Hi all, I need your graphic art knowledge for my current spare time project: I'm writing a raytracer, and while dealing with texture coordinates generation I came up with two different options for meshes: the first is a bit cleaner and transparent to the rest of the system, but allows for only one uv set per mesh. The second requires a specific texture handler, but allows for a different set per texture applied. The difference in terms of lines of code and complexity is quite low, but requires to move some code from the intersection routine to the texturing code, so I ask: do you graphic guys use multiple uv sets per mesh or do you simply create the textures over the single uv set available? What do other applications? Would you find usefull to can attach a uv set for each texture? Thank you!

Most artists' in my experience get by quite fine with a single uv set. Normally multiple uv's are used for special trickery, like blending baked gi maps over an already textured geometry.
But, for a raytracer i am not sure if such trickery would be needed. So, i'd venture a guess that one set of uv's would be plenty. But then again, it never hurts to give an artist a bit more to play with.

Thank you for your answer. If there is even a small usefulness for more us coords sets then I will most probably support this... I will think a bit more about the architecture in order to minimize the difference between meshes and other primitives.