placeholder copy — to be reworked
Two numbers go in — your X / Y control position — and a whole fistful of parameters comes out. You can't picture a dozen dimensions at once, so picture them stacked: each slab is one parameter, drawn as a 3-D landscape whose height at any spot is that parameter's value. One probe reads every slab at the same control position.
Seen straight from above the slabs overlap into a single flat map — it looks 2-D. Tilt it and the reveal lands: it was a stack of 3-D surfaces all along. Add parameters to grow the sandwich.
Every surface here is the real NISPS network sampled across the input space. 👍 keep teaches the net that this sound belongs at this spot and the landscape bends to honour it; 👎 either pushes the mapping away from the current sound or (in explore mode) re-rolls a fresh candidate to audition. nudge and randomise perturb the net; undo steps back.