Dev Builds » 20210723-1647

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 06:55:55 571078 4020 536 1532 1952 -87.91 ± 5.13 33 1047 815 113 2 -181.75 ± 11.92
ncm-dbt-02 06:55:30 580129 4008 511 1516 1981 -89.02 ± 5.07 29 1058 806 111 0 -184.89 ± 11.98
ncm-dbt-03 06:55:42 578588 3996 510 1530 1956 -90.69 ± 5.02 33 1050 819 96 0 -188.08 ± 11.86
ncm-dbt-04 06:55:38 565141 4002 480 1519 2003 -92.31 ± 5.07 26 1098 766 111 0 -193.75 ± 12.31
ncm-dbt-05 06:55:29 580744 3974 491 1516 1967 -91.68 ± 5.2 41 1053 785 106 2 -189.1 ± 12.15
20000 2528 7613 9859 -90.32 ± 2.28 162 5306 3991 537 4 -187.48 ± 5.38

Test Detail

ID Host Base NPS Games WLD Standard Elo Ptnml(0-2) Gamepair Elo CLI PGN
418281 ncm-dbt-03 582193 42 3 17 22 -120.38 ± 39.67 0 14 7 0 0 -279.54 ± 142.86
418280 ncm-dbt-05 584076 42 7 14 21 -58.43 ± 64.67 1 9 7 4 0 -102.11 ± 131.14
418279 ncm-dbt-02 586223 54 5 18 31 -85.31 ± 48.15 0 16 8 3 0 -182.37 ± 126.06
418278 ncm-dbt-04 568991 50 8 17 25 -63.22 ± 44.25 0 11 12 2 0 -130.94 ± 101.68
418277 ncm-dbt-01 576496 60 8 22 30 -82.58 ± 37.11 0 15 14 1 0 -175.73 ± 92.83
418276 ncm-dbt-03 584832 500 70 200 230 -92.45 ± 14.05 5 130 105 10 0 -190.85 ± 33.16
418275 ncm-dbt-05 578218 500 60 210 230 -107.54 ± 14.05 7 143 93 7 0 -226.0 ± 35.35
418274 ncm-dbt-04 568792 500 58 201 241 -102.22 ± 15.37 7 145 82 16 0 -211.87 ± 37.79
418273 ncm-dbt-02 586731 500 79 197 224 -83.57 ± 14.74 3 130 99 18 0 -172.78 ± 34.4
418272 ncm-dbt-01 571471 500 62 197 241 -96.19 ± 13.16 4 132 109 5 0 -202.15 ± 32.11
418271 ncm-dbt-05 574337 500 70 181 249 -78.44 ± 14.31 3 120 113 13 1 -162.35 ± 31.95
418270 ncm-dbt-04 571913 500 52 181 267 -91.71 ± 13.09 1 135 106 8 0 -196.45 ± 32.87
418269 ncm-dbt-03 583950 500 69 189 242 -85.04 ± 14.49 5 124 107 14 0 -172.78 ± 32.96
418268 ncm-dbt-02 587070 500 59 185 256 -89.48 ± 13.53 2 132 106 10 0 -189.0 ± 32.98
418267 ncm-dbt-01 578177 500 67 192 241 -88.74 ± 14.43 2 137 95 16 0 -187.16 ± 35.14
418266 ncm-dbt-05 582485 500 59 199 242 -99.95 ± 14.15 7 134 101 8 0 -206.01 ± 33.8
418265 ncm-dbt-04 569509 500 66 194 240 -90.96 ± 13.72 3 132 105 10 0 -190.85 ± 33.16
418264 ncm-dbt-03 582694 500 65 176 259 -78.43 ± 14.6 3 123 106 18 0 -160.64 ± 33.2
418263 ncm-dbt-01 577110 500 72 196 232 -88.0 ± 14.12 4 128 106 12 0 -181.7 ± 33.07
418262 ncm-dbt-02 586985 500 52 195 253 -102.22 ± 14.18 5 143 92 10 0 -215.85 ± 35.68
418261 ncm-dbt-05 582694 500 65 192 243 -90.22 ± 15.03 6 131 97 16 0 -183.51 ± 34.76
418260 ncm-dbt-01 577356 500 73 194 233 -85.78 ± 15.49 8 122 104 15 1 -171.02 ± 33.51
418259 ncm-dbt-03 586604 500 75 190 235 -81.36 ± 14.68 3 127 102 18 0 -167.53 ± 33.88
418258 ncm-dbt-04 569549 500 61 178 261 -82.83 ± 14.86 1 136 92 21 0 -174.55 ± 35.63
418257 ncm-dbt-02 583112 500 67 188 245 -85.78 ± 14.51 3 131 100 16 0 -178.11 ± 34.22
418256 ncm-dbt-05 610571 500 58 178 264 -85.04 ± 14.78 4 129 100 17 0 -174.55 ± 34.22
418255 ncm-dbt-03 583070 500 61 196 243 -96.19 ± 14.71 6 136 95 13 0 -198.34 ± 35.12
418254 ncm-dbt-01 576496 500 66 195 239 -91.71 ± 14.34 5 131 102 12 0 -189.0 ± 33.78
418253 ncm-dbt-04 570589 500 56 178 266 -86.52 ± 14.81 3 134 95 18 0 -179.9 ± 35.13
418252 ncm-dbt-02 585843 500 70 188 242 -83.57 ± 14.01 3 125 109 13 0 -172.78 ± 32.58
418251 ncm-dbt-02 530974 500 57 192 251 -96.19 ± 14.11 3 141 94 12 0 -204.07 ± 35.31
418250 ncm-dbt-05 548714 500 74 184 242 -77.7 ± 14.86 4 121 106 19 0 -157.24 ± 33.21
418249 ncm-dbt-03 536086 500 55 200 245 -103.73 ± 13.22 4 142 99 5 0 -221.9 ± 33.98
418248 ncm-dbt-01 528214 500 69 193 238 -88.0 ± 14.27 3 132 101 14 0 -183.51 ± 34.01
418247 ncm-dbt-04 525789 500 59 197 244 -98.44 ± 15.03 5 144 85 16 0 -206.01 ± 37.14
418246 ncm-dbt-05 584076 432 42 172 218 -107.89 ± 16.25 7 126 74 8 1 -227.08 ± 39.92
418245 ncm-dbt-02 588004 454 70 176 208 -82.64 ± 15.6 4 114 93 16 0 -168.11 ± 35.49
418244 ncm-dbt-03 585464 454 50 165 239 -89.96 ± 14.41 2 121 94 10 0 -189.83 ± 35.15
418243 ncm-dbt-04 569949 452 57 173 222 -91.2 ± 14.66 3 120 93 10 0 -190.85 ± 35.35
418242 ncm-dbt-01 575883 460 52 160 248 -83.12 ± 15.31 1 124 87 18 0 -175.09 ± 36.71
418241 ncm-dbt-02 586223 500 52 177 271 -88.74 ± 14.57 6 126 105 13 0 -179.9 ± 33.27
418240 ncm-dbt-05 581527 500 56 186 258 -92.45 ± 14.2 2 140 94 14 0 -196.45 ± 35.33
418239 ncm-dbt-04 571190 500 63 200 237 -97.69 ± 13.82 3 141 96 10 0 -207.95 ± 34.86
418238 ncm-dbt-03 582402 500 62 197 241 -96.19 ± 13.8 5 133 104 8 0 -200.24 ± 33.24
418237 ncm-dbt-01 578506 500 67 183 250 -82.1 ± 15.79 6 126 97 20 1 -165.8 ± 34.73

Commit

Commit ID d957179df7285f8d032803661b378b7f1a80382e
Author Vizvezdenec
Date 2021-07-23 16:47:30 UTC
Prune illegal moves in qsearch earlier The main idea is that illegal moves influencing search or qsearch obviously can't be any sort of good. The only reason why initially legality checks for search and qsearch were done after they actually can influence some heuristics is because legality check is expensive computationally. Eventually in search it was moved to the place where it makes sure that illegal moves can't influence search. This patch shows that the same can be done for qsearch + it passed STC with elo-gaining bounds + it removes 3 lines of code because one no longer needs to increment/decrement movecount on illegal moves. passed STC with elo-gaining bounds https://tests.stockfishchess.org/tests/view/60f20aefd1189bed71812da0 LLR: 2.94 (-2.94,2.94) <-0.50,2.50> Total: 61512 W: 4688 L: 4492 D: 52332 Ptnml(0-2): 139, 3730, 22848, 3874, 165 The same version functionally but with moving condition ever earlier passed LTC with simplification bounds. https://tests.stockfishchess.org/tests/view/60f292cad1189bed71812de9 LLR: 2.98 (-2.94,2.94) <-2.50,0.50> Total: 60944 W: 1724 L: 1685 D: 57535 Ptnml(0-2): 11, 1556, 27298, 1597, 10 closes https://github.com/official-stockfish/Stockfish/pull/3618 bench 4709569
Copyright 2011–2025 Next Chess Move LLC