Dev Builds » 20211108-1203

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 06:51:38 582860 4000 712 1287 2001 -50.29 ± 5.13 9 777 996 216 2 -101.46 ± 10.76
ncm-dbt-02 06:52:28 586363 4000 730 1276 1994 -47.72 ± 5.12 8 755 1016 217 4 -96.57 ± 10.64
ncm-dbt-03 06:53:12 584889 4000 689 1339 1972 -56.96 ± 5.11 6 840 952 202 0 -116.0 ± 11.02
ncm-dbt-04 06:51:27 567326 4000 691 1338 1971 -56.7 ± 5.21 15 827 949 208 1 -113.87 ± 11.04
ncm-dbt-05 06:53:33 579547 4000 716 1304 1980 -51.45 ± 5.14 8 791 983 217 1 -103.92 ± 10.83
20000 3538 6544 9918 -52.62 ± 2.3 46 3990 4896 1060 8 -106.32 ± 4.85

Test Detail

ID Host Base NPS Games WLD Standard Elo Ptnml(0-2) Gamepair Elo CLI PGN
435571 ncm-dbt-03 585421 500 88 177 235 -62.51 ± 15.5 2 115 103 30 0 -126.18 ± 33.63
435570 ncm-dbt-05 579537 500 97 157 246 -41.89 ± 15.62 3 92 117 38 0 -80.63 ± 31.58
435569 ncm-dbt-01 578918 500 89 153 258 -44.72 ± 13.55 1 85 141 23 0 -89.48 ± 28.14
435568 ncm-dbt-02 587028 500 90 161 249 -49.67 ± 13.83 1 92 134 23 0 -99.95 ± 29.1
435567 ncm-dbt-04 568554 500 86 158 256 -50.38 ± 14.28 1 96 127 26 0 -101.46 ± 30.14
435566 ncm-dbt-03 585295 500 82 162 256 -56.07 ± 14.56 0 107 116 27 0 -115.23 ± 31.71
435565 ncm-dbt-02 586308 500 84 165 251 -56.78 ± 13.77 1 99 130 20 0 -115.23 ± 29.58
435564 ncm-dbt-04 568275 500 94 171 235 -53.93 ± 14.32 1 100 124 25 0 -109.07 ± 30.54
435563 ncm-dbt-05 577520 500 80 155 265 -52.51 ± 14.11 0 100 125 25 0 -107.54 ± 30.4
435562 ncm-dbt-01 582152 500 86 149 265 -44.01 ± 13.79 0 89 135 26 0 -89.48 ± 29.04
435561 ncm-dbt-03 581652 500 86 164 250 -54.64 ± 14.22 1 100 125 24 0 -110.6 ± 30.39
435560 ncm-dbt-02 586774 500 88 156 256 -47.54 ± 14.66 1 96 123 30 0 -95.44 ± 30.74
435559 ncm-dbt-01 579951 500 101 161 238 -41.9 ± 14.74 0 93 125 31 1 -86.52 ± 30.48
435558 ncm-dbt-05 577520 500 94 160 246 -46.13 ± 14.18 0 94 128 28 0 -93.95 ± 30.03
435557 ncm-dbt-04 567403 500 81 155 264 -51.8 ± 14.75 2 98 122 28 0 -102.97 ± 30.86
435556 ncm-dbt-03 587494 500 93 161 246 -47.54 ± 14.53 1 95 125 29 0 -95.44 ± 30.46
435555 ncm-dbt-04 565312 500 91 184 225 -65.38 ± 14.14 1 111 118 20 0 -134.15 ± 31.33
435554 ncm-dbt-02 586139 500 86 154 260 -47.55 ± 14.66 3 88 135 22 2 -95.44 ± 28.99
435553 ncm-dbt-01 583489 500 80 165 255 -59.64 ± 15.25 2 110 109 29 0 -119.89 ± 32.73
435552 ncm-dbt-05 582861 500 86 160 254 -51.8 ± 14.88 1 102 117 30 0 -104.49 ± 31.58
435551 ncm-dbt-04 567324 500 83 171 246 -61.79 ± 14.27 2 105 122 21 0 -124.6 ± 30.76
435550 ncm-dbt-03 585759 500 81 163 256 -57.5 ± 13.66 0 102 128 20 0 -118.33 ± 29.86
435549 ncm-dbt-02 586646 500 105 153 242 -33.46 ± 15.73 0 93 113 43 1 -68.99 ± 32.07
435548 ncm-dbt-01 585885 500 92 168 240 -53.22 ± 14.15 2 95 130 23 0 -106.01 ± 29.66
435547 ncm-dbt-05 580323 500 82 170 248 -61.79 ± 13.85 1 105 125 19 0 -126.18 ± 30.27
435546 ncm-dbt-04 567759 500 87 169 244 -57.5 ± 15.03 0 112 108 30 0 -118.33 ± 32.88
435545 ncm-dbt-03 587962 500 92 180 228 -61.79 ± 13.85 0 108 122 20 0 -127.76 ± 30.73
435544 ncm-dbt-02 586646 500 99 159 242 -41.89 ± 13.94 0 88 134 28 0 -85.04 ± 29.21
435543 ncm-dbt-05 578794 500 92 163 245 -49.67 ± 14.1 1 94 130 25 0 -99.95 ± 29.7
435542 ncm-dbt-01 583028 500 88 167 245 -55.36 ± 14.66 0 107 115 28 0 -113.68 ± 31.86
435541 ncm-dbt-04 565233 500 91 170 239 -55.36 ± 15.94 7 95 119 28 1 -104.49 ± 31.29
435540 ncm-dbt-03 580323 500 72 180 248 -76.25 ± 13.96 2 119 114 15 0 -157.24 ± 31.83
435539 ncm-dbt-05 581402 500 99 173 228 -51.8 ± 14.49 1 99 123 27 0 -104.49 ± 30.71
435538 ncm-dbt-01 583028 500 83 164 253 -56.78 ± 15.25 3 103 117 26 1 -113.68 ± 31.56
435537 ncm-dbt-02 587749 500 88 171 241 -58.21 ± 14.11 0 105 124 20 1 -121.46 ± 30.47
435536 ncm-dbt-04 568752 500 78 160 262 -57.5 ± 15.16 1 110 109 30 0 -116.78 ± 32.73
435535 ncm-dbt-03 585211 500 95 152 253 -39.78 ± 15.0 0 94 119 37 0 -80.63 ± 31.31
435534 ncm-dbt-01 586435 500 93 160 247 -46.84 ± 14.62 1 95 124 30 0 -93.95 ± 30.6
435533 ncm-dbt-05 578424 500 86 166 248 -56.07 ± 14.83 1 105 118 25 1 -115.23 ± 31.41
435532 ncm-dbt-02 583614 500 90 157 253 -46.84 ± 14.88 2 94 123 31 0 -92.46 ± 30.75

Commit

Commit ID 7b278aab9f61620b9dba31896b38aeea1eb911e2
Author Joost VandeVondele
Date 2021-11-08 12:03:52 UTC
Reduce use of lazyEval In case the evaluation at root is large, discourage the use of lazyEval. This fixes https://github.com/official-stockfish/Stockfish/issues/3772 or at least improves it significantly. In this case, poor play with large odds can be observed, in extreme cases leading to a loss despite large advantage: r1bq1b1r/ppp3p1/3p1nkp/n3p3/2B1P2N/2NPB3/PPP2PPP/R3K2R b KQ - 5 9 With this patch the poor move is only considered up to depth 13, in master up to depth 28. The patch did not pass at LTC with Elo gainer bounds, but with slightly positive Elo nevertheless (95% LOS). STC: LLR: 2.94 (-2.94,2.94) <0.00,2.50> Total: 40368 W: 10318 L: 10041 D: 20009 Ptnml(0-2): 103, 4493, 10725, 4750, 113 https://tests.stockfishchess.org/tests/view/61800ad259e71df00dcc420d LTC: LLR: -2.94 (-2.94,2.94) <0.50,3.00> Total: 212288 W: 52997 L: 52692 D: 106599 Ptnml(0-2): 112, 22038, 61549, 22323, 122 https://tests.stockfishchess.org/tests/view/618050d959e71df00dcc426d closes https://github.com/official-stockfish/Stockfish/pull/3780 Bench: 7127040
Copyright 2011–2025 Next Chess Move LLC