I can't get the kinetic painter app out of my head. It's been swimming around in my brain for the last six months. That's the one with the voice activation, internal classification, and 36 points of contact. A sort of high precision finger painting that I'm going to build on a game engine.
The most exciting piece of it is going to be the way you can layer and lock time.
Imagine you're using watercolors, for example. You put the perfect amount of paint down, for a perfect bloom, and for a split second your brain clicks, and says, "God, that's beautiful" just before it melts and goes somewhere... less interesting.
With kinetic painter, there is no undo button.
Instead, you get a layered timeline where you can use your hands and words to traverse causality. Want to run the timeline back 400 milliseconds? No problem.
Want to add another pair of hands that interact with the first pair of hands to spread the bloom around in an interesting way? Cool.
What if you did that 30 more times?
Also, no problem.
And then there's the game physics, and how everything acts like real materials, until they don't. Paint that dries, water that spills, fire that burns.
One question artists ask a lot is, "When did we ever ask for AI?"
And granted, it's going to take seven kinds of machine learning to pull this off.
But I can't think of anyone who's ever painted in a serious way that wouldn't want something like this.
Part of the reason I can tell you about it, is because it's by far the biggest thing I've ever tried to develop. Right now, my biggest single project is the resurrection of Flash, which is an absurdly large and expensive project that was dumb to take on.
This one's easily twice as complex.
And I don't think it could be casually reverse engineered.
So there is that.
I keep getting closer. Right now, I'm researching fluid and powder simulations.
It's going to be a great product when it's done.
Long live precision finger painting!!!