Maya Rigging: Single Skeleton vs Multiple Joint Chains

Recursive Sweatpants · May 2016

I really enjoy making and tinkering with rigs inside of Maya, but something has been bugging me lately. Most of my rigging knowledge comes from this tutorial from the Maya Learning Channel:

https://www.youtube.com/watch?v=MV4XRgmTynY
It advocates using separate joint chains for different bodyparts, as opposed to keeping it all in one skeleton hierarchy.

However, I've also seen videos such as the following rigging demo reel which shows a fantastic rig but also points out that the mesh is bound to a single skeleton:

https://www.youtube.com/watch?v=zvUfiKQI5jQ

I would like to know more about the pros and cons of having a single skeleton versus keeping my joint chains separate. I know some basic things, such as keeping joint chains separate allows for modularity and easy space switching, while using a single skeleton is good for when you want to export the character to a game engine or an animation to Sketchfab (which I have learned the hard way that it does NOT like multiple skeletons), but what other advantages and disadvantages have you guys found with using one method or the other?

GlowingPotato · May 2016

Well, for games, you will (most of the time) want a single chain, not sure about movies and such.

m4dcow · May 2016

I mean for stuff like games, you have the skeleton that is bound to the model, that will make it into game. You can still use your preferred modular setup though, except to drive the export skeleton that's what IK/FK setups do, and then you bake the keys down to your export skeleton. Admittedly I haven't done any rigging in years and with all the new technology that might not be the way it is done these days.

Add3r · May 2016

The idea behind a single skeleton is that some engines will use the origin as a tag point for FX and other content. It really depends on engine, the pipeline for engine, etc. Some engines can take advantage of the benefits of a multi chain rig, but a lot of in-house engines will usually require a single chain with specific Bone0 locations to use as an "origin" for the mesh attached and the rig itself (including anims). Example of engines that require a single chain would be the publicly available Radiant engines used for Call of Duty development (though modifications can be made on a per asset basis if something specific is needed). I have, however, witnessed a multi chain rig work really well for a modular character system.

Like @m4dcow said, I am also a bit behind on times with current rig tech, primarily working with an older system like Radiant over the past years, so read up on documentation for the engine that this rig would be used with

Seems to be flexible, and very circumstantially based.

Recursive Sweatpants · May 2016

I see. So maybe I could bind the mesh(s) to a single skeleton hierarchy, but drive those joints by using a modular multi-skeleton setup on top of it. That way, I could have all of the convenience and utility of animating with complex rigs in Maya, but still be able to export animations and skinned characters to other platforms by baking all of the animation onto the bind joints. Do you guys think this is a good idea, or are there drawbacks to it as well?

Maya Rigging: Single Skeleton vs Multiple Joint Chains

Replies