Paperman and the Future of 2D Animation

Neat! I've had a long-term interest in "toon shading". The problem with a lot of the traditional real-time methods of creating the outlines is that they're too literal, and cause unpleasant artifacts at some joints and seams. Also, you typically only have one parameter to tune, for the whole scene: line thickness. The problem is finding a value or range of this one parameter that works for a whole scene. E.g. if you use the same thickness for close and far objects, then the farther objects will appear more black since a greater proportion of their area is devoted to outline. Another problem (for VR) is that current VR systems don't deal well with thin lines. You really want to avoid aliasing (crawlies) for high-quality VR. Would love to see something like Paperman implemented in real-time: ▶ this video

I think there's a good argument to be made that the sooner we make this tech boring, the better. I want this stuff to be off-the-shelf, got-it-by-default now. When it's fully proliferated and cheap with loads of Stack Overflow-like answers and tutorials for it (like phone accelerometers and motion capture software), the sooner we'll really start developing and digging into the cool stuff. And get the crufty stuff out of the way. But I fundamentally disaggree that it's less interesting. I'm going to assume you're just jaded by the awesome we're constantly surrounded by instead of just taking the easy, boring interpretation. Even 2D animation is advancing where you wouldn't expect (I found the technical work giving the artists better control on The Paperman surprising [0]). New rendering philosophies, networking algorithms, AI/bots, social integration, all of this leads to a refinement that's well worth the effort. It took a long, long time to get from oscilloscopes [1] to the Occulus, and hopefully in a couple generations we'll have even cooler stuff. Expect to see demos and then have to wait a while for the hard, brutal engineering work to make it actually usable [2]. For the Occulus, the real win is getting the feedback loop between a just-good-enough-monitor and the accelerometer down to few enough milliseconds the brain can kinda fall for it. Like a cartoon, but for motion feedback. Expect other companies to say "I can do better" and let the races begin. The Occulus is a big deal, and better stuff is just a little bit away. Don't be jaded! Real cool stuff is in the pipe! [0] ▶ this video [1] link [2] link