72afde7115
Two things this iteration, both of which change what is worth doing next. The order is DETERMINISTIC: two independent boots give byte-for-byte identical permutations for both live screens. That kills the hypothesis that the child list is built in I/O-completion order — which mattered, because a run-dependent list would have made deriving a rule pointless. It is a pure function of the bundle. And the 7-element bundle is identified by matching its pivots against every 7-element build on the disc: it is the developer-logo splash (GAME ARTS / SETA / studio anima, GP_TITLE entries 11/14). With names attached, its paint order 0 2 4 6 1 3 5 reads as the full-screen .prm, then all three _eff glows, then all three base logos — glow behind, logo on top. That refines the earlier "grouped by sprite" reading and partly withdraws it: here every element has its own sprite, so the grouping is by ROLE, not sprite identity. The title build could not tell the two apart because its repeated logo instances share both. Still not derived: where the role comes from. No decoded field carries it, and sorting on a "_eff" name suffix would be an odd thing for a loader to do.