Interesting. Well that is great to know.
I have been playing more with my problem and made some discoveries. I changed all the code as you suggested and removed all the java math operators.
What I had originally was, for example, A head composition made up of 6 frames and 8 layers. Each frame had a different head composition ie.front,3/4,side…each taking up there own layer. The top two layers had two eye compositions. The eye compositions consisted of an eye white, a pupil and an eyelid layer. The pupil was coded to follow a little square in the root composition.
What I was trying to do was control the timing of each composition through code. So if the slider was set to 3/4 view the head comp would be on frame three, while the eye comp would be on frame 6 and the mouth comp would be on frame 25.
It worked when I was on frame 1. I had a perfectly interactive little puppet controlled by sliders and everything. But once a single keyframe was applied, everything stopped working.
What I have found is that my head and eye compositions only work if they have as many frames as the main compositions. It seems that without frames to play in, the code doesn’t work….
I don’t know if this is making any sense.
Anyways, thanks for the help so far. It is too bad the code can’t function independent of time. If you have any thoughts on how this might be possible, or maybe I am missing the boat entirely, that would be great.
Thanks Dan