Dev Builds » 20200811-0617

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 06:33:26 585156 3992 52 2586 1354 -260.36 ± 6.63 611 1312 73 0 0 -691.94 ± 40.53
ncm-dbt-02 06:32:06 585470 4000 78 2613 1309 -259.76 ± 6.98 639 1258 102 1 0 -629.43 ± 34.11
ncm-dbt-03 06:32:22 584543 4000 58 2611 1331 -262.39 ± 6.81 632 1291 75 2 0 -678.31 ± 39.95
ncm-dbt-04 06:32:31 567743 4020 50 2621 1349 -263.15 ± 6.8 639 1294 76 1 0 -681.45 ± 39.69
ncm-dbt-05 06:33:04 581493 3988 67 2633 1288 -265.44 ± 7.08 663 1240 91 0 0 -652.67 ± 36.16
20000 305 13064 6631 -262.21 ± 3.07 3184 6395 417 4 0 -665.33 ± 16.72

Test Detail

ID Host Base NPS Games WLD Standard Elo Ptnml(0-2) Gamepair Elo CLI PGN
461531 ncm-dbt-04 568395 20 1 9 10 -147.12 ± 52.02 0 8 2 0 0 -381.52 ± 492.07
461530 ncm-dbt-05 581652 488 9 329 150 -272.83 ± 20.88 86 148 10 0 0 -671.74 ± 124.52
461529 ncm-dbt-01 584453 492 7 318 167 -258.81 ± 19.03 75 161 10 0 0 -673.18 ± 124.53
461528 ncm-dbt-02 584537 500 12 338 150 -270.57 ± 21.1 89 148 13 0 0 -629.41 ± 104.8
461527 ncm-dbt-03 584958 500 9 321 170 -254.16 ± 19.31 75 163 11 1 0 -629.43 ± 115.29
461526 ncm-dbt-04 565312 500 8 320 172 -254.16 ± 19.82 77 159 13 1 0 -603.86 ± 103.96
461525 ncm-dbt-05 584453 500 8 321 171 -255.29 ± 18.66 74 165 11 0 0 -659.13 ± 116.77
461524 ncm-dbt-02 585717 500 7 316 177 -250.76 ± 18.08 70 169 11 0 0 -659.13 ± 116.77
461523 ncm-dbt-01 584579 500 6 326 168 -263.41 ± 18.55 77 166 7 0 0 -739.05 ± 163.27
461522 ncm-dbt-03 584579 500 6 329 165 -266.96 ± 19.28 81 161 8 0 0 -715.51 ± 146.53
461521 ncm-dbt-04 569390 500 3 336 161 -279.16 ± 19.54 87 159 4 0 0 -837.28 ± 240.94
461520 ncm-dbt-05 582611 500 7 336 157 -274.22 ± 20.15 87 155 8 0 0 -715.51 ± 146.53
461519 ncm-dbt-02 584790 500 16 338 146 -265.78 ± 22.05 90 143 16 1 0 -571.11 ± 91.87
461518 ncm-dbt-01 584579 500 5 318 177 -255.29 ± 17.25 69 175 6 0 0 -766.17 ± 188.02
461517 ncm-dbt-03 585801 500 5 332 163 -271.78 ± 19.57 84 159 7 0 0 -739.05 ± 163.27
461516 ncm-dbt-04 570029 500 8 328 164 -263.41 ± 19.12 79 162 9 0 0 -694.7 ± 134.18
461515 ncm-dbt-02 587240 500 9 333 158 -268.16 ± 21.07 88 148 14 0 0 -616.18 ± 100.06
461514 ncm-dbt-01 583656 500 8 334 158 -270.57 ± 20.56 87 152 11 0 0 -659.13 ± 116.77
461513 ncm-dbt-05 581943 500 9 321 170 -254.15 ± 19.57 77 158 15 0 0 -603.84 ± 95.91
461512 ncm-dbt-03 583321 500 10 319 171 -250.76 ± 18.88 73 163 14 0 0 -616.18 ± 100.06
461511 ncm-dbt-04 569869 500 5 329 166 -268.16 ± 19.71 83 158 9 0 0 -694.7 ± 134.18
461510 ncm-dbt-02 585632 500 9 328 163 -262.24 ± 20.33 83 153 14 0 0 -616.18 ± 100.06
461509 ncm-dbt-01 586266 500 6 318 176 -254.15 ± 18.24 72 168 10 0 0 -676.04 ± 124.55
461508 ncm-dbt-05 583908 500 6 327 167 -264.59 ± 19.27 80 161 9 0 0 -694.7 ± 134.18
461507 ncm-dbt-03 583531 500 5 314 181 -250.76 ± 17.25 67 175 8 0 0 -715.51 ± 146.53
461506 ncm-dbt-04 567363 500 5 322 173 -259.9 ± 19.24 78 161 11 0 0 -659.13 ± 116.77
461505 ncm-dbt-02 584201 500 9 313 178 -245.19 ± 18.41 69 166 15 0 0 -603.84 ± 95.91
461504 ncm-dbt-05 580572 500 12 322 166 -251.89 ± 19.02 74 162 14 0 0 -616.18 ± 100.06
461503 ncm-dbt-01 586985 500 6 330 164 -268.16 ± 19.99 84 156 10 0 0 -676.04 ± 124.55
461502 ncm-dbt-03 583698 500 6 353 141 -297.27 ± 22.89 105 137 8 0 0 -715.51 ± 146.53
461501 ncm-dbt-04 565233 500 6 323 171 -259.9 ± 18.69 76 165 9 0 0 -694.7 ± 134.18
461500 ncm-dbt-02 585928 500 9 324 167 -257.59 ± 18.95 76 163 11 0 0 -659.13 ± 116.77
461499 ncm-dbt-05 578465 500 9 330 161 -264.59 ± 20.09 83 155 12 0 0 -643.67 ± 110.29
461498 ncm-dbt-03 585421 500 10 317 173 -248.52 ± 18.85 71 166 12 1 0 -616.2 ± 109.19
461497 ncm-dbt-01 587367 500 8 318 174 -251.89 ± 17.95 70 170 10 0 0 -676.04 ± 124.55
461496 ncm-dbt-04 565744 500 3 327 170 -268.16 ± 19.13 81 162 7 0 0 -739.05 ± 163.27
461495 ncm-dbt-02 585717 500 7 323 170 -258.74 ± 18.26 74 168 8 0 0 -715.51 ± 146.53
461494 ncm-dbt-03 585042 500 7 326 167 -262.24 ± 18.41 76 167 7 0 0 -739.05 ± 163.27
461493 ncm-dbt-05 578342 500 7 347 146 -288.06 ± 22.97 102 136 12 0 0 -643.67 ± 110.29
461492 ncm-dbt-04 568355 500 11 327 162 -258.74 ± 19.36 78 160 12 0 0 -643.67 ± 110.29
461491 ncm-dbt-01 583363 500 6 324 170 -261.07 ± 18.83 77 164 9 0 0 -694.7 ± 134.18

Commit

Commit ID ea6220f3813e5b76b444a02905eaf2c556bdb368
Author Guy Vreuls
Date 2020-08-11 06:17:03 UTC
This commit enables a mixed bench, to improve CI and allow for PGO (profile-build) of the NNUE part of the code. Joint work gvreuls / vondele * Download the default NNUE net in AppVeyor * Download net in travis CI `make net` * Adjust tests to cover more archs, speedup instrumented testing * Introduce 'mixed' bench as default, with further options: classical, NNUE, mixed. mixed (default) and NNUE require the default net to be present, which can be obtained with ``` make net ``` Further examples (first is equivalent to `./stockfish bench`): ``` ./stockfish bench 16 1 13 default depth mixed ./stockfish bench 16 1 13 default depth classical ./stockfish bench 16 1 13 default depth NNUE ``` The net is now downloaded automatically if needed for `profile-build` (usual `build` works fine without net present) PGO gives a nice speedup on fishtest: passed STC: LLR: 2.93 (-2.94,2.94) {-0.50,1.50} Total: 3360 W: 469 L: 343 D: 2548 Ptnml(0-2): 20, 246, 1030, 356, 28 https://tests.stockfishchess.org/tests/view/5f31b5499081672066537569 passed LTC: LLR: 2.97 (-2.94,2.94) {0.25,1.75} Total: 8824 W: 609 L: 502 D: 7713 Ptnml(0-2): 8, 430, 3438, 519, 17 https://tests.stockfishchess.org/tests/view/5f31c87b908167206653757c closes https://github.com/official-stockfish/Stockfish/pull/2931 fixes https://github.com/official-stockfish/Stockfish/issues/2907 requires fishtest updates before commit Bench: 4290577
Copyright 2011–2026 Next Chess Move LLC