| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
 |
World Community Grid Forums
Category: Support
Forum: GPU Support Forum
Thread: Graphics Card Performance |
| No member browsing this thread |
|
Thread Status: Active Total posts in this thread: 254
|
|
| Author |
|
|
Jim1348
Veteran Cruncher USA Joined: Jul 13, 2009 Post Count: 1066 Status: Offline Project Badges: 
|
Still, one thing is interesting with FAH, going by the status-page the GROGPU2 max handles 1254 atoms while OPENMMGPU max handles 2262 atoms. If not mistaken the GPUGPU2 is CUDA while OPENMMGPU is Amd... I am current running P8044 and P8045 on a GTX 560 and GTX 560 Ti, and they are both listed as OPENMMGPU. I don't know about your other speculations, but you are hardly the first to claim that the points are skewed one way or the other. But they are based on the scientific value that Pande Group gives them; you can choose your own as you like. [Edit 1 times, last edit by Jim1348 at Oct 18, 2012 6:11:05 PM] |
||
|
|
Ingleside
Veteran Cruncher Norway Joined: Nov 19, 2005 Post Count: 974 Status: Offline Project Badges: 
|
I am current running P8044 and P8045 on a GTX 560 and GTX 560 Ti, and they are both listed as OPENMMGPU. Ah, this is for not having any Nvidia-card.  "I make so many mistakes. But then just think of all the mistakes I don't make, although I might." |
||
|
|
cristipurdel
Senior Cruncher Joined: Dec 13, 2008 Post Count: 158 Status: Offline Project Badges: 
|
Should I also add that Intel is also supporting OpenCL?
----------------------------------------Once lunatics guys start porting the seti app to Intel gpu, thinks may look pretty bad for cuda's support in boinc :) |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
yes met to say 7870 there is no 7880 duh!!!
|
||
|
|
Jim1348
Veteran Cruncher USA Joined: Jul 13, 2009 Post Count: 1066 Status: Offline Project Badges:
|
Should I also add that Intel is also supporting OpenCL? Once lunatics guys start porting the seti app to Intel gpu, thinks may look pretty bad for cuda's support in boinc :) You are more than welcome to. It has been said often enough on the Folding forum, and they always point out that until it gets big enough, they won't try to get Folding to run on it, and it might not work very well even then. But as I have said elsewhere, it appears that Nvidia will be going to OpenCl on their next-generation cards (Maxwell), though I have seen no official announcement on that, if it makes you feel better. It is always surprising to me what some people take comfort in. |
||
|
|
Hypernova
Master Cruncher Audaces Fortuna Juvat ! Vaud - Switzerland Joined: Dec 16, 2008 Post Count: 1908 Status: Offline Project Badges: 
|
I see that the Nvidia GTX 680 is way worse than a GTX 580 or 560. Why such a big difference ?
---------------------------------------- |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Maybe, just maybe we need to think in terms of hyperthreading concepts. The times... well what's the difference between e.g. a 580 running 1 and someone with a 680 running 3 or 4 simultaneous [seems some are doing this on various powercards]
|
||
|
|
Jim1348
Veteran Cruncher USA Joined: Jul 13, 2009 Post Count: 1066 Status: Offline Project Badges:
|
I see that the Nvidia GTX 680 is way worse than a GTX 580 or 560. Why such a big difference ? It may have to do with the shader speeds. The shader clocks on the Fermis (GTX 560, 580) are frequency-doubled as compared to the core clock rate, whereas on the Keplers (GTX 680) they run at the same rate as the core clock. Nvidia did this to save power, and apparently when running games it is compensated for by the larger number of shaders on Kepler. But it hurts performance on a lot of crunching projects. |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
running 3 or 4 simultaneous how do you do that? |
||
|
|
Hypernova
Master Cruncher Audaces Fortuna Juvat ! Vaud - Switzerland Joined: Dec 16, 2008 Post Count: 1908 Status: Offline Project Badges:
|
Can anyone explain what is the "app_info" function that could allow to run multiple GPU tasks.
----------------------------------------Will this be possible to run 2 or 3 tasks on GTX 580 and GTX 680 boards? |
||
|
|
|