Folding@home

GPU accelerated computing is the latest trend to speed up computationally intense applications. One of the first projects to implement such a GPGPU method is Folding@home. Participants install a small program on their computer, the "Folding Client", which downloads work units from a server to be processed on the system.
In the past folding has been performed on on the CPU. Recently new clients have been developed for Folding on PlayStation 3, ATI and NVIDIA CUDA.

We used version 6.20r1 to download a work unit and fold it using whatever GPU acceleration is available. Due to the different GPU designs, different types of work unit were used. However, for most Folding users the PPD (Points Per Day) metric is the most important because that's what determines their ranking in the system.

Please note that we will not start multiple client instances. If a manufacturer chooses to implement native support for multiple GPUs, our client will benefit from that however. For comparison: a PlayStation 3 gets about 900 PPD, a Core 2 Duo E8400 about 550 PPD per processor core.

If you want to join the good cause, use team number 50711, which is TechPowerUps own Folding Team.

It seems something is wrong with the score of the HIS HD 4870, I would expect a result of around 3600 PPD. A possible explanation is that our Folding client ended up with a different work unit than when all the other cards were tested. I will look into this and fix the graphs accordingly.