Keith Whitwell wrote: >> I'd guess it has to do with the _ae_loopback_array_elt function not >> being called anymore, the single-vertex-emit fallback really had a >> high-impact (mostly because of the too large arrays) - for some reason >> this fallback seemed faster in mesa 5, but with vbo it's gone (memcpy >> shows up rather prominently with oprofile now...). I'd have thought >> though that with swtnl it would have been more math limited rather than >> just limited by function calls... > > It's possible that the little vertex cache in the split_copy routines > helps in this case as well. I wonder if disabling that makes the > speedup go away. This seems to help somewhat indeed, though only very small sizes seem to make a difference - from ELT_TABLE_SIZE 4 upwards performance pretty much seems to stay the same (for the ut2k3 flyby, I tried up to size 256). Performance with ELT_TABLE_SIZE 4 and larger is around 80fps (this is btw a bit more than yesterday's result, not exactly sure why, maybe I had some background app running). This goes down to about 67fps with the vertex cache code disabled. Even a size of 1 already helps.
Roland ------------------------------------------------------------------------- Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT & business topics through brief surveys - and earn cash http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV _______________________________________________ Mesa3d-dev mailing list Mesa3d-dev@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/mesa3d-dev