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 14. 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 09:57:56 1210450 3344 1428 292 1624 +122.91 ± 5.39 0 41 505 1075 51 +268.72 ± 15.16
ncm-dbt-02 09:50:25 1239624 3322 1402 297 1623 +120.14 ± 5.43 0 56 481 1087 37 +265.18 ± 15.55
ncm-dbt-03 09:54:26 1236085 3336 1400 286 1650 +120.65 ± 5.36 0 51 488 1093 36 +267.14 ± 15.44
ncm-dbt-04 09:56:02 1235873 3356 1428 281 1647 +123.72 ± 5.31 0 41 494 1098 45 +273.45 ± 15.33
ncm-dbt-05 09:47:01 1225494 3306 1397 267 1642 +123.73 ± 5.38 0 44 478 1088 43 +273.97 ± 15.59
ncm-dbt-06 09:58:24 1230695 3336 1435 313 1588 +121.58 ± 5.37 0 43 505 1075 45 +266.78 ± 15.16
20000 8490 1736 9774 +122.12 ± 2.19 0 276 2951 6516 257 +269.19 ± 6.27

Test Detail

ID Host Base NPS Games WLD Standard Elo Ptnml(0-2) Gamepair Elo CLI PGN
192449 ncm-dbt-05 1231414 306 126 21 159 +124.25 ± 15.76 0 2 45 105 1 +287.89 ± 51.42
192448 ncm-dbt-02 1226707 322 139 23 160 +131.04 ± 16.21 0 4 39 116 2 +306.89 ± 55.67
192447 ncm-dbt-03 1224347 336 151 28 157 +133.37 ± 18.29 0 5 44 110 9 +287.14 ± 52.23
192446 ncm-dbt-06 1242319 336 141 27 168 +122.74 ± 16.32 0 4 49 112 3 +275.89 ± 49.37
192445 ncm-dbt-01 1218412 344 143 28 173 +120.79 ± 16.41 0 4 53 111 4 +266.58 ± 47.35
192444 ncm-dbt-04 1215023 356 158 34 164 +126.3 ± 15.96 0 3 53 117 5 +280.76 ± 47.31
192437 ncm-dbt-05 1222634 500 212 47 241 +119.11 ± 13.86 0 7 77 160 6 +261.07 ± 39.14
192436 ncm-dbt-02 1250605 500 213 49 238 +118.33 ± 14.52 0 10 73 160 7 +256.44 ± 40.21
192435 ncm-dbt-03 1232833 500 211 26 263 +134.95 ± 13.03 0 3 66 174 7 +309.64 ± 42.33
192434 ncm-dbt-06 1231966 500 217 48 235 +122.24 ± 14.17 0 7 75 160 8 +265.78 ± 39.69
192433 ncm-dbt-04 1233981 500 209 35 256 +126.17 ± 12.3 0 1 78 167 4 +288.06 ± 38.42
192432 ncm-dbt-01 1200503 500 209 43 248 +119.89 ± 14.02 0 5 83 153 9 +256.44 ± 37.51
192425 ncm-dbt-05 1212952 500 205 48 247 +112.91 ± 14.38 0 8 85 149 8 +238.66 ± 37.17
192424 ncm-dbt-02 1233122 500 205 42 253 +117.55 ± 14.69 0 11 72 160 7 +254.16 ± 40.45
192423 ncm-dbt-06 1209168 500 211 53 236 +113.68 ± 13.21 0 7 80 161 2 +254.16 ± 38.36
192422 ncm-dbt-03 1231777 500 204 57 239 +105.25 ± 14.66 0 16 73 159 2 +230.16 ± 39.91
192421 ncm-dbt-04 1225309 500 206 42 252 +118.33 ± 14.36 0 10 72 162 6 +258.75 ± 40.49
192420 ncm-dbt-01 1178156 500 225 41 234 +134.15 ± 12.67 0 3 65 177 5 +312.48 ± 42.67
192413 ncm-dbt-05 1230585 500 202 39 259 +117.55 ± 13.18 0 5 81 160 4 +261.07 ± 38.02
192412 ncm-dbt-02 1209752 500 201 44 255 +112.91 ± 13.39 0 7 82 158 3 +249.64 ± 37.86
192411 ncm-dbt-03 1239724 500 214 39 247 +126.97 ± 13.76 0 7 67 170 6 +285.49 ± 42.08
192410 ncm-dbt-06 1240684 500 227 44 229 +133.34 ± 14.34 0 6 66 167 11 +293.29 ± 42.41
192409 ncm-dbt-04 1258324 500 208 42 250 +119.89 ± 14.02 0 7 77 159 7 +261.07 ± 39.14
192408 ncm-dbt-01 1194633 500 208 39 253 +122.24 ± 13.66 0 6 75 163 6 +270.57 ± 39.67
192401 ncm-dbt-05 1226356 500 212 37 251 +126.97 ± 14.61 0 8 69 163 10 +275.45 ± 41.43
192400 ncm-dbt-02 1235715 500 215 46 239 +122.24 ± 13.12 0 7 69 172 2 +280.42 ± 41.44
192399 ncm-dbt-03 1232078 500 212 37 251 +126.97 ± 13.41 0 4 74 165 7 +282.94 ± 39.86
192398 ncm-dbt-06 1234231 500 209 47 244 +116.77 ± 14.21 0 4 91 144 11 +243.0 ± 35.57
192397 ncm-dbt-04 1217672 500 218 34 248 +134.15 ± 14.15 0 7 61 173 9 +301.33 ± 44.13
192396 ncm-dbt-01 1234850 500 214 52 234 +116.77 ± 14.37 0 11 71 163 5 +256.44 ± 40.73
192389 ncm-dbt-05 1219578 500 219 40 241 +130.14 ± 14.4 0 9 61 172 8 +290.66 ± 44.0
192388 ncm-dbt-02 1231028 500 218 42 240 +127.76 ± 13.92 0 8 64 172 6 +288.06 ± 43.03
192387 ncm-dbt-03 1248540 500 206 52 242 +110.6 ± 13.39 0 8 82 158 2 +245.2 ± 37.88
192386 ncm-dbt-04 1251039 500 219 53 228 +119.89 ± 14.68 0 8 78 154 10 +254.16 ± 38.89
192385 ncm-dbt-06 1226516 500 220 50 230 +123.02 ± 14.16 0 9 68 167 6 +273.0 ± 41.7
192384 ncm-dbt-01 1206546 500 215 40 245 +126.97 ± 13.94 0 5 74 162 9 +277.93 ± 39.92
192377 ncm-dbt-05 1234945 500 221 35 244 +135.76 ± 13.2 0 5 60 179 6 +315.35 ± 44.57
192376 ncm-dbt-02 1290445 500 211 51 238 +115.22 ± 14.85 0 9 82 149 10 +240.82 ± 37.9
192375 ncm-dbt-03 1243301 500 202 47 251 +111.37 ± 13.56 0 8 82 157 3 +245.2 ± 37.88
192374 ncm-dbt-04 1249765 500 210 41 249 +122.24 ± 13.12 0 5 75 166 4 +275.45 ± 39.63
192373 ncm-dbt-01 1240055 500 214 49 237 +119.11 ± 15.0 0 7 84 146 13 +245.2 ± 37.37
192372 ncm-dbt-06 1229984 500 210 44 246 +119.89 ± 13.33 0 6 76 164 4 +268.17 ± 39.39

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