Finished
87Recklesssingular-do-shallowerdiff8.0+0.08
LLR: -2.53 (-2.25, 2.89) [0.00, 3.00]
Games: 20326 W: 5042 L: 5146 D: 10138
Ptnml(0-2): 53, 2448, 5257, 2360, 45
SapherRecklesssee-queendiff8.0+0.08
LLR: -2.33 (-2.25, 2.89) [0.00, 3.00]
Games: 11690 W: 2919 L: 3023 D: 5748
Ptnml(0-2): 17, 1288, 3349, 1164, 27
Take 1
ProtonRecklessps_queenthreats102diff8.0+0.08
LLR: -1.09 (-2.25, 2.89) [0.00, 3.00]
Games: 3876 W: 945 L: 1002 D: 1929
Ptnml(0-2): 12, 501, 972, 438, 15
1k
JoshkaRecklessjoshka/optimize-avx2-move-emissiondiff4.0+0.04
LLR: -2.54 (-2.25, 2.89) [0.00, 3.00]
Games: 18500 W: 4683 L: 4792 D: 9025
Ptnml(0-2): 88, 2104, 4964, 2017, 77
Prototype AVX2 bulk move emission Replace scalar setwise move emission with an AVX2 chunked path on x86 while preserving move ordering and move encoding behavior. This is a non-functional optimization experiment for move generation.
JoshkaRecklessjoshka/reduce-quiet-scoring-lookupsdiff4.0+0.04
LLR: -2.62 (-2.25, 2.89) [0.00, 3.00]
Games: 9422 W: 2377 L: 2518 D: 4527
Ptnml(0-2): 62, 1188, 2342, 1067, 52
Reduce repeated work in `score_quiet` by hoisting shared history/lookups out of the per-move scoring loop while keeping the scoring formula unchanged.
ProtonRecklessps_skipquiets101diff8.0+0.08
LLR: 2.90 (-2.25, 2.89) [-2.75, 0.25]
Games: 18510 W: 4783 L: 4651 D: 9076
Ptnml(0-2): 46, 1974, 5084, 2104, 47
Merge branch 'main' into ps_skipquiets101
JoshkaRecklessjoshka/optimize-x86-move-picker-scandiff4.0+0.04
LLR: -2.25 (-2.25, 2.89) [0.00, 3.00]
Games: 7824 W: 1988 L: 2103 D: 3733
Ptnml(0-2): 32, 923, 2123, 796, 38
Optimize x86 move picker scan Prototype an x86-first move ordering experiment by adding a faster best-entry selection path in MovePicker. Keep the change isolated so it can be validated locally before any OpenBench run.
JoshkaRecklessjoshka/setwise-queen-threat-generationdiff4.0+0.04
LLR: -2.31 (-2.25, 2.89) [0.00, 3.00]
Games: 16508 W: 4169 L: 4268 D: 8071
Ptnml(0-2): 67, 1894, 4429, 1799, 65
Use setwise queen threat generation Replace the scalar queen threat accumulation in update_threats with the existing setwise rook and bishop threat helpers. Keep behavior the same while simplifying and broadening the vectorized threat path.
Cj5716Recklesslimit-cutnode-lmrdiff8.0+0.08
LLR: -2.60 (-2.25, 2.89) [0.00, 3.00]
Games: 19288 W: 4788 L: 4901 D: 9599
Ptnml(0-2): 69, 2369, 4885, 2248, 73
If we have a real chance of a fail low we take this cutnode more seriously
WorldRecklessrfp-nmp-margins-baseline-100kdiff40.0+0.40
LLR: -1.35 (-2.25, 2.89) [0.00, 3.00]
Games: 13126 W: 3197 L: 3244 D: 6685
Ptnml(0-2): 4, 1479, 3643, 1434, 3
spec
WorldRecklessaggressive-scoresdiffN=25000
Elo: -1.22 +- 2.03 (95%) [N=40000]
Games: 40052 W: 11255 L: 11396 D: 17401
Ptnml(0-2): 575, 4930, 9143, 4817, 561
WorldRecklessaggressive-scoresdiff8.0+0.08
LLR: -2.25 (-2.25, 2.89) [0.00, 3.00]
Games: 10954 W: 2710 L: 2822 D: 5422
Ptnml(0-2): 54, 1367, 2725, 1299, 32
WorldRecklessrfp-nmp-margins-baseline-100kdiff8.0+0.08
LLR: -2.26 (-2.25, 2.89) [0.00, 3.00]
Games: 62056 W: 15696 L: 15693 D: 30667
Ptnml(0-2): 193, 7423, 15791, 7430, 191
WorldRecklessrfp-nmp-margins-baselinediff40.0+0.40
Tuning 15 Parameters
13262/25000 Iterations
106096/200000 Games Played
87Recklesssplit-pawn-histdiff8.0+0.08
LLR: -2.29 (-2.25, 2.89) [0.00, 3.00]
Games: 15454 W: 3894 L: 3997 D: 7563
Ptnml(0-2): 46, 1930, 3886, 1811, 54
DanielRecklessasymcorr1diff8.0+0.08
LLR: -2.27 (-2.25, 2.89) [0.00, 3.00]
Games: 72014 W: 18306 L: 18281 D: 35427
Ptnml(0-2): 238, 8641, 18220, 8674, 234
changes the way the symmetry in correction updates is fixed didn't work ;(
ProtonRecklessps_scoring116diff40.0+0.40
LLR: 3.04 (-2.25, 2.89) [-2.75, 0.25]
Games: 16800 W: 4242 L: 4102 D: 8456
Ptnml(0-2): 7, 1838, 4567, 1984, 4
Merge branch 'main' into ps_scoring116
Cj5716Recklesspcb-higher-upperdiff8.0+0.08
LLR: -2.36 (-2.25, 2.89) [0.00, 3.00]
Games: 23464 W: 5874 L: 5960 D: 11630
Ptnml(0-2): 47, 2768, 6196, 2666, 55
87Recklesslog2-halfmove-clock-zobristdiff40.0+0.40
LLR: -2.26 (-2.25, 2.89) [0.00, 3.00]
Games: 48638 W: 12003 L: 12028 D: 24607
Ptnml(0-2): 26, 5502, 13281, 5491, 19
PeregrRecklessqs-promodiff8.0+0.08
LLR: -2.36 (-2.25, 2.89) [0.00, 3.00]
Games: 22940 W: 5739 L: 5827 D: 11374
Ptnml(0-2): 60, 2731, 5975, 2645, 59
PeregrRecklesspc-upperdiff8.0+0.08
LLR: -2.46 (-2.25, 2.89) [0.00, 3.00]
Games: 34710 W: 8866 L: 8936 D: 16908
Ptnml(0-2): 137, 4174, 8776, 4158, 110
No upperbound Probcut
PeregrRecklessrazor-fail-lowdiff40.0+0.40
LLR: -0.43 (-2.25, 2.89) [0.00, 3.00]
Games: 690 W: 165 L: 188 D: 337
Ptnml(0-2): 0, 94, 180, 71, 0
spec LTC since the problem is very rare
PeregrRecklessrazor-fail-lowdiff8.0+0.08
LLR: -2.38 (-2.25, 2.89) [0.00, 3.00]
Games: 68662 W: 17444 L: 17433 D: 33785
Ptnml(0-2): 244, 8110, 17612, 8121, 244
ProtonRecklessps_skipquiets100diff8.0+0.08
LLR: -1.72 (-2.25, 2.89) [-2.75, 0.25]
Games: 2024 W: 459 L: 565 D: 1000
Ptnml(0-2): 7, 294, 513, 194, 4
simplify dance around skip_quiets.
87Recklesslog2-halfmove-clock-zobristdiff8.0+0.08
LLR: 2.96 (-2.25, 2.89) [0.00, 3.00]
Games: 80918 W: 20342 L: 19992 D: 40584
Ptnml(0-2): 189, 9393, 20984, 9665, 228