Dev Builds » 20250724-0808

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 06:43:01 581747 4000 1653 462 1885 +106.68 ± 4.78 1 63 701 1214 21 +233.06 ± 12.84
ncm-dbt-02 06:42:10 585164 4000 1652 435 1913 +109.16 ± 4.76 2 63 666 1254 15 +241.91 ± 13.19
ncm-dbt-03 06:43:21 584119 4000 1637 436 1927 +107.63 ± 4.68 0 57 701 1226 16 +236.78 ± 12.83
ncm-dbt-04 06:44:05 568386 4018 1657 495 1866 +103.43 ± 4.79 1 65 736 1185 22 +223.92 ± 12.52
ncm-dbt-05 06:43:45 578664 3982 1643 453 1886 +107.1 ± 4.71 0 58 702 1214 17 +234.99 ± 12.82
20000 8242 2281 9477 +106.79 ± 2.12 4 306 3506 6093 91 +234.06 ± 5.74

Test Detail

ID Host Base NPS Games WLD Standard Elo Ptnml(0-2) Gamepair Elo CLI PGN
424149 ncm-dbt-04 567324 18 7 2 9 +99.09 ± 86.45 0 1 2 6 0 +217.63 ± 581.07
424148 ncm-dbt-05 579413 482 192 55 235 +101.55 ± 13.15 0 6 93 141 1 +222.06 ± 35.23
424147 ncm-dbt-03 582569 500 218 54 228 +118.33 ± 13.35 0 4 84 156 6 +258.75 ± 37.2
424146 ncm-dbt-01 580613 500 209 52 239 +112.91 ± 13.22 0 6 84 157 3 +249.64 ± 37.33
424145 ncm-dbt-02 585548 500 207 49 244 +113.68 ± 13.89 2 4 81 160 3 +256.44 ± 38.07
424144 ncm-dbt-04 567126 500 203 66 231 +97.69 ± 13.5 0 8 100 139 3 +207.95 ± 33.99
424143 ncm-dbt-05 581277 500 214 54 232 +115.22 ± 13.2 0 7 78 163 2 +258.75 ± 38.88
424142 ncm-dbt-03 584537 500 209 63 228 +104.49 ± 13.05 0 8 88 154 0 +232.26 ± 36.48
424141 ncm-dbt-01 580613 500 202 55 243 +105.25 ± 13.22 0 8 88 153 1 +232.26 ± 36.48
424140 ncm-dbt-02 583824 500 205 57 238 +106.01 ± 13.88 0 11 82 155 2 +232.26 ± 37.9
424139 ncm-dbt-04 567918 500 196 59 245 +97.69 ± 13.01 0 7 100 142 1 +211.87 ± 33.93
424138 ncm-dbt-05 580613 500 214 52 234 +116.77 ± 13.36 0 7 77 163 3 +261.07 ± 39.14
424137 ncm-dbt-03 584958 500 190 44 266 +104.49 ± 13.39 0 8 90 150 2 +228.08 ± 36.04
424136 ncm-dbt-01 581444 500 201 67 232 +95.44 ± 13.94 0 11 97 139 3 +202.15 ± 34.7
424135 ncm-dbt-02 585126 500 203 68 229 +96.19 ± 14.26 0 13 92 142 3 +204.07 ± 35.72
424134 ncm-dbt-04 566257 500 207 61 232 +104.49 ± 12.71 0 4 98 146 2 +228.08 ± 34.09
424133 ncm-dbt-05 585000 500 206 50 244 +112.14 ± 13.22 0 6 85 156 3 +247.41 ± 37.09
424132 ncm-dbt-03 584076 500 197 67 236 +92.45 ± 13.74 0 12 97 140 1 +198.34 ± 34.72
424131 ncm-dbt-01 581361 500 204 68 228 +96.94 ± 13.65 0 10 96 142 2 +207.95 ± 34.86
424130 ncm-dbt-02 586350 500 215 56 229 +114.45 ± 13.55 0 8 78 161 3 +254.16 ± 38.89
424129 ncm-dbt-04 570068 500 204 63 233 +100.7 ± 14.01 0 11 90 146 3 +215.85 ± 36.11
424128 ncm-dbt-05 577520 500 213 65 222 +106.01 ± 13.05 0 7 89 153 1 +234.38 ± 36.22
424127 ncm-dbt-02 584327 500 208 44 248 +118.33 ± 12.26 0 4 78 168 0 +273.0 ± 38.74
424126 ncm-dbt-01 582694 500 214 52 234 +116.77 ± 13.01 0 3 87 155 5 +256.44 ± 36.38
424125 ncm-dbt-03 584495 500 203 48 249 +111.37 ± 13.56 0 9 79 160 2 +247.41 ± 38.64
424124 ncm-dbt-04 571873 500 208 64 228 +102.97 ± 13.54 1 6 93 148 2 +226.0 ± 35.35
424123 ncm-dbt-05 575476 500 201 57 242 +102.97 ± 13.71 0 8 94 144 4 +219.87 ± 35.19
424122 ncm-dbt-02 586858 500 205 40 255 +119.11 ± 13.17 0 7 73 168 2 +270.57 ± 40.25
424121 ncm-dbt-01 580613 500 207 48 245 +114.45 ± 12.85 0 5 83 160 2 +256.44 ± 37.51
424120 ncm-dbt-03 584579 500 210 49 241 +116.0 ± 12.47 0 3 85 160 2 +261.07 ± 36.86
424119 ncm-dbt-04 568514 500 213 55 232 +113.68 ± 13.04 0 7 79 163 1 +256.44 ± 38.62
424118 ncm-dbt-02 585042 500 202 68 230 +95.44 ± 12.81 0 7 102 141 0 +207.95 ± 33.55
424117 ncm-dbt-05 574094 500 201 59 240 +101.46 ± 13.04 0 7 95 147 1 +221.9 ± 34.93
424116 ncm-dbt-03 582569 500 203 64 233 +99.2 ± 13.02 0 7 98 144 1 +215.85 ± 34.33
424115 ncm-dbt-04 566770 500 211 65 224 +104.49 ± 14.66 0 12 86 146 6 +219.87 ± 36.98
424114 ncm-dbt-01 583070 500 207 64 229 +102.22 ± 14.33 0 13 84 150 3 +219.87 ± 37.42
424113 ncm-dbt-02 584243 500 207 53 240 +110.6 ± 13.56 0 9 80 159 2 +245.2 ± 38.39
424112 ncm-dbt-05 575924 500 202 61 237 +100.7 ± 13.69 0 10 91 147 2 +217.85 ± 35.89
424111 ncm-dbt-03 585169 500 207 47 246 +115.22 ± 13.03 0 6 80 162 2 +258.75 ± 38.32
424110 ncm-dbt-04 569629 500 208 60 232 +106.01 ± 13.88 0 9 88 149 4 +228.08 ± 36.51
424109 ncm-dbt-01 583572 500 209 56 235 +109.83 ± 13.73 1 7 82 158 2 +245.2 ± 37.88

Commit

Commit ID 8c2d21f91a5840a67c36267e5043070ffad06860
Author 87
Date 2025-07-24 08:08:35 UTC
Speedup movegen with VBMI2 Passed STC LLR: 2.96 (-2.94,2.94) <0.00,2.00> Total: 166720 W: 43191 L: 42701 D: 80828 Ptnml(0-2): 348, 18567, 45069, 18999, 377 https://tests.stockfishchess.org/tests/view/686ae98dfe0f2fe354c0c867 Refactor movegen to emit to a vector with 16-bit elements, which enables a speedup with AVX512-VBMI2 when writing moves to the move list. Very crude timing measurements of perft via timing ./stockfish "go perft 7" demonstrates approximately 17% perft speedup: Summary ./Stockfish-dev/src/stockfish 'go perft 7' ran 1.17 ± 0.04 times faster than ./Stockfish-base/src/stockfish 'go perft 7' Estimated overall nps increase of 0.4% via speedtest: 33605229 -> 33749825 (many thanks JonathanHallstrom). The corresponding arch is avx512icl as it is a good baseline for consumer avx-512 feature set support; Intel Ice Lake was the first consumer AVX-512 CPU and it is a decent subset of what AMD Zen 4 supports. closes https://github.com/official-stockfish/Stockfish/pull/6153 No functional change
Copyright 2011–2025 Next Chess Move LLC