Path: chuka.playstation.co.uk!news From: "Craig Graham" Newsgroups: scee.yaroze.programming.2d_graphics Subject: Re: Caches Date: 17 Aug 1998 09:29:01 GMT Organization: PlayStation Net Yaroze (SCEE) Lines: 41 Message-ID: <01bdc9c0$c2507b00$230b0a0a@Angela1.intelligent-group.com> References: <35C81A30.4ACB@club-internet.fr> <35C82DDB.3E12E37C@scee.sony.co.uk> <01bdc061$2c186980$f30b0a0a@Angela1.intelligent-group.com> <35D2E01D.5CA1@mdx.ac.uk> <6r0vps$n7i6@chuka.playstation.co.uk> <35D43404.282A@mdx.ac.uk> <35D438D2.908C3316@ndirect.co.uk> <35D43F08.36BA@mdx.ac.uk> <35D44CBE.1FA95769@ndirect.co.uk> <35D46843.5DF5@mdx.ac.uk> <35D4750C.A4613436@ndirect.co.uk> <35D47AA8.4E80@mdx.ac.uk> <35D48242.CC86BE26@ndirect.co.uk> <35d51fb9.8182565@news.scea.sony.com> <35D7D900.39D7@mdx.ac.uk> NNTP-Posting-Host: d1-s42-74-telehouse.mistral.co.uk X-Newsreader: Microsoft Internet News 4.70.1155 Robert Swan wrote in article <35D7D900.39D7@mdx.ac.uk>... > Patrick Bureau wrote: > > And I noticed something quite interesting last week when I was playing > > with the automatic subdivision flag. I've got a big 8-bit 256x256 > > rectangular poly among other things in my scene. Here's the results > > by applying the GsDiv flag on the rectangular poly: > > Division amount Frame rate(in hsync) > > None 215 > > 1:2x2 212 > > 2:4x4 198 > > 3:8x8 181 > > 4:16x16 191 > > 5:32x32 270 > > Blimey, those are extremely wierd results :) Ill have a look as well > though, but must admit Im so sceptical that subdividing something into > 64 bits is quicker than not doing so! I dont care if it works though > about having to increase packet space, Id live with that for at 10%ish > increase in speed. If this is always the case, maybe James Russell can > come up with one of his clear and concise answers. Depends on whether you were GPU, CPU, or GTE bound in the first place. Think like this....assume you were running out or GPU (draw) time. Because of the way the GPU works, poly's that go over the edge or the draw area cause a big performance hit. But maybe you had a fair amount of GTE/CPU time spare. Setting subdivision would make more of the display drop out in the trivial reject (all points of a poly outside draw area after local->screen xform). So you'd quite possibly improve performance in some cases (not all by any means) by subdividing. > Rob Craig (possibly waffling, don't quote me, James can answer if he likes (maybe ask Colin?))