11 Sep 99
Posts: 6534
Credit: 196,805,888
RAC: 57
United States
Message 1264066 - Posted: 24 Jul 2012, 13:16:44 UTC - in response to Message 1263719.  

I've noticed that some WUs are getting low credits, is there a problem somewhere? It's not happening only with the latest app, it also happened with previous versions, not sure when I first noticed it.

http://setiathome.berkeley.edu/results.php?hostid=6442255&offset=0&show_names=0&state=3&appid=12

Do you mean the ones with 0.13 & 0.08 credit? Which are 100% blanked.
SETI@home classic workunits: 93,865 CPU time: 863,447 hours
Join the [url=http://tinyurl.com/8y46zvu]BP6/VP6 User Group[
ID: 1264066 · Report as offensive
Profile Mike Special Project $75 donor
Volunteer tester
Avatar

Send message
Joined: 17 Feb 01
Posts: 34253
Credit: 79,922,639
RAC: 80
Germany
Message 1264080 - Posted: 24 Jul 2012, 13:55:25 UTC - in response to Message 1264024.  

About 1 hour 20 for the operating time on my gtx 470


Both results are looking fine.



With each crime and every kindness we birth our future.
ID: 1264080 · Report as offensive
JohnDK Crowdfunding Project Donor*Special Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 28 May 00
Posts: 1222
Credit: 451,243,443
RAC: 1,127
Denmark
Message 1264102 - Posted: 24 Jul 2012, 14:46:56 UTC - in response to Message 1264066.  

I've noticed that some WUs are getting low credits, is there a problem somewhere? It's not happening only with the latest app, it also happened with previous versions, not sure when I first noticed it.

http://setiathome.berkeley.edu/results.php?hostid=6442255&offset=0&show_names=0&state=3&appid=12

Do you mean the ones with 0.13 & 0.08 credit? Which are 100% blanked.

No there's some that only gives around 150-250 and not around 600-800 as normal. Right now there's only one with 140.64 Cr on my finished WUs list, last when I wrote there was 3-4 WUs with the low credit.
ID: 1264102 · Report as offensive
Profile Fred J. Verster
Volunteer tester
Avatar

Send message
Joined: 21 Apr 04
Posts: 3252
Credit: 31,903,643
RAC: 0
Netherlands
Message 1264128 - Posted: 24 Jul 2012, 21:50:36 UTC - in response to Message 1262988.  

I had an abortion because the download is very long ...

In my preferences I put yes to everything in CPU / GPU


Well, DownLoads are very slow at SETI, comaired to projects with
fewer users and hosts, GPUgrid hits 20MBit, Einstein is also faster.
Milkyway, WCG, etc..
Problem is SETI is constant overloaded/saturated and in too much cases
the cause of a DownLoad Errors as in your case.

How long for a MB (~360KBytes) and AstroPulse (8MBytes) workunit?



ID: 1264128 · Report as offensive
JohnDK Crowdfunding Project Donor*Special Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 28 May 00
Posts: 1222
Credit: 451,243,443
RAC: 1,127
Denmark
Message 1264139 - Posted: 24 Jul 2012, 22:39:00 UTC - in response to Message 1264102.  

I've noticed that some WUs are getting low credits, is there a problem somewhere? It's not happening only with the latest app, it also happened with previous versions, not sure when I first noticed it.

http://setiathome.berkeley.edu/results.php?hostid=6442255&offset=0&show_names=0&state=3&appid=12

Do you mean the ones with 0.13 & 0.08 credit? Which are 100% blanked.

No there's some that only gives around 150-250 and not around 600-800 as normal. Right now there's only one with 140.64 Cr on my finished WUs list, last when I wrote there was 3-4 WUs with the low credit.

Here's one with only 175.27 Cr
http://setiathome.berkeley.edu/result.php?resultid=2525075894

Another one, same blanking, almost same run time, but 697.07 Cr
http://setiathome.berkeley.edu/result.php?resultid=2525071993
ID: 1264139 · Report as offensive
Profile [AF>EDLS] Polynesia
Volunteer tester
Avatar

Send message
Joined: 1 Apr 09
Posts: 54
Credit: 5,361,172
RAC: 0
France
Message 1264162 - Posted: 25 Jul 2012, 18:53:06 UTC - in response to Message 1264128.  
Last modified: 25 Jul 2012, 18:54:04 UTC

I had an abortion because the download is very long ...

In my preferences I put yes to everything in CPU / GPU


Well, DownLoads are very slow at SETI, comaired to projects with
fewer users and hosts, GPUgrid hits 20MBit, Einstein is also faster.
Milkyway, WCG, etc..
Problem is SETI is constant overloaded/saturated and in too much cases
the cause of a DownLoad Errors as in your case.

How long for a MB (~360KBytes) and AstroPulse (8MBytes) workunit?




FFor astro: more than 3 hours in download time but I have caused to manually restart the download dozens of times

For the rest, I can not remember ...


Alliance Francophone
ID: 1264162 · Report as offensive
Wedge009
Volunteer tester
Avatar

Send message
Joined: 3 Apr 99
Posts: 451
Credit: 431,396,357
RAC: 553
Australia
Message 1265643 - Posted: 29 Jul 2012, 21:52:46 UTC
Last modified: 29 Jul 2012, 21:53:39 UTC

Just some more feedback for Raistmer and any other AP OpenCL developers:

Getting a lot of invalid results on my old HD 4670, mostly due to 30 single pulses, 30 repeating pulses, or both. This seems to happen regardless of what combination of parameters I try. The confusing thing is that, sporadically, I still get the occasional valid result.

I think I'll have to revert to r555 even though it has much lower performance (especially with regards to CPU usage), it's still bound to be much better than CPU-only processing on the ancient machine it's on. There were still invalid results, but they were limited to 30 repeating pulses and there were far more valid results than invalid ones. I put all this down to the 'beta-level' OpenCL support on the HD 4000 series - I ceased all GPU processing on my HD 4850 a long time ago because of similar issues presumably arising from this.

Other GPUs (HD 5000 series and later, as well as GeForce cards) continue to work very well on r1316. I'm aware there is a r1363 release, but I don't understand the parameters enough to try it. Doesn't seem as though there's as large a performance increase either.
Soli Deo Gloria
ID: 1265643 · Report as offensive
Profile Mike Special Project $75 donor
Volunteer tester
Avatar

Send message
Joined: 17 Feb 01
Posts: 34253
Credit: 79,922,639
RAC: 80
Germany
Message 1265647 - Posted: 29 Jul 2012, 22:14:24 UTC


Have you tried lower unroll ?

I´d suggest -unroll 4 -ffa_block 2048 -ffa_block_fetch 1024

Mike



With each crime and every kindness we birth our future.
ID: 1265647 · Report as offensive
Wedge009
Volunteer tester
Avatar

Send message
Joined: 3 Apr 99
Posts: 451
Credit: 431,396,357
RAC: 553
Australia
Message 1265675 - Posted: 29 Jul 2012, 23:16:46 UTC

Yes. As I mentioned, the both valid and invalid results occur for all manner of parameters. I've tried unroll of 4, 8, 10 with r1316. With r555, unroll of 4 always resulted in invalid results, whereas 10 was the only one that consistently returned valid results.
Soli Deo Gloria
ID: 1265675 · Report as offensive
Profile Jim Holly
Volunteer tester

Send message
Joined: 14 May 99
Posts: 11
Credit: 3,363,801
RAC: 0
United States
Message 1265960 - Posted: 30 Jul 2012, 18:51:26 UTC

I would be delighted to just download some AstroPulse files that would run! Every one shows completion in the Transfers tab, but the Tasks list always says 'Download failed'.

Running all other WU's in Seti and Einstein with no problems.

Windows 7 Pro 64-bit; i7-2600 quad core with 8 Gb RAM.
ID: 1265960 · Report as offensive
Profile BilBg
Volunteer tester
Avatar

Send message
Joined: 27 May 07
Posts: 3720
Credit: 9,385,827
RAC: 0
Bulgaria
Message 1266045 - Posted: 31 Jul 2012, 2:17:50 UTC - in response to Message 1265960.  


Your problem is with the file ap_graphics_6.01_windows_intelx86.exe
http://setiathome.berkeley.edu/result.php?resultid=2544369687

Very probably your antivirus is interfering:
http://setiathome.berkeley.edu/forum_thread.php?id=68629&postid=1257824#1257824

Get the file manually:
http://boinc2.ssl.berkeley.edu/sah/download_fanout/ap_graphics_6.01_windows_intelx86.exe

(Probably your antivirus will say something nasty and wrong. If it does: complain to the antivirus company)


If you can get the file from the above link (I can, I get the file just now and the file is the same as the one I have):
put it in the ....\BOINC\projects\setiathome.berkeley.edu\ and then start BOINC

In your case of Windows 7 the exact path is most likely:
C:\ProgramData\BOINC\projects\setiathome.berkeley.edu\


 


- ALF - "Find out what you don't do well ..... then don't do it!" :)
 
ID: 1266045 · Report as offensive
Profile Raistmer
Volunteer developer
Volunteer tester
Avatar

Send message
Joined: 16 Jun 01
Posts: 6325
Credit: 106,370,077
RAC: 121
Russia
Message 1266127 - Posted: 31 Jul 2012, 9:25:40 UTC - in response to Message 1265643.  
Last modified: 31 Jul 2012, 9:26:03 UTC

Just some more feedback for Raistmer and any other AP OpenCL developers:

Getting a lot of invalid results on my old HD 4670, mostly due to 30 single pulses, 30 repeating pulses, or both. This seems to happen regardless of what combination of parameters I try. The confusing thing is that, sporadically, I still get the occasional valid result.

I think I'll have to revert to r555 even though it has much lower performance (especially with regards to CPU usage), it's still bound to be much better than CPU-only processing on the ancient machine it's on. There were still invalid results, but they were limited to 30 repeating pulses and there were far more valid results than invalid ones. I put all this down to the 'beta-level' OpenCL support on the HD 4000 series - I ceased all GPU processing on my HD 4850 a long time ago because of similar issues presumably arising from this.

Other GPUs (HD 5000 series and later, as well as GeForce cards) continue to work very well on r1316. I'm aware there is a r1363 release, but I don't understand the parameters enough to try it. Doesn't seem as though there's as large a performance increase either.


Actually I'm trying to maintain compatibility with HD4xxx though has no hardware installed now. But HD46xx low-end ones AFAIK and have some different issues.
Special LHD4 build for them still in debugging. From the other side I would not expect same failures on HD4850. Can you replace HD46xx with HD48xx on that host ?
ID: 1266127 · Report as offensive
Profile Jim Holly
Volunteer tester

Send message
Joined: 14 May 99
Posts: 11
Credit: 3,363,801
RAC: 0
United States
Message 1266310 - Posted: 1 Aug 2012, 0:29:17 UTC

Thanks BilBg. Trend did try to block it, and I placed it on the 'approved' list. I replaced the one in the folder which was only 9kB. Time will tell, once I get another AP download.
ID: 1266310 · Report as offensive
Wedge009
Volunteer tester
Avatar

Send message
Joined: 3 Apr 99
Posts: 451
Credit: 431,396,357
RAC: 553
Australia
Message 1266319 - Posted: 1 Aug 2012, 1:41:28 UTC - in response to Message 1266127.  
Last modified: 1 Aug 2012, 1:45:54 UTC

Special LHD4 build for them still in debugging. From the other side I would not expect same failures on HD4850. Can you replace HD46xx with HD48xx on that host ?

Sorry, no, the HD 4850 I put in the machine for my brother and besides, HD 4850 is PCIe, HD 4670 is AGP. For HD 4850's issues, I think it's driver issues more than anything else.

After getting yet another 30 single pulses result on the HD 4670, I reverted to r555 and re-processed the task. Hasn't validated yet, but it seems to be giving more reasonable results again.

As mentioned previously, the puzzling thing is that there are still some valid results produced with r1316. These two were validated recently, so they're still present in the database at time of writing:

2543227294

Oops, no, scratch that. Second one just got purged as I was writing...

Edit: If you want a tester for your special HD 4000 build, I'm willing to help. Will just take a while as 0% blanking AP WUs take about 7.5 hours with r1316, just under 10 hours with r555.
Soli Deo Gloria
ID: 1266319 · Report as offensive
Profile Raistmer
Volunteer developer
Volunteer tester
Avatar

Send message
Joined: 16 Jun 01
Posts: 6325
Credit: 106,370,077
RAC: 121
Russia
Message 1266353 - Posted: 1 Aug 2012, 6:21:40 UTC - in response to Message 1266319.  


Edit: If you want a tester for your special HD 4000 build, I'm willing to help.

Currently I'm busy with MultiBeam, but will keep your proposal for future, thanks.
ID: 1266353 · Report as offensive
Wedge009
Volunteer tester
Avatar

Send message
Joined: 3 Apr 99
Posts: 451
Credit: 431,396,357
RAC: 553
Australia
Message 1266365 - Posted: 1 Aug 2012, 6:54:07 UTC
Last modified: 1 Aug 2012, 6:54:36 UTC

I'm in no hurry, I only offered because you mentioned developing a special build.

Best wishes for MB development.
Soli Deo Gloria
ID: 1266365 · Report as offensive
Profile Jim Holly
Volunteer tester

Send message
Joined: 14 May 99
Posts: 11
Credit: 3,363,801
RAC: 0
United States
Message 1266502 - Posted: 1 Aug 2012, 19:59:23 UTC

AstroPulse WU's are running fine again. :)
ID: 1266502 · Report as offensive
Profile dskagcommunity
Volunteer tester
Avatar

Send message
Joined: 24 Feb 11
Posts: 43
Credit: 2,901,049
RAC: 0
Austria
Message 1266716 - Posted: 2 Aug 2012, 15:10:42 UTC

Only as Info, Application runs without errors on a low low lowest end HD4350 ^^ ~24h for 1 WU
ID: 1266716 · Report as offensive
Previous · 1 . . . 6 · 7 · 8 · 9 · 10 · 11 · Next

Message boards : Number crunching : New AstroPulse for GPU ( ATi & NV) released (r1316)


 
©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.