On Fri, Oct 18, 2002 at 11:49:46AM -0600, Brian Paul wrote: | | IIRC, the test draws extremely small (subpixel) triangles.
Yeah, the original version of the test did that. The current version uses a full-window-size quad, so I think it's much better with respect to the problems you mentioned. In the current version there might be some precision problems involved in establishing the minimum resolvable difference. For example, the NVIDIA drivers fail, but only by about half the least-significant depth buffer bit. I've been curious to see whether other implementations passed, or if the test should be loosened up to allow one more bit of slack. Allen ------------------------------------------------------- This sf.net email is sponsored by:ThinkGeek Welcome to geek heaven. http://thinkgeek.com/sf _______________________________________________ Dri-devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/dri-devel
