Save Selector Opcode 0x20 Sample Order Effects

route map1_01a -> map2_02d; sample source -; active slot count offset 0x0010; active order offset 0x0011; current frontier selector 2:0; current frontier sample covered False; proofFound False; promotion status blocked.

The public sample saves make the opcode 0x20 count/order bytes concrete: selector 0:0 and 1:0 samples use count 1 with descriptor order [0], and progressed selector 22:0 samples use count 3 with [0,1,2]. Applying those orders to descriptor+4 static effects leaves only pointer-dword low-byte collisions in the last shaped context+0xa8 setter rows; there is no non-pointer context+0xa8 setter in the active descriptors covered by the public samples. None of the public samples covers the current frontier selector 2:0, so this remains calibration evidence rather than promotion proof for map1_01a -> map2_02d.

Failed Gates

current-selector-2-0-sample
runtime-active-order
non-pointer-context-a8-effect
route-promotion-proof

Missing Evidence

public or captured sample covering current selector 2:0 active order
runtime active order/count on the current route path
non-pointer context+0xa8 effect in active descriptors
route-promotion proof linking sample order to map1_01a -> map2_02d

Evidence Refs

data/savedata_sample_party_order.json: public savedat active count/order byte samples
out/save_selector_opcode20_descriptor_scripts.json: descriptor+4 static effects applied to public sample orders

Final Context+0xa8 Base Histogram

kind	base expression	sample count

Samples

sample	selector	tile	count	active descriptors	final shaped context+0xa8 base	final non-pointer context+0xa8 base	covers current selector