Dev Builds » 20200928-2042

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 06:40:20 582187 4000 275 1848 1877 -144.41 ± 4.79 91 1415 470 24 0 -331.0 ± 15.72
ncm-dbt-02 06:41:39 587204 3996 292 1874 1830 -145.5 ± 4.79 90 1428 454 26 0 -335.47 ± 16.0
ncm-dbt-03 06:41:41 585524 4014 253 1888 1873 -150.23 ± 4.83 114 1422 457 13 1 -344.76 ± 15.92
ncm-dbt-04 06:41:02 566911 4000 239 1865 1896 -149.89 ± 4.84 101 1452 420 26 1 -348.6 ± 16.65
ncm-dbt-05 06:42:15 581920 3990 236 1877 1877 -151.88 ± 4.63 89 1482 405 19 0 -361.38 ± 16.95
20000 1295 9352 9353 -148.37 ± 2.14 485 7199 2206 108 2 -343.99 ± 7.25

Test Detail

ID Host Base NPS Games WLD Standard Elo Ptnml(0-2) Gamepair Elo CLI PGN
447938 ncm-dbt-03 587452 14 0 6 8 -159.07 ± 56.3 0 6 1 0 0 -445.23 ± 480.27
447937 ncm-dbt-05 580945 490 26 233 231 -156.58 ± 13.13 11 188 43 3 0 -381.7 ± 53.09
447936 ncm-dbt-02 587282 496 49 229 218 -132.1 ± 13.74 6 175 60 7 0 -302.43 ± 44.5
447935 ncm-dbt-04 566414 500 26 231 243 -151.34 ± 12.36 8 191 49 2 0 -370.41 ± 49.56
447934 ncm-dbt-01 579661 500 35 237 228 -148.85 ± 14.09 17 169 63 1 0 -330.23 ± 43.2
447933 ncm-dbt-03 587325 500 30 232 238 -148.84 ± 13.31 14 174 62 0 0 -339.63 ± 43.41
447932 ncm-dbt-05 581194 500 33 237 230 -150.51 ± 13.64 14 178 56 2 0 -346.12 ± 46.15
447931 ncm-dbt-02 588728 500 37 235 228 -145.54 ± 14.39 15 172 59 4 0 -324.17 ± 44.97
447930 ncm-dbt-04 567324 500 32 246 222 -158.93 ± 14.04 18 180 50 2 0 -366.78 ± 49.02
447929 ncm-dbt-01 582068 500 36 230 234 -142.25 ± 13.94 12 174 60 4 0 -321.19 ± 44.57
447928 ncm-dbt-03 586858 500 37 248 215 -156.39 ± 13.77 14 187 45 4 0 -370.41 ± 51.77
447927 ncm-dbt-05 582611 500 33 229 238 -143.89 ± 12.92 8 183 56 3 0 -339.63 ± 46.2
447926 ncm-dbt-02 587325 500 31 237 232 -152.18 ± 12.95 12 183 54 1 0 -359.68 ± 46.95
447925 ncm-dbt-01 582193 500 30 236 234 -152.18 ± 13.36 12 185 50 3 0 -359.68 ± 49.05
447924 ncm-dbt-04 568991 500 27 229 244 -148.85 ± 13.51 11 183 54 1 1 -352.8 ± 47.05
447923 ncm-dbt-03 584790 500 32 233 235 -148.02 ± 14.3 17 169 62 2 0 -327.18 ± 43.69
447922 ncm-dbt-02 587834 500 46 233 221 -136.56 ± 12.59 7 174 68 1 0 -315.35 ± 41.44
447921 ncm-dbt-05 579744 500 32 232 236 -147.19 ± 12.36 7 188 53 2 0 -356.21 ± 47.52
447920 ncm-dbt-01 583950 500 35 230 235 -143.07 ± 13.73 12 174 61 3 0 -324.17 ± 44.15
447919 ncm-dbt-04 568116 500 32 235 233 -149.68 ± 14.24 16 174 57 3 0 -336.46 ± 45.77
447918 ncm-dbt-03 582861 500 32 238 230 -152.18 ± 12.74 9 190 50 0 1 -374.11 ± 48.94
447917 ncm-dbt-02 585295 500 31 233 236 -148.85 ± 14.46 16 174 56 4 0 -333.32 ± 46.21
447916 ncm-dbt-05 580323 500 24 230 246 -152.18 ± 13.36 14 179 56 1 0 -352.8 ± 46.04
447915 ncm-dbt-01 581610 500 50 228 222 -129.35 ± 14.07 11 160 75 4 0 -280.42 ± 39.57
447914 ncm-dbt-04 567205 500 37 230 233 -141.44 ± 14.33 11 178 54 7 0 -321.19 ± 46.91
447913 ncm-dbt-03 583824 500 25 231 244 -152.18 ± 13.36 13 182 53 2 0 -356.21 ± 47.52
447912 ncm-dbt-04 565233 500 30 225 245 -143.07 ± 14.1 13 173 60 4 0 -321.19 ± 44.57
447911 ncm-dbt-02 587494 500 36 229 235 -141.44 ± 12.61 7 181 60 2 0 -333.32 ± 44.47
447910 ncm-dbt-05 580738 500 33 241 226 -153.86 ± 12.87 11 188 49 2 0 -370.41 ± 49.56
447909 ncm-dbt-01 584327 500 34 229 237 -143.07 ± 12.34 5 188 54 3 0 -346.12 ± 47.1
447908 ncm-dbt-03 586520 500 32 244 224 -157.23 ± 13.93 19 174 57 0 0 -356.21 ± 45.43
447907 ncm-dbt-04 564644 500 25 232 243 -153.02 ± 13.92 14 183 49 4 0 -356.21 ± 49.54
447906 ncm-dbt-02 587409 500 29 244 227 -159.78 ± 13.8 17 183 48 2 0 -374.11 ± 50.1
447905 ncm-dbt-03 584327 500 26 236 238 -155.54 ± 13.61 15 182 51 2 0 -363.2 ± 48.51
447904 ncm-dbt-05 580779 500 29 241 230 -157.23 ± 12.9 13 187 49 1 0 -377.87 ± 49.47
447903 ncm-dbt-01 579909 500 24 221 255 -144.71 ± 12.68 8 183 57 2 0 -342.85 ± 45.71
447902 ncm-dbt-04 567363 500 30 237 233 -153.02 ± 12.91 10 190 47 3 0 -370.41 ± 50.67
447901 ncm-dbt-02 586266 500 33 234 233 -148.02 ± 13.54 10 186 49 5 0 -349.43 ± 49.46
447900 ncm-dbt-03 585759 500 39 220 241 -131.74 ± 14.2 13 158 76 3 0 -282.94 ± 39.21
447899 ncm-dbt-01 583782 500 31 237 232 -152.18 ± 13.96 14 182 50 4 0 -352.8 ± 49.02
447898 ncm-dbt-05 589027 500 26 234 240 -153.86 ± 13.49 11 191 43 5 0 -370.41 ± 52.84

Commit

Commit ID 5af09cfda5b71f9470ef233298e0f4233651337d
Author Stéphane Nicolet
Date 2020-09-28 20:42:26 UTC
Include pawns in NNUE scaling We now include the total pawn count in the scaling factor for the output of the NNUE evaluation network. This should have the effect of trying to keep more pawns when SF has the advantage, but exchange them when she is defending. Thanks to Alexander Pagel (Lolligerhans) for the idea of using the value of pawns to ease the comparison with the rest of the material estimation. Passed STC: LLR: 2.93 (-2.94,2.94) {-0.25,1.25} Total: 15072 W: 1700 L: 1539 D: 11833 Ptnml(0-2): 65, 1202, 4845, 1355, 69 https://tests.stockfishchess.org/tests/view/5f7235a63b22d6afa50699b3 Passed LTC: LLR: 2.93 (-2.94,2.94) {0.25,1.25} Total: 25880 W: 1270 L: 1124 D: 23486 Ptnml(0-2): 23, 980, 10788, 1126, 23 https://tests.stockfishchess.org/tests/view/5f723b483b22d6afa5069a99 closes https://github.com/official-stockfish/Stockfish/pull/3164 Bench: 3776081
Copyright 2011–2026 Next Chess Move LLC