Im having a desktop with two nVidia display devices.
One GeForce 8500 GT which is used as a primary display device and another GeForce 8800 GTX which i would like to use for my computations.
Is there anyway to identify device that is used as primary display device and automatically tell my application that the dedicated device can be used for computations?
On my setup I right click on the desktop, click properties. Then I choose the settings tab. Under display choose the graphics card you don’t want to use to display and unselect “Extend my Windows desktop onto this monitor”
Thank you very much for your response.
I want to know programatically which is primary device and secondary device(not used for display) and want to use the secondary device for my computations, say for example, if device count is 2 and deviceNo = 1 is the secondary device, i can call ‘cudaSetDevice(1)’ to use secondary device for my Kernel execution.
- cudaGetDeviceCount gives you device count
- cudaGetDeviceProperties in .name returns name of the device
8800 will contain “8800” substring in name :)