In this question "CUDA: How many concurrent threads in total?", it says:
The GTX 580 can have 16 * 48 concurrent warps (32 threads each) running at a time. That is 16 multiprocessors (SMs) * 48 resident warps per SM * 32 threads per warp = 24,576 thread
I am curious: with OpenGL and the same GPU, can we say: the same number of threads (and in my understanding, one thread processes one pixel in the texture) concurrently?
If not, what is the number? How many pixels in a texture can be processed at the same time?
And if the OpenGL on this GPU can have say, 10000 concurrent threads. If we have 10000 pixels in one texture, and 20000 pixels in another texture, can we say: The texture of 20000 pixels need approximately 2 times of processing time than that for the texture of 10000 pixels (regardless of overhead of loading shaders etc)?
