This is a bit of an odd question, but I'll try and explain it best I can When you see all those quake3, Unreal Tourn... etc. models, how do the artists know how to draw the textures to map onto the models? What I mean is, how do the programmers know to map the 'hand' part of the texture onto the models hand, etc? I thought of one way: If you take, say, a cube, you could 'unwrap' all of the points onto a plane, like so: (please esxcuse the sketchy ASCII art)

        ********
        *      *
        *      *
        *      *
************************
*       *      *       *
*       *      *       *
*       *      *       *
************************
        *      *
        *      *
        *      *
        ********
        *      *
        *      *
        *      *
        ********

 

The artist could then use this as a template to draw, so when the texture coords are mapped, everthing appears in the right place. I know that you can 'squash' models onto a single plane and I think it has something to do with the OpenGL Projection matrix Thanks in advance to all who reply Ferrit Edited by - ferrit on March 25, 2002 2:00:42 PM

Us programmers leave that up to the artists :o). Basically they have a program that they use to map the U,V coordinates to their model. The 2 ways are... A.) Wrap a texture around the model, and it gives you a template to work on (outlines each triangle on the 2d texture map) B.) You make the texture map and set the coordinates on the model (ussually gives you some 2d view with the triangles layed out to place on the texture, so you simply click/drag them in 2d to their proper texture coordinates). Different artists/programs have different ways of doing it, but us programmers tend to leave it up to them and load the texture coordinates from the model file :o).

Billy - BillyB@mrsnj.com

Ferrit, one popular program for doing what you describe is lithunwrap, available at http://www.geocities.com/lithunwrap/ Check it out!