Dev Builds » 20190104-1323

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 06:27:43 584467 4000 40 2757 1203 -287.58 ± 7.58 773 1172 54 1 0 -739.1 ± 47.42
ncm-dbt-02 06:29:08 586667 4000 39 2772 1189 -290.17 ± 7.64 786 1162 51 1 0 -748.8 ± 48.86
ncm-dbt-03 06:27:19 585848 4000 37 2785 1178 -292.63 ± 7.62 793 1162 45 0 0 -777.57 ± 52.22
ncm-dbt-04 06:28:26 567445 4022 40 2813 1169 -294.25 ± 7.8 817 1139 55 0 0 -743.23 ± 46.98
ncm-dbt-05 06:28:25 583934 3978 37 2774 1167 -293.31 ± 7.73 797 1143 49 0 0 -761.63 ± 49.92
20000 193 13901 5906 -291.58 ± 3.43 3966 5778 254 2 0 -753.51 ± 21.47

Test Detail

ID Host Base NPS Games WLD Standard Elo Ptnml(0-2) Gamepair Elo CLI PGN
474074 ncm-dbt-04 567126 22 0 13 9 -235.77 ± 62.44 2 9 0 0 0 -1199.83 ± 162.17
474073 ncm-dbt-05 584243 478 4 339 135 -301.89 ± 23.42 102 131 6 0 0 -758.26 ± 187.89
474072 ncm-dbt-01 585084 500 4 331 165 -271.78 ± 20.15 86 155 9 0 0 -694.7 ± 134.18
474071 ncm-dbt-02 588813 500 4 347 149 -291.96 ± 21.96 100 143 7 0 0 -739.05 ± 163.27
474070 ncm-dbt-03 586266 500 10 350 140 -288.06 ± 22.97 102 136 12 0 0 -643.67 ± 110.29
474069 ncm-dbt-04 564017 500 5 345 150 -288.06 ± 22.1 99 142 9 0 0 -694.7 ± 134.18
474068 ncm-dbt-05 583908 500 6 341 153 -281.67 ± 20.75 92 151 7 0 0 -739.05 ± 163.27
474067 ncm-dbt-01 584495 500 3 349 148 -295.93 ± 21.16 99 148 3 0 0 -887.58 ± 222.18
474066 ncm-dbt-03 586943 500 6 345 149 -286.77 ± 21.05 95 149 6 0 0 -766.17 ± 188.02
474065 ncm-dbt-02 586646 500 8 352 140 -293.28 ± 22.42 102 140 8 0 0 -715.51 ± 146.53
474064 ncm-dbt-04 568196 500 6 354 140 -298.61 ± 22.43 104 140 6 0 0 -766.17 ± 188.02
474063 ncm-dbt-05 582903 500 2 360 138 -312.47 ± 23.02 111 136 3 0 0 -887.58 ± 222.18
474062 ncm-dbt-01 584579 500 5 340 155 -281.67 ± 20.45 91 153 6 0 0 -766.17 ± 188.02
474061 ncm-dbt-02 584664 500 8 347 145 -286.77 ± 21.95 98 143 9 0 0 -694.7 ± 134.18
474060 ncm-dbt-03 582611 500 4 346 150 -290.65 ± 20.88 96 150 4 0 0 -837.28 ± 240.94
474059 ncm-dbt-04 568832 500 5 364 131 -313.9 ± 24.53 116 127 7 0 0 -739.05 ± 163.27
474058 ncm-dbt-05 581985 500 8 361 131 -305.44 ± 24.47 113 127 10 0 0 -676.04 ± 124.55
474057 ncm-dbt-01 585632 500 6 348 146 -290.66 ± 22.41 100 143 6 1 0 -715.55 ± 177.2
474056 ncm-dbt-02 588260 500 2 350 148 -298.61 ± 21.13 100 148 2 0 0 -958.31 ± 196.57
474055 ncm-dbt-03 587707 500 1 352 147 -302.69 ± 22.25 105 141 4 0 0 -837.28 ± 240.94
474054 ncm-dbt-04 568713 500 4 347 149 -291.96 ± 21.34 98 147 5 0 0 -798.18 ± 231.48
474053 ncm-dbt-03 586139 500 5 342 153 -284.21 ± 20.44 92 153 5 0 0 -798.18 ± 231.48
474052 ncm-dbt-01 582360 500 8 351 141 -291.96 ± 22.86 103 137 10 0 0 -676.04 ± 124.55
474051 ncm-dbt-02 586477 500 4 367 129 -319.71 ± 24.18 117 129 4 0 0 -837.28 ± 240.94
474050 ncm-dbt-05 584706 500 6 332 162 -270.57 ± 20.0 85 156 9 0 0 -694.7 ± 134.18
474049 ncm-dbt-04 566296 500 4 354 142 -301.32 ± 23.68 109 132 9 0 0 -694.7 ± 134.18
474048 ncm-dbt-03 586308 500 5 352 143 -297.27 ± 22.89 105 137 8 0 0 -715.51 ± 146.53
474047 ncm-dbt-02 589155 500 4 334 162 -275.45 ± 20.87 89 153 7 1 0 -694.74 ± 156.93
474046 ncm-dbt-01 585970 500 4 335 161 -276.68 ± 20.45 89 153 8 0 0 -715.51 ± 146.53
474045 ncm-dbt-05 586054 500 5 348 147 -291.96 ± 21.34 98 147 5 0 0 -798.18 ± 231.48
474044 ncm-dbt-04 567759 500 4 343 153 -286.77 ± 20.74 94 151 5 0 0 -798.18 ± 231.48
474043 ncm-dbt-01 585759 500 6 346 148 -288.06 ± 21.5 97 146 7 0 0 -739.05 ± 163.27
474042 ncm-dbt-03 583321 500 4 346 150 -290.65 ± 20.88 96 150 4 0 0 -837.28 ± 240.94
474041 ncm-dbt-02 584159 500 3 330 167 -271.78 ± 19.28 83 161 6 0 0 -766.17 ± 188.02
474040 ncm-dbt-05 580821 500 1 345 154 -293.28 ± 20.53 96 152 2 0 0 -958.31 ± 196.57
474039 ncm-dbt-04 568275 500 6 344 150 -285.48 ± 20.9 94 150 6 0 0 -766.17 ± 188.02
474038 ncm-dbt-01 581860 500 4 357 139 -305.44 ± 22.89 108 137 5 0 0 -798.18 ± 231.48
474037 ncm-dbt-03 587494 500 2 352 146 -301.32 ± 21.43 102 146 2 0 0 -958.31 ± 196.57
474036 ncm-dbt-05 586858 500 5 348 147 -291.96 ± 21.96 100 143 7 0 0 -739.05 ± 163.27
474035 ncm-dbt-04 567799 500 6 349 145 -291.96 ± 22.27 101 141 8 0 0 -715.51 ± 146.53
474034 ncm-dbt-02 585169 500 6 345 149 -286.77 ± 21.65 97 145 8 0 0 -715.51 ± 146.53

Commit

Commit ID 3c576efa77f431cf3687881b8fd6a728e87ed97d
Author Marco Costalba
Date 2019-01-04 13:23:14 UTC
Delay castling legality check Delay legality check of castling moves at search time, just before making the move, as is the standard with all the other move types. This should avoid an useless and not trivial legality check when the castling is then not tried later. For instance due to a previous cut-off. The patch is also a big simplification and allows to entirely remove generate_castling() Bench changes due to a different move sequence out of MovePicker. STC: LLR: 2.95 (-2.94,2.94) [-3.00,1.00] Total: 45073 W: 9918 L: 9843 D: 25312 http://tests.stockfishchess.org/tests/view/5c2f176f0ebc596a450bdfb3 LTC: LLR: 3.15 (-2.94,2.94) [-3.00,1.00] Total: 10156 W: 1707 L: 1560 D: 6889 http://tests.stockfishchess.org/tests/view/5c2e7dfd0ebc596a450bcdf4 Verified with perft both in standard and Chess960 cases. Closes https://github.com/official-stockfish/Stockfish/pull/1929 Bench: 3559104
Copyright 2011–2026 Next Chess Move LLC