I will soon start to work on a project whose aim is to develop a CUDA program for simulation on Nvidia GTX 980 card. I'm new to CUDA programming, so before starting with the project I wanted to find out if there are any factors which would affect the development or subsequent execution of the program.
As far as I could see, Nvidia has provided drivers, development environment & debugging tools for both Windows and Linux. But in case of program execution, I came across some discussion where it was mentioned that CUDA program runs a bits slower on windows compared to LINUX.
Am I right in assuming that ease of development should be the same for both OS but CUDA program execution being faster on Linux? Are there any other difference between CUDA programming on windows vs Linux?
Any and all advice are deeply appreciated.
Aucun commentaire:
Enregistrer un commentaire