TheDarkHorizon said:I've been folding as TheDarkHorizon for almost a week now. I'm not sure why I never thought to start folding in the past, but I'm here now. I've got a few computers running it in the background.
SmokeRings said:Checkin in, didn't read the whole thread though.
I added over 10GHz equiv p4 processing about 2 months ago after SETI blew it with thier new software.
I hope we can get our Graphics cards working soon (or did I miss it?) because I'm pretty sure our team has a lot of gaming rigs. I use this site mostly for Radeon support, tweaks, etc.
-Marc
Excellent & welcome!Dem said:Hi, I just started folding few days ago, better late then never!
Rig is in Sig....Also putting together 2 of my older systems;3200+(2.2Ghz)XP and AMD K7 1400 mhz non XP.
Welcome! Actually, there is a GPU client in the works:CommanderROR said:Actually it's a real pity that you can't use ure GPU for folding as well...![]()
GROMACS ON GPUs
GPUs have the possibility to perform an enormous number of Floating Point OPerations (FLOPs). However, they achieve this high performance by losing generality -- there are only certain types of calculations which would be well-suited to GPUs. The calculations in FAH could possibly take advantage of GPUs and we have been investigating this with our collaborators Prof. Pat Hanrahan (Stanford University, Computer Science Dept) and Prof. Eric Darve (Stanford University, Mechanical Engineering Dept) and their groups. Prof. Hanrahan's group has pioneered methods for programming GPUs with the Brook streaming computer language and Prof. Darve's group has written many applications using Brook. It is unclear which GPUs will be supported and we don't want to specific list any GPUs at the moment. However, it will likely require the very latest GPUs from NVIDIA or ATI. There is also a possibility that only NVIDIA GPUs will work (they use 32 bit floating point, whereas ATI uses 24 bit).
July 2005 We have a working version of GROMACS on Brook, but are tuning performance.
August 2005 Vishal has made great progress in rewriting the GROMACS inner loops in order to take advantage of more memory. Now, we're tweaking for performance.
Lupine said:Excellent & welcome!
BTW, you're going to want to shorten that siggy to fit the 10 line limit.![]()
pm sent.Dem said:Hi and thanks. How do you make the Sig go wider so it don't take up so many lines? Got it down to 12![]()
Lupine said:Welcome! Actually, there is a GPU client in the works:
GROMACS ON GPUs
GPUs have the possibility to perform an enormous number of Floating Point OPerations (FLOPs). However, they achieve this high performance by losing generality -- there are only certain types of calculations which would be well-suited to GPUs. The calculations in FAH could possibly take advantage of GPUs and we have been investigating this with our collaborators Prof. Pat Hanrahan (Stanford University, Computer Science Dept) and Prof. Eric Darve (Stanford University, Mechanical Engineering Dept) and their groups. Prof. Hanrahan's group has pioneered methods for programming GPUs with the Brook streaming computer language and Prof. Darve's group has written many applications using Brook. It is unclear which GPUs will be supported and we don't want to specific list any GPUs at the moment. However, it will likely require the very latest GPUs from NVIDIA or ATI. There is also a possibility that only NVIDIA GPUs will work (they use 32 bit floating point, whereas ATI uses 24 bit).
July 2005 We have a working version of GROMACS on Brook, but are tuning performance.
August 2005 Vishal has made great progress in rewriting the GROMACS inner loops in order to take advantage of more memory. Now, we're tweaking for performance.
http://folding.stanford.edu/FAQ-highperformance.html
One of these day ...
OT: are ATi GPUs still 24bit? How about the upcoming R520?
I posted a thread in the ATi forum asking that question. Please post anything in regard to ATi 32bit support within the following thread so we can keep this thread on topic. Lets also resist turning that thread Folding specific.
http://www.rage3d.com/board/showthread.php?p=1333882564#post1333882564
EDIT: apparently R3xx / R4xx are all 24bit, but the R520 will be 32bit. So once the R5xx series is released, Pande Group will need to update that FAQ!![]()