Dev Builds » 20210425-1116

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 08:21:48 1962118 3352 2707 18 627 +383.84 ± 13.6
ncm-et-4 08:22:02 1962601 3332 2698 10 624 +388.28 ± 13.61
ncm-et-9 08:21:43 1963894 3361 2712 14 635 +384.35 ± 13.51
ncm-et-10 08:21:39 1952570 3271 2623 14 634 +379.41 ± 13.51
ncm-et-13 08:21:57 1960674 3348 2694 14 640 +382.16 ± 13.45
ncm-et-15 08:21:47 1958199 3336 2644 18 674 +369.65 ± 13.11
20000 16078 88 3834 +381.21 ± 5.49

Test Detail

ID Host Started (UTC) Duration Base NPS Games Wins Losses Draws Elo CLI PGN
146246 ncm-et-10 2021-04-26 13:44 00:41:41 1953739 271 217 1 53 +378.87 ± 47.49
146245 ncm-et-4 2021-04-26 13:36 00:50:07 1972971 332 270 1 61 +391.81 ± 44.18
146244 ncm-et-15 2021-04-26 13:35 00:50:59 1947665 336 257 3 76 +342.82 ± 39.45
146243 ncm-et-3 2021-04-26 13:33 00:52:46 1970498 352 276 4 72 +356.84 ± 40.63
146242 ncm-et-13 2021-04-26 13:33 00:53:02 1958317 348 285 2 61 +394.85 ± 44.27
146241 ncm-et-9 2021-04-26 13:32 00:53:06 1962154 361 298 0 63 +407.81 ± 43.38
146240 ncm-et-10 2021-04-26 12:26 01:17:18 1952681 500 399 2 99 +375.98 ± 34.46
146239 ncm-et-15 2021-04-26 12:20 01:14:11 1964004 500 410 2 88 +397.72 ± 36.65
146238 ncm-et-4 2021-04-26 12:19 01:15:54 1959116 500 390 0 110 +363.2 ± 32.48
146237 ncm-et-13 2021-04-26 12:18 01:14:00 1977641 500 423 1 76 +429.05 ± 39.52
146236 ncm-et-3 2021-04-26 12:18 01:14:25 1962611 500 408 3 89 +391.57 ± 36.46
146235 ncm-et-9 2021-04-26 12:17 01:15:14 1962462 500 404 1 95 +387.57 ± 35.16
146234 ncm-et-10 2021-04-26 11:08 01:16:49 1954957 500 401 2 97 +379.78 ± 34.83
146233 ncm-et-15 2021-04-26 11:04 01:14:48 1961542 500 391 3 106 +359.68 ± 33.28
146232 ncm-et-4 2021-04-26 11:04 01:14:31 1968470 500 405 3 92 +385.59 ± 35.84
146231 ncm-et-3 2021-04-26 11:02 01:14:45 1961587 500 406 1 93 +391.57 ± 35.55
146230 ncm-et-13 2021-04-26 11:02 01:15:28 1957563 500 396 2 102 +370.41 ± 33.92
146229 ncm-et-9 2021-04-26 11:01 01:14:25 1965282 500 406 5 89 +383.64 ± 36.48
146228 ncm-et-10 2021-04-26 09:52 01:16:20 1943871 500 387 5 108 +349.43 ± 33.01
146227 ncm-et-15 2021-04-26 09:49 01:14:27 1962248 500 398 4 98 +370.41 ± 34.7
146226 ncm-et-13 2021-04-26 09:48 01:13:36 1964310 500 396 2 102 +370.41 ± 33.92
146225 ncm-et-3 2021-04-26 09:47 01:14:20 1964314 500 402 4 94 +377.87 ± 35.46
146224 ncm-et-4 2021-04-26 09:47 01:16:09 1953588 500 411 1 88 +401.92 ± 36.6
146223 ncm-et-9 2021-04-26 09:47 01:13:47 1969903 500 398 4 98 +370.41 ± 34.7
146222 ncm-et-10 2021-04-26 08:34 01:16:28 1954965 500 413 1 86 +406.2 ± 37.04
146221 ncm-et-15 2021-04-26 08:34 01:14:58 1960770 500 396 2 102 +370.41 ± 33.92
146220 ncm-et-13 2021-04-26 08:33 01:14:27 1960165 500 398 2 100 +374.11 ± 34.27
146219 ncm-et-9 2021-04-26 08:32 01:14:53 1955151 500 397 0 103 +375.98 ± 33.63
146218 ncm-et-3 2021-04-26 08:31 01:15:28 1958721 500 404 1 95 +387.57 ± 35.16
146217 ncm-et-4 2021-04-26 08:31 01:15:54 1956369 500 397 2 101 +372.25 ± 34.1
146216 ncm-et-10 2021-04-26 07:18 01:15:45 1952673 500 404 0 96 +389.56 ± 34.9
146215 ncm-et-15 2021-04-26 07:17 01:15:56 1957724 500 399 0 101 +379.77 ± 33.98
146214 ncm-et-3 2021-04-26 07:16 01:14:49 1960443 500 407 3 90 +389.56 ± 36.25
146213 ncm-et-9 2021-04-26 07:16 01:15:20 1969086 500 408 2 90 +393.6 ± 36.22
146212 ncm-et-13 2021-04-26 07:15 01:16:39 1940268 500 394 2 104 +366.78 ± 33.58
146211 ncm-et-4 2021-04-26 07:15 01:14:58 1960619 500 408 1 91 +395.65 ± 35.96
146210 ncm-et-10 2021-04-26 06:00 01:17:18 1955110 500 402 3 95 +379.78 ± 35.24
146209 ncm-et-4 2021-04-26 06:00 01:14:29 1967076 500 417 2 81 +412.8 ± 38.27
146208 ncm-et-9 2021-04-26 06:00 01:14:58 1963225 500 401 2 97 +379.78 ± 34.83
146207 ncm-et-3 2021-04-26 06:00 01:15:15 1956657 500 404 2 94 +385.59 ± 35.4
146206 ncm-et-15 2021-04-26 06:00 01:16:28 1953440 500 393 4 103 +361.43 ± 33.81
146205 ncm-et-13 2021-04-26 06:00 01:14:45 1966459 500 402 3 95 +379.78 ± 35.24

Commit

Commit ID b748b46714d5f8e0acca0a042ede1fc95e4f5190
Author Tomasz Sobczyk
Date 2021-04-25 11:16:30 UTC
Cleanup and simplify NNUE code. A lot of optimizations happend since the NNUE was introduced and since then some parts of the code were left unused. This got to the point where asserts were have to be made just to let people know that modifying something will not have any effects or may even break everything due to the assumptions being made. Removing these parts removes those inexisting "false dependencies". Additionally: * append_changed_indices now takes the king pos and stateinfo explicitly, no more misleading pos parameter * IndexList is removed in favor of a generic ValueList. Feature transformer just instantiates the type it needs. * The update cost and refresh requirement is deferred to the feature set once again, but now doesn't go through the whole FeatureSet machinery and just calls HalfKP directly. * accumulator no longer has a singular dimension. * The PS constants and the PieceSquareIndex array are made local to the HalfKP feature set because they are specific to it and DO differ for other feature sets. * A few names are changed to more descriptive Passed STC non-regression: https://tests.stockfishchess.org/tests/view/608421dd95e7f1852abd2790 LLR: 2.95 (-2.94,2.94) <-2.50,0.50> Total: 180008 W: 16186 L: 16258 D: 147564 Ptnml(0-2): 587, 12593, 63725, 12503, 596 closes https://github.com/official-stockfish/Stockfish/pull/3441 No functional change
Copyright 2011–2024 Next Chess Move LLC