diff --git a/handouts/src/29-finish-him.md b/handouts/src/29-finish-him.md index edda049bb7f72a091bcaa9390abe77b7d9736744..28f4ce7c43758dca0ad41e3a19ff8cf433ee39e0 100644 --- a/handouts/src/29-finish-him.md +++ b/handouts/src/29-finish-him.md @@ -24,8 +24,8 @@ are some possible tracks that you may want to explore: worthwhile, and we should test this. Stop using samplers for input images, and instead try both if-testing in the shader and a strip of zeros on the edge of the GPU image. -- While we are at it, are we actually sure that images are actually faster than - manual buffer management, as they intuitively should be? It would be a good +- While we are at it, are we actually sure that 2D images are faster than manual + 2D indexing of a 1D buffer, as they intuitively should be? It would be a good idea to do another version that uses buffers, and manually caches data using local memory. This will likely change the optimal work-group size, so we will want to make this parameter easily tunable via specialization constants, then