It’s been a while since I last posted, mostly because I’ve had a lot of work and been doing different personal projects at night, but since the very first moment I finished my last animation I was thinking about posting a breakdown of the process that’s taken to animate it from planning to the final render.
In the first place, I compiled my animation demo reel for this year about a month and a half ago, and after showing it to different animators, I was told by a fellow, Alex Kong, that my demo reel was short in motion mechanics and a cartoonish style. So he suggested me the audio from May on 11second club to get that covered and also put me as an example the very first place of that month.
- 1. Becoming one with the audio
So I get started by hearing the audio many many times, until I get used to the general timing, accents and rhythm, then start writing down ideas of what it could be, what can I done with that audio. Many ideas popped out (I can’t recall many right now :P) and in the end I thought that the approach of an elevator gag would fit great.
In this point, even though it is not related to animation at all,once I decided what would my characters be doing, in parallel I start making the scene, picking one of the many free rigs that are all over the internet (this is very important to me, because I get to know the rig, its benefits and limitations) , customizing them, modeling the background environment, adding materials and a few textures, getting the scene lighted and the first teaser of the render. I love doing that process, it gets me to imagine what it would be looking like when it’s done.
This is what I’ve got
- 2. Video Reference
I started making video references, the first couple ones just trying to follow the audio but not thinking too much about it, just letting it flow naturally, then I watched what I had and start gathering some movements that fit well together, and when I have a couple of structured ideas, I recorded new ones, trying to see if they could really fit the audio and keep the performance believable.
- 3. Thumbnails
Next step was putting those ideas down on paper (in this case… digital paper), so I begin to draw the key poses on a free version of Digicel FlipBook and make them hit the accents on the audio. This was a pretty straight forward process, because I already have the main idea on my head. And I like to make it on that software because it has many advantages: you can see and hear the audio along the way and you have a digital light table as a 2d traditional animator artist. It’s not necessary say that I’m not the best 2d animator in the world, and the idea is not to have a perfect and constant proportion of the characters, but for me, just have the main idea of what will be going on.
This is what it looked like at this point.
It was very important along all the process to ask for feedback to as many fellow animators as I could. So every time I’ve got a new step done, I sending the video everywhere. It’s a key thing on doing animation, because some times you get used to your own ideas and maybe there are things that could be improved.
Once I have the general idea through thumbnails, I proceed to transfer it form thumbnails to keyposes on Maya, this is not a copy paste of what I have already done, instead I just use it as a guide and as I go forward taking some changes from the previous feedback or what I think could feet better with what I’ve got on stage. Here take place many big changes, first of all, I removed the other guys that climbed up to the elevator, in behalf of focusing myself on only one character as Alex Kong’s feedback suggested. Also I found that some thins looked good on 2d, but will look unrealistic when putting them on 3d, they simply don’t fit any more, those include some little steps, goings and comings of the character.
I have to say that I really like the animation on stepped mode, it’s way too snappy and it let you hit the audio accents very precisely.
Here it is what it look like at this moment.
(Asked More Feedback) Getting from stepped to splines could be a difficult task to tackle, because just when you change from one to another your holds get lost, the accents are blurred and all the timing is messed up. I have found a way to lower the impact of the initial transfer from one to another, but that will be covered in a next post.
What I’ve focused here is to preserve the best that I can the timing and that the accents are done in the right moment. And of course making sure that the arcs well done, fixing some weight and balance issues.
At this moment.
- 6. Polish
(Even more feedback) At this point it is almost done, and what I’ve focused to do on this last part, is to start making some subtle changes and adjustments, like the position of the fingers, checking that they slide and touch properly, refine the follow through and overlapping action, making the knees and arms bending in arcs.
And at last, batching the render…