Hi everyone,
I have a Ubuntu Linux server with a Tesla P40 card that I use for CUDA workloads under CUDA 10.0. I’m considering adding a Quadro RTX card to my system which would give a great performance boost to my calculations, however I’m not 100% sure how the driver and CUDA Toolkit would behave under such circumstances. As far as I understand it should be fine, but just to be completely sure… would that work?
Thanks for your comment - how much of “should” is “will”? :) Are you aware of any reason specifically they should or shouldn’t work in the same machine? I couldn’t find anything on the NVIDIA website regarding this specific question. I have used Tesla cards from different generations on the same machine successfully in the past, and I wonder if there’s any difference when considering Quadros as well.