Well from the way I understand it, I've always been using box modelling. I use Lightwave (LW. Box modelling in the general term, is where you start out with a very basic shape (not always a box) and you cut, smoothshift(LW), and move points around building outward as you go, and then detailing. The Sub-D Surf that to a final model Bill Flemming (from what I've read in articles; one here even) draws the model in polygon's 1-dimensionally. Then he pulls those vertices out into 3D space. Kinda like if you took half your model and flattened the point values to 0 on one axis. Seems like a complicated way to go about it, and hard to get multi-limbed figures his way; without a lot of planning and welding, or to build things made for animation. Some people also use the per-polygon method. There are examples of that for LW owners in Dan Alban's books. This is where you build sections of a face poly by poly, then weld those sections together. Its good I think for areas that are going to ahve a lot of detail, and need a certain style of topology. I've used it before. ML