Santa Clara (CA) – During Nvidia Editor's Day, we learned that Nvidia and the Folding@Home research group led by Vijay Pande are making final preparation to launch the first version of the Folding@Home client for Nvidia graphics processors.

The unveiling of the client is set for the next week as part of the launch of Nvidia’s  GT200 GPU series. Owning such a card will have its benefits in Folding@Home and will outrun Radeon 3870 cards. The new GeForce cards are expected to hit more than 650 nanoseconds of protein simulation in a single day, while the Radeon HD 3870 is stuck at about 170 ns. The Playstation 3 is able to produce "only" 100 ns of simulation, while a quad-core CPU creates an output of just four nanoseconds. For those who are keeping count: The GeForce GPU will be about 163 times faster than a quad-core processor in this specific application.

Nvidia founded Team “Whoopass”, which consists only of several computers that are running the Folding@Home GPU client. Even with just 4-5 test machines, the team quickly moved into the top 5% of all contributors by sheer processing power. Dr. Vijay told us that if only 1% of all CUDA-capable users would start using Folding@Home in their spare time, the Folding@Home machine would quickly be considered the fastest performing HPC computer in the whole world – hitting about 60-80 Peta FLOPS of processing power.

Folding@Home for Nvidia CUDA-capable graphics cards (GeForce 8 and above) should become available next week. The codename for this client is GPU2/NVIDIA. The GPU1 client was retired, while GPU2 client will continue to be updated for both Nvidia and ATI cards.

ATI was first with a client for the Folding@Home project, which was released back in September of 2006 for the X1900 series of cards. Back then, the cards topped out at 375 GFlops. The next GPU generation should provide more than double the horsepower.

just pointing out...

Jun 13, 2008 07:57     

Also consider that the GPUs cannot do all of the protein foldings that a CPU can. There is a very limited range of data that a GPU can work with but what it does do, it does magnitudes better than if CPUs did it. Same goes for the cell but its somewhere inbetween a CPU and a GPU.
Mattcrwi report abuse vote down vote up Votes: +0

Great, now all they need to do is code an ATI clien...

Jun 13, 2008 12:05     

Great, now all they need to do is code an ATI client as well so everyone can contribute. Great nVidia marketing material, though, when you can say "nVidia is five times faster than ATI," ignoring the fine print. :P Combining multiple computing strategies -- stream processors on graphics cards doing all operations they're best suited for, properly utilizing multi-core processor resources and letting them do whatever operations they're best suited for, maybe even PCI-Express drop-in additional parallel computing power not specifically designed for rasterization -- I can hardly wait. We'll have tons of parallel processing in small desktop boxes.

VitriolicV report abuse vote down vote up Votes: -6

The whole project started on ATI VitrolicV...

Jun 13, 2008 12:54     

Um, Vit, when did you wake up from the coma? Glad to see you're with us now. Go check out http://folding.stanford.edu and you'll see the ATI client is currently in it's 2nd itteration. They've been folding on ATI for a couple years now. That's because ATI is much better at supporting other types of programming. That's also why nVidia has been advertising the piddle out of their CUDA b.s. I can't wait for the nVidia client, because I've got a nice SLI machine that would fold quite nicely. I can't play games on it because the drivers are so darn buggy, so it's nice that I can now do something with it. I hate having to stop my ATI based machine from folding so much because I want to play a game on it. The nVidia machine can fold 7x24 because it's worthless for anything else.

Narg report abuse vote down vote up Votes: -2

Re: Great, now all they need to do is code an ATI clien...

Jun 13, 2008 13:24     

The ATI client already have years in existence. The news is that now, Nvidia also is in the game. Good. I was wonder for years on why all that horsepower was ignored. Now, we need GPGPU compressión/Decompressión. That would allow accelerate hard disk read/write, overcoming the hard disk bottleneck. Once that is achieved, GPUs gonna be a serious contender for Intel.
mm report abuse vote down vote up Votes: +0

RE: The whole project started on ATI VitrolicV...

Jun 13, 2008 14:12     

Right, Folding@Home has had a client for several years now. You can't play games on your nVidia SLI machine because their drivers are too buggy? Wasn't that the whole point of your building your expensive nVidia gaming machine in the first place? What do you use your nVidia SLI machine for, a paperweight? LOL
Cypherdude report abuse vote down vote up Votes: +0

RE: Re: Great, now all they need to do is code an ATI clien...

Jun 13, 2008 14:32     

"Now, we need GPGPU compressión/Decompressión. That would allow accelerate hard disk read/write, overcoming the hard disk bottleneck." I'm not exactly sure what you meant by "compressión/Decompressión" and how a GPGPU will "overcome the hard disk bottleneck." By "compressión/Decompressión" did you mean audiovideo or did you mean readingwriting actual compressed or encrypted data off the hard drive? If you meant audiovideo, whether not RAM is utilized is dependent of numerous factors. The programmer chooses how much RAM is used, the amount of RAM available, and the size of the audiovideo segment. In most cases, regardless of what processor, the hard drive is used anyway. It wouldn't make a difference. If you meant readingwriting actual compressed or encrypted data off the hard drive, there is usually only a small percentage of CPU power required for this. CPU's are so much faster now. Perhaps you were not clear.

Cypherdude report abuse vote down vote up Votes: +0

RE:The whole project started on ATI VitrolicV...

Jun 13, 2008 18:00     

Oh quit being such a moron fanboy... Everyone knows nVidia's top-of-the-line Geforce 8/9 series are significantly faster than anything ATI has. The GT280 will continue the trend. And 'CUDE' is "bs"? Is that why nearly every major professional research-level GPGPU project uses Nvidia's GPUs. If you do any reading it all, you will see how many computational scientists have clearly stated that Nvidia's CUDA is more powerful and easier to use than it's ATI, "CTM" or whatever it is called.

lol report abuse vote down vote up Votes: -2

It's about damn time.

Jun 14, 2008 13:56     

Wow, it's taken them forever. Great to hear though.
surprised report abuse vote down vote up Votes: +0

nvidia

Jun 14, 2008 20:05     

To the ignorant turd who wrote "That's also why nVidia has been advertising the piddle out of their CUDA b.s. " try reading actual publications from academia and industry such as this: http://fgb.informatik.unibas.ch/people/oschenk/publications/references/gpu-workshop.pdf Many universities around the world are teaching courses on CUDA -> http://www.nvidia.com/object/cuda_universities.html http://www.nvidia.com/object/cuda_develop.html

somebody report abuse vote down vote up Votes: +0