Sunday, November 06, 2011

Squash and Stretch (part 3) and Creating a Believeable World

This post about Squash and Stretch is more of a conceptual discussion. I talk about this sometimes when I'm giving talks to students, and it's a very important concept in my mind, but I get the feeling that people usually think I'm overstating the case or just crazy. But in my mind this is a big deal.

My first year at CalArts, I made a short animated film where I gave every object a lot of squash and stretch. Every object was bouncy and pliable. It was a lot of fun to animate (Early animated cartoons were often this way too - in early Mickey Mouse cartoons, Mickey could easily stretch out the tail of an airplane and make it longer. An airplane or a car was animated as though it was made out of the same material as a cow's udder or an animal's tongue).

As a worked at learning to draw and animate, I thought more and more about what I was trying to accomplish as an animator. I realized that I wanted to animate fun characters, but gradually I also realized that I wanted to tell stories that made people feel real emotions. I wanted to make films as funny as Chuck Jones' Warner Brothers shorts, and I also wanted to make films that could make you worry over the fate of the characters, like "Pinocchio", or "Lady and the Tramp", and make you get choked up or maybe even cry over what happened to the characters, like "Bambi" or "Dumbo". Those things seemed like great magic tricks, and I was always searching for answers about how that was possible using only drawings.

One day when I was at CalArts the animator Glen Keane gave a lecture and a handout on "Dynamics of Animated Drawing" (full handout can be seen here) and it contained a sentence that I didn't really take notice of at the time, but over the years I've realized that there's a lot of wisdom contained within that thought.



It says, simply: "Your character is bound by natural laws - we can fudge and cheat these to a certain degree, but the audience relates to these laws."

I think this has a larger implication that can really make or break an animated film and can be the difference between a film that can tug at your heartstrings and give you a satisfying emotional experience, and one that is fun to watch but doesn't really have deeper, more involving emotions and characters (and there's nothing wrong with just making a fun film, if that's your goal).

In order for an audience to really engage with your characters and feel empathy for them - even thought they're just drawings, or zeros and ones in a computer - the audience has to be able to believe that they're real characters in a real world. It's a contract that your audience can enter with you if you've done your job right, and I think a big part of it is portraying a world that your audience can believe actually exists. Intellectually, I think it's impossible for people to care about characters if they are conscious while they are watching the movie that the characters are too weird to be real, or live in a world that's too weird to actually exist, or if objects in that world have too much squash and stretch and don't react like the type of objects we experience in our world every day that make our world feel real to us.

 So I'm not saying that you can't caricature materials or give objects a little more squash and stretch than they have in our everyday world. Certainly that can be a lot of fun, and people have pulled it off just fine without destroying the credibility of the world they're creating. But be conscious of this and I think you'll find that if you don't overdo it it'll help the believeability of the world you're creating and will give people "anchors" to relate to that help them get immersed in your world and help them believe in your world as a real place and your characters as living breathing entities. Because that's really a great feat when you can pull it off, and probably the best magic trick you can accomplish in animation!

3 comments:

Quentin Lebegue said...

Great post - wise words to keep in mind !

Smurfswacker said...

Your statement, "I think it's impossible for people to care about characters if they are conscious while they are watching the movie that the characters are too weird to be real," points up the problem I have with CGI animated movies which try to look "real."

Even on the biggest budget shows (e.g. the Tintin movie), CGI people still look wrong. Their movements may be detailed and realistic as hell, yet they seem to lack weight. I presume it's connected with motion capture. Raw mocap figures are also weightless, just more so. I speculate that the animated figures need some unrealistic squash and stretch, as well as exaggerated follow-through, to look real.

On the other hand, maybe it's the result of how my eye was trained. I grew up understanding animated characters using the old hand-drawn paradigm. Perhaps younger audiences who've grown up watching mo-capped characters actually see them differently. To them the movements might seem real. I wonder.

the Flying Animator said...

Bashing my head against the Squash and Stretch dilema I thought of 2 things:

1. Looking at old cartoons, I learned that this was not just to show the elasticity of materials - at 12 frames per second, squash and stretch took the place of motion blur. Fast action just looks strobbbed without it.

2. When I look at my own work, my students' work, and all the motion capture stuff, I think the "non-realistic" feel in CGI comes from not picking the frames yourself - you let the computer decide where to put the inbetweens. It might be mathematically correct, calculate the easing splines perfectly, but human vision is still faster.
What I think is, that a human brain would pick better frames to describe an action.

The reason motion capture looks strange remains a mystery to me. If I look at a video running - it looks great. But if I play it frame by frame - it's wrong again.
More than that - almost every frame will look blurred.
The only way I can think of to explain this, is that our brain translates squash and stretch acording to CONTEXT.
And that is what makes the difference.

Any thoughts?
Netta