I can not see a real advantage to use lightmapping.
I textured a square prop in my Poser 4.
I added the lights, I rendered it an finally I exported the .bmp file, to be used as a normal texture,instead of multitexturing in the opengl engine.
It should be better.
Am I wrong?
nothing wrong with that, if you want to have a different texture for every single polygon in the whole damn scene.
lightmaps were designed because 1) they take up bugger-all space in memory (most lightmaps are 64x64), and secondly, you only need one texture which can be shoved on to of any other texture in the scene (wood, metal, etc).
twice in the same day i end up posting this link:
somewhere in there is a couple of cool demos on lightmapping. read them, and be amazed at the amount of textures that are re-used, the color-changing lights, and more importantly, moving lights.
you might as well pre-render the whole scene from every concievable angle and position possible.
oh, and no-one said you had to multi-texture. look at the tuts to see what i mean.
MENTAL explained what light mapping was, now I''ll go a bit further. Light mapping is *ONLY* for static objects/scene ambience-type stuff. What you want to do, is figure out a way to precalculate all the lighting for a scene (using a Radiosity solution is the best way I can think of), and then use light mapping to accomplish the actual lighting. There are ways to make lightmapping work for dynamic objects, but overall, its not worth it. You''ll have to figure out your own way to light those objects (like using OGL''s 8 hardware dynamic lights).
If you wanna research Radiosity a bit, check out the reference section on this site...