Would it be possible to have encoding processes be evenly distributed between the two NVENC chips in a card with 2 or more NVENCs? Or does it already do it automatically when you set and initialize the device? I guess another question would be if it would have to be explicitly set to use both NVENCs on the card.
The NVENC devices on each GPU would need to be set up separately and managed individually, on a card that has multiple GPUs on it.
Thanks for clarifying!
I’m experiencing an issue where I have multiple GPUs (two Quadro P5000s) and I’m running into this problem where I try to use GPU 1 (second gpu) to do all the processing for (encoding/decoding), but all of the processes keep going to GPU 0 (first gpu) by default. Would there be a way to set GPU 1 to run all the processes. I’ve tried cudaSetDevice() and it still goes to GPU 0.
This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.