I''m currently writing a Billard game with a more Sci-Fi-ish look to it (glass, reflection, ball are a little liquidious (is that a word?) etc.) But I''ve got a very simple question: What would be faster, a sphere in whole as an vertex array, or an 1/8 of a sphere translated a few times? My first guess would be the sphere in whole as an vertex array, but in my mind I always see the same vertices a little translated used in one array. Maybe you even have an idea to speed it up even more (Display lists are not an option)? Thnx in advance! SNAFU - Sitation normal; All fucked up

I would suggest you to do it however you want first.. and simple is better usually.. make it a single vertex array, finish your project, if it is slow, look for the bottlenecks afterwards.. because it looks like you are trying to optimize sth that won''t be a problem at all. (of course if you are not planning to use a mesh with 100000 faces for a sphere)

hope that helps..