I am attempting to start learning CUDA using my old laptop with a hybrid Nvidia GT650M/Intel iGPU setup. I have tried following the instructions on a couple resources[1,2,3]. Those resources helped tremendously and gave me a working
nvcc. However, while most of the samples compiled, none of the samples that use OpenGL work at all. I assume this is because my installed OpenGL drivers are for my Intel GPU, but I was forced to not install the OpenGL drivers during the Nvidia driver installation to avoid overwriting the old drivers (for my intel card) that I need.
Is it possible to install the nvidia opengl drivers alongside the ones I have installed already? There seems to be a flag in the NVIDIA*.run script that allows me to change the installation directory for opengl files, but I don't know if this will mess things up or not work at all.