The next time you do a radeon vs Catalyst comparison, please include GPU usage for every benchmark. It'd be the first online resource showing clearly the pain points (where Catalyst manages to be GPU-limited but mesa is CPU-limited). With it one can easily direct attention to where needed.
I'd also like to see a test with the hotness kernel+mesa, but I'm not sure if that'd bring a lot of visitors. A single high-end card would be enough, like 6870, as it would complement my own low-to-mid-end testing. Even at full VRAM, confirming that there are no regressions on a high-end card would be helpful, and a few rounds with different limited VRAM amounts would be nice in addition.
I'd also like to see a test with the hotness kernel+mesa, but I'm not sure if that'd bring a lot of visitors. A single high-end card would be enough, like 6870, as it would complement my own low-to-mid-end testing. Even at full VRAM, confirming that there are no regressions on a high-end card would be helpful, and a few rounds with different limited VRAM amounts would be nice in addition.
Comment