I have two systems running folding@home. One has two GTX980Tis, and one has 2 x Tesla M40s. I believe these should be roughly equivalent in compute power, but for some reason, the Teslas don't seem to get work units near as often. I noticed a difference in the fields in GPUS.txt:
0x10de:0x17c8:2:6:GM200 [GeForce GTX 980 Ti] 5632
0x10de:0x17fd:2:5:GM200GL [Tesla M40]
Anyone know why?
Nvidia Tesla M40
Moderators: Site Moderators, FAHC Science Team
-
toTOW
- Site Moderator
- Posts: 6309
- Joined: Sun Dec 02, 2007 10:38 am
- Location: Bordeaux, France
- Contact:
Re: Nvidia Tesla M40
Because the guy who enters data in the list don't know all GPUs (especially the professional ones we rarely see) ... 
On NV, species has no real meaning ... it's usually related to the generation of the GPU, and sometimes to attempt to classify some slow/faster ones ... but it doesn't always work well.
Anyway, I set the professional GM200 (Tesla and Quadro) to the same species (6) as GeForce equivalents.
On NV, species has no real meaning ... it's usually related to the generation of the GPU, and sometimes to attempt to classify some slow/faster ones ... but it doesn't always work well.
Anyway, I set the professional GM200 (Tesla and Quadro) to the same species (6) as GeForce equivalents.
Folding@Home beta tester since 2002. Folding Forum moderator since July 2008.
-
twistdroach
- Posts: 2
- Joined: Wed Mar 25, 2020 1:47 pm
Re: Nvidia Tesla M40
Nice - ok no problem - thanks for the quick response! Have a good day.