Graph Editor Tricks that I have found my self using. One is for when I am animating in stepped and the Other trick is for when I'm animating Layered. (stepped is when your graph editor does not allow your splines to curve and keeps looking like steps, layered is a way of animating when you FEEL the animation and its timing from the root out)
I'm not here to debate which way of animating is better, I'm not about to regress into a school girl type of shouting match. I'm just going to give two tricks I find my self using often and the help me out a lot. Actually.. they've saved my ass more time than I care to admit :)
So.. lets say you are animating Layered, so happy that you are getting results super fast and you even dare to think you might actually finish before the deadline. But soon you perfect start disappears and soon you animation is muddy, your timing is wrong, your takes melt into each other and nothing is left of your initial timing! You were almost done but now you don't know what to do. There are hitches in the arms, the nose arcs and you even discover you have problems in your root! What are you going to do?
What I do is:
|I do hope this makes sense. The numbers at the bottom are frames on the time line.|
So I kind of start over but keep my 'blue print' of what my animation once was. I use the buffer curves (ghost curves) to guide me. I go through each curve pretty fast and double check its right and working well with the curves before it on the hierarchy. Doing this process saves more time than actually trying to fix everything while all the body parts are moving. And most of the time you don't even know that the curves you are fixing are responsible for the pops or bad timing in the first place. So that's why once you remove all the curves from sight and slide them back one by one, you easily come across the bad curves that were messing everything up. Its a bit tedious but it goes by fast. and you know once you are done, you have done everything you could have done and that leaves you feeling satisfied.
The other situation where I use one of my tricks (well I wouldn't call it a trick, its more a piece of advice) is when I'm animating in stepped. Usually I'm pretty far along with my animation and I'm ready to start splining and making my curves smooth when I realize when that making my curves smooth is LOOSING all my timing etc. Now I'm faced with a choice: Do I continue smoothing out my stepped curves and FIGURE it out like how us good little animators were taught?? Or do i say fuck it and just keep animating in stepped?
I say Fuck it!... bad choice of words danny.. bad choice..
BUT DANNY why would you animate like that? I don't make this decision out of laziness all the while as I am adding breakdowns I have my eye on the graph editor watching the curves FORM as breakdowns and poses are added and added. once I can see the curves develop more defined, then I will start to clean them up (all the while in stepped) Eventually my animation will be on 2's 3's and 4's here and there. This method keeps my timing and I spend more time animating than fighting with the graph editor. I never allow my graph editor to become messy and I never ignore my curves. I baby sit them the whole way still using as little controls as I can while getting the most out of them.
These are my methods, they are not wrong, they are not right, but they allow me to finish my animation. It works for me and I encourage that if these methods do not seem to help you, you will search for a way that does. Many times I see a lot of great blocking and Ideas turn into mush and then becomes forever lost in the graph editor.
Internship deadlines are coming fast in the spring, I wish all of you who are applying to places the best of luck and happy holidays! I guess I should mention here that I will be trying out tutoring after the new years. Not online tutoring but a real legit person to person hands on tutoring. I'll see how it goes. It might be cool it might actually be a utter complete business failure :) I'll do a post in a few days with more details about my plans for this... experiment!