True. I think this is what's bugging me: drawing a lit version implies information about 3D shape; in particular, the artist has to reason/see in 3D. So, in a sense, it is a new method of entering 3D information, that works especially well for traditional 2D art styles.

Why not have the artist enter height directly, instead of shading from several different angles? It seems like less work (because there's less information to input); though possibly doesn't mesh as well with how 2D artists work... whereas shading is part of the tradition.

There's something I'm not getting here (that might lead to a better way of doing it, or not).