Dev Builds » 20200816-1910

Use this dev build

NCM plays each Stockfish dev build 20,000 times against Stockfish 15. This yields an approximate Elo difference and establishes confidence in the strength of the dev builds.

Summary

Host Duration Avg Base NPS Games WLD Standard Elo Ptnml(0-2) Gamepair Elo
ncm-dbt-01 06:33:55 583609 4000 59 2653 1288 -268.46 ± 7.16 681 1233 85 1 0 -661.2 ± 37.47
ncm-dbt-02 06:33:42 586785 3996 77 2612 1307 -260.13 ± 6.9 632 1272 93 1 0 -645.38 ± 35.77
ncm-dbt-03 06:34:24 586584 4000 63 2568 1369 -255.44 ± 6.64 595 1317 86 2 0 -655.17 ± 37.23
ncm-dbt-04 06:33:44 568006 4024 76 2684 1264 -268.23 ± 7.32 701 1206 105 0 0 -628.79 ± 33.6
ncm-dbt-05 06:34:54 582385 3980 56 2583 1341 -260.45 ± 6.86 626 1276 87 1 0 -656.27 ± 37.02
20000 331 13100 6569 -262.51 ± 3.12 3235 6304 456 5 0 -648.96 ± 15.99

Test Detail

ID Host Base NPS Games WLD Standard Elo Ptnml(0-2) Gamepair Elo CLI PGN
449232 ncm-dbt-04 570188 24 1 16 7 -254.6 ± 128.08 5 5 2 0 0 -416.38 ± 473.37
449231 ncm-dbt-05 582360 480 11 311 158 -254.72 ± 20.48 76 148 16 0 0 -584.94 ± 92.2
449230 ncm-dbt-02 586477 496 11 337 148 -273.76 ± 21.61 91 144 13 0 0 -627.97 ± 104.79
449229 ncm-dbt-03 587792 500 6 319 175 -255.3 ± 18.66 73 168 8 1 0 -676.08 ± 142.43
449228 ncm-dbt-01 585970 500 13 341 146 -272.99 ± 22.2 94 140 16 0 0 -592.27 ± 92.24
449227 ncm-dbt-04 567047 500 5 339 156 -280.42 ± 20.9 92 150 8 0 0 -715.51 ± 146.53
449226 ncm-dbt-05 583363 500 7 322 171 -257.59 ± 18.95 76 163 11 0 0 -659.13 ± 116.77
449225 ncm-dbt-02 586308 500 5 319 176 -256.44 ± 17.69 71 172 7 0 0 -739.05 ± 163.27
449224 ncm-dbt-01 584579 500 3 342 155 -286.77 ± 21.05 95 149 6 0 0 -766.17 ± 188.02
449223 ncm-dbt-03 587028 500 7 317 176 -251.89 ± 18.5 72 166 12 0 0 -643.67 ± 110.29
449222 ncm-dbt-04 569749 500 9 326 165 -259.9 ± 20.3 82 153 15 0 0 -603.84 ± 95.91
449221 ncm-dbt-05 580655 500 8 318 174 -251.89 ± 18.76 73 164 13 0 0 -629.41 ± 104.8
449220 ncm-dbt-02 587155 500 15 334 151 -262.24 ± 22.33 90 140 19 1 0 -543.25 ± 83.19
449219 ncm-dbt-01 581860 500 10 333 157 -266.96 ± 20.39 85 153 12 0 0 -643.67 ± 110.29
449218 ncm-dbt-03 585421 500 10 314 176 -245.19 ± 18.41 69 166 15 0 0 -603.84 ± 95.91
449217 ncm-dbt-04 568792 500 9 334 157 -269.36 ± 20.41 86 153 11 0 0 -659.13 ± 116.77
449216 ncm-dbt-05 580075 500 5 325 170 -263.41 ± 19.4 80 160 10 0 0 -676.04 ± 124.55
449215 ncm-dbt-01 580240 500 7 328 165 -264.59 ± 20.36 84 153 13 0 0 -629.41 ± 104.8
449214 ncm-dbt-03 583363 500 9 318 173 -250.76 ± 19.13 73 164 12 1 0 -616.2 ± 109.19
449213 ncm-dbt-02 587664 500 8 333 159 -269.36 ± 19.85 84 157 9 0 0 -694.7 ± 134.18
449212 ncm-dbt-04 565666 500 11 344 145 -279.16 ± 22.43 97 139 14 0 0 -616.18 ± 100.06
449211 ncm-dbt-05 582861 500 4 327 169 -266.96 ± 18.99 80 163 7 0 0 -739.05 ± 163.27
449210 ncm-dbt-02 587113 500 9 316 175 -248.52 ± 17.52 67 173 10 0 0 -676.04 ± 124.55
449209 ncm-dbt-03 589411 500 4 333 163 -274.22 ± 19.57 85 159 6 0 0 -766.17 ± 188.02
449208 ncm-dbt-01 585000 500 6 336 158 -275.44 ± 20.59 89 152 9 0 0 -694.7 ± 134.18
449207 ncm-dbt-04 569151 500 9 331 160 -265.77 ± 20.51 85 152 13 0 0 -629.41 ± 104.8
449206 ncm-dbt-02 587367 500 6 325 169 -262.24 ± 18.41 76 167 7 0 0 -739.05 ± 163.27
449205 ncm-dbt-05 584664 500 5 328 167 -266.96 ± 18.69 79 165 6 0 0 -766.17 ± 188.02
449204 ncm-dbt-03 586477 500 11 313 176 -243.0 ± 17.34 64 174 12 0 0 -643.67 ± 110.29
449203 ncm-dbt-04 565272 500 8 328 164 -263.41 ± 19.95 82 156 12 0 0 -643.67 ± 110.29
449202 ncm-dbt-01 583237 500 4 323 173 -262.24 ± 18.7 77 165 8 0 0 -715.51 ± 146.53
449201 ncm-dbt-02 584411 500 10 321 169 -253.02 ± 19.42 76 159 15 0 0 -603.84 ± 95.91
449200 ncm-dbt-03 584285 500 7 325 168 -261.07 ± 18.83 77 164 9 0 0 -694.7 ± 134.18
449199 ncm-dbt-05 584076 500 11 340 149 -274.22 ± 22.61 95 140 14 1 0 -592.29 ± 99.42
449198 ncm-dbt-01 583112 500 7 326 167 -262.24 ± 19.26 79 161 10 0 0 -676.04 ± 124.55
449197 ncm-dbt-04 568156 500 8 344 148 -282.94 ± 21.49 95 146 9 0 0 -694.7 ± 134.18
449196 ncm-dbt-02 587792 500 13 327 160 -256.44 ± 19.34 77 160 13 0 0 -629.41 ± 104.8
449195 ncm-dbt-04 568037 500 16 322 162 -247.41 ± 20.18 77 152 21 0 0 -543.23 ± 78.65
449194 ncm-dbt-01 584874 500 9 324 167 -257.59 ± 19.75 78 160 11 1 0 -629.43 ± 115.29
449193 ncm-dbt-03 588899 500 9 329 162 -263.41 ± 19.95 82 156 12 0 0 -643.67 ± 110.29
449192 ncm-dbt-05 581028 500 5 312 183 -248.52 ± 17.52 67 173 10 0 0 -676.04 ± 124.55

Commit

Commit ID 72dc7a5c54554a8c7c4bf68aa7de2d4de05f3294
Author syzygy1
Date 2020-08-16 19:10:26 UTC
Assume network file is in little-endian byte order This patch fixes the byte order when reading 16- and 32-bit values from the network file on a big-endian machine. Bytes are ordered in read_le() using unsigned arithmetic, which doesn't need tricks to determine the endianness of the machine. Unfortunately the compiler doesn't seem to be able to optimise the ordering operation, but reading in the weights is not a time-critical operation and the extra time it takes should not be noticeable. Big endian systems are still untested with NNUE. fixes #3007 closes https://github.com/official-stockfish/Stockfish/pull/3009 No functional change.
Copyright 2011–2026 Next Chess Move LLC