Pose to Pose vs Layering
I have only recently begun tinkering with Maya and animating in 3D, but I have immediately noticed my strategies for animating have to change. In 2D, animation is usually created pose-to-pose or straight ahead, while these exist in 3D another mainstream (maybe easier) method is called “layering.” I had noticed my pose to pose method wasn’t working as well as it usually had in the past, but it took a guest speaker presentation to spell it out for me.
Let me back up for a second… pose-to-pose animation, is when the artist creates keyframes to establish the movement of the character before meticulously filling in inbetweens. This is done to make sure the movement is correct and the character’s poses read well. The best example of this is the walk cycle, the animator would “key” the important poses (shown here) then when the movement looks correct they go back and add the inbetween drawings (the lighter colored circles in the image represent these)
“Straight ahead” animation is when the artist creates the animation straight through without the use of keyframes. When animating longer movements this can result in characters accidentally growing and shrinking throughout their performance. I would us this method primarily for effects such as water or sparks or any small objects that quickly dart around the screen where keyframes are not particularly beneficial.
Layering is almost exclusive to rigged characters. It has to do with animating the primary form of the character (sometimes called the “root”) through its action, then going back and animating its extremities. Think of animating a fish, the artist would animate the body along the path it travels first, before going back and animating the fins and the tail. The same applies to a person walking, the main body of the character would be animated first (the path through the scene along with the bounce of every step) then the artist would go back and move the legs, then go back and move the arms.
The benefits of layering, is after roughing out the movement, the majority of the animation is complete, all that remains to be done is details and secondary movement. This is different from animating rigged characters pose-to-pose because the computer tends to try to create the inbetweens for you, which generally results in bizarre movements as the limbs try to find their way from point A to point B along with poor timing. Not that this cannot be cleaned up later… One of the drawbacks to layering however is the poses tend to be less dynamic and may require further tweaking after the shot is finished.
These are just some observations I felt were worth jotting down. Hopefully they provide a new strategy that can be applied in future projects when a rigged character isn’t animating quite right.