Measured while building #16: at 10 000 enemies the grid rebuild costs
0.057 ms against 0.021 ms for move_enemies — 2.7x more, not less. ADR-0004
§3 counted arithmetic and missed what dominates: move_enemies is a linear
table walk, the rebuild writes into ~7000 scattered buckets.
The decision does not change. It never rested on the comparison but on the
absolute bound of 0.10 ms, and 0.057 ms is 0.34 % of the frame budget; an
incremental update would pay the same scattered write on every cell
crossing. Revision trigger 1 and the acceptance criterion of spec feature 3
are restated on that bound, which is what the runnable check now asserts.
Also records the second deviation the build produced: a cell holds
(Entity, Vec2) rather than Entity, because soft separation writes Transform
while reading its neighbours' positions.
Enemies spawn on the arena rim, walk radially at the tower and stop at the
attack range. The spawner is cap-driven — it tops the field up to the target
population instead of placing N enemies on a schedule — so the target
population is the binding quantity the difficulty curve will stand on.
The spatial grid of ADR-0004 lands with them, because it now has its first
consumer and #17 needs it unchanged for target selection and projectile
collision. One full rebuild per frame is the only pass over the enemy set;
every consumer asks the grid.
Soft separation is a clamped positional offset, not a force: overlap stays
allowed, and capping the offset below one movement step is what keeps the
tower ring from pulsing.
Closes#16.
Schließt die /wayfinder-Reise ab (Karte: Issue #1). Enthält CONTEXT.md,
die Vertical-Slice-Spec, vier ADRs, den In-Run-HUD-Entwurf, den gesammelten
Wiki-Primärtext und den Perf-Spike, der das Entity-Budget gemessen hat.
Kein Produktionscode: src/main.rs bleibt das Bevy-Scaffold.
