Am 25.10.2013 um 14:12 schrieb Henri Verbeet hverbeet@gmail.com:
This kind of construction doesn't generally make me very happy, but on the other hand I'm not sure I can come up with something that's much better either. I do think we should mark the alternative results as broken. Some of the Evergreen results in later patches in this series are a bit surprising to me. I wrote the test on Windows 7 / AMD Cayman, which should be very close to Evergreen hardware.
I was thinking about marking some of those results broken, but in the end I couldn't find an argument why the results from GPU A are more correct than the results from GPU B, especially when considering that the differences between ddraw versions on the same GPU are just as big and strange as the differences between GPUs.