Nvidia GPUs responsible for 42% of Folding@home output

Posted at: 6:13pm 26th August 2008 by Ben Hardwidge

The green team’s graphics chips are contributing 1,428 TFLOPS of processing power, according to the latest stats

Members of Custom PC’s Folding@home team may have noticed a major increase in our production rate recentlys, and there’s little doubt that much of this can be attributed to Stanford’s new GPU2 client for Nvidia and ATI GPUs. However, it looks as though Nvidia GPUs are the king of the folding castle at the moment, as they’re currently responsible for a whopping 42 per cent of the project’s total output.

This is despite the fact that over three times as many active PlayStation 3s as there are Nvidia GPUs contributing to the project, not to mention 16 times as many CPUs folding under Windows. There are currently 12,982 Nvidia GPUs actively folding, out of a total of 323,424 active processors in total. As such, the fact that Nvidia GPUs make up such a colossal percentage of the total output is pretty incredible.

You can see the current statistics on the next page, showing that Nvidia GPUs are currently churning out 1,428 TFLOPS out of a total of 3,372 TFLOPS for the whole project. Meanwhile, ATI GPUs are responsible for 404 TFLOPS, but even this is double the output from Windows CPU clients, and is an impressive figure considering that there are only 3,677 ATI GPUs actively folding at the moment.

Speaking to Custom PC, the director and founder of the Folding@home project, Vijay Pande, explained that ‘in terms of raw performance, Nvidia is doing a bit better, but there are different metrics for performance. In terms of performance overall, the GTX 280 is a very powerful board, but it’s also a very expensive board. So, in terms of the amount of calculations per dollar, you could buy one GTX 280 or three ATI 4870s, or maybe four, at the same cost.’

Pande is especially pleased with the results from Nvidia GPUs, though, saying ‘as these statistics show, the impact of Nvidia GPUs on protein folding simulations has been extraordinary. Teams that are folding with Nvidia GPUs are seeing huge boosts to their production and this is helping to accelerate the project significantly.’

Nvidia is also chuffed with the results. Nvidia’s general manager of visual consumer solution, Michael Steele, said that ‘I know everyone at Nvidia has been closely tracking the progress of the Folding@home project since the release of the CUDA port for our GPUs, and we are delighted to see them making such a significant and meaningful contribution to what is extremely valuable work.’

If you’ve got an Nvidia GeForce 8000-series GPU or above, or an ATI Radeon HD 2000-series GPU or above, then you could download Stanford’s GPU2 client and get it folding for team Custom PC, which is currently at number 7 on the world leaderboard.