Hi Andreas,
I'm not meaning to tell you that you should use CL, I'm just trying to make sure you base your decision on accurate information.
[...]
MS have released a GPGPU language, it's called DirectCompute, and mainly for use with DirectX in games. (Were they planning on releasing another one?)
Remind me next time I research A vs B vs C... that I use Google's new "Past year/month" filter :) Thanks for the heads up. I'll at least consider OpenCL now. I heard horror stories of 10x slower, but I probably need a larger sample size ;) Actually, just one thing I'd like clarified: Maybe it doesn't matter that I can't really redistribute nvcc with my program, because using PyCUDA, I can utilize run-time generated code anyway. The important question is then how much slower would that be compared to precompiled GPGPU code, as if nvcc were always present to use. Regards, Daniel _______________________________________________ PyCUDA mailing list PyCUDA@tiker.net http://lists.tiker.net/listinfo/pycuda