I haven’t written much technical about Maya in a while and this question by T. I. Burbage on Steven’s Creating and Managing Assets was one that stood out to me as a good place I could offer some insight.

At a high level, Character Sets (aka Character Nodes) and Assets are the same. They both provide a simplified interface to other nodes.

The big complaint I hear about Character Sets (CS) is how the connections are made. Take this example of a cube. You get connections from the transform to the CS and back. In the Channel Box, the transform attributes on pCube1 all have connections so it can be tricky to tell what kind of animation is happening (Constraint? Keys? Expression?). You have to look at the Inputs, the CS, to tell (in this case TX is keyed and TY has an expression).

The main difference, and advantage with Assets over Character Sets, is the flexibility and additional controls Maya makes available for controlling your set-up.

Now, if we look at the same example of a cube, but as an Asset, we can see some differences.

It may be arguable but I think the graph looks a bit cleaner with Assets, especially if you have the Container collapsed (I’ve left it open here for clarity). The big thing is that the connections are different; the connections don’t go two ways like they do with a CS and this means that if you look at the Channel Box, it’s clear what kind of data is associated with each channel (Keys on TX, Expression on TY). I also like that when adding a node to a container, the shape nodes go by default - again, I think this makes the graph a bit cleaner. In terms of editing the attributes, Assets and CS both allow you to change values on either side of the connection, but the Asset really keeps the display more clear.

You should notice that the Container has the cube as an input – it’s opposite with the CS example as the CS is the input. The container is actually an output of the cube but Maya has a Channel Box option, under the Show menu, to put Assets at the top. Think of this as a way to reinforce  that an Asset is a high level view of more complex data.

One of the things I’ve done with the Asset, is to publish the Translate attributes’ names as Move. This is much easier to do with Assets than with CS. And if you don’t like the name Move, you can easily change it to something else after the fact.

Where the real power comes in with Assets is the Template files. As you build up the interface, you can save that to a file and reuse it. When you’re rigging a hand, you’ll usually do it the same so why not reuse what you’ve done before?