Steve Jobs, Ed Catmull, and John Lasseter said, in effect, “The only thing we’re afraid of is complacency—feeling like we have it all figured out. We want you to come shake things up. We will give you a good argument if we think what you’re doing doesn’t make sense, but if you can convince us, we’ll do things a different way.” For a company that has had nothing but success to invite a guy who had just come off a failure and say, “Go ahead, mess with our heads, shake it up”—when do you run into that?

The Quarterly: How did your first project at Pixar—The Incredibles—shake things up?

Brad Bird:The Incredibles was everything that computer-generated animation had trouble doing. It had human characters, it had hair, it had water, it had fire, it had a massive number of sets. The creative heads were excited about the idea of the film, but once I showed story reels of exactly what I wanted, the technical teams turned white. They took one look and thought, “This will take ten years and cost $500 million. How are we possibly going to do this?”

So I said, “Give us the black sheep. I want artists who are frustrated. I want the ones who have another way of doing things that nobody’s listening to. Give us all the guys who are probably headed out the door.” A lot of them were malcontents because they saw different ways of doing things, but there was little opportunity to try them, since the established way was working very, very well.

We gave the black sheep a chance to prove their theories, and we changed the way a number of things are done here. For less money per minute than was spent on the previous film, Finding Nemo, we did a movie that had three times the number of sets and had everything that was hard to do. All this because the heads of Pixar gave us leave to try crazy ideas.

The Quarterly: What sorts of things did you do differently?

Brad Bird: There are purists in computer graphics who are brilliant but don’t have the urgency about budgets and scheduling that responsible filmmakers do. I had to shake the purist out of them—essentially frighten them into realizing I was ready to use quick and dirty “cheats” to get something on screen if they took too long to achieve it in the computer. I’d say, “Look, I don’t have to do the water through a computer simulation program. If we can’t get a program to work, I’m perfectly content to film a splash in a swimming pool and just composite the water in.” This absolutely horrified them. Or I’d say, “You can build a flying saucer, or you can take a pie plate and fling it across the screen. If the audience only sees the pie plate very briefly and you throw it just right, they will buy it as a flying saucer.”

I never did film the pool splash or throw the pie plate, but talking this way helped everyone understand that we didn’t have to make something that would work from every angle. Not all shots are created equal. Certain shots need to be perfect, others need to be very good, and there are some that only need to be good enough to not break the spell.

We also made superelaborate storyboards. We even emulated camera movement in them, so everyone knew that “We only need to make things work between here and there.” Once I was able to commit to the camera angles, we could be very specific about how we built things. Something would look beautiful from one position, but if you moved five feet to the right, the image would disintegrate. I gave up the flexibility to move within a set, but in exchange I bought size and scope.