Dev Builds » 20230824-0611

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 06:54:14 583096 4000 1367 692 1941 +59.2 ± 5.11 3 183 960 844 10 +120.67 ± 10.96
ncm-dbt-02 06:55:08 586136 4016 1371 701 1944 +58.51 ± 5.22 5 198 939 854 12 +119.18 ± 11.1
ncm-dbt-03 06:52:44 583290 4000 1367 678 1955 +60.45 ± 5.03 3 163 989 832 13 +122.83 ± 10.76
ncm-dbt-04 06:54:51 568037 3984 1300 713 1971 +51.57 ± 5.11 1 207 998 776 10 +103.79 ± 10.73
ncm-dbt-05 06:54:48 582527 4000 1371 707 1922 +58.21 ± 5.09 2 188 961 842 7 +118.92 ± 10.95
20000 6776 3491 9733 +57.59 ± 2.29 14 939 4847 4148 52 +117.05 ± 4.87

Test Detail

ID Host Base NPS Games WLD Standard Elo Ptnml(0-2) Gamepair Elo CLI PGN
423305 ncm-dbt-02 585717 16 6 2 8 +88.72 ± 93.48 0 1 2 5 0 +190.85 ± 457.95 ↓
423304 ncm-dbt-04 566691 484 160 94 230 +47.67 ± 14.91 0 30 116 96 0 +97.22 ± 31.7 ↓
423303 ncm-dbt-05 578218 500 173 87 240 +60.36 ± 14.35 0 23 119 107 1 +123.02 ± 31.23 ↓
423302 ncm-dbt-01 581735 500 154 89 257 +45.42 ± 14.81 0 32 122 95 1 +90.97 ± 30.89 ↓
423301 ncm-dbt-03 582277 500 169 81 250 +61.79 ± 13.42 1 14 131 104 0 +129.35 ± 29.24 ↓
423300 ncm-dbt-02 585126 500 176 80 244 +67.54 ± 14.23 0 19 118 111 2 +137.37 ± 31.31 ↓
423299 ncm-dbt-04 569390 500 173 93 234 +56.07 ± 14.7 0 25 123 99 3 +110.6 ± 30.69 ↓
423298 ncm-dbt-05 582068 500 176 94 230 +57.5 ± 14.36 0 25 118 107 0 +118.33 ± 31.4 ↓
423297 ncm-dbt-01 581818 500 176 106 218 +48.96 ± 15.0 1 29 120 99 1 +99.95 ± 31.16 ↓
423296 ncm-dbt-03 582903 500 176 81 243 +66.83 ± 14.2 1 18 116 115 0 +140.62 ± 31.61 ↓
423295 ncm-dbt-02 586139 500 180 86 234 +66.1 ± 14.31 0 20 118 110 2 +134.15 ± 31.33 ↓
423294 ncm-dbt-04 569230 500 171 78 251 +65.38 ± 13.72 0 17 124 108 1 +134.15 ± 30.35 ↓
423293 ncm-dbt-05 580199 500 177 73 250 +73.34 ± 13.86 1 12 121 114 2 +152.18 ± 30.64 ↓
423292 ncm-dbt-01 584874 500 179 100 221 +55.36 ± 14.26 1 21 127 100 1 +113.68 ± 30.06 ↓
423291 ncm-dbt-03 584622 500 174 79 247 +66.82 ± 13.92 0 16 126 105 3 +134.15 ± 30.02 ↓
423290 ncm-dbt-02 585042 500 157 92 251 +45.42 ± 15.32 1 32 120 95 2 +90.97 ± 31.17 ↓
423289 ncm-dbt-04 564409 500 156 87 257 +48.25 ± 14.57 0 28 127 93 2 +95.44 ± 30.17 ↓
423288 ncm-dbt-01 580945 500 186 84 230 +71.88 ± 13.81 0 15 120 113 2 +147.19 ± 30.88 ↓
423287 ncm-dbt-05 578794 500 163 96 241 +46.84 ± 14.22 0 27 130 92 1 +93.95 ± 29.74 ↓
423286 ncm-dbt-03 580157 500 167 82 251 +59.64 ± 14.32 0 22 123 103 2 +119.89 ± 30.63 ↓
423285 ncm-dbt-02 585801 500 155 81 264 +51.8 ± 15.27 0 33 111 105 1 +104.49 ± 32.42 ↓
423284 ncm-dbt-04 568236 500 166 90 244 +53.22 ± 14.56 0 28 118 104 0 +109.07 ± 31.43 ↓
423283 ncm-dbt-01 583698 500 168 84 248 +58.93 ± 15.6 0 31 107 109 3 +116.78 ± 33.01 ↓
423282 ncm-dbt-05 577930 500 171 79 250 +64.66 ± 14.53 0 22 116 110 2 +130.94 ± 31.66 ↓
423281 ncm-dbt-03 586266 500 166 91 243 +52.51 ± 14.66 0 27 123 98 2 +104.49 ± 30.71 ↓
423280 ncm-dbt-02 586181 500 175 89 236 +60.36 ± 15.28 0 29 108 111 2 +121.46 ± 32.88 ↓
423279 ncm-dbt-04 569909 500 152 92 256 +41.89 ± 14.34 0 30 131 88 1 +83.57 ± 29.65 ↓
423278 ncm-dbt-01 582736 500 164 71 265 +65.38 ± 14.42 1 19 117 112 1 +135.76 ± 31.48 ↓
423277 ncm-dbt-05 582778 500 162 85 253 +53.94 ± 14.73 1 26 118 105 0 +112.14 ± 31.42 ↓
423276 ncm-dbt-03 581693 500 173 91 236 +57.5 ± 14.36 0 24 121 104 1 +116.78 ± 30.96 ↓
423275 ncm-dbt-02 585295 500 175 81 244 +66.1 ± 14.31 1 17 121 109 2 +135.76 ± 30.83 ↓
423274 ncm-dbt-04 568156 500 162 99 239 +44.01 ± 14.47 1 27 131 90 1 +89.48 ± 29.62 ↓
423273 ncm-dbt-01 582318 500 169 81 250 +61.79 ± 14.41 0 23 117 109 1 +126.18 ± 31.53 ↓
423272 ncm-dbt-05 579496 500 175 107 218 +47.55 ± 14.26 0 28 126 96 0 +96.94 ± 30.31 ↓
423271 ncm-dbt-03 585253 500 177 88 235 +62.51 ± 14.58 0 21 123 102 4 +123.02 ± 30.61 ↓
423270 ncm-dbt-02 587962 500 176 87 237 +62.51 ± 14.17 2 15 126 106 1 +130.94 ± 30.05 ↓
423269 ncm-dbt-03 583154 500 165 85 250 +56.07 ± 14.29 1 21 126 101 1 +115.23 ± 30.2 ↓
423268 ncm-dbt-04 568275 500 160 80 260 +56.07 ± 14.15 0 22 128 98 2 +112.14 ± 29.92 ↓
423267 ncm-dbt-05 600735 500 174 86 240 +61.79 ± 14.68 0 25 113 111 1 +126.18 ± 32.14 ↓
423266 ncm-dbt-01 586646 500 171 77 252 +66.1 ± 13.0 0 13 130 107 0 +137.37 ± 29.29 ↓
423265 ncm-dbt-02 587962 500 171 103 226 +47.55 ± 15.18 1 32 115 102 0 +98.44 ± 31.86 ↓

Commit

Commit ID 4c4cb185aaaa0b3175ca35ab6473f17e9ec64055
Author Stéphane Nicolet
Date 2023-08-24 06:11:17 UTC
Play turbulent when defending, simpler when attacking This patch decays a little the evaluation (up to a few percent) for positions which have a large complexity measure (material imbalance, positional compensations, etc). This may have nice consequences on the playing style, as it modifies the search differently for attack and defense, both effects being desirable: - to see the effect on positions when Stockfish is defending, let us suppose for instance that the side to move is Stockfish and the nnue evaluation on the principal variation is -100 : this patch will decay positions with an evaluation of -103 (say) to the same level, provided they have huge material imbalance or huge positional compensation. In other words, chaotic positions with an evaluation of -103 are now comparable in our search tree to stable positions with an evaluation of -100, and chaotic positions with an evaluation of -102 are now preferred to stable positions with an evaluation of -100. - the effect on positions when Stockfish is attacking is the opposite. Let us suppose for instance that the side to move is Stockfish and the nnue evaluation on the principal variation is +100 : this patch will decay the evaluation to +97 if the positions on the principal variation have huge material imbalance or huge positional compensation. In other words, stable positions with an evaluation of +97 are now comparable in our search tree to chaotic positions with an evaluation of +100, and stable positions with an evaluation of +98 are now preferred to chaotic positions with an evaluation of +100. So the effect of this small change of evaluation on the playing style is that Stockfish should now play a little bit more turbulent when defending, and choose slightly simpler lines when attacking. passed STC: LLR: 2.93 (-2.94,2.94) <0.00,2.00> Total: 268448 W: 68713 L: 68055 D: 131680 Ptnml(0-2): 856, 31514, 68943, 31938, 973 https://tests.stockfishchess.org/tests/view/64e252bb99700912526653ed passed LTC: LLR: 2.94 (-2.94,2.94) <0.50,2.50> Total: 141060 W: 36066 L: 35537 D: 69457 Ptnml(0-2): 71, 15179, 39522, 15666, 92 https://tests.stockfishchess.org/tests/view/64e4447a9009777747553725 closes https://github.com/official-stockfish/Stockfish/pull/4762 Bench: 1426295
Copyright 2011–2025 Next Chess Move LLC
Feedback & Support Terms of Service Privacy Policy Desktop Version