Dev Builds » 20230425-0619

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 09:28:37 1956644 3369 2870 5 494 +436.93 ± 15.32
ncm-et-4 09:28:34 1949200 3304 2826 4 474 +441.65 ± 15.64
ncm-et-9 09:29:04 1954490 3320 2861 3 456 +450.48 ± 15.95
ncm-et-10 09:28:44 1944522 3312 2849 8 455 +446.43 ± 15.99
ncm-et-13 09:28:50 1961399 3339 2887 2 450 +454.81 ± 16.05
ncm-et-15 09:28:47 1958587 3356 2930 3 423 +466.28 ± 16.57
20000 17223 25 2752 +449.22 ± 6.48

Test Detail

ID Host Started (UTC) Duration Base NPS Games Wins Losses Draws Elo CLI PGN
185378 ncm-et-4 2023-04-26 22:40 00:52:38 1950556 304 256 2 46 +419.07 ± 51.33
185377 ncm-et-10 2023-04-26 22:39 00:53:53 1943742 312 282 3 27 +501.23 ± 67.7
185376 ncm-et-9 2023-04-26 22:37 00:55:25 1952691 320 267 0 53 +417.74 ± 47.47
185375 ncm-et-13 2023-04-26 22:35 00:57:16 1968634 339 294 1 44 +455.18 ± 52.58
185374 ncm-et-15 2023-04-26 22:33 00:59:21 1966014 356 305 1 50 +441.42 ± 49.14
185373 ncm-et-3 2023-04-26 22:30 01:02:01 1960621 369 318 0 51 +451.75 ± 48.54
185366 ncm-et-4 2023-04-26 21:13 01:25:49 1945551 500 437 0 63 +468.95 ± 43.54
185365 ncm-et-10 2023-04-26 21:11 01:26:32 1954514 500 435 1 64 +460.32 ± 43.24
185364 ncm-et-9 2023-04-26 21:11 01:25:41 1959586 500 430 1 69 +446.7 ± 41.57
185363 ncm-et-15 2023-04-26 21:09 01:24:05 1963545 500 440 0 60 +477.98 ± 44.67
185362 ncm-et-13 2023-04-26 21:09 01:26:08 1957712 500 438 0 62 +471.92 ± 43.9
185361 ncm-et-3 2023-04-26 21:05 01:24:14 1955333 500 432 0 68 +454.76 ± 41.82
185358 ncm-et-4 2023-04-26 19:47 01:25:33 1954181 500 430 0 70 +449.35 ± 41.19
185354 ncm-et-9 2023-04-26 19:46 01:24:38 1948442 500 430 0 70 +449.35 ± 41.19
185352 ncm-et-10 2023-04-26 19:45 01:26:16 1950698 500 436 0 64 +466.03 ± 43.18
185351 ncm-et-13 2023-04-26 19:42 01:25:28 1958248 500 421 0 79 +426.65 ± 38.66
185350 ncm-et-15 2023-04-26 19:42 01:25:34 1944413 500 442 1 57 +481.1 ± 45.96
185349 ncm-et-3 2023-04-26 19:39 01:25:44 1951494 500 414 0 86 +410.58 ± 36.97
185347 ncm-et-9 2023-04-26 18:20 01:24:59 1957436 500 431 0 69 +452.04 ± 41.5
185346 ncm-et-4 2023-04-26 18:20 01:26:43 1936855 500 429 0 71 +446.7 ± 40.89
185344 ncm-et-10 2023-04-26 18:18 01:26:05 1942233 500 424 0 76 +433.94 ± 39.45
185342 ncm-et-13 2023-04-26 18:17 01:24:27 1965435 500 440 0 60 +477.98 ± 44.67
185341 ncm-et-15 2023-04-26 18:16 01:25:41 1955574 500 434 1 65 +457.52 ± 42.89
185337 ncm-et-3 2023-04-26 18:15 01:23:40 1966772 500 422 2 76 +424.28 ± 39.56
185336 ncm-et-9 2023-04-26 16:54 01:25:21 1955726 500 432 0 68 +454.76 ± 41.82
185335 ncm-et-4 2023-04-26 16:53 01:26:22 1957275 500 435 0 65 +463.15 ± 42.83
185334 ncm-et-10 2023-04-26 16:52 01:25:19 1939378 500 434 2 64 +454.76 ± 43.27
185333 ncm-et-3 2023-04-26 16:52 01:22:15 1966489 500 429 2 69 +441.5 ± 41.61
185332 ncm-et-13 2023-04-26 16:51 01:26:07 1956033 500 429 1 70 +444.09 ± 41.26
185330 ncm-et-15 2023-04-26 16:49 01:26:05 1958183 500 435 0 65 +463.15 ± 42.83
185324 ncm-et-9 2023-04-26 15:27 01:26:05 1951975 500 440 0 60 +477.98 ± 44.67
185323 ncm-et-10 2023-04-26 15:27 01:24:30 1937638 500 417 1 82 +415.05 ± 37.98
185322 ncm-et-3 2023-04-26 15:26 01:25:22 1956048 500 429 0 71 +446.7 ± 40.89
185321 ncm-et-4 2023-04-26 15:25 01:27:04 1946685 500 426 1 73 +436.43 ± 40.36
185320 ncm-et-15 2023-04-26 15:24 01:24:41 1955763 500 442 0 58 +484.24 ± 45.47
185319 ncm-et-13 2023-04-26 15:24 01:25:57 1958032 500 426 0 74 +438.95 ± 40.01
185312 ncm-et-4 2023-04-26 14:00 01:24:25 1953299 500 413 1 86 +406.2 ± 37.04
185311 ncm-et-15 2023-04-26 14:00 01:23:20 1966618 500 432 0 68 +454.76 ± 41.82
185310 ncm-et-13 2023-04-26 14:00 01:23:27 1965702 500 439 0 61 +474.93 ± 44.28
185309 ncm-et-10 2023-04-26 14:00 01:26:09 1943451 500 421 1 78 +424.28 ± 38.99
185308 ncm-et-3 2023-04-26 14:00 01:25:21 1939756 500 426 1 73 +436.43 ± 40.36
185307 ncm-et-9 2023-04-26 14:00 01:26:55 1955575 500 431 2 67 +446.7 ± 42.25

Commit

Commit ID 41f50b2c83a0ba36a2b9c507c1783e57c9b13485
Author Linmiao Xu
Date 2023-04-25 06:19:00 UTC
Update default net to nn-e1fb1ade4432.nnue Created by retraining nn-dabb1ed23026.nnue with a dataset composed of: * The previous best dataset (nn-1ceb1a57d117.nnue dataset) * Adding de-duplicated T80 data from feb2023 and the last 10 days of jan2023, filtered with v6-dd Initially trained with the same options as the recent master net (nn-1ceb1a57d117.nnue). Around epoch 890, training was manually stopped and max epoch increased to 1000. ``` python3 easy_train.py \ --experiment-name leela96-dfrc99-T60novdec-v2-T80augsep-v6-T80junjuloctnovjanfebT79aprmayT78jantosepT77dec-v6dd \ --training-dataset /data/leela96-dfrc99-T60novdec-v2-T80augsep-v6-T80junjuloctnovjanfebT79aprmayT78jantosepT77dec-v6dd.binpack \ --nnue-pytorch-branch linrock/nnue-pytorch/misc-fixes \ --start-from-engine-test-net True \ --early-fen-skipping 30 \ --start-lambda 1.0 \ --end-lambda 0.7 \ --max_epoch 900 \ --lr 4.375e-4 \ --gamma 0.995 \ --tui False \ --gpus "0," \ --seed $RANDOM ``` The same v6-dd filtering and binpack minimizer was used for preparing the recent nn-1ceb1a57d117.nnue dataset. ``` python3 interleave_binpacks.py \ leela96-filt-v2.binpack \ dfrc99-filt-v2.binpack \ T60-nov2021-12tb7p-eval-filt-v2.binpack \ T60-dec2021-12tb7p-eval-filt-v2.binpack \ filt-v6/test80-aug2022-16tb7p-filter-v6.min-mar2023.binpack \ filt-v6/test80-sep2022-16tb7p-filter-v6.min-mar2023.binpack \ filt-v6-dd/test80-jun2022-16tb7p-filter-v6-dd.min-mar2023.binpack \ filt-v6-dd/test80-jul2022-16tb7p-filter-v6-dd.binpack \ filt-v6-dd/test80-oct2022-16tb7p-filter-v6-dd.binpack \ filt-v6-dd/test80-nov2022-16tb7p-filter-v6-dd.binpack \ filt-v6-dd/test80-jan2022-3of3-16tb7p-filter-v6-dd.min-mar2023.binpack \ filt-v6-dd/test80-feb2023-16tb7p-filter-v6-dd.min-mar2023.binpack \ filt-v6-dd/test79-apr2022-16tb7p-filter-v6-dd.binpack \ filt-v6-dd/test79-may2022-16tb7p-filter-v6-dd.binpack \ filt-v6-dd/test78-jantomay2022-16tb7p-filter-v6-dd.min-mar2023.binpack \ filt-v6-dd/test78-juntosep2022-16tb7p-filter-v6-dd.binpack \ filt-v6-dd/test77-dec2021-16tb7p-filter-v6-dd.binpack \ /data/leela96-dfrc99-T60novdec-v2-T80augsep-v6-T80junjuloctnovjanfebT79aprmayT78jantosepT77dec-v6dd.binpack ``` Links for downloading the training data components can be found at: https://robotmoon.com/nnue-training-data/ Local elo at 25k nodes per move: nn-epoch919.nnue : 2.6 +/- 2.8 Passed STC vs. nn-dabb1ed23026.nnue https://tests.stockfishchess.org/tests/view/644420df94ff3db5625f2af5 LLR: 2.94 (-2.94,2.94) <0.00,2.00> Total: 125960 W: 33898 L: 33464 D: 58598 Ptnml(0-2): 351, 13920, 34021, 14320, 368 Passed LTC vs. nn-1ceb1a57d117.nnue https://tests.stockfishchess.org/tests/view/64469f128d30316529b3dc46 LLR: 2.95 (-2.94,2.94) <0.50,2.50> Total: 24544 W: 6817 L: 6542 D: 11185 Ptnml(0-2): 8, 2252, 7488, 2505, 19 closes https://github.com/official-stockfish/Stockfish/pull/4546 bench 3714847
Copyright 2011–2024 Next Chess Move LLC