Path: chuka.playstation.co.uk!news From: "Alex Herbert" Newsgroups: scee.yaroze.programming.3d_graphics Subject: Re: Maximum efficient size of TIM's to be texture mapped Date: Sun, 7 Nov 1999 14:22:12 -0000 Organization: PlayStation Net Yaroze (SCEE) Lines: 34 Message-ID: <804179$5f54@chuka.playstation.co.uk> References: <01bf25a2$8d6ca320$f1638cd4@paladin> NNTP-Posting-Host: 212.56.110.67 X-Priority: 3 X-MSMail-Priority: Normal X-Newsreader: Microsoft Outlook Express 5.00.2314.1300 X-MimeOLE: Produced By Microsoft MimeOLE V5.00.2314.1300 Hi Daniel, Yep, it's better to stick to those maximum sizes if you want any performance out of the GPU. If the texture doesn't fit into the GPU cache, the GPU will have to keep reading from VRAM which is much slower than the cache. I don't know where the 30% figure came from as it really depends on the size of the texture, the size of the polygon being drawn, and the extent the texture needs to be rotated. It could be less than 30%, but it could also be much more. I seem to remember that there's a fair amount of info in the Programming FAQ on the NY site. Herbs Paladin wrote in message news:01bf25a2$8d6ca320$f1638cd4@paladin... > > I think I remember reading somewhere that textures to be mapped onto > polygons should be: > > 16 bit = 32 * 32 Maximum > 8 bit = 64 * 32 Maximum > 4 bit = 64 * 64 Maximum > > otherwise the operation will take 30% longer, since the texture doesn't fit > within the GPU's cache or something, does anyone have information on this? > > thanks in advance Daniel J. Dobson