Advanced search

Message boards : Number crunching : non- GIANNI WUs starting with e1s

Author Message
Profile Beyond
Avatar
Send message
Joined: 23 Nov 08
Posts: 1112
Credit: 6,162,416,256
RAC: 0
Level
Tyr
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 38834 - Posted: 5 Nov 2014 | 23:30:02 UTC

I've had 5 failed WUs today starting with "e1s" (both SDOERR and NOELIA). They don't run on other peoples' GPUs either. The GIANNI ones are fine. I'm starting to abort the bad ones.

Also have received 1 e2s NOELIA WU. It failed for the 1st user but completed on my 750Ti, but took 2x the normal time and yielded 1/3 of the credits. Think there may also be a problem with "e2s".

Other than that I've had no recent failures except for 2 caused by a total power failure in the neighborhood.

Jim1348
Send message
Joined: 28 Jul 12
Posts: 819
Credit: 1,591,285,971
RAC: 0
Level
His
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 38840 - Posted: 6 Nov 2014 | 4:13:15 UTC - in response to Message 38834.
Last modified: 6 Nov 2014 | 4:19:00 UTC

The only "e1s" that I have gotten was e1s6_1-SDOERR_ubievss3-0-1-RND7940_1, which failed right away with Computation error (-98,) on a GTX 660 Ti. This is in the problematic "ubi" series mentioned here: http://www.gpugrid.net/forum_thread.php?id=3915

I also got an "e2s" NOELIA, which took a very long time (22 1/2 hours) to complete on a GTX 660, and to add insult to injury yielded only 63,000 points. It failed on one other machine, so they must not be easy. http://www.gpugrid.net/workunit.php?wuid=10250836

I have had no problems with the GIANNI "e2s" (have not gotten an "e1s") on either a GTX 660 Ti or a GTX 750 Ti.

I have not gotten any of the "e1s" or "e2s" (SDOERR or NOELIA) on my GTX 750 Tis yet, so I can not say about that.

Stefan
Project administrator
Project developer
Project tester
Project scientist
Send message
Joined: 5 Mar 13
Posts: 348
Credit: 0
RAC: 0
Level

Scientific publications
wat
Message 38851 - Posted: 6 Nov 2014 | 10:11:13 UTC

Hey, thanks for pointing out the problem with the Ubi simulations. It should be fixed now.

But to clarify stuff a bit, the eXsY_blabla simulations are not the same system. It's just a new method that we are working with that names them that way (e1s6 is epoch 1 simulation 6). So for a bit more clarity, mention which group/system they belong to like ubievss or AGluGlu.

I will mention the credit thing to Noelia to see if it's a mistake.
Many thanks.

Profile Beyond
Avatar
Send message
Joined: 23 Nov 08
Posts: 1112
Credit: 6,162,416,256
RAC: 0
Level
Tyr
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 38854 - Posted: 6 Nov 2014 | 15:07:14 UTC

Thanks for fixing this so promptly. Just got a new e1s9_1-SDOERR_ubievsslong and it seems to be running fine.

Post to thread

Message boards : Number crunching : non- GIANNI WUs starting with e1s

//