Best offers
Exclusive Interview: Nvidia's Ian Buck Talks GPGPU
With Snow Leopard and Windows 7 both offering GPGPU capabilities, we wanted to talk to Nvidia's Ian Buck. Not only is he one of the fathers of Brook, the programming language ultimately adopted by AMD/ATI, but the head of Nvidia's CUDA group as well. Read More
-
Beamforming: The Best WiFi You’ve Never Seen
Forget 802.11n Draft 2.0. The future of video-capable WiFi depends on a signal-boosting technique called beamforming. We put the pioneers in this frontier through some real-world testing to find out which technology is going to change the wireless world. Read More
-
Exclusive Interview: Going Three Levels Beyond Kernel Rootkits
Today we have the pleasure of chatting with Joanna Rutkowska, one of the top computing security innovators in the world. She is the founder and CEO of Invisible Things Lab (ITL), a boutique computer security consulting and research firm. Read More
Partners
The Games selection
violent :
More Mindless Violence
Basic shooting game, but still so powerful! Use the mouse to take aim and shoot at the little beasties before they get to you. Use Space to reload....
|
action :
Yoyo the Star
Yoyo is a young girl who recently graduated and dreams to become a movie star (don't we all). You'll have to guide her on the path to stardom,...
|
Sponsored links
Nvidia GPUs Approach 1.5 PFlops In Folding@Home Project
Next news- Email |
- Print |
- Comments (8) |
- Share
Santa Clara (CA) - Nvidia GPUs have become the leading processing platform within Folding@Home and continue to grow quickly: GeForce Processors are likely to become the first technology to break the 1.5 PFlops barrier.
Nvidia GPUs currently represent 42% of the total processing power of Folding@Home - or 1428 TFlops of a total of 3372 TFlops. While it appears that Nvidia can increase its overall share in this environment only slowly, it has become the leading technology and is now close to be hitting 1.5 PFlops of overall performance. The current performance is achieved with 12,982 processing units, according to the statistics released by Folding@Home.
The second most powerful platform is Sony’s PS3 with 1251 TFlops (44,379 processors), followed by ATI Radeon graphics cards with 404 TFlops (3677 processors). At least in comparison with ATI/AMD, Nvidia’s advantage is currently based solely on the sheer number of GPUs on the network. Broken down to each unit, both Nvidia and ATI processors deliver 110 GFlops on average. PS3s are measured at 28 GFlops, while Windows CPUs, which represent the majority in absolute numbers on the network (211,978 active units), provide about 9 GFlops each.
"Applications like Folding@Home are just the beginning, every day we are seeing more and more examples of computing problems that are benefitting from CUDA and our GPU technologies," said Michael Steele, general manager of visual consumer solutions at Nvidia, in a prepared statement.
Source : Tom's Hardware US
I run F@H on normal CPUs, and ATI GPUs. I attempted to run the nVidia F@H, but the driver installation required for this was a mess. Crashed all the time too. If they got the power, great, but they don't have the programming expertese to make this worthwhile. Oh well. More wasted time and money. I'll stick to the easy and still powerfull ATI solution.
Driver installation was a mess? It's a normal driver you install.... You uninstall it, restart the pc, clean out the old drivers then install the new ones. Not hard.
Stop being a ATI fanboi. You probably aren't even using the supported cards to run it in the first place.
No problem with Nvidia F@H here
Even if it was a mess, it is still worth it. What, you don't think an eventual opportunity to LIVE LONGER is worth it? You do realize that this is partially about extending lifespan, right?
I ran the FOH client on my 8600GT a couple of times, but it kept crashing at about 40% of the way through each unit.....
You guys do realize that the GPU2 client is still in BETA right?
You have to use this with the Nvidia CUDA drivers as well (which are still in beta), you can't just pop over to Nvidia.com and download the latest driver...google nvidia cuda drivers and click on the first link.
^Correct.