This is a presentation I gave at Microsoft Studios concerning my experiences as a character rigger
both before and after joining Microsoft. It describes the unique challenges faced on a publishing team with several
offsite rigging projects and how I overcame them.
Evolution of a robust rigging toolset
Creating better characters faster
Topics:
My background and experience with rigging
Automatic solutions vs. manual rigging
How to get the best of both worlds
Projects that prompted my new toolset
Looking forward
Philosophy on Character Art:
A well-designed and implemented rig can have as much impact as a great sculpt or a beautiful concept drawing.
Everything is Interconnected:
Design
Anatomy
Kinesiology
Topology
Deformation
Sculpture
Making a Rig into a Character:
Gather input from Art Directors, Concept Artists and Animators
Design the rig for acting and emotion, not just motion
Test!
Must easily hit the characteristic poses of the design for the shot, cut scene, or game animation
Good Character rigs:
Are based on real life anatomy
Are designed to emote
Have layers of control
Are fast
Are FUN!
Fun???
A fun rig is:
Inspiring
Encouraging of experimentation
Easy to use
Previous experience:
rigger on commercials and television series
Created rigs by hand
Customized to shot/job
Takes long time
Prone to user error
Requires a very experienced rigger
Inconsistent for animators
Scripted automatic rigs
Very fast to create
Typically limited to one type of character
Eliminates user error
Anyone can use it
Consistent
Microsoft Games Studios (MGS) Projects:
Questions:
How do I help a studio choose between a manual or off-the-shelf rigging pipeline?
How do I deal with the constraints of a short production deadline?
How can I lend my experience to an off-site staff?
Case Study: "Animals" Rigs
Problems with one of our partner's quadruped rigs:
Their first foreleg attempt: Can't create a complete gesture from a single control
Their second foreleg attempt: Used biped, which is not built for animal locomotion
Their face rig: No rig heirarchy and no pose library
"Animals" Research:
Anatomy of a quadruped
Scapula slides over ribs
Walks on Toes
No clavicles
Forearms twist
Flexible body
Can walk in crouch
"Animals" Solutions:
Write scripts to rig quadruped components:
Proper animal leg set up: Gesture is complete from scapula to toes
Fun! Acts like a cat with few controls, encouraging play and experimentation
Layered facial controls: Elastic influences allow for natural stretching and motion
"Animals" Take-away:
Needed the advantages of a manual rig because biped was not built to handle this specific situation
I do not have enough time and resources to rig and support characters for multiple projects at the same time
I have to create a rigging toolset!
Case Study: "Alan Wake" Rigs
Dog Rig
Reuse a lot of the "Animals" rig scripts
Raven Rig
Not a quadruped rig, still reused some scripts
"Alan Wake" Take-aways:
Anatomy between a dog and cat is similar and could have used a similar rig
Anatomy between a raven and a dog is dissimilar, but I still reused some scripts from Animals
My toolset should be modular!
Toolset:
Modular
Customizable
Flexible enough to be manual or automatic
Breakdown:
One maxscript that contains all functions used to rig a character
One macroscript that contains a rigging UI and macro recorder
Usage:
To generate a rig macro:
Using the provided UI, create rig components manually
As you use the UI, macro commands are created in the UI macro window that can be edited
Save the rig macro for future use
In this example, I create a simple finger rig, recording as it is created.
Then, I duplicate the recorded macro for the other fingers of the same hand.
I am then able to duplicate and mirror the entire hand to the opposite side.
To create a new rig from an existing macro:
Load it
Run it!
In this example, I load a rig template and a rig macro.
I run the macro, which creates the character rig.
I can then undo the rig creation and edit the template to fit a new character.
The rig macro rebuilds the rig to the new proportions.
At the very end, I show a dog rig example using the same toolset.
Advantages:
Custom rig macros for a specific job/studio can be reused, edited or built upon later
New rig components can be easily created through the macro recorder, no scripting required
Rig templates can also be created, recorded and saved using this toolset
Avoids reliance on external plugins or specific software versions
Easily used by riggers of any skill level
Allows for quick iteration and experimentation
Summary:
How do I help a studio choose between a
manual or automatic rigging pipeline?
Why choose? One toolset can support both pipelines.
How do I deal with the constraints of a short production deadline?
I can easily and quickly combine existing rig components into a new customized rig
How can I lend my experience to an off-site staff?
I can provide rig macros that they can apply to their cast of characters without my further involvement
