The shots where I get cocky and think "aw - I totally know how to animate that, I'll just sit down at the computer and start animating it" are the ones I'm the most disappointed in when I look back on them. Sure, they got the job done at the time, and they were "good enough" to final for the film I guess, but looking back - some of those are big wasted opportunities and they certainly aren't the shots that would ever make it onto my reel.
For me, the best thing I can do is thumbnail a ton of ideas (to get away from the cliched ideas. The first thing you think of is also the first thing the audience will think of, thus it's the most expected and boring route to take), then bust out the video camera and film a ton of takes. Often I'll do over 50 takes, so I can at least get to the point where I'm no longer thinking about the actual words in the dialogue or certain poses I want to hit, but instead I can just feel the emotion and react as naturally to it as possible.
I know I've talked a lot about planning on cg-char over the years, so I won't bore anyone with the whole process again, (except to reiterate that it isn't a good idea to copy your video stuff frame-for-frame, you should be layering the principles of animation on top of the posing and timing you discover in your takes), but basically my goal is to answer these two questions:
1. What are my key poses, my breakdowns, and the inbetweens necessary to know my arcs and weight shifts.
and
2. What frames do I want those poses, breakdowns, and inbetweens on?
For me, if I put the time and work into the planning process to get me to the point where I can answer those two questions, the computer stage is extremely easy - I'm just slapping the poses into the frames I've already chosen, and saving a key on everything on those frames. At that point, the shot should be 90% done and ready for polishing...
The time-saving is happening because it's a hell of a lot easier to draw 100 thumbnails (little stick-guys are fine - just something that shows the angle of the hips, the angle of the shoulders, and the direction of the head) than it is to try 100 poses in the computer. It's a lot easier to film 50 takes of your performance searching for something true than it is to do 50 versions of your scene in the computer.
The more you plan, the better off you'll be - no matter what the scene is. That's totally just my opinion, but I've seen it proven time and time again, not only in my own work, but in other animator's work (and student work) as well...
shawn
