Can I do all my smartskinning in one window? Are there drawbacks or advantages to doing it that way?

I usually work in a test ACTION. I move things around, testing the deformation, weighting, etc. When I get to an angle where I want to adjust the mesh, I select the bone that is driving it, right-click, and Add or Edit the Smartskin relationship.

I have a video tutorial about it, somewhere....I used to have it on my online store, but I think it is on one of the Siggraph demos too.

You have to do a separate smartskin for each bone that needs smartskin, if that's what you mean.

Yeah, that's the question, but do I need to create a new relationship for each bone/smartskin?

Each bone (that needs one) will have it's own smartskin relationship(s).

Edit: Upon re-reading your post I can see that this is not what you meant however, I'll leave it anyway for comments/questions.

I think you mean dividing the smart skinning up of the same bone into smart skin 1, smart skin 2, etc.... I have done this to keep rotational axes, touch ups, and,d the keying of CPs and the keying of bones separate (note: placing bone rotation/translation in a smart skin did not mirror... at least in version 14). It allows you to organize these things as well as turn off and adjust the percentage of enforcement on grouped items however, I don't really know if there are advantages/disadvantages beyond this. I guess it would be easy to test whether or not multiple smart skin keys affect each other differently when in a new smart skin.

r