j146 TFIL: the bullet's own heat becomes tunable (TR_TFIL_BULLET_CORE/AURA, default 10/5 = today) and the field-shape A/B is pre-registered

2026-09-26 22:16:06 +02:00
parent 5dd921c26b
commit 298ea6d586
5 changed files with 164 additions and 3 deletions
@@ -0,0 +1,43 @@
+# ─────────────────────────────────────────────────────────────────────────────
+# arms_tfil_shape.txt — j146: the TFIL FIELD SHAPE, five arms on the frozen
+# movement panel (tools/ab/panel_movement.txt), all with TR_MOVEMENT=tfil pinned
+# EXPLICITLY and the j144 fixes ON, so only the SHAPE varies.
+#
+# WHY. The strafe sweep (j119 batch 4) already measured the field SHAPE, not the
+# engine: the shape tfil SHIPS (corridor 20, wall 30/10) was -0.29 wins/run
+# (p=0.039) and field_off was -0.47 wins/run (p=0.0063); the champion was the
+# MIDDLE shape (bullet core 20 / aura 10, corridor 10, wall 15 / radiance 5).
+# tfil has never been run on the middle shape. On top of that, j145 found 59% of
+# tfil's picks have no safe set to tie-break in, because corridor 20 is TWICE
+# PathDangerThreshold 10 and tfil's own bullet heat (core 10) is exactly the
+# threshold, so a bullet is never dangerous by itself.
+#
+# Every arm carries `TR_TFIL_COMMIT_ARRIVAL=1 TR_TFIL_NOREV_SPEED=4` (j144's
+# recommended .env) so the shape is isolated ON TOP OF THE CURRENT BEST TFIL.
+# The pillar stays OFF (shipped default 0/0) in every arm.
+#
+# Columns: CORRIDOR / WALL_HOTNESS / WALL_RADIANCE / BULLET_CORE / BULLET_AURA
+#
+# Pre-registered in docs/movement_campaign.md ("Batch 7 — TFIL field shape")
+# BEFORE any of these battles ran. Reference is `shape_shipped`.
+#
+# Format: name | ENV=value ENV=value | label
+# ─────────────────────────────────────────────────────────────────────────────
+
+# 1. THE ARM TO BEAT — the field tfil ships today (20/30/10/10/5), on j144.
+shape_shipped | TR_MOVEMENT=tfil TR_TFIL_COMMIT_ARRIVAL=1 TR_TFIL_NOREV_SPEED=4 TR_TFIL_CORRIDOR_HEAT=20 TR_TFIL_WALL_HOTNESS=30 TR_TFIL_WALL_RADIANCE=10 TR_TFIL_BULLET_CORE=10 TR_TFIL_BULLET_AURA=5 | today's shape, on j144 (control / reference)
+
+# 2. THE CANDIDATE — the shape strafe measured as the champion.
+shape_middle | TR_MOVEMENT=tfil TR_TFIL_COMMIT_ARRIVAL=1 TR_TFIL_NOREV_SPEED=4 TR_TFIL_CORRIDOR_HEAT=10 TR_TFIL_WALL_HOTNESS=15 TR_TFIL_WALL_RADIANCE=5 TR_TFIL_BULLET_CORE=20 TR_TFIL_BULLET_AURA=10 | the middle shape strafe won on (10/15/5/20/10)
+
+# 3. "corridor alone was the problem" — only the corridor is halved.
+shape_corr10 | TR_MOVEMENT=tfil TR_TFIL_COMMIT_ARRIVAL=1 TR_TFIL_NOREV_SPEED=4 TR_TFIL_CORRIDOR_HEAT=10 TR_TFIL_WALL_HOTNESS=30 TR_TFIL_WALL_RADIANCE=10 TR_TFIL_BULLET_CORE=10 TR_TFIL_BULLET_AURA=5 | shipped, corridor 20->10 only
+
+# 4. "bullets made dangerous themselves" — only the bullet's own heat is raised.
+shape_bullets | TR_MOVEMENT=tfil TR_TFIL_COMMIT_ARRIVAL=1 TR_TFIL_NOREV_SPEED=4 TR_TFIL_CORRIDOR_HEAT=20 TR_TFIL_WALL_HOTNESS=30 TR_TFIL_WALL_RADIANCE=10 TR_TFIL_BULLET_CORE=20 TR_TFIL_BULLET_AURA=10 | shipped, bullet core/aura 10/5 -> 20/10 only
+
+# 5. THE "field off" CONTROL — corridors, walls and pillar all zero, so the
+#    bullets' own heat is the only bullet threat. strafe measured this shape as
+#    the WORST of the three; it is here so a middle-shape win cannot be a false
+#    winner (it proves the win is the middle, not simply "less lava").
+shape_nofield | TR_MOVEMENT=tfil TR_TFIL_COMMIT_ARRIVAL=1 TR_TFIL_NOREV_SPEED=4 TR_TFIL_CORRIDOR_HEAT=0 TR_TFIL_WALL_HOTNESS=0 TR_TFIL_WALL_RADIANCE=0 TR_TFIL_BULLET_CORE=20 TR_TFIL_BULLET_AURA=10 | no corridor/wall field, bullets only (0/0/0/20/10)