Dev Builds » 20220419-1759

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 11:20:26 1961010 3360 2870 4 486 +440.19 ± 15.45
ncm-et-4 11:20:24 1923046 3255 2809 9 437 +449.64 ± 16.32
ncm-et-9 11:20:36 1964400 3360 2882 7 471 +443.64 ± 15.71
ncm-et-10 11:13:38 1959959 3314 2811 7 496 +431.62 ± 15.3
ncm-et-13 11:20:32 1960776 3375 2875 7 493 +436.15 ± 15.35
ncm-et-15 11:20:39 1957008 3336 2834 7 495 +433.23 ± 15.31
20000 17081 41 2878 +438.95 ± 6.34

Test Detail

ID Host Started (UTC) Duration Base NPS Games Wins Losses Draws Elo CLI PGN
158131 ncm-et-4 2022-04-20 21:01 00:50:55 1961849 255 214 2 39 +414.34 ± 55.97
158130 ncm-et-10 2022-04-20 20:48 01:04:06 1960625 314 280 0 34 +496.91 ± 60.27
158129 ncm-et-15 2022-04-20 20:43 01:09:07 1951468 336 297 0 39 +484.13 ± 55.99
158128 ncm-et-3 2022-04-20 20:40 01:11:48 1969863 360 307 0 53 +439.93 ± 47.55
158127 ncm-et-9 2022-04-20 20:39 01:12:55 1966486 360 298 2 60 +404.29 ± 44.67
158126 ncm-et-13 2022-04-20 20:37 01:15:36 1960310 375 321 0 54 +444.08 ± 47.1
158125 ncm-et-4 2022-04-20 19:18 01:42:32 1958326 500 448 0 52 +504.32 ± 48.17
158124 ncm-et-10 2022-04-20 19:05 01:39:25 1960616 500 422 3 75 +421.93 ± 39.85
158123 ncm-et-15 2022-04-20 19:00 01:42:30 1955876 500 417 0 83 +417.31 ± 37.67
158122 ncm-et-3 2022-04-20 19:00 01:39:54 1959543 500 430 2 68 +444.09 ± 41.92
158121 ncm-et-9 2022-04-20 18:57 01:42:11 1965999 500 428 1 71 +441.5 ± 40.95
158120 ncm-et-13 2022-04-20 18:56 01:39:52 1964003 500 423 1 76 +429.05 ± 39.52
158119 ncm-et-4 2022-04-20 17:33 01:44:42 1933960 500 434 2 64 +454.76 ± 43.27
158118 ncm-et-10 2022-04-20 17:21 01:43:04 1954959 500 418 0 82 +419.61 ± 37.91
158117 ncm-et-15 2022-04-20 17:18 01:41:43 1958034 500 430 0 70 +449.35 ± 41.19
158116 ncm-et-3 2022-04-20 17:17 01:42:18 1952251 500 414 1 85 +408.38 ± 37.27
158115 ncm-et-9 2022-04-20 17:16 01:40:21 1962920 500 421 0 79 +426.65 ± 38.66
158114 ncm-et-13 2022-04-20 17:15 01:40:54 1959861 500 415 2 83 +408.38 ± 37.78
158113 ncm-et-4 2022-04-20 15:49 01:43:04 1936477 500 426 1 73 +436.43 ± 40.36
158112 ncm-et-10 2022-04-20 15:39 01:41:04 1969421 500 433 1 66 +454.76 ± 42.55
158111 ncm-et-3 2022-04-20 15:36 01:40:20 1961846 500 431 0 69 +452.04 ± 41.5
158110 ncm-et-15 2022-04-20 15:34 01:43:04 1951304 500 421 0 79 +426.65 ± 38.66
158109 ncm-et-9 2022-04-20 15:34 01:41:13 1966772 500 441 2 57 +474.93 ± 45.97
158108 ncm-et-13 2022-04-20 15:32 01:42:14 1958013 500 426 1 73 +436.43 ± 40.36
158107 ncm-et-4 2022-04-20 13:54 01:54:45 1750465 500 413 1 86 +406.2 ± 37.04
158106 ncm-et-3 2022-04-20 13:53 01:42:04 1966153 500 423 0 77 +431.48 ± 39.18
158105 ncm-et-15 2022-04-20 13:52 01:40:53 1960003 500 426 3 71 +431.48 ± 41.0
158104 ncm-et-10 2022-04-20 13:52 01:42:55 1951921 500 419 3 78 +415.05 ± 39.05
158103 ncm-et-13 2022-04-20 13:50 01:40:41 1960768 500 421 1 78 +424.28 ± 38.99
158102 ncm-et-9 2022-04-20 13:50 01:43:15 1959695 500 435 0 65 +463.15 ± 42.83
158101 ncm-et-4 2022-04-20 12:12 01:41:13 1961695 500 439 0 61 +474.93 ± 44.28
158100 ncm-et-3 2022-04-20 12:11 01:41:33 1957875 500 426 1 73 +436.43 ± 40.36
158099 ncm-et-10 2022-04-20 12:10 01:41:57 1959395 500 416 0 84 +415.04 ± 37.43
158098 ncm-et-13 2022-04-20 12:09 01:40:27 1960308 500 436 1 63 +463.16 ± 43.6
158097 ncm-et-9 2022-04-20 12:09 01:40:08 1966150 500 434 0 66 +460.32 ± 42.48
158096 ncm-et-15 2022-04-20 12:08 01:43:25 1953128 500 416 0 84 +415.04 ± 37.43
158095 ncm-et-10 2022-04-20 10:28 01:41:07 1962778 500 423 0 77 +431.48 ± 39.18
158094 ncm-et-4 2022-04-20 10:28 01:43:13 1958552 500 435 3 62 +454.76 ± 43.98
158093 ncm-et-13 2022-04-20 10:28 01:40:48 1962170 500 433 1 66 +454.76 ± 42.55
158092 ncm-et-3 2022-04-20 10:28 01:42:29 1959543 500 439 0 61 +474.93 ± 44.28
158091 ncm-et-9 2022-04-20 10:28 01:40:33 1962781 500 425 2 73 +431.48 ± 40.4
158090 ncm-et-15 2022-04-20 10:28 01:39:57 1969246 500 427 4 69 +431.48 ± 41.59

Commit

Commit ID 6e0680efa0d0653686b8eb8753dc49718b95cb2b
Author Joost VandeVondele
Date 2022-04-19 17:59:04 UTC
Update default net to nn-d0b74ce1e5eb.nnue train a net using training data with a heavier weight on positions having 16 pieces on the board. More specifically, with a relative weight of `i * (32-i)/(16 * 16)+1` (where i is the number of pieces on the board). This is done with the trainer branch https://github.com/glinscott/nnue-pytorch/pull/173 The command used is: ``` python train.py $datafile $datafile $restarttype $restartfile --gpus 1 --threads 4 --num-workers 12 --random-fen-skipping=3 --batch-size 16384 --progress_bar_refresh_rate 300 --smart-fen-skipping --features=HalfKAv2_hm^ --lambda=1.00 --max_epochs=$epochs --seed $RANDOM --default_root_dir exp/run_$i ``` The datafile is T60T70wIsRightFarseerT60T74T75T76.binpack, the restart is from the master net. passed STC: LLR: 2.94 (-2.94,2.94) <0.00,2.50> Total: 22728 W: 6197 L: 5945 D: 10586 Ptnml(0-2): 105, 2453, 6001, 2695, 110 https://tests.stockfishchess.org/tests/view/625cf944ff677a888877cd90 passed LTC: LLR: 2.94 (-2.94,2.94) <0.50,3.00> Total: 35664 W: 9535 L: 9264 D: 16865 Ptnml(0-2): 30, 3524, 10455, 3791, 32 https://tests.stockfishchess.org/tests/view/625d3c32ff677a888877d7ca closes https://github.com/official-stockfish/Stockfish/pull/3989 Bench: 7269563
Copyright 2011–2024 Next Chess Move LLC