Benchmark: Folding@home
 |
HEXUS Folding@home team ID: 69776 | Guide: setting up folding@home |
HEXUS team statistics: |
 |
Folding@home is a distributed computing project, that very simply stated, studies protein folding and misfolding with the aim of helping scientists better understand various related diseases.
People from around the world are able to contribute to the folding@home project from just about any PC, but high-performance hardware is seen as key to the success of the project. With this in mind, highly-optimised clients developed specifically for GPUs have been made available to achieve a 20x to 40x speed increase over comparable CPU code for certain types of calculations.
For the purpose of our benchmark, each graphics card's folding performance - measured in points per day - is calculated using the latest-generation Folding@home GPU clients; GPU3 for NVIDIA and GPU2 for AMD.
You've asked for it, so here it is; folding@home performance for all six tested GPUs. All cards were tested on the same day, ensuring that the one project was used for all NVIDIA cards, and another was used for all AMD cards.
The results speak for themselves - NVIDIA's hardware, at present, offers superior folding performance. Why is there a dramatic gap between the Radeon and GeForce GPUs? We suspect it isn't all down to the hardware, as software optimisation has made the folding@home client particularly suited to NVIDIA's CUDA cores.
