i’m curious, what’s the difference between the streamed and non-streamed kernels? do you immediately start working on chunks of image as they come in over PCIe instead of blocking on the full transfer?
i’m curious, what’s the difference between the streamed and non-streamed kernels? do you immediately start working on chunks of image as they come in over PCIe instead of blocking on the full transfer?
(it looks super cozy fwiw)
(it looks super cozy fwiw)