Dev Builds » 20210621-2116

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 08:22:37 1959951 3333 2812 7 514 +426.16 ± 15.02
ncm-et-4 08:22:23 1963157 3358 2813 12 533 +417.46 ± 14.76
ncm-et-9 08:22:34 1961429 3348 2815 10 523 +421.71 ± 14.9
ncm-et-10 08:22:35 1955260 3311 2748 13 550 +408.42 ± 14.53
ncm-et-13 08:20:35 1907983 3315 2769 11 535 +415.02 ± 14.73
ncm-et-15 08:22:48 1955972 3335 2844 4 487 +438.41 ± 15.43
20000 16801 57 3142 +421.0 ± 6.07

Test Detail

ID Host Started (UTC) Duration Base NPS Games Wins Losses Draws Elo CLI PGN
149288 ncm-et-13 2021-06-22 06:08 00:31:47 1520709 160 134 1 25 +414.2 ± 71.0
149287 ncm-et-10 2021-06-22 05:53 00:47:17 1952980 311 265 0 46 +439.06 ± 51.19
149286 ncm-et-15 2021-06-22 05:49 00:50:41 1959395 335 281 0 54 +422.87 ± 47.03
149285 ncm-et-3 2021-06-22 05:49 00:50:47 1957422 333 279 2 52 +414.86 ± 48.14
149284 ncm-et-9 2021-06-22 05:48 00:51:28 1973127 348 287 2 59 +400.83 ± 45.05
149283 ncm-et-4 2021-06-22 05:46 00:53:39 1963840 358 294 0 64 +403.22 ± 43.01
149282 ncm-et-13 2021-06-22 05:43 00:23:21 1961546 155 129 0 26 +415.32 ± 69.19
149281 ncm-et-10 2021-06-22 04:36 01:16:16 1955898 500 408 3 89 +391.57 ± 36.46
149280 ncm-et-15 2021-06-22 04:33 01:15:37 1959101 500 413 0 87 +408.38 ± 36.75
149279 ncm-et-3 2021-06-22 04:33 01:15:43 1956082 500 435 0 65 +463.15 ± 42.83
149278 ncm-et-9 2021-06-22 04:32 01:15:43 1960017 500 417 2 81 +412.8 ± 38.27
149277 ncm-et-4 2021-06-22 04:32 01:13:39 1965854 500 414 2 84 +406.2 ± 37.55
149276 ncm-et-13 2021-06-22 04:29 01:13:10 1967081 500 421 1 78 +424.28 ± 38.99
149275 ncm-et-10 2021-06-22 03:19 01:15:44 1953746 500 407 0 93 +395.65 ± 35.48
149274 ncm-et-4 2021-06-22 03:17 01:14:18 1970017 500 417 2 81 +412.8 ± 38.27
149273 ncm-et-9 2021-06-22 03:17 01:14:36 1964160 500 411 1 88 +401.92 ± 36.6
149272 ncm-et-15 2021-06-22 03:17 01:15:58 1952762 500 433 0 67 +457.52 ± 42.15
149271 ncm-et-3 2021-06-22 03:15 01:16:47 1958506 500 429 1 70 +444.09 ± 41.26
149270 ncm-et-13 2021-06-22 03:13 01:15:35 1953598 500 410 2 88 +397.72 ± 36.65
149269 ncm-et-10 2021-06-22 02:03 01:15:09 1959544 500 408 2 90 +393.6 ± 36.22
149268 ncm-et-9 2021-06-22 02:02 01:14:16 1959862 500 419 2 79 +417.32 ± 38.77
149267 ncm-et-15 2021-06-22 02:01 01:15:12 1953046 500 445 1 54 +490.71 ± 47.28
149266 ncm-et-4 2021-06-22 02:00 01:15:57 1958185 500 416 0 84 +415.04 ± 37.43
149265 ncm-et-3 2021-06-22 02:00 01:14:55 1962624 500 423 0 77 +431.48 ± 39.18
149264 ncm-et-13 2021-06-22 01:57 01:14:56 1962917 500 412 3 85 +399.81 ± 37.34
149263 ncm-et-10 2021-06-22 00:47 01:16:01 1955117 500 418 4 78 +410.58 ± 39.05
149262 ncm-et-15 2021-06-22 00:46 01:13:36 1960318 500 428 0 72 +444.08 ± 40.59
149261 ncm-et-9 2021-06-22 00:46 01:15:30 1953738 500 422 1 77 +426.65 ± 39.25
149260 ncm-et-4 2021-06-22 00:45 01:14:38 1960328 500 422 2 76 +424.28 ± 39.56
149259 ncm-et-3 2021-06-22 00:43 01:16:02 1955123 500 420 1 79 +421.93 ± 38.73
149258 ncm-et-13 2021-06-22 00:42 01:14:25 1966305 500 420 2 78 +419.61 ± 39.03
149257 ncm-et-15 2021-06-21 23:31 01:15:08 1951002 500 419 2 79 +417.32 ± 38.77
149256 ncm-et-10 2021-06-21 23:31 01:15:34 1954349 500 427 2 71 +436.43 ± 40.99
149255 ncm-et-9 2021-06-21 23:29 01:15:51 1962309 500 424 2 74 +429.05 ± 40.11
149254 ncm-et-4 2021-06-21 23:29 01:15:17 1963710 500 433 4 63 +446.7 ± 43.57
149253 ncm-et-3 2021-06-21 23:28 01:14:53 1964181 500 417 2 81 +412.8 ± 38.27
149252 ncm-et-13 2021-06-21 23:27 01:14:11 1964318 500 431 2 67 +446.7 ± 42.25
149251 ncm-et-4 2021-06-21 22:14 01:14:55 1960170 500 417 2 81 +412.8 ± 38.27
149250 ncm-et-15 2021-06-21 22:13 01:16:36 1956186 500 425 1 74 +433.94 ± 40.08
149249 ncm-et-9 2021-06-21 22:13 01:15:10 1956796 500 435 0 65 +463.15 ± 42.83
149248 ncm-et-13 2021-06-21 22:13 01:13:10 1967391 500 412 0 88 +406.2 ± 36.53
149247 ncm-et-10 2021-06-21 22:13 01:16:34 1955192 500 415 2 83 +408.38 ± 37.78
149246 ncm-et-3 2021-06-21 22:13 01:13:30 1965721 500 409 1 90 +397.72 ± 36.17

Commit

Commit ID 9b82414b67aa7d1279e1cc99a6970ab766025bfa
Author MichaelB7
Date 2021-06-21 21:16:55 UTC
Make net nn-190f102a22c3.nnue the default net. Trained with the pytorch trainer: https://github.com/glinscott/nnue-pytorch 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_17 --resume-from-model ./pt/nn-75980ca503c6.pt This run is thus started from the previous master net. 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 - making one large Wrong_NNUE 2 binpack and one large Training so the were approximately equal in size. They were then interleaved together. The idea was to give Wrong_NNUE.binpack closer to equal weighting with the Training_Data binpack passed LTC https://tests.stockfishchess.org/tests/view/60d09f52b4c17000d679517f LLR: 2.93 (-2.94,2.94) <0.50,3.50> Total: 32184 W: 1100 L: 970 D: 30114 Ptnml(0-2): 10, 878, 14193, 994, 17 passed STC https://tests.stockfishchess.org/tests/view/60d086c02114332881e7368e LLR: 2.93 (-2.94,2.94) <-0.50,2.50> Total: 11360 W: 1056 L: 906 D: 9398 Ptnml(0-2): 25, 735, 4026, 853, 41 closes https://github.com/official-stockfish/Stockfish/pull/3576 Bench: 4631244
Copyright 2011–2024 Next Chess Move LLC