2 ms·

Is NVidia's (CUDA) GPU library development aimed at anything other than proliferating their proprietary lock-in?