mirror of
https://github.com/Druthulu/BFM-decomp
synced 2026-09-26 21:36:06 -04:00
Drew T
605a4b77c8
Drew asked whether waves cracked better before MAXTOK went 8000 -> 16000. Recording both halves of the answer so next session does not relitigate it from memory: CLEAN: raising to 16k did cause a real regression — draft completion 84-89% (8k) -> 41% on wave bt, 69% on bu — but the cause was a harness interaction, not the model. A 16k generation runs ~530 s at ~30 tok/s while STRAGGLER_GRACE was 120 s, so agents were cut off mid-thought with no draft. Grace at 700 s fixed it; completion has run 97-99% since. CONFOUNDED: on banks per draft the 8k era looks better (S59: 1,335 of 2,996 = 44.6%; today's best 16k waves dd 34.7%, de 29.7%) — but the populations differ completely. 8k waves had never-drafted work; today's draw from skeletons that refused six times. Budget and exhaustion moved together, so neither figure isolates the other. Neither should be cited as evidence about the budget. AGAINST the simple story: truncated-turn rate is INVERSELY correlated with bank rate (cx 8.7% trunc/43.9% bank, dd 8.3%/51.4% vs dl 1.3%/0.5%, ej 0.6%/0%). Budget exhaustion driving the decline would produce the opposite relationship. THE A/B: split ONE wave's card pool — half the shards at 8k, half at 16k, same generation mix, same binaries, same gate, same tree, grace 700 s in both arms. Compare banks per DRAFT and per GATE MINUTE. Holding the population constant is the whole point; every historical comparison fails exactly there. If 8k matches 16k, the cheaper budget also buys more agents per unit time.
phase-ends/ — the living record
PhaseEnd_Phase[N].md— written at each phase boundary (format defined inPROJECT_CONTEXT.md). Append-only history: build log, deviations, commit message, rules added, changelog. Never deleted.CURRENT_PHASE.md— the in-phase autonomous log: approved phase plan, per-task checkboxes, current task pointer, blockers. Created at phase start, updated after every task, absorbed into the PhaseEnd file and deleted at phase close. This is the crash/compaction recovery point.- Sessions read
PROJECT_CONTEXT.md, then everyPhaseEnd_*.mdin numeric order, thenCURRENT_PHASE.md(if present) — in that order, every session.