Dev Builds » 20241208-1852

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 06:50:49 583687 4000 1570 513 1917 +94.04 ± 4.86 1 88 777 1121 13 +201.43 ± 12.19
ncm-dbt-02 06:49:18 587326 4000 1576 550 1874 +91.15 ± 4.82 1 83 819 1083 14 +193.87 ± 11.84
ncm-dbt-03 06:50:44 587005 4000 1550 510 1940 +92.46 ± 4.85 0 85 808 1089 18 +195.98 ± 11.93
ncm-dbt-04 06:49:39 569870 4000 1568 502 1930 +94.88 ± 4.88 1 88 770 1126 15 +203.11 ± 12.25
ncm-dbt-05 06:50:10 583794 4000 1531 535 1934 +88.37 ± 4.98 2 104 806 1072 16 +186.7 ± 11.98
20000 7795 2610 9595 +92.18 ± 2.18 5 448 3980 5491 76 +196.17 ± 5.38

Test Detail

ID Host Base NPS Games WLD Standard Elo Ptnml(0-2) Gamepair Elo CLI PGN
390146 ncm-dbt-01 581985 500 206 64 230 +101.46 ± 13.69 0 10 90 148 2 +219.87 ± 36.1
390145 ncm-dbt-04 570949 500 200 50 250 +107.54 ± 14.05 1 10 78 160 1 +240.82 ± 38.87
390144 ncm-dbt-03 585421 500 195 66 239 +91.71 ± 12.93 0 6 111 131 2 +194.57 ± 31.84
390143 ncm-dbt-05 580157 500 186 75 239 +78.43 ± 13.88 0 15 110 124 1 +164.07 ± 32.48
390142 ncm-dbt-02 587792 500 201 59 240 +101.46 ± 12.87 0 6 97 146 1 +221.9 ± 34.46
390141 ncm-dbt-01 584874 500 192 67 241 +88.74 ± 14.13 0 14 99 135 2 +187.16 ± 34.38
390140 ncm-dbt-04 569829 500 203 74 223 +91.71 ± 13.73 0 9 107 130 4 +190.85 ± 32.75
390139 ncm-dbt-03 585211 500 197 55 248 +101.46 ± 13.53 0 9 92 147 2 +219.87 ± 35.65
390138 ncm-dbt-05 579702 500 196 74 230 +86.52 ± 13.93 0 14 101 134 1 +183.51 ± 34.01
390137 ncm-dbt-02 584243 500 183 72 245 +78.43 ± 13.88 0 15 110 124 1 +164.07 ± 32.48
390136 ncm-dbt-01 584706 500 193 70 237 +87.26 ± 14.1 0 13 104 130 3 +181.7 ± 33.45
390135 ncm-dbt-04 570629 500 198 58 244 +99.95 ± 14.0 0 12 88 148 2 +215.85 ± 36.55
390134 ncm-dbt-03 588345 500 195 73 232 +86.52 ± 13.93 0 11 110 125 4 +178.11 ± 32.33
390133 ncm-dbt-05 582277 500 187 64 249 +87.26 ± 13.64 0 11 107 130 2 +183.51 ± 32.85
390132 ncm-dbt-02 586604 500 201 66 233 +96.19 ± 14.56 0 15 88 144 3 +204.07 ± 36.53
390131 ncm-dbt-01 586139 500 196 73 231 +87.26 ± 12.69 0 6 116 127 1 +185.33 ± 30.99
390130 ncm-dbt-04 570268 500 182 64 254 +83.57 ± 14.31 0 17 99 133 1 +176.33 ± 34.4
390129 ncm-dbt-03 587155 500 195 57 248 +98.44 ± 13.34 0 7 101 139 3 +209.91 ± 33.74
390128 ncm-dbt-05 585632 500 190 60 250 +92.45 ± 14.35 0 14 95 138 3 +194.57 ± 35.13
390127 ncm-dbt-02 590868 500 200 68 232 +93.95 ± 12.96 0 7 105 137 1 +202.15 ± 32.99
390126 ncm-dbt-01 585084 500 195 51 254 +102.97 ± 12.87 0 6 95 148 1 +226.0 ± 34.87
390125 ncm-dbt-04 568116 500 186 63 251 +87.26 ± 13.8 0 11 108 128 3 +181.7 ± 32.68
390124 ncm-dbt-03 585843 500 195 58 247 +97.69 ± 13.5 0 7 103 136 4 +206.01 ± 33.36
390123 ncm-dbt-05 587834 500 191 63 246 +90.96 ± 13.72 0 11 102 135 2 +192.71 ± 33.75
390122 ncm-dbt-02 586731 500 200 77 223 +87.26 ± 13.33 0 10 108 131 1 +185.33 ± 32.63
390121 ncm-dbt-01 583112 500 197 70 233 +90.22 ± 13.23 0 10 103 137 0 +194.57 ± 33.53
390120 ncm-dbt-03 586097 500 186 67 247 +84.3 ± 14.33 0 16 101 131 2 +176.33 ± 34.04
390119 ncm-dbt-04 570629 500 194 64 242 +92.45 ± 13.59 0 12 96 142 0 +200.24 ± 34.91
390118 ncm-dbt-05 586097 500 201 65 234 +96.94 ± 14.57 1 13 87 147 2 +209.91 ± 36.75
390117 ncm-dbt-02 585759 500 206 63 231 +102.22 ± 13.54 0 9 91 148 2 +221.9 ± 35.86
390116 ncm-dbt-01 578259 500 200 60 240 +99.95 ± 15.05 1 16 77 154 2 +217.85 ± 38.88
390115 ncm-dbt-04 569430 500 206 72 222 +95.44 ± 13.63 0 12 92 146 0 +207.95 ± 35.71
390114 ncm-dbt-03 588345 500 189 68 243 +85.78 ± 13.61 0 13 103 134 0 +183.51 ± 33.63
390113 ncm-dbt-05 583992 500 197 74 229 +87.26 ± 14.25 1 10 108 127 4 +181.7 ± 32.68
390112 ncm-dbt-02 589796 500 197 73 230 +88.0 ± 13.03 0 7 114 127 2 +185.33 ± 31.41
390111 ncm-dbt-04 569111 500 199 57 244 +101.46 ± 13.2 0 5 102 139 4 +215.85 ± 33.4
390110 ncm-dbt-01 585337 500 191 58 251 +94.69 ± 14.09 0 13 93 142 2 +202.15 ± 35.52
390109 ncm-dbt-05 584664 500 183 60 257 +87.26 ± 14.25 0 16 96 137 1 +185.33 ± 34.95
390108 ncm-dbt-02 586816 500 188 72 240 +82.1 ± 14.56 1 14 106 126 3 +171.02 ± 33.15
390107 ncm-dbt-03 589625 500 198 66 236 +93.95 ± 14.38 0 16 87 146 1 +202.15 ± 36.72

Commit

Commit ID 713000c517c63e6926bdbe1071e647280bc3da32
Author pb00067
Date 2024-12-08 18:52:16 UTC
Same weight for black and white nonPawnCorrection history Since we don't have color dependent parameters in NNUE eval, it also has no sense IMO to have color dependent parameters in correction histories. Ideally a fixed depth search on a single thread should be determistic, so delivering the same result (move) if we just flip colors on the board. Patch replaces 2 parameters (122 and 185) with just one value 154 (= the avg of the two). Passed STC-non regression https://tests.stockfishchess.org/tests/view/6740a63286d5ee47d953f656 LLR: 2.95 (-2.94,2.94) <-1.75,0.25> Total: 122336 W: 31499 L: 31372 D: 59465 Ptnml(0-2): 336, 14535, 31301, 14658, 338 Passed LTC-non regression https://tests.stockfishchess.org/tests/view/67419bae86d5ee47d953f7b6 LLR: 2.94 (-2.94,2.94) <-1.75,0.25> Total: 101400 W: 25870 L: 25731 D: 49799 Ptnml(0-2): 78, 11109, 28166, 11290, 57 closes https://github.com/official-stockfish/Stockfish/pull/5698 Bench: 1215483
Copyright 2011–2024 Next Chess Move LLC