This is an exercise I came up with to work on my timing.
I’ve had a problem visualizing timing and found that I spend a lot of time fiddling with shifting around keys. This eats up my time while animating. It causes a problem where the more I look at something in a transitional phase the more my original conception of the shot fades away and I start thinking the transition looks good. Which is bad.
To work on it, I decided to take up Pete Paquette’s challenge to playblast less, working as if we were traditional animators in the days when pencil tests had to be sent to the lab to be developed. I figured a good starting point would be to get a better feel for what the length of frame transitions looked like, using the info from a well known chart.
A simple left to right head turn with a ball head, solid oval eyes, and a cylinder nose. Phear my l337 drawing skillz.
The idea is I’d do the beginning and end pose (a mechanical flip-on-axis. Seriously – phear) and then do a straight inbetween. I’d continue with straight inbetweens until I got around 12 drawings total. Then, I’d read the description of the inbetween number on the chart and watch my example.
When I started doing the inbetweens, I couldn’t resist putting in a bit of drag on the nose and a blink in the eyes. I arced the nose down. I also couldn’t figure out how to do a mechanical axis flip on the key so I had to draw it out. I also did it with a mouse during my breaks at work, so I had extra fun lines to sort out before I started watching. I filled in the nose red to differentiate it from the face easily.
What I had was a head turn on 1s. I went through and looked at the various inbetweens and got a vague idea of the speeds of things. But I got bored. So, I started changing WHICH inbetweens I used. This made a TON of difference. Let’s say I started with a 3 inbetween turn (5f total):
Where K=Key, i=inbetween used and .=inbetween created, but not used and (#,#,#) are the numbers of the inbetweens for clarity
What I tried next was using different variations of inbetweens:
As the number of inbetweens I used got higher, obviously I couldn’t play around as much.
What I’ve represented here is sort of a primitive translation of what charting shows. Keith Sintay had shown my class traditional charts and emphasized how they showed spacing. It was something which I was struggling to really ‘get’ until now.
These all have the same timing (5 frames on 1s), but the feel of the turns are very different. I found the ones I liked rarely used the direct inbetween of my two keys (4), and that if I used 4, it slowed down the perceived speed. It also resembled some of the problems I’ve been having in CG. I also found that a large spacing gap still looked okay if I used 1 or 7. In fact, a middle inbetween was hardly necessary as long as it eased in or out. Easing out of the first key and snapping to the second was less effective than snapping to the ease in of the second key.
For me it was a total success because it illustrated a way in which my spacing could be improved to correct my timing, like making sure the pose directly inbetween my two keys is offset one way or another, or keying a gap in spacing i.e. two keys on successive frames that create the proper spacing for a move. This would in essence allow the computer to do what it does best – fiddly incremental inbetweens – and keep the decision making on the spacing still in my hands. I’d also avoid getting the smack-dab-in-the-middle inbetween which I’d have to eliminate by curve fiddling or brute forcing the inbetween.
Hypothetically, at least. I’ve yet to test it out, but I have high hopes. Anyway, try this little experiment for yourself. Now, Pencil has its problems, like being unable to differentiate the keys once created which makes the experiment a little tedious, organization-wise, but you can’t beat the price. Check it out and let me know if you find out anything cool!