Dev Builds » 20230614-1636

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 09:24:35 1949128 3341 2869 1 471 +447.26 ± 15.68
ncm-et-4 09:27:16 1955032 3348 2880 5 463 +447.66 ± 15.84
ncm-et-9 09:28:06 1953355 3330 2857 8 465 +443.51 ± 15.81
ncm-et-10 09:21:48 1954566 3282 2844 6 432 +455.75 ± 16.41
ncm-et-13 09:28:43 1954004 3367 2851 3 513 +431.31 ± 15.02
ncm-et-15 09:26:28 1951923 3332 2873 3 456 +451.16 ± 15.95
20000 17174 26 2800 +445.91 ± 6.43

Test Detail

ID Host Started (UTC) Duration Base NPS Games Wins Losses Draws Elo CLI PGN
192455 ncm-et-10 2023-06-15 12:25 00:48:18 1955615 282 244 0 38 +456.47 ± 56.67
192454 ncm-et-9 2023-06-15 12:18 00:55:23 1959245 330 274 1 55 +409.78 ± 46.68
192453 ncm-et-15 2023-06-15 12:18 00:55:52 1959391 332 286 1 45 +447.27 ± 51.94
192452 ncm-et-3 2023-06-15 12:17 00:56:03 1962466 341 292 0 49 +444.48 ± 49.54
192451 ncm-et-4 2023-06-15 12:14 00:59:23 1947907 348 304 0 44 +468.31 ± 52.49
192450 ncm-et-13 2023-06-15 12:12 01:01:38 1949332 367 317 0 50 +454.43 ± 49.05
192443 ncm-et-10 2023-06-15 10:58 01:26:35 1952101 500 446 1 53 +494.03 ± 47.75
192442 ncm-et-3 2023-06-15 10:54 01:23:11 1960778 500 428 0 72 +444.08 ± 40.59
192441 ncm-et-9 2023-06-15 10:53 01:24:59 1953154 500 417 1 82 +415.05 ± 37.98
192440 ncm-et-15 2023-06-15 10:52 01:25:09 1954268 500 436 1 63 +463.16 ± 43.6
192439 ncm-et-4 2023-06-15 10:48 01:25:23 1963080 500 424 1 75 +431.48 ± 39.8
192438 ncm-et-13 2023-06-15 10:46 01:25:03 1956060 500 422 0 78 +429.05 ± 38.92
192431 ncm-et-10 2023-06-15 09:31 01:25:55 1954350 500 424 1 75 +431.48 ± 39.8
192430 ncm-et-3 2023-06-15 09:28 01:25:00 1922742 500 432 0 68 +454.76 ± 41.82
192429 ncm-et-9 2023-06-15 09:26 01:26:15 1944663 500 437 1 62 +466.04 ± 43.97
192428 ncm-et-15 2023-06-15 09:25 01:26:07 1935787 500 429 0 71 +446.7 ± 40.89
192427 ncm-et-4 2023-06-15 09:24 01:23:30 1965095 500 425 0 75 +436.43 ± 39.73
192426 ncm-et-13 2023-06-15 09:20 01:26:06 1944639 500 425 1 74 +433.94 ± 40.08
192419 ncm-et-10 2023-06-15 08:05 01:25:24 1948763 500 439 2 59 +468.96 ± 45.15
192418 ncm-et-3 2023-06-15 08:02 01:25:32 1952237 500 445 0 55 +494.02 ± 46.77
192417 ncm-et-9 2023-06-15 08:00 01:25:25 1954671 500 441 1 58 +477.99 ± 45.54
192416 ncm-et-4 2023-06-15 07:59 01:24:07 1961855 500 436 1 63 +463.16 ± 43.6
192415 ncm-et-15 2023-06-15 07:58 01:26:37 1944349 500 432 0 68 +454.76 ± 41.82
192414 ncm-et-13 2023-06-15 07:56 01:23:13 1965077 500 421 1 78 +424.28 ± 38.99
192407 ncm-et-10 2023-06-15 06:39 01:25:18 1957708 500 425 0 75 +436.43 ± 39.73
192406 ncm-et-3 2023-06-15 06:36 01:24:50 1958166 500 429 1 70 +444.09 ± 41.26
192405 ncm-et-9 2023-06-15 06:34 01:25:16 1952113 500 435 1 64 +460.32 ± 43.24
192404 ncm-et-4 2023-06-15 06:34 01:25:00 1953453 500 427 2 71 +436.43 ± 40.99
192403 ncm-et-15 2023-06-15 06:32 01:24:52 1954982 500 437 1 62 +466.04 ± 43.97
192402 ncm-et-13 2023-06-15 06:30 01:24:39 1958239 500 439 1 60 +471.92 ± 44.73
192395 ncm-et-10 2023-06-15 05:15 01:23:59 1957402 500 436 1 63 +463.16 ± 43.6
192394 ncm-et-3 2023-06-15 05:11 01:24:17 1933912 500 423 0 77 +431.48 ± 39.18
192393 ncm-et-4 2023-06-15 05:08 01:24:42 1937952 500 428 1 71 +441.5 ± 40.95
192392 ncm-et-15 2023-06-15 05:08 01:23:27 1955605 500 427 0 73 +441.5 ± 40.29
192391 ncm-et-9 2023-06-15 05:08 01:24:57 1954980 500 419 1 80 +419.61 ± 38.47
192390 ncm-et-13 2023-06-15 05:06 01:23:17 1959725 500 419 0 81 +421.93 ± 38.15
192383 ncm-et-10 2023-06-15 03:48 01:26:19 1956029 500 430 1 69 +446.7 ± 41.57
192382 ncm-et-3 2023-06-15 03:45 01:25:42 1953599 500 420 0 80 +424.27 ± 38.4
192381 ncm-et-15 2023-06-15 03:43 01:24:24 1959084 500 426 0 74 +438.95 ± 40.01
192380 ncm-et-4 2023-06-15 03:42 01:25:11 1955887 500 436 0 64 +466.03 ± 43.18
192379 ncm-et-9 2023-06-15 03:42 01:25:51 1954659 500 434 2 64 +454.76 ± 43.27
192378 ncm-et-13 2023-06-15 03:41 01:24:47 1944962 500 408 0 92 +397.72 ± 35.69

Commit

Commit ID 32d3284df5b2fd395504efa5319d64856902fef1
Author AndrovT
Date 2023-06-14 16:36:39 UTC
Permute master net weights to increase sparsity Activation data collection using https://github.com/AndrovT/Stockfish/commit/ac468039ab544b03ad9a22c859a4217729c10a77 run as bench 16 1 13 varied_1000.epd depth NNUE log.bin on FENs from https://gist.github.com/AndrovT/7eae6918eb50764227e2bafe7938953c. Permutation found using https://gist.github.com/AndrovT/359c831b7223c637e9156b01eb96949e. Uses a greedy algorithm that goes sequentially through the output positions and chooses a neuron for that position such that the number of nonzero quartets is the smallest. Net weights permuted using https://gist.github.com/AndrovT/9e3fbaebb7082734dc84d27e02094cb3. Benchmark: Result of 100 runs of 'bench 16 1 13 default depth NNUE' ======================================================== base (...kfish-master) = 885869 +/- 7395 test (./stockfish ) = 895885 +/- 7368 diff = +10016 +/- 2984 speedup = +0.0113 P(speedup > 0) = 1.0000 Passed STC: https://tests.stockfishchess.org/tests/view/648866c4713491385c804728 LLR: 2.94 (-2.94,2.94) <0.00,2.00> Total: 126784 W: 34003 L: 33586 D: 59195 Ptnml(0-2): 283, 13001, 36437, 13358, 313 closes https://github.com/official-stockfish/Stockfish/pull/4620 No functional change.
Copyright 2011–2024 Next Chess Move LLC