SETI@home v8 beta to begin on Tuesday

Message boards : News : SETI@home v8 beta to begin on Tuesday
Message board moderation

To post messages, you must log in.

Previous · 1 . . . 22 · 23 · 24 · 25 · 26 · 27 · 28 . . . 99 · Next

AuthorMessage
Profile Jimbocous
Volunteer tester
Avatar

Send message
Joined: 9 Jan 16
Posts: 51
Credit: 1,038,205
RAC: 0
United States
Message 55830 - Posted: 10 Jan 2016, 20:46:21 UTC

Getting quite a few video driver restarts (Nvidia 355.82) on VLAR GPU tasks. Leaves the tasks waiting to run for later.
If I can help out by testing something, please let me know.
Available hardware and software is listed in my profile here.
ID: 55830 · Report as offensive
Profile Eric J Korpela
Volunteer moderator
Project administrator
Project developer
Project scientist
Avatar

Send message
Joined: 15 Mar 05
Posts: 1547
Credit: 27,183,456
RAC: 0
United States
Message 55831 - Posted: 10 Jan 2016, 20:46:51 UTC - in response to Message 55828.  

Message from server: NVIDIA GPU: Upgrade to the latest driver to process tasks using your computer's GPU


That's a long standing problem in BOINC. That message is gerenated whenever any GPU version requires a higher or lower driver revision, regardless of whether any versions can be run.
ID: 55831 · Report as offensive
Grumpy Swede
Volunteer tester
Avatar

Send message
Joined: 10 Mar 12
Posts: 1700
Credit: 13,216,373
RAC: 0
Sweden
Message 55832 - Posted: 10 Jan 2016, 21:12:35 UTC
Last modified: 10 Jan 2016, 21:37:39 UTC

Still working through the VLAR CUDAs on my GTX 980. Slow as he** (about 2 hours each, when running 3 at a time), no problems though at all, no driver restarts and no other issues.

Examples here, the ones with over 7000 seconds run time:
https://setiweb.ssl.berkeley.edu/beta/results.php?hostid=75292&offset=0&show_names=0&state=4&appid=38

EDIT: And of course, running these VLARs at the same time as "normal" AR's or VHARs, makes those run extremely slow too. About 1 hour for something that usually only takes 10 minutes or so.
ID: 55832 · Report as offensive
Profile Jeff Buck
Volunteer tester

Send message
Joined: 11 Dec 14
Posts: 96
Credit: 1,240,941
RAC: 0
United States
Message 55833 - Posted: 10 Jan 2016, 23:44:44 UTC - in response to Message 55829.  

Got an inconclusive on a Cuda50 versus a Cuda50. Mine is a GTX750Ti but the other one is a 9800GT and exited with a -9. work unit

That host with the 9800GT, 77071, seems to have gone off the rails. It appears to be producing nothing but errors and overflows leading to Inconclusives and Invalids, for at least the last 12 hours or so, regardless of which app is running. I think several of the current Inconclusives in my own task lists are from pairings with that one.
ID: 55833 · Report as offensive
Profile Mr. Kevvy
Volunteer moderator
Volunteer tester
Avatar

Send message
Joined: 21 Apr 13
Posts: 23
Credit: 2,253,909
RAC: 0
Canada
Message 55834 - Posted: 11 Jan 2016, 0:02:15 UTC - in response to Message 55833.  

That host with the 9800GT, 77071, seems to have gone off the rails. It appears to be producing nothing but errors and overflows leading to Inconclusives and Invalids, for at least the last 12 hours or so, regardless of which app is running. I think several of the current Inconclusives in my own task lists are from pairings with that one.


Ditto... checked my queue and all of the last 8 Inconclusives were paired with that host.

Driver is 341.44; there is a known issue with 9800s and drivers later than 340.42. See this thread.

I'll take the liberty of writing to the owner.
ID: 55834 · Report as offensive
Profile Jimbocous
Volunteer tester
Avatar

Send message
Joined: 9 Jan 16
Posts: 51
Credit: 1,038,205
RAC: 0
United States
Message 55835 - Posted: 11 Jan 2016, 0:29:19 UTC - in response to Message 55834.  
Last modified: 11 Jan 2016, 0:30:33 UTC

Disregard

If I can help out by testing something, please let me know.
Available hardware and software is listed in my profile here.
ID: 55835 · Report as offensive
Richard Haselgrove
Volunteer tester

Send message
Joined: 3 Jan 07
Posts: 1451
Credit: 3,272,268
RAC: 0
United Kingdom
Message 55836 - Posted: 11 Jan 2016, 0:35:26 UTC - in response to Message 55834.  

That host with the 9800GT, 77071, seems to have gone off the rails. It appears to be producing nothing but errors and overflows leading to Inconclusives and Invalids, for at least the last 12 hours or so, regardless of which app is running. I think several of the current Inconclusives in my own task lists are from pairings with that one.


Ditto... checked my queue and all of the last 8 Inconclusives were paired with that host.

Driver is 341.44; there is a known issue with 9800s and drivers later than 340.42. See this thread.

I'll take the liberty of writing to the owner.

That problem was specifically OpenCL, and was fixed by a later hotfix driver. I'll check the exact details in the morning.
ID: 55836 · Report as offensive
Urs Echternacht
Volunteer tester
Avatar

Send message
Joined: 18 Jan 06
Posts: 1038
Credit: 18,734,730
RAC: 0
Germany
Message 55837 - Posted: 11 Jan 2016, 3:47:03 UTC

Linux/x86_64 8.04 (opencl_ati5_cat132) 10 Jan 2016, 1:29:37 UTC 0 GigaFLOPS
Linux/x86_64 8.04 (opencl_ati5_nocal) 10 Jan 2016, 1:29:37 UTC 0 GigaFLOPS
Linux/x86_64 8.04 (opencl_ati5_sah) 10 Jan 2016, 1:29:37 UTC 0 GigaFLOPS
Linux/x86_64 8.04 (opencl_atiapu_sah) 10 Jan 2016, 1:29:37 UTC 21 GigaFLOPS
Linux/x86_64 8.04 (opencl_ati_cat132) 10 Jan 2016, 1:29:37 UTC 0 GigaFLOPS
Linux/x86_64 8.04 (opencl_ati_nocal) 10 Jan 2016, 1:29:37 UTC 0 GigaFLOPS
Linux/x86_64 8.04 (opencl_ati_sah) 10 Jan 2016, 1:29:37 UTC 0 GigaFLOPS

Should we be worried about the scheduler ?
_\|/_
U r s
ID: 55837 · Report as offensive
zoom314
Volunteer tester

Send message
Joined: 29 Nov 14
Posts: 12
Credit: 63,078
RAC: 0
United States
Message 55838 - Posted: 11 Jan 2016, 5:49:23 UTC
Last modified: 11 Jan 2016, 5:51:17 UTC

Ok even though I'm using an old nvidia driver(266.58 whql), on Windows 7 Pro sp1 x64, I'm now doing v8 cuda32 for Seti-Beta, any problems?

GALACTICA

542	SETI@home Beta Test	1/10/2016 9:30:37 PM	Started download of setiathome_8.00_windows_intelx86__cuda32.exe	
543	SETI@home Beta Test	1/10/2016 9:30:37 PM	Started download of mbcuda-8.00-cuda32.cfg.sample	
544	SETI@home Beta Test	1/10/2016 9:30:37 PM	Started download of mbcuda-8.00-cuda32.cfg	
545	SETI@home Beta Test	1/10/2016 9:30:37 PM	Started download of setiathome-8.00-cuda32_AUTHORS.txt	
546	SETI@home Beta Test	1/10/2016 9:30:37 PM	Started download of setiathome-8.00-cuda32_COPYING.txt	
547	SETI@home Beta Test	1/10/2016 9:30:37 PM	Started download of setiathome-8.00-cuda32_COPYRIGHT.txt	
548	SETI@home Beta Test	1/10/2016 9:30:37 PM	Started download of setiathome-8.00-cuda32_README.txt	
549	SETI@home Beta Test	1/10/2016 9:30:37 PM	Started download of setiathome-8.00-cuda32_README_x41zh.txt	
550	SETI@home Beta Test	1/10/2016 9:30:37 PM	Started download of 06ap11ag.29420.5793.10.44.205	
551	SETI@home Beta Test	1/10/2016 9:30:39 PM	Finished download of mbcuda-8.00-cuda32.cfg.sample	
552	SETI@home Beta Test	1/10/2016 9:30:39 PM	Finished download of mbcuda-8.00-cuda32.cfg	
553	SETI@home Beta Test	1/10/2016 9:30:39 PM	Finished download of setiathome-8.00-cuda32_AUTHORS.txt	
554	SETI@home Beta Test	1/10/2016 9:30:39 PM	Finished download of setiathome-8.00-cuda32_COPYRIGHT.txt	
555	SETI@home Beta Test	1/10/2016 9:30:39 PM	Finished download of setiathome-8.00-cuda32_README.txt	
556	SETI@home Beta Test	1/10/2016 9:30:39 PM	Finished download of setiathome-8.00-cuda32_README_x41zh.txt	
557	SETI@home Beta Test	1/10/2016 9:30:41 PM	Finished download of setiathome-8.00-cuda32_COPYING.txt	
558	SETI@home Beta Test	1/10/2016 9:30:41 PM	Finished download of 06ap11ag.29420.5793.10.44.205	
559	SETI@home Beta Test	1/10/2016 9:30:45 PM	Finished download of setiathome_8.00_windows_intelx86__cuda32.exe	

[snip]

568	SETI@home Beta Test	1/10/2016 9:37:53 PM	Started download of 06ap11ag.15132.6611.10.44.86	
569	SETI@home Beta Test	1/10/2016 9:37:55 PM	Finished download of 06ap11ag.15132.6611.10.44.86	
570	SETI@home Beta Test	1/10/2016 9:37:55 PM	Starting task 06ap11ag.15132.6611.10.44.86_0	


0.15C + 1NV is being used. I just hope the card can do this in it's dirty state, or I'll wake up and find the screen has gone Black again. I do need to change the card out to clean it since it's not possible to clean the radiator in the PC, cause some dufus put the water pump on the extremely thin radiator...
ID: 55838 · Report as offensive
Richard Haselgrove
Volunteer tester

Send message
Joined: 3 Jan 07
Posts: 1451
Credit: 3,272,268
RAC: 0
United Kingdom
Message 55839 - Posted: 11 Jan 2016, 9:34:29 UTC - in response to Message 55836.  

That host with the 9800GT, 77071, seems to have gone off the rails. It appears to be producing nothing but errors and overflows leading to Inconclusives and Invalids, for at least the last 12 hours or so, regardless of which app is running. I think several of the current Inconclusives in my own task lists are from pairings with that one.


Ditto... checked my queue and all of the last 8 Inconclusives were paired with that host.

Driver is 341.44; there is a known issue with 9800s and drivers later than 340.42. See this thread.

I'll take the liberty of writing to the owner.

That problem was specifically OpenCL, and was fixed by a later hotfix driver. I'll check the exact details in the morning.

From the linked NVidia forum thread:

I have confirmed that the recently-released R340 driver version 341.44, fixes the R340 OpenCL silent data errors that we were having on pre-Fermi hardware running on prior R340 drivers.

Regards,
Jacob

Posted 02/27/2015 02:44 AM

Jacob Klein is a BOINC alpha tester: the 9800GT mentioned in the SETI thread was mine, and I also tested 341.44 and confirmed my findings with Jacob before he made that post. Whatever the problem is with 77071, he's using a valid driver.
ID: 55839 · Report as offensive
William
Volunteer tester
Avatar

Send message
Joined: 14 Feb 13
Posts: 606
Credit: 588,843
RAC: 0
Message 55840 - Posted: 11 Jan 2016, 10:16:35 UTC - in response to Message 55827.  

Reiterating for all:
The userid link only works for that particular user.
If you want to link to your task lists, please make sure you picked the hostid link.

In the example quoted below, the first link is fine, the second, though well meant, is unavailable to all but Tru - and he's got to be logged in.

Ran out of tasks on this machine....

Dayly quota of 70 tasks reached....I've done alot more then that....

http://setiweb.ssl.berkeley.edu/beta/results.php?hostid=72388

driver 13.4 win vista 32-bit ATI HD5 .r3330

No more errors today. 0700 and forward.
Pending not included.

On all boxes.
http://setiweb.ssl.berkeley.edu/beta/hosts_user.php?userid=14842

A person who won't read has no advantage over one who can't read. (Mark Twain)
ID: 55840 · Report as offensive
William
Volunteer tester
Avatar

Send message
Joined: 14 Feb 13
Posts: 606
Credit: 588,843
RAC: 0
Message 55841 - Posted: 11 Jan 2016, 10:19:11 UTC

Note to all:

If you run into errors or problems that don't look like they can be resolved within a few posts, make an initial post here and duplicate the post into a dedicated thread on the seti_enhanced forum, so troubleshooting can occur there and any widely applicable solution posted back here.
If it concerns Eric's domain - scheduler, CPU apps and so on, stay here, so he gets all the info in one place.

This makes sure this thread doesn't get overloaded with troubleshooting while still having all information regarding the testing.
A person who won't read has no advantage over one who can't read. (Mark Twain)
ID: 55841 · Report as offensive
Profile Mr. Kevvy
Volunteer moderator
Volunteer tester
Avatar

Send message
Joined: 21 Apr 13
Posts: 23
Credit: 2,253,909
RAC: 0
Canada
Message 55844 - Posted: 11 Jan 2016, 11:27:16 UTC - in response to Message 55839.  


Jacob Klein is a BOINC alpha tester: the 9800GT mentioned in the SETI thread was mine, and I also tested 341.44 and confirmed my findings with Jacob before he made that post. Whatever the problem is with 77071, he's using a valid driver.


Nice... thank you for the update. :^)
ID: 55844 · Report as offensive
MarkJ
Volunteer tester

Send message
Joined: 18 Oct 09
Posts: 48
Credit: 73,283
RAC: 0
Australia
Message 55846 - Posted: 11 Jan 2016, 12:09:38 UTC - in response to Message 55839.  

Whatever the problem is with 77071, he's using a valid driver.

Given the compute capability of the 9800GT ideally it should be running Cuda23 tasks. It would probably require all sorts of BOINC scheduler changes for it to use the compute capability. Something else to add to the BOINC wish list...
ID: 55846 · Report as offensive
Richard Haselgrove
Volunteer tester

Send message
Joined: 3 Jan 07
Posts: 1451
Credit: 3,272,268
RAC: 0
United Kingdom
Message 55847 - Posted: 11 Jan 2016, 12:20:31 UTC - in response to Message 55846.  

Whatever the problem is with 77071, he's using a valid driver.

Given the compute capability of the 9800GT ideally it should be running Cuda23 tasks. It would probably require all sorts of BOINC scheduler changes for it to use the compute capability. Something else to add to the BOINC wish list...

cuda32 is also valid on those cards, and in some cases faster - it's a bit of a moot point (he says, having been running three of them for five years). But not worth going higher.
ID: 55847 · Report as offensive
William
Volunteer tester
Avatar

Send message
Joined: 14 Feb 13
Posts: 606
Credit: 588,843
RAC: 0
Message 55851 - Posted: 11 Jan 2016, 15:38:24 UTC

Eric, something strange with the scheduler.

my wingman got a
SETI@home v8 v8.05 (opencl_ati_cat132) windows_intelx86 at 11 Jan 2016, 11:47:20 UTC
despite the app having been updated to
8.06 at 10 Jan 2016, 1:29:37 UTC
(task 21795895 host 56715)

why did he get 8.05 and not 8.06?
A person who won't read has no advantage over one who can't read. (Mark Twain)
ID: 55851 · Report as offensive
Profile Eric J Korpela
Volunteer moderator
Project administrator
Project developer
Project scientist
Avatar

Send message
Joined: 15 Mar 05
Posts: 1547
Credit: 27,183,456
RAC: 0
United States
Message 55853 - Posted: 11 Jan 2016, 15:50:24 UTC - in response to Message 55851.  

Might have been a resend of lost work. I'll see if I can figure it out from the server logs.
ID: 55853 · Report as offensive
William
Volunteer tester
Avatar

Send message
Joined: 14 Feb 13
Posts: 606
Credit: 588,843
RAC: 0
Message 55855 - Posted: 11 Jan 2016, 16:17:27 UTC - in response to Message 55853.  

Might have been a resend of lost work. I'll see if I can figure it out from the server logs.

it was split 11 Jan 2016, 10:36:12 UTC - can't have been a resend?
A person who won't read has no advantage over one who can't read. (Mark Twain)
ID: 55855 · Report as offensive
zoom314
Volunteer tester

Send message
Joined: 29 Nov 14
Posts: 12
Credit: 63,078
RAC: 0
United States
Message 55856 - Posted: 11 Jan 2016, 16:31:08 UTC - in response to Message 55838.  

I got a few v8 gpu wu's last night, none errored out, all reported just fine.
Ok even though I'm using an old nvidia driver(266.58 whql), on Windows 7 Pro sp1 x64, I'm now doing v8 cuda32 for Seti-Beta, any problems?

GALACTICA

542	SETI@home Beta Test	1/10/2016 9:30:37 PM	Started download of setiathome_8.00_windows_intelx86__cuda32.exe	
543	SETI@home Beta Test	1/10/2016 9:30:37 PM	Started download of mbcuda-8.00-cuda32.cfg.sample	
544	SETI@home Beta Test	1/10/2016 9:30:37 PM	Started download of mbcuda-8.00-cuda32.cfg	
545	SETI@home Beta Test	1/10/2016 9:30:37 PM	Started download of setiathome-8.00-cuda32_AUTHORS.txt	
546	SETI@home Beta Test	1/10/2016 9:30:37 PM	Started download of setiathome-8.00-cuda32_COPYING.txt	
547	SETI@home Beta Test	1/10/2016 9:30:37 PM	Started download of setiathome-8.00-cuda32_COPYRIGHT.txt	
548	SETI@home Beta Test	1/10/2016 9:30:37 PM	Started download of setiathome-8.00-cuda32_README.txt	
549	SETI@home Beta Test	1/10/2016 9:30:37 PM	Started download of setiathome-8.00-cuda32_README_x41zh.txt	
550	SETI@home Beta Test	1/10/2016 9:30:37 PM	Started download of 06ap11ag.29420.5793.10.44.205	
551	SETI@home Beta Test	1/10/2016 9:30:39 PM	Finished download of mbcuda-8.00-cuda32.cfg.sample	
552	SETI@home Beta Test	1/10/2016 9:30:39 PM	Finished download of mbcuda-8.00-cuda32.cfg	
553	SETI@home Beta Test	1/10/2016 9:30:39 PM	Finished download of setiathome-8.00-cuda32_AUTHORS.txt	
554	SETI@home Beta Test	1/10/2016 9:30:39 PM	Finished download of setiathome-8.00-cuda32_COPYRIGHT.txt	
555	SETI@home Beta Test	1/10/2016 9:30:39 PM	Finished download of setiathome-8.00-cuda32_README.txt	
556	SETI@home Beta Test	1/10/2016 9:30:39 PM	Finished download of setiathome-8.00-cuda32_README_x41zh.txt	
557	SETI@home Beta Test	1/10/2016 9:30:41 PM	Finished download of setiathome-8.00-cuda32_COPYING.txt	
558	SETI@home Beta Test	1/10/2016 9:30:41 PM	Finished download of 06ap11ag.29420.5793.10.44.205	
559	SETI@home Beta Test	1/10/2016 9:30:45 PM	Finished download of setiathome_8.00_windows_intelx86__cuda32.exe	

[snip]

568	SETI@home Beta Test	1/10/2016 9:37:53 PM	Started download of 06ap11ag.15132.6611.10.44.86	
569	SETI@home Beta Test	1/10/2016 9:37:55 PM	Finished download of 06ap11ag.15132.6611.10.44.86	
570	SETI@home Beta Test	1/10/2016 9:37:55 PM	Starting task 06ap11ag.15132.6611.10.44.86_0	


0.15C + 1NV is being used. I just hope the card can do this in it's dirty state, or I'll wake up and find the screen has gone Black again. I do need to change the card out to clean it since it's not possible to clean the radiator in the PC, cause some dufus put the water pump on the extremely thin radiator...
ID: 55856 · Report as offensive
Profile Jeff Buck
Volunteer tester

Send message
Joined: 11 Dec 14
Posts: 96
Credit: 1,240,941
RAC: 0
United States
Message 55857 - Posted: 11 Jan 2016, 17:02:43 UTC
Last modified: 11 Jan 2016, 17:23:09 UTC

Just noticed this morning that one of my hosts running stock Cuda hasn't been able to get any new tasks for a little over 2 hours now. My Anonymous platform hosts are still getting them, but I see they're all currently VLARs. Have VLARs now been turned off for stock Cuda?

EDIT: Finally just got a new task for that host a few minutes ago, a non-VLAR.
ID: 55857 · Report as offensive
Previous · 1 . . . 22 · 23 · 24 · 25 · 26 · 27 · 28 . . . 99 · Next

Message boards : News : SETI@home v8 beta to begin on Tuesday


 
©2021 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.