I have just started programming with CUDA after a go at opengl and cg. When I made a simple opengl programme, I found I was unable to run two different applications using opengl at the same time. Does this sound right? It may have been a setting that I was unaware of. If so, does CUDA have this limitation?