Dev Builds » 20200610-1107

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 02:19:38 580974 1450 27 940 483 -257.39 ± 11.26 223 467 35 0 0 -642.67 ± 59.62
ncm-dbt-02 02:19:45 587693 1426 25 938 463 -263.56 ± 11.85 235 443 35 0 0 -639.69 ± 59.62
ncm-dbt-03 02:19:36 585322 1444 24 917 503 -251.0 ± 10.71 203 488 30 1 0 -657.87 ± 64.78
ncm-dbt-04 02:19:06 566193 1432 22 935 475 -261.99 ± 11.31 224 466 25 1 0 -686.53 ± 71.59
ncm-dbt-05 02:18:31 583897 1412 14 921 477 -264.8 ± 11.26 223 461 22 0 0 -720.22 ± 77.08
7164 112 4651 2401 -259.67 ± 5.04 1108 2325 147 2 0 -666.77 ± 28.31

Test Detail

ID Host Base NPS Games WLD Standard Elo Ptnml(0-2) Gamepair Elo CLI PGN
450676 ncm-dbt-05 581777 412 3 267 142 -263.87 ± 20.21 63 138 5 0 0 -764.18 ± 230.48
450675 ncm-dbt-02 586901 426 6 274 146 -257.07 ± 21.02 66 136 11 0 0 -630.63 ± 116.56
450674 ncm-dbt-04 567522 432 4 291 137 -278.14 ± 21.49 76 135 5 0 0 -772.51 ± 230.75
450673 ncm-dbt-01 580862 450 11 293 146 -255.67 ± 20.99 71 140 14 0 0 -597.32 ± 99.95
450672 ncm-dbt-03 581902 444 8 276 160 -242.78 ± 18.17 56 156 10 0 0 -654.96 ± 124.39
450671 ncm-dbt-05 585042 500 3 321 176 -261.07 ± 17.97 74 170 6 0 0 -766.17 ± 188.02
450670 ncm-dbt-02 587537 500 6 337 157 -276.68 ± 20.45 89 153 8 0 0 -715.51 ± 146.53
450669 ncm-dbt-04 564683 500 11 322 167 -253.02 ± 19.17 75 161 14 0 0 -616.18 ± 100.06
450668 ncm-dbt-03 588558 500 3 326 171 -266.97 ± 19.27 80 164 5 1 0 -739.1 ± 208.89
450667 ncm-dbt-01 579909 500 7 326 167 -262.24 ± 18.7 77 165 8 0 0 -715.51 ± 146.53
450666 ncm-dbt-05 584874 500 8 333 159 -269.36 ± 20.41 86 153 11 0 0 -659.13 ± 116.77
450665 ncm-dbt-04 566375 500 7 322 171 -257.59 ± 18.4 73 170 6 1 0 -715.55 ± 177.2
450664 ncm-dbt-02 588643 500 13 327 160 -256.44 ± 20.12 80 154 16 0 0 -592.27 ± 92.24
450663 ncm-dbt-01 582152 500 9 321 170 -254.15 ± 19.05 75 162 13 0 0 -629.41 ± 104.8
450662 ncm-dbt-03 585506 500 13 315 172 -243.0 ± 18.13 67 168 15 0 0 -603.84 ± 95.91

Commit

Commit ID 3af083a7cd9be1659f1d8a39a65e33b87608f762
Author Stéphane Nicolet
Date 2020-06-10 11:07:12 UTC
Improve the anti-shuffling policy We replace the current decrease of the complexity term in initiative when shuffling by a direct damping of the evaluation. This scheme may have two benefits over the initiative approach: a) the damping effect is more brutal for fortresses with heavy pieces on the board, because the initiative term is almost an endgame term; b) the initiative implementation had a funny side effect, almost a bug, in the rare positions where mg > 0, eg < 0 and the tampered eval returned a positive value (ie with heavy pieces still on the board): sending eg to zero via shuffling would **increase** the tampered eval instead of decreasing it, which is somewhat illogical. This patch avoids this phenomenon. STC: LLR: 2.94 (-2.94,2.94) {-0.50,1.50} Total: 43072 W: 8373 L: 8121 D: 26578 Ptnml(0-2): 729, 4954, 9940, 5162, 751 https://tests.stockfishchess.org/tests/view/5ee008ebf29b40b0fc95ade2 LTC: LLR: 2.94 (-2.94,2.94) {0.25,1.75} Total: 37376 W: 4816 L: 4543 D: 28017 Ptnml(0-2): 259, 3329, 11286, 3508, 306 https://tests.stockfishchess.org/tests/view/5ee03b06f29b40b0fc95ae0c Closes https://github.com/official-stockfish/Stockfish/pull/2727 Bench: 4757174
Copyright 2011–2026 Next Chess Move LLC