Eric Pierce, Infinity Ward's CoD4 Rigging Tools

Hey folks,

I was just watching the Autodesk GDC 08 video from Infinity Ward, and they showed all the cool rigging and animation management tools that Eric Pierce wrote for Call of Duty 4. The slide on the screenshot is pretty small, I can’t make out much other than the main panels there, and it looks really interesting.

Was wondering if anyone knows him, or if he reads these forums already… and would he be able to comment on those tools at all or break them down?

Additionally, has anyone else written animation tools similar to this, and would they be willing to share thoughts on how they went about it, and what sorts of things they found really important and useful?

Cheers,
-MoP

Well, minus the moCap portion, I wrote some pretty similar pose tools at the studio I was at previously. I’ll have to dig them up so I can show some screen shots if you want.

The pose system was separated into face and body UIs. The face UI consisted of a main section that held all of the controls for the face. Off to the side were a camera view which had focus on the face at all times, and a pose icon which showed a preview of the currently selected pose. Below were a number of list controls which held the poses (full face, mouth, eyes).

The body UI was basically the same minus the face controls… with a camera, a pose preview icon, and a number of list boxes for various parts of the body (full body, upper, arms, hands, etc…).

Each UI had a slider that controlled the blend amount of the loaded pose. I think the blending played a much bigger roll in the facial poses than the body. It made more sense to load an expression or phoneme at say 50%, than to load a walking or running pose at some arbitrary value. But it was available for both, just in case. :slight_smile:

One important feature was having the preview image that would update based on the selected pose in the list. It was up to the animators to hopefully give the poses descriptive names when saving, but that didn’t always work as planned, so the image added a secondary description without having to actually load a pose to see what it was.

Other important features were the ability to completely flip a full body pose, and to save a pose to one side of the body and be able to load it to the opposite side. I think these assisted most when it came to creating walk and run cycles, and other navigational animations.

Later on in the tool, I added the ability for the user to choose to work locally or off of a predetermined network folder. This allowed the animators in their day to day work to save any and all poses they wanted. If needed though, a lead could set up an official set of poses and save them to a network folder. This would assure that all the animators were working off of and loading the proper poses. When the user would switch to the network folder, saving was disabled in the tool and only loading allowed.

So yeah, that’s the basics of it. I’d love to hear what other people have done with similar tools!

I do not think so, the truth often lies in the hands of a minority, to the future we will think this is wrong.