Why would Pixar want lens flares?

A computer camera has no glass, so it never flares. Pixar's math came from "formulas they'd copied out of the American Cinematographer Manual", and still, in Jeremy Lasky's words, "our cameras never look as we expect them to". So they tested a real film camera with Panavision lenses: "We looked at lens flares and how to focus lights in the background." Andrew Stanton: "They didn't match at all". A flare is light bouncing between the glass surfaces inside a lens. Every pair of surfaces makes its own ghost: for n surfaces, n(n-1)/2 of them, 91 for a 14-surface lens (Hullin et al., SIGGRAPH 2011), and in a round lens they sit on the line from the light through the centre of the frame. Pixar rebuilt WALL-E's camera with oval highlights, breathing focus and flares. Stanton: "Life is nothing but imperfection and the computer likes perfection".

Our own lights, lens diagram and ghosts; no frames from the film, and the robot never appears.

Next time you see a flare, can you find the centre of the frame?

Sources: Bill Desowitz, "Hello, WALL-E!", Animation World Network, 2008; Iain Stasukevich, "Expert Eyes Enhance WALL-E", American Cinematographer, July 2008; Barbara Robertson, "Rampart Risk-Taking", Computer Graphics World, July 2008; Hullin, Eisemann, Seidel and Lee, "Physically-Based Real-Time Lens Flare Rendering", SIGGRAPH 2011.
Sounds: our own, made in code (earshot, CC0).