SirStone 94ffc63160 j153: the open-loop ruler for "hold when trapped" + the A/B proposal (NOT run) ...

Measures, on the recorded fixtures and with NO counterfactual replay: how long
until a safe tile appears at a forced pick, whether the enemy had just fired,
whether our own tile is already hot, and all three split by distance. Registers
nothing and changes no default: the knob TR_TFIL_HOLD_WHEN_TRAPPED and its
implementation live in the concurrently edited working tree.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>

2026-09-27 10:16:53 +02:00

..

adr

docs(adr): Evo_Bot neuroevolution gun architecture (#63)

2026-08-24 00:02:47 +02:00

agents

chore: add agent skills configuration

2026-08-15 19:26:43 +02:00

papers

feat(ModularBot): 6 guns, pattern matcher, melee modules, adversarial bots

2026-09-20 00:59:53 +02:00

research

docs(research): portability audit of DrussGT 3.1.4159

2026-09-20 23:44:42 +02:00

bitbrain_algorithm.md

docs: reverse-engineer the BitBrain algorithm from its C source; assess as a gun

2026-09-24 20:17:08 +02:00

bitbrain_campaign.md

j142 retire the ADE+SBC gun (rack id 17): the owner watched it, it does not learn, throw it away

2026-09-26 19:20:23 +02:00

bitbrain_counted_sbc.md

docs: correct counted-SBC decay amortised cost figure

2026-09-25 08:43:48 +02:00

bitbrain_gate_test.md

BitBrain gate test: fine-grained aim correction vs naive + Pattern (offline)

2026-09-24 21:23:29 +02:00

bitbrain_gun_verdict.md

BitBrain verdict: clean negative at 30 runs/arm; analyzer gets MC + Mann-Whitney + MDE

2026-09-24 23:02:59 +02:00

bitbrain_vs_tmhorizon_ab.md

BitBrain vs TMHorizon vs Pattern: live A/B on shipped TFIL (null result)

2026-09-25 23:55:21 +02:00

drussgt_dodge_vs_power.md

DrussGT dodge vs fired power: no movement response once range is controlled

2026-09-24 21:22:38 +02:00

env_reference.md

j147: the 1-tick fire-detection lag is OURS — measure it, then back-date it (TR_FIRE_LAG)

2026-09-26 23:08:57 +02:00

feature_ab_results.md

Ram: finisher-only by default, and the bullet-rain abort now measures real energy

2026-09-22 08:14:03 +02:00

gauntlet_bitbrain_vs_pattern.md

gauntlet: BitBrain vs Pattern across 32 legacy opponents (does not generalize)

2026-09-26 00:55:57 +02:00

gun_campaign.md

j142 retire the ADE+SBC gun (rack id 17): the owner watched it, it does not learn, throw it away

2026-09-26 19:20:23 +02:00

gun_mix_disruption.md

Gun mixing (TMHorizon+BitBrain) vs DrussGT: clean negative, no dodge disruption

2026-09-24 23:07:05 +02:00

gun_rack_analysis.md

selector: arrival-accuracy tie-break measured NEGATIVE; randomness is load-bearing

2026-09-22 01:07:12 +02:00

gun_rack_summary.md

selector: arrival-accuracy tie-break measured NEGATIVE; randomness is load-bearing

2026-09-22 01:07:12 +02:00

headon_longrange_live.md

HeadOn (no-lead) vs Pattern LIVE at long range: clean negative, offline ruler killed

2026-09-24 23:52:49 +02:00

lead_capture_by_range.md

lead capture by range: we under-lead (0.40->0.135) — a RANGE effect, not a power one

2026-09-24 22:47:18 +02:00

melee_bitbrain_ab.md

melee A/B doc: correct the per-arm [bb-reset] counts (perRound 68.75, retained 68.19, decay 66.75)

2026-09-26 00:43:04 +02:00

melee_vs_1v1_racks.md

Melee vs 1v1 racks: mechanism built, but NO gun-level payoff - do not split

2026-09-22 08:08:43 +02:00

movement_campaign.md

j148: bound the corridor by speed x ticks (TR_{TFIL,STRAFE}_CORRIDOR_TICKS, default 0 = to the wall); UNTESTED

2026-09-27 08:23:16 +02:00

offline_harness_trust.md

Offline harness: audited, calibrated against live, and one real bug fixed

2026-09-24 21:40:38 +02:00

ramming_negative_result.md

Ram: finisher-only by default, and the bullet-rain abort now measures real energy

2026-09-22 08:14:03 +02:00

rl-algorithm-choice.md

chore: rename CLAUDE.md → AGENTS.md, move research doc to docs/

2026-08-27 18:11:56 +02:00

selector_negative_value.md

SHIPPED: the default rack is now Pattern-only (+49% hit rate, +66% damage on the boss)

2026-09-22 02:10:07 +02:00

spinner_and_melee_claims.md

j124: spinner + fair-melee results doc; analyzer convergence tests

2026-09-26 05:05:09 +02:00

state_window_gate.md

State-window gate: a temporal window of wave-relative states does NOT beat a single state

2026-09-25 08:43:36 +02:00

surfer_wiring_ab.md

docs: add per-run damage/wins series and liveness line to surfer_wiring_ab

2026-09-26 00:29:34 +02:00

tfil_heat_pillar_ab.md

tfil heat/pillar A/B: record the owner's decision to keep the virtual pillar removed

2026-09-25 22:24:47 +02:00

tfil_hold_when_trapped_ab.md

j153: the open-loop ruler for "hold when trapped" + the A/B proposal (NOT run)

2026-09-27 10:16:53 +02:00

tmcomposites_gate.md

TMComposites gate: per-gun confidence faithful for 3 guns; no pair composes

2026-09-25 22:04:39 +02:00

tracker_death_events.md

tracker: corpses do not exist - revert the fix and retire the workaround

2026-09-21 22:41:11 +02:00