I learned something interesting about quick timing the other day I’d like to share. I don’t have a visual example (which prevented me from posting about it before) but I thought I’d post the thought anyway before I delve into it deeper at some point. In the meantime, here’s what I got:
The time in which an action occurs stays constant, regardless of how snappy or realistic the timing is. I.e. the combination of anticipation, movement, and follow through of an action will be the same number of frames no matter how realistic or snap-cartoony of a motion it is.
Stick with me now while I give some background.
At CTN I picked up a great DVD of some old WWII instructional shorts, the Snafu series. They were made by Warner Bros. Directed by Fritz Freling, Chuck Jones, and other notables. One of the producers was Ted Geisel aka Dr. Seuss! They’re pretty entertaining (and far more risqué than your typical fare – only soldiers were going to see them so the normal kid-proof rules didn’t apply) and fun, but more importantly they have that classic WB timing and were made quick enough that the flaws in the craftsmanship help reveal the tools the animators used to put them together. They also have commentaries from Eric Goldberg amongst others that are pretty insightful.
Anyway, I was studying the shorts for the cartoony style and counting frames for the timing. In one short a Japanese soldier and Snafu are in a tree and finally see each other. They sloooowwly anticipate moving, then pop out of frame leaving behind streaks. I counted the frames for the streaks and found out that although their bodies went off-screen in a frame, the streaks stayed behind for a full 8 frames. I looked at other examples and saw the same thing – the streaks remained behind, dissolving out of view for about how much time it would’ve taken to get them off-screen quickly in a more realistic style. The more the move was anticipated, the quicker the move could be. When I counted the frames of anticipation and follow-thru for a really quick move like a salute, the streak lines caught up with the hand in what would be the normal time for a quick move in a realistic style. I was taking into account that these shorts were mostly on 2s unless it was absolutely necessary to go to 1s and comparing that to a style that would be mostly on 1s like CG. The time of the move was conserved even if the action happened too rapidly for real life.
Since the project I was working on was mostly in a cartoony style, I tried it out. Before a quick move I really anticipated it and after I did more follow through. It made my timing more snappy but kept the overall time of the actions the same as it was before. When I was directed to snap things up, it made it a lot easier to make the adjustment, almost mechanical – cut the transition time and increase the antic and follow thru.
This was a bit of a revelation for me. I had learned off of a John K post several years ago that anticipation and follow through could actually eliminate the need for an inbetween or transition frame of any kind. He had posted an example of work from one of his animators where the character was doing a dance cycle. Basically the character had his arms in a circle and was swinging them left and right. John K was impressed with the snappy-ness and how cartoony it was. I was impressed that there was no transition frame at all – just the antic of the swing and the follow thru after.
I think this is an exciting concept, and I want to attach a scientific name to it like the Conservation of Motion, but I haven’t explored it enough yet. Just sort of buckshot brainstorming the idea, I would guess that the eye inherently understands the physics of moving from one place to another. We have an ingrained notion of gravity, mass, and inertia. I imagine this kept us alive when estimating how long it would take a predator to run across the plain and snarf down Aunt Sue. Since the eye understands it so innately, we have to give it something to follow to compensate for the impossible movement in order to make it believable. This may be why the Disney animators found out that you can’t watch an 80 minute film of cartoony animation – it breaks the rules too much to hold attention at much more than a novelty (that is, curiosity about something new) threshold which would be between 5 and 10 minutes. Some off-hand articles I’ve read about concentration and breaking up learning into digestible chunks seem to peg the time in the same 10 minute area, so I’d guess that it’s related.
Anyway, some interesting tidbits to chaw on for a while. Cheers!