Dev Builds » 20221105-0815

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 06:59:24 585238 4000 1143 937 1920 +17.91 ± 5.2 4 353 1082 555 6 +35.56 ± 10.31
ncm-dbt-02 06:59:43 587443 3992 1122 916 1954 +17.94 ± 5.13 5 338 1104 544 5 +35.99 ± 10.18
ncm-dbt-03 06:58:43 587790 3994 1124 911 1959 +18.55 ± 5.28 6 358 1055 573 5 +37.37 ± 10.46
ncm-dbt-04 06:59:10 569265 4014 1128 904 1982 +19.41 ± 5.14 6 336 1096 566 3 +39.47 ± 10.23
ncm-dbt-05 06:58:06 583656 4000 1156 886 1958 +23.49 ± 5.02 0 308 1120 566 6 +46.13 ± 10.08
20000 5673 4554 9773 +19.46 ± 2.31 21 1693 5457 2804 25 +38.9 ± 4.58

Test Detail

ID Host Base NPS Games WLD Standard Elo Ptnml(0-2) Gamepair Elo CLI PGN
396840 ncm-dbt-04 570749 14 2 4 8 -49.96 ± 94.42 0 3 3 1 0 -102.11 ± 234.06 ↓
396839 ncm-dbt-03 585211 494 138 117 239 +14.78 ± 14.31 1 40 144 61 1 +29.61 ± 28.0 ↓
396838 ncm-dbt-02 586012 492 141 104 247 +26.18 ± 15.35 0 44 122 79 1 +51.21 ± 30.91 ↓
396837 ncm-dbt-01 586223 500 138 112 250 +18.08 ± 13.89 0 40 144 66 0 +36.26 ± 28.05 ↓
396836 ncm-dbt-05 586054 500 126 104 270 +15.3 ± 13.79 0 41 146 63 0 +30.65 ± 27.79 ↓
396835 ncm-dbt-04 571511 500 137 125 238 +8.34 ± 14.13 1 45 145 59 0 +18.08 ± 27.96 ↓
396834 ncm-dbt-03 587834 500 156 111 233 +31.35 ± 15.6 0 45 116 88 1 +61.79 ± 31.68 ↓
396833 ncm-dbt-01 583866 500 142 104 254 +26.46 ± 15.52 1 43 125 79 2 +51.8 ± 30.53 ↓
396832 ncm-dbt-02 586774 500 143 118 239 +17.38 ± 13.83 0 39 148 62 1 +33.46 ± 27.51 ↓
396831 ncm-dbt-05 582277 500 139 106 255 +22.96 ± 14.66 0 41 137 70 2 +43.3 ± 28.97 ↓
396830 ncm-dbt-04 569989 500 139 116 245 +15.99 ± 13.85 0 41 145 64 0 +32.05 ± 27.93 ↓
396829 ncm-dbt-03 590868 500 141 106 253 +24.36 ± 14.63 0 42 131 77 0 +48.96 ± 29.75 ↓
396828 ncm-dbt-01 586731 500 141 121 238 +13.9 ± 14.85 0 49 133 67 1 +26.46 ± 29.52 ↓
396827 ncm-dbt-02 587579 500 136 105 259 +21.57 ± 15.06 2 41 131 76 0 +46.13 ± 29.76 ↓
396826 ncm-dbt-05 588217 500 153 112 235 +28.55 ± 14.79 0 41 127 82 0 +57.5 ± 30.27 ↓
396825 ncm-dbt-04 567720 500 131 105 264 +18.08 ± 14.67 2 40 138 70 0 +39.08 ± 28.85 ↓
396824 ncm-dbt-03 588728 500 138 130 232 +5.56 ± 15.41 1 57 125 67 0 +12.51 ± 30.53 ↓
396823 ncm-dbt-01 585717 500 152 122 226 +20.87 ± 15.01 1 43 132 73 1 +41.89 ± 29.63 ↓
396822 ncm-dbt-02 587579 500 144 110 246 +23.66 ± 14.05 0 36 146 66 2 +44.72 ± 27.74 ↓
396821 ncm-dbt-05 582152 500 151 117 232 +23.66 ± 14.05 0 37 143 69 1 +46.13 ± 28.15 ↓
396820 ncm-dbt-04 569071 500 147 105 248 +29.25 ± 15.09 1 39 128 81 1 +58.93 ± 30.13 ↓
396819 ncm-dbt-03 588472 500 143 119 238 +16.69 ± 14.44 0 44 139 66 1 +32.05 ± 28.73 ↓
396818 ncm-dbt-01 580489 500 146 113 241 +22.96 ± 14.53 0 42 133 75 0 +46.13 ± 29.5 ↓
396817 ncm-dbt-02 586097 500 147 123 230 +16.69 ± 14.69 1 43 138 67 1 +33.46 ± 28.86 ↓
396816 ncm-dbt-05 583363 500 140 115 245 +17.39 ± 13.42 0 37 151 62 0 +34.86 ± 27.08 ↓
396815 ncm-dbt-04 568037 500 132 107 261 +17.39 ± 15.12 0 50 125 75 0 +34.86 ± 30.53 ↓
396814 ncm-dbt-03 586943 500 131 107 262 +16.69 ± 15.07 2 43 135 69 1 +34.86 ± 29.25 ↓
396813 ncm-dbt-01 585717 500 139 118 243 +14.6 ± 14.91 2 43 138 66 1 +30.65 ± 28.87 ↓
396812 ncm-dbt-02 588132 500 138 115 247 +15.99 ± 13.99 0 42 143 65 0 +32.05 ± 28.2 ↓
396811 ncm-dbt-05 579248 500 160 109 231 +35.56 ± 14.35 0 33 134 82 1 +70.44 ± 29.29 ↓
396810 ncm-dbt-04 566730 500 143 113 244 +20.87 ± 14.88 1 42 134 72 1 +41.89 ± 29.37 ↓
396809 ncm-dbt-03 586562 500 146 108 246 +26.46 ± 14.9 1 40 129 80 0 +54.65 ± 30.01 ↓
396808 ncm-dbt-02 587622 500 134 112 254 +15.3 ± 14.45 1 43 139 67 0 +32.05 ± 28.73 ↓
396807 ncm-dbt-01 585970 500 131 120 249 +7.64 ± 14.59 0 52 135 63 0 +15.3 ± 29.27 ↓
396806 ncm-dbt-05 584285 500 140 110 250 +20.87 ± 13.57 0 36 148 66 0 +41.89 ± 27.47 ↓
396805 ncm-dbt-04 569829 500 141 102 257 +27.15 ± 14.18 0 36 140 73 1 +53.22 ± 28.53 ↓
396804 ncm-dbt-03 587707 500 131 113 256 +12.51 ± 14.87 1 47 136 65 1 +25.06 ± 29.14 ↓
396803 ncm-dbt-05 583656 500 147 113 240 +23.66 ± 14.83 0 42 134 72 2 +44.72 ± 29.37 ↓
396802 ncm-dbt-01 587197 500 154 127 219 +18.78 ± 14.21 0 41 142 66 1 +36.26 ± 28.32 ↓
396801 ncm-dbt-02 589753 500 139 129 232 +6.95 ± 14.66 1 50 137 62 0 +15.3 ± 29.02 ↓
396800 ncm-dbt-04 569749 500 156 127 217 +20.18 ± 14.45 1 40 138 71 0 +41.89 ± 28.84 ↓

Commit

Commit ID ad2aa8c06f438de8b8bb7b7c8726430e3f2a5685
Author Joost VandeVondele
Date 2022-11-05 08:15:53 UTC
Normalize evaluation Normalizes the internal value as reported by evaluate or search to the UCI centipawn result used in output. This value is derived from the win_rate_model() such that Stockfish outputs an advantage of "100 centipawns" for a position if the engine has a 50% probability to win from this position in selfplay at fishtest LTC time control. The reason to introduce this normalization is that our evaluation is, since NNUE, no longer related to the classical parameter PawnValueEg (=208). This leads to the current evaluation changing quite a bit from release to release, for example, the eval needed to have 50% win probability at fishtest LTC (in cp and internal Value): June 2020 : 113cp (237) June 2021 : 115cp (240) April 2022 : 134cp (279) July 2022 : 167cp (348) With this patch, a 100cp advantage will have a fixed interpretation, i.e. a 50% win chance. To keep this value steady, it will be needed to update the win_rate_model() from time to time, based on fishtest data. This analysis can be performed with a set of scripts currently available at https://github.com/vondele/WLD_model fixes https://github.com/official-stockfish/Stockfish/issues/4155 closes https://github.com/official-stockfish/Stockfish/pull/4216 No functional change
Copyright 2011–2025 Next Chess Move LLC
Feedback & Support Terms of Service Privacy Policy Desktop Version