Travis and Beyond Myth are right. Personally, I like to add the details in my nurbs and then convert to the lowest number of polys that will allow me to keep the majority of the modled detail. This is the one drawback and bonus to nurbs...you can convert your nurb object to however dense a poly you want BUT you have no control over where those polys are placed. I think Travis forgot that. :) Sub-Division Surfaces allows you to place details only where you need them, thus resulting in a lower poly model with maximum detail. We (meaning Rhino users) don't have that choice, we have no control over where the polys are placed really. That's why it's important that our nurb model be as clean as possible, so that when we convert to poly the detail isn't lost as much and the poly count remains as low as possible. In the end, it depends on what the model is being used for. As Travis pointed out, in animation, you need as light a model as you can get which means all the work is done with textures. If you're making a still image, however, there's nothing wrong with a highly detailed model. Especially if it's the center of attention for the image. If you're making models for Poser then you are maki ng models for animation and must keep your poly count as low as you can...this means ddamn good texture/bump work will be needed for all those details you can't modle into it but still want.