Rigamajig: Elevating Rigging Workflows with Modular Tools

Rigamajig2 is a modular rigging system I developed over the past year. As a technical artist, I think it's essential to help find ways to iterate quickly and maintain flexibility.

In this post, I'll share a bit about my process of developing rigamajig2 and share some specific examples of how it has benefited my rigging workflow from projects this year.

Rigamajig

While the powerhouse of the tool comes from its library of components (pre-built rig parts, like arms, legs, lips, etc.), at its very core Rigamajig is a sequential script runner. Each rig is comprised of data saved in JSON files and rebuilt in Maya to construct a rig.

Case Study: The Last Dungeon

Responsible for: Rigging | Modeling | Pipeline

The last Dungeon was a collaborative project I worked on that utilized the unreal engine. From a rigging perspective, this posed several challenges. Not only were there five characters to rig, but the rig also had to be primarily joint-based!

Note: I was responsible for rigging the Paladin, Skeleton, and Barbarian Face. However, all rigs were made using Rigamajig2 and the pipeline I developed.

Using Rigamajig, we could make a new archetype for the project. Since everything is saved externally, everything is sharable between characters. From there, we only needed to adjust the guides to fit each model.

I wrote some new joint-based components for the face. This included a new eyelid, eyebrow, and lip setup.

The images below show some range of motion tests using the rig.

Additionally, unlike traditional renders, Unreal doesn't have any options to subdivide at render-time. Using Rigamajig allowed our team to create two versions of each rig:

The rig-deliver was built and published as a Maya file for animators.

The actor-ue was nearly identical, but upon publishing. We would subdivide the model and transfer all the deformations to the new model.

By keeping all the joint, mesh, and blend shape names the same, once imported into unreal, we can apply animation clips from the rig-deliver onto the skeletal mesh of the actor-ue .

This became our solution to 'subdividing' the models inside unreal.

You can see the difference in the slider below. The differences are subtle but noticeable around the edges of objects.

Within Rigamajig we created two '.rig' files. Below, you can see the rig files for the rig-deliver and actor-ue. Notice that only a few lines are different to create the smoothed actor.

Case Study: Goro Goro

Responsible for: Rigging | Modeling | Pipeline | Lighting

Goro Goro was another project I completed this year using Rigamajig2. One of the challenges of this project was maintaining appealing shapes through a wide variety of extreme shapes.

One of the biggest challenges involved shaping the character for different poses. For this, I developed the Deformation cage, a post-rig deformation system that not only allowed for lattice-like deformation of the characters. The system is optimized for Maya's parallel evaluation, it added almost no performance hit to the rig. (60-70 FPS)

Finally, here is my most rigging demo reel showing Rigamajig used in production!