Lightmapping

check the device caps, if it supports multitexturing, then use it, otherwise have a backup method that uses 2 passes.

Well, I have implemented both solutions and so far I see the multitexturing one is the fastest.
For testing purposes, and since there are no exporters I know of that can output two coordinates sets form 3DS MAX to an .X file, I am loading two meshes (one textured normally and the other with the lightmap) and making another mesh in memory by merging the other two.
But now I have a problem. If there is only one lightmap for the whole mesh, it''s easy - I just set the lightmap texture in the second stage and render using a SetTexture() - DrawSubset() loop as always. But what happens if I have more than one textures for lightmapping? How do I go about applying the textures in both stages and calling DrawSubset()? Both textures group different polygons!