Dev Builds » 20230203-1918

Use this dev build

NCM plays each Stockfish dev build 20,000 times against Stockfish 15. This yields an approximate Elo difference and establishes confidence in the strength of the dev builds.

Summary

Host Duration Avg Base NPS Games WLD Standard Elo Ptnml(0-2) Gamepair Elo
ncm-dbt-01 07:00:57 584316 4014 1248 814 1952 +37.71 ± 5.12 1 267 1041 693 5 +75.61 ± 10.52
ncm-dbt-02 07:00:08 586689 3984 1236 854 1894 +33.42 ± 5.14 2 275 1063 643 9 +66.2 ± 10.39
ncm-dbt-03 06:59:30 587941 4000 1230 871 1899 +31.27 ± 5.13 2 291 1055 650 2 +63.05 ± 10.44
ncm-dbt-04 07:00:19 571163 4002 1167 839 1996 +28.54 ± 5.17 1 312 1049 636 3 +57.11 ± 10.48
ncm-dbt-05 07:01:07 584332 4000 1211 824 1965 +33.72 ± 5.13 3 274 1062 655 6 +67.55 ± 10.4
20000 6092 4202 9706 +32.93 ± 2.3 9 1419 5270 3277 25 +65.89 ± 4.67

Test Detail

ID Host Base NPS Games WLD Standard Elo Ptnml(0-2) Gamepair Elo CLI PGN
375746 ncm-dbt-04 570029 2 1 1 0 0.0 ± 34.02 0 0 1 0 0 0.0 ± 30.47 ↓
375745 ncm-dbt-01 582027 14 5 3 6 +49.96 ± 94.42 0 1 3 3 0 +102.11 ± 234.06 ↓
375744 ncm-dbt-02 587579 484 153 113 218 +28.78 ± 14.55 0 35 133 73 1 +56.48 ± 29.35 ↓
375743 ncm-dbt-05 583866 500 163 91 246 +50.38 ± 13.72 0 22 135 92 1 +101.46 ± 28.94 ↓
375742 ncm-dbt-03 587919 500 154 109 237 +31.36 ± 14.47 1 34 134 81 0 +64.66 ± 29.31 ↓
375741 ncm-dbt-04 572114 500 151 113 236 +26.46 ± 14.26 0 38 136 76 0 +53.22 ± 29.08 ↓
375740 ncm-dbt-01 581985 500 141 94 265 +32.75 ± 14.95 0 39 126 84 1 +64.66 ± 30.39 ↓
375739 ncm-dbt-02 587792 500 156 98 246 +40.48 ± 14.39 0 30 134 84 2 +79.17 ± 29.25 ↓
375738 ncm-dbt-05 586816 500 154 106 240 +33.46 ± 14.86 0 38 127 84 1 +66.1 ± 30.25 ↓
375737 ncm-dbt-03 588132 500 150 111 239 +27.15 ± 14.44 0 39 133 78 0 +54.65 ± 29.48 ↓
375736 ncm-dbt-01 585126 500 169 104 227 +45.42 ± 14.55 0 30 126 93 1 +90.97 ± 30.33 ↓
375735 ncm-dbt-04 573446 500 148 104 248 +30.65 ± 14.94 0 40 127 82 1 +60.36 ± 30.26 ↓
375734 ncm-dbt-02 586054 500 162 105 233 +39.78 ± 14.48 0 32 130 87 1 +79.17 ± 29.81 ↓
375733 ncm-dbt-05 586816 500 159 99 242 +41.89 ± 13.66 0 26 138 86 0 +85.04 ± 28.63 ↓
375732 ncm-dbt-04 568236 500 142 110 248 +22.26 ± 14.86 0 44 131 74 1 +43.3 ± 29.76 ↓
375731 ncm-dbt-03 588984 500 151 99 250 +36.26 ± 14.53 0 35 128 87 0 +73.34 ± 30.1 ↓
375730 ncm-dbt-01 583740 500 152 101 247 +35.56 ± 15.0 0 37 127 84 2 +68.99 ± 30.25 ↓
375729 ncm-dbt-02 582110 500 151 117 232 +23.66 ± 14.83 0 43 131 75 1 +46.13 ± 29.76 ↓
375728 ncm-dbt-05 583824 500 137 99 264 +26.45 ± 14.26 0 37 139 73 1 +51.8 ± 28.68 ↓
375727 ncm-dbt-04 569629 500 139 97 264 +29.25 ± 14.46 0 38 132 80 0 +58.93 ± 29.6 ↓
375726 ncm-dbt-03 588856 500 151 110 239 +28.55 ± 14.54 0 39 131 80 0 +57.5 ± 29.74 ↓
375725 ncm-dbt-01 586435 500 156 88 256 +47.55 ± 14.13 1 24 131 94 0 +98.44 ± 29.57 ↓
375724 ncm-dbt-02 587962 500 151 112 237 +27.15 ± 14.83 0 40 133 75 2 +51.8 ± 29.48 ↓
375723 ncm-dbt-05 582569 500 149 119 232 +20.87 ± 14.24 1 37 144 67 1 +41.89 ± 28.03 ↓
375722 ncm-dbt-04 570709 500 154 98 248 +39.08 ± 14.17 0 31 132 87 0 +79.17 ± 29.53 ↓
375721 ncm-dbt-03 587325 500 155 116 229 +27.16 ± 15.33 1 43 122 84 0 +56.07 ± 30.91 ↓
375720 ncm-dbt-01 584453 500 153 105 242 +33.46 ± 14.74 0 38 126 86 0 +67.55 ± 30.39 ↓
375719 ncm-dbt-02 586520 500 148 100 252 +33.46 ± 14.48 0 35 133 81 1 +66.1 ± 29.44 ↓
375718 ncm-dbt-03 588600 500 155 108 237 +32.75 ± 14.3 0 34 136 79 1 +64.66 ± 29.04 ↓
375717 ncm-dbt-04 573123 500 155 118 227 +25.76 ± 14.21 0 37 140 72 1 +50.38 ± 28.55 ↓
375716 ncm-dbt-05 581444 500 137 104 259 +22.96 ± 14.65 1 39 137 72 1 +46.13 ± 28.97 ↓
375715 ncm-dbt-01 585885 500 166 105 229 +42.6 ± 14.12 0 29 131 90 0 +86.52 ± 29.64 ↓
375714 ncm-dbt-02 588771 500 160 109 231 +35.56 ± 14.61 2 29 136 82 1 +73.34 ± 29.0 ↓
375713 ncm-dbt-04 570428 500 141 107 252 +23.66 ± 14.83 1 41 131 77 0 +48.96 ± 29.75 ↓
375712 ncm-dbt-03 587070 500 161 100 239 +42.6 ± 14.52 0 31 128 90 1 +85.04 ± 30.07 ↓
375711 ncm-dbt-05 584369 500 162 103 235 +41.19 ± 15.46 0 39 114 96 1 +82.1 ± 31.97 ↓
375710 ncm-dbt-01 584874 500 152 102 246 +34.86 ± 14.31 0 33 135 81 1 +68.99 ± 29.16 ↓
375709 ncm-dbt-02 586731 500 155 100 245 +38.37 ± 14.13 0 31 133 86 0 +77.71 ± 29.4 ↓
375708 ncm-dbt-04 572759 500 136 91 273 +31.35 ± 15.23 0 43 119 88 0 +63.23 ± 31.3 ↓
375707 ncm-dbt-01 584327 500 154 112 234 +29.25 ± 14.19 0 36 136 78 0 +58.93 ± 29.06 ↓
375706 ncm-dbt-03 586646 500 153 118 229 +24.36 ± 13.84 0 36 143 71 0 +48.96 ± 28.14 ↓
375705 ncm-dbt-05 584958 500 150 103 247 +32.76 ± 14.82 1 36 128 85 0 +67.55 ± 30.12 ↓

Commit

Commit ID 8d3457a9966f8c744ab7f8536be408196ccd8af9
Author pb00067
Date 2023-02-03 19:18:50 UTC
Improve excluded move logic PR consists of 2 improvements on nodes with excludeMove: 1. Remove xoring the posKey with make_key(excludedMove) Since we never call tte->save anymore with excludedMove, the unique left purpose of the xoring was to avoid a TT hit. Nevertheless on a normal bench run this produced ~25 false positives (key collisions) To avoid that we now forbid early TT cutoff's with excludeMove Maybe these accesses to TT with xored key caused useless misses in the CPU caches (L1, L2 ...) Now doing the probe with the same key as the enclosing search does, should hit the CPU cache. 2. Don't probe Tablebases with excludedMove. This can't be tested on fishtest, but it's obvious that tablebases don't deliver any information about suboptimal moves. Side note: Very surprisingly it looks like we cannot use static eval's from TT since they slightly differ over time due to changing optimism. Attempts to use static eval's from TT did loose about 13 ELO. This is something about to investigate. LTC: https://tests.stockfishchess.org/tests/view/63dc0f8de9d4cdfbe672d0c6 LLR: 2.94 (-2.94,2.94) <0.50,2.50> Total: 44736 W: 12046 L: 11733 D: 20957 Ptnml(0-2): 12, 4212, 13617, 4505, 22 An analogue of this passed STC & LTC see PR #4374 (thanks Dubslow for reviewing!) closes https://github.com/official-stockfish/Stockfish/pull/4380 Bench: 4758694
Copyright 2011–2024 Next Chess Move LLC
Feedback & Support Terms of Service Privacy Policy