SirStone
8efa627c05
The repo's first multi-opponent gun measurement. Adds tools/ab/gauntlet_run.sh (per-opponent A/B over the legacy roster, subject = frozen ModularBot), tools/ab/gauntlet_analyze.py (paired per-opponent deltas, cross-opponent sign test, style split, MDE) and the arm/opponent fixtures. Result: BitBrain does NOT generalize beyond DrussGT. 32 opponents x 2 arms x 3 runs x 5 rounds = 192 battles / 960 rounds, 0 failed, 0 retries: damage/run 214.5 (pattern) vs 210.9 (bb), sign-flip p=0.53; round wins 237/480 vs 239/480, p=0.91. Sign test: bb better on 13/32 opponents (damage). The DrussGT-only penalty does not carry. The owner's 'killer vs regular movers' sub-claim is not supported: regular bucket +1.3 dmg/run vs dodgers -0.2 (MW p=0.85), and the measured movement predictability does not correlate with the delta.
9 lines
473 B
Plaintext
9 lines
473 B
Plaintext
# BitBrain vs Pattern gauntlet — two arms only, so the per-opponent sample is
|
|
# as large as the battle budget allows.
|
|
#
|
|
# pattern = shipped default (onlyPattern rack), NO env reference
|
|
# bb = Pattern off, BitBrain on, the learned-gain config the owner
|
|
# most likely ran (gains 1.0..2.0, decay memory)
|
|
pattern |
|
|
bb | TR_RACK_PATTERN=off TR_RACK_BITBRAIN=both TR_BITBRAIN_GAINS=1.0,1.25,1.5,2.0 TR_BITBRAIN_MEM=decay | BitBrain learned gain
|