Advanced search

Message boards : Graphics cards (GPUs) : From the evening of April 14 (UTC+3) on one of my machines constant errors

Author Message
ElleSolomina
Avatar
Send message
Joined: 22 Mar 14
Posts: 43
Credit: 527,549,678
RAC: 18,056
Level
Lys
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwat
Message 46953 - Posted: 16 Apr 2017 | 7:09:25 UTC

Good day!

I'm trying to understand where the problem is. Such imprinting is that the problem is in the errors in the tasks.

Link to one of the problematic tasks:
http://www.gpugrid.net/workunit.php?wuid=12497229

Or the problem with my computer? Help to understand, please.

Link to the computer:
http://www.gpugrid.net/results.php?hostid=170640
____________

ElleSolomina
Avatar
Send message
Joined: 22 Mar 14
Posts: 43
Credit: 527,549,678
RAC: 18,056
Level
Lys
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwat
Message 46955 - Posted: 16 Apr 2017 | 7:36:19 UTC

The second machine also started to have problems. :((

http://www.gpugrid.net/results.php?hostid=170784
____________

Profile Retvari Zoltan
Avatar
Send message
Joined: 20 Jan 09
Posts: 2343
Credit: 16,201,255,749
RAC: 7,520
Level
Trp
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 46958 - Posted: 16 Apr 2017 | 11:14:38 UTC

The CUDA 6.5 client has been expired, that caused the immediate computational errors.
Right now there's no CUDA 6.5 client present, and the present CUDA 8.0 client (v9.18) does not work with Compute Capability 3.0 cards, but hopefully it would be fixed soon.
See the all WUs downloaded recently produce "computation error" right away thread for details.

ElleSolomina
Avatar
Send message
Joined: 22 Mar 14
Posts: 43
Credit: 527,549,678
RAC: 18,056
Level
Lys
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwat
Message 46959 - Posted: 16 Apr 2017 | 11:22:05 UTC - in response to Message 46958.
Last modified: 16 Apr 2017 | 11:28:42 UTC

OK. While I stopped receiving jobs from the project for these glitches are not acceptable. Write, please, when the work is right. I had to reboot one of the hosts for after errors in GPUGrid jobs the frequency on the video card did not rise above 500 MHz instead of the normal 1200 MHz.

P.S. Please, in the future, warn about such changes (about stopping support for CUDA versions) through messages in the client in advance ;) I have no desire to just warm up the air, and then fight against glitches. If I wanted to fight against glitches - I would choose beta version ;)
____________

Post to thread

Message boards : Graphics cards (GPUs) : From the evening of April 14 (UTC+3) on one of my machines constant errors

//