Just a quick question - I have a 640x480 24bit RGB webcam feed I'm trying to process with some OpenCL. I'm trying to use RGBA OpenCL 2DImages, but I'm a bit confused how to go about this. Let's say I copy my 640*480*3 byte image to an opencl RGBA 2DImage with a RGBA, CL_UNSIGNED_INT8 format. Isn't the indexing messed up now? (The fourth byte is technically the red channel of the second pixel, but it's indexed as the alpha channel of the first pixel.) Then, if I take the normal approach and get global x and y IDs in my kernel to do pixel-by-pixel processing, I'm actually grabbing a pixel and a third, right? I feel like this must be a common problem, so has anybody had to deal with this before?
Thanks in advance,