Dev Builds » 20210618-2150

You are viewing an old NCM Stockfish dev build test. You may find the most recent dev build tests using Stockfish 15 as the baseline here.

Use this dev build

NCM plays each Stockfish dev build 20,000 times against Stockfish 7. This yields an approximate Elo difference and establishes confidence in the strength of the dev builds.

Summary

Host Duration Avg Base NPS Games Wins Losses Draws Elo
ncm-et-3 08:20:25 1970411 3384 2789 12 583 +402.59 ± 14.1
ncm-et-4 08:20:25 1961869 3352 2775 10 567 +407.16 ± 14.3
ncm-et-9 08:20:28 1954587 3319 2758 12 549 +409.87 ± 14.54
ncm-et-10 08:20:20 1951205 3303 2764 9 530 +417.42 ± 14.79
ncm-et-13 08:20:20 1958498 3325 2780 15 530 +414.57 ± 14.81
ncm-et-15 08:20:24 1956455 3317 2722 7 588 +400.35 ± 14.03
20000 16588 65 3347 +408.54 ± 5.88

Test Detail

ID Host Started (UTC) Duration Base NPS Games Wins Losses Draws Elo CLI PGN
149155 ncm-et-10 2021-06-20 20:36 00:46:29 1956025 303 261 1 41 +446.82 ± 54.56
149154 ncm-et-13 2021-06-20 20:34 00:48:11 1961236 325 266 1 58 +397.08 ± 45.37
149153 ncm-et-9 2021-06-20 20:34 00:48:12 1962928 319 264 2 53 +403.32 ± 47.64
149152 ncm-et-15 2021-06-20 20:34 00:48:40 1946174 317 258 1 58 +392.3 ± 45.36
149151 ncm-et-4 2021-06-20 20:28 00:54:03 1950796 352 303 0 49 +450.41 ± 49.56
149150 ncm-et-3 2021-06-20 20:25 00:57:35 1973891 384 313 0 71 +396.78 ± 40.75
149149 ncm-et-10 2021-06-20 19:20 01:14:56 1959389 500 425 0 75 +436.43 ± 39.73
149148 ncm-et-13 2021-06-20 19:19 01:15:16 1957095 500 427 1 72 +438.95 ± 40.66
149147 ncm-et-15 2021-06-20 19:18 01:15:07 1959405 500 402 2 96 +381.7 ± 35.02
149146 ncm-et-9 2021-06-20 19:18 01:16:02 1928226 500 424 2 74 +429.05 ± 40.11
149145 ncm-et-4 2021-06-20 19:14 01:13:26 1970328 500 417 2 81 +412.8 ± 38.27
149144 ncm-et-3 2021-06-20 19:11 01:13:07 1969126 500 400 3 97 +375.98 ± 34.86
149143 ncm-et-13 2021-06-20 18:03 01:14:33 1967850 500 419 2 79 +417.32 ± 38.77
149142 ncm-et-10 2021-06-20 18:03 01:16:48 1947410 500 420 1 79 +421.93 ± 38.73
149141 ncm-et-15 2021-06-20 18:02 01:15:43 1952990 500 429 0 71 +446.7 ± 40.89
149140 ncm-et-9 2021-06-20 18:01 01:15:37 1958174 500 416 3 81 +408.38 ± 38.29
149139 ncm-et-4 2021-06-20 18:00 01:13:58 1965723 500 420 2 78 +419.61 ± 39.03
149138 ncm-et-3 2021-06-20 17:57 01:13:35 1970790 500 419 2 79 +417.32 ± 38.77
149137 ncm-et-9 2021-06-20 16:47 01:13:27 1961865 500 424 1 75 +431.48 ± 39.8
149136 ncm-et-10 2021-06-20 16:47 01:15:01 1959405 500 413 3 84 +401.92 ± 37.57
149135 ncm-et-13 2021-06-20 16:45 01:17:22 1948603 500 418 5 77 +408.38 ± 39.29
149134 ncm-et-15 2021-06-20 16:45 01:15:52 1957883 500 403 1 96 +385.59 ± 34.96
149133 ncm-et-4 2021-06-20 16:45 01:14:12 1972497 500 423 2 75 +426.65 ± 39.83
149132 ncm-et-3 2021-06-20 16:42 01:13:41 1963110 500 422 1 77 +426.65 ± 39.25
149131 ncm-et-9 2021-06-20 15:30 01:16:07 1954209 500 422 3 75 +421.93 ± 39.85
149130 ncm-et-10 2021-06-20 15:30 01:16:13 1948575 500 421 0 79 +426.65 ± 38.66
149129 ncm-et-4 2021-06-20 15:30 01:14:34 1957248 500 409 1 90 +397.72 ± 36.17
149128 ncm-et-13 2021-06-20 15:29 01:15:15 1960007 500 412 1 87 +404.05 ± 36.82
149127 ncm-et-15 2021-06-20 15:29 01:15:06 1959556 500 414 1 85 +408.38 ± 37.27
149126 ncm-et-3 2021-06-20 15:27 01:14:24 1975760 500 415 2 83 +408.38 ± 37.78
149125 ncm-et-13 2021-06-20 14:14 01:14:34 1963841 500 418 1 81 +417.32 ± 38.22
149124 ncm-et-15 2021-06-20 14:14 01:14:23 1967862 500 412 0 88 +406.2 ± 36.53
149123 ncm-et-4 2021-06-20 14:14 01:15:14 1957102 500 400 2 98 +377.87 ± 34.64
149122 ncm-et-10 2021-06-20 14:13 01:16:13 1936930 500 410 0 90 +401.92 ± 36.1
149121 ncm-et-9 2021-06-20 14:13 01:16:30 1951481 500 388 1 111 +357.94 ± 32.39
149120 ncm-et-3 2021-06-20 14:13 01:14:07 1970482 500 412 2 86 +401.92 ± 37.09
149119 ncm-et-13 2021-06-20 12:59 01:15:09 1950856 500 420 4 76 +415.05 ± 39.57
149118 ncm-et-3 2021-06-20 12:58 01:13:56 1969721 500 408 2 90 +393.6 ± 36.22
149117 ncm-et-9 2021-06-20 12:58 01:14:33 1965226 500 420 0 80 +424.27 ± 38.4
149116 ncm-et-10 2021-06-20 12:58 01:14:40 1950705 500 414 4 82 +401.92 ± 38.05
149115 ncm-et-15 2021-06-20 12:58 01:15:33 1951321 500 404 2 94 +385.59 ± 35.4
149114 ncm-et-4 2021-06-20 12:58 01:14:58 1959394 500 403 1 96 +385.59 ± 34.96

Commit

Commit ID adfb23c029e54c7522aadca1adf3e0b15fdcebcd
Author Joost VandeVondele
Date 2021-06-18 21:50:26 UTC
Make net nn-50144f835024.nnue the default trained with the Python command c:\nnue>python train.py i:/bin/all.binpack i:/bin/all.binpack --gpus 1 --threads 4 --num-workers 30 --batch-size 16384 --progress_bar_refresh_rate 300 --smart-fen-skipping --random-fen-skipping 3 --features=HalfKAv2^ --lambda=1.0 --max_epochs=440 --seed %random%%random% --default_root_dir exp/run_8 --resume-from-model ./pt/nn-6ad41a9207d0.pt ` all.binpack equaled 4 parts Wrong_NNUE_2.binpack https://drive.google.com/file/d/1seGNOqcVdvK_vPNq98j-zV3XPE5zWAeq/view?usp=sharing plus two parts of Training_Data.binpack https://drive.google.com/file/d/1RFkQES3DpsiJqsOtUshENtzPfFgUmEff/view?usp=sharing Each set was concatenated together - make one large Wrong_NNUE 2 binpack and one large Training_Data of approximate size. They were then interleaved together. The idea was to give Wrong_NNUE.binpack closer to equal weighting with the Training _Data binpack . nn-6ad41a9207d0.pt was derived from a net vondele ran which passed STC quickly, but faltered in LTC. https://tests.stockfishchess.org/tests/view/60cba666457376eb8bcab443 STC: LLR: 2.95 (-2.94,2.94) <-0.50,2.50> Total: 18792 W: 2068 L: 1889 D: 14835 Ptnml(0-2): 82, 1480, 6117, 1611, 106 https://tests.stockfishchess.org/tests/view/60ccda8b457376eb8bcab568 LTC: LLR: 2.94 (-2.94,2.94) <0.50,3.50> Total: 11376 W: 574 L: 454 D: 10348 Ptnml(0-2): 4, 412, 4747, 510, 15 https://tests.stockfishchess.org/tests/view/60ccf952457376eb8bcab58d closes https://github.com/official-stockfish/Stockfish/pull/3568 Bench: 4900906
Copyright 2011–2024 Next Chess Move LLC