What about multi-gpu support?
Yes, add the "-gpu N" flag (N starts at 0) to the extra parameters setting in the advanced page of the Systray client, or to the advanced settings, extra parameters of the console client. Again N starts with 0 not 1, so the primary display is 0, the next is 1, etc. When running more then one client, each needs a different -gpu N, a different MachineID, and a different working directory. Please follow the instructions for multiple clients.
To support Tesla and other non-consumer level cards, use the "-forcegpu" flag. The flag will instruct the client to ignore the client's automatic hardware detection. But the switch will not add support for GPU architecture not in the correct family of GPUs. This flag has several options to match the supported GPU families.
Currently these are ati_r600, ati_r700, ati_r800, nvidia_g80, and nvidia_fermi. This flag may also help for headless multi-gpu setups on which the client refuses to acknowledge the presence of a specific GPU, but isn't necessary with the latest drivers. The -forcegpu flag can be use in conjunction with the -gpu N flag to force the client to use the given GPUID. For example:
Folding@home.exe -gpu 2 -forcegpu nvidia_g80
Would force the client to try and run the NVIDIA core on the 3rd (2 + 1) cuda enabled device in the system. Similarly:
Folding@home.exe -gpu 3 -forcegpu ati_r600
Would force the client to try and run the ATI core on the 4th (3 + 1) CAL enabled device in the system.
For a more detailed set of instructions, please see these links
- FAQ on the forum
- Windows GPU Guide
We will update our main FAQ in time with these details, but we are also working on ways to handle this from an installer directly.
.
