Dev Builds » 20230706-2103

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 09:56:42 1189056 3344 1432 291 1621 +123.5 ± 5.18 0 38 488 1113 33 +277.11 ± 15.42
ncm-dbt-02 09:51:25 1239908 3306 1401 284 1621 +122.19 ± 5.32 3 41 476 1102 31 +274.72 ± 15.63
ncm-dbt-03 09:56:02 1236588 3336 1416 276 1644 +123.7 ± 5.45 1 47 479 1093 48 +272.56 ± 15.58
ncm-dbt-04 09:57:35 1233502 3340 1386 299 1655 +117.34 ± 5.36 0 46 531 1053 40 +255.84 ± 14.78
ncm-dbt-05 09:50:14 1228411 3318 1420 263 1635 +126.46 ± 5.32 1 39 463 1114 42 +283.39 ± 15.85
ncm-dbt-06 09:56:05 1225719 3356 1447 307 1602 +122.9 ± 5.38 0 47 489 1097 45 +270.91 ± 15.42
20000 8502 1720 9778 +122.67 ± 2.18 5 258 2926 6572 239 +272.27 ± 6.29

Test Detail

ID Host Base NPS Games WLD Standard Elo Ptnml(0-2) Gamepair Elo CLI PGN
195816 ncm-dbt-02 1229299 306 134 29 143 +124.25 ± 16.93 0 2 48 99 4 +275.53 ± 49.6
195815 ncm-dbt-05 1231626 318 142 28 148 +130.34 ± 17.44 0 5 39 111 4 +295.82 ± 55.56
195814 ncm-dbt-03 1240228 336 148 37 151 +119.25 ± 18.17 0 6 52 103 7 +251.36 ± 47.86
195813 ncm-dbt-04 1237977 340 140 34 166 +112.04 ± 17.19 0 6 56 104 4 +240.82 ± 46.03
195812 ncm-dbt-01 1175755 344 151 29 164 +128.81 ± 15.28 0 2 49 118 3 +295.85 ± 49.22
195811 ncm-dbt-06 1225999 356 151 23 182 +130.76 ± 16.45 0 8 36 132 2 +306.75 ± 56.97
195810 ncm-dbt-02 1222869 500 213 38 249 +126.97 ± 12.85 0 4 71 171 4 +290.66 ± 40.76
195809 ncm-dbt-05 1234161 500 225 46 229 +130.14 ± 13.34 0 4 70 169 7 +293.29 ± 41.07
195808 ncm-dbt-03 1232450 500 212 37 251 +126.97 ± 14.45 0 9 65 168 8 +280.42 ± 42.65
195807 ncm-dbt-04 1240472 500 204 45 251 +114.45 ± 13.89 0 7 83 154 6 +247.41 ± 37.61
195806 ncm-dbt-01 1184412 500 208 43 249 +119.11 ± 12.99 0 5 78 164 3 +268.17 ± 38.8
195805 ncm-dbt-06 1213894 500 211 52 237 +114.45 ± 14.21 0 9 79 156 6 +247.41 ± 38.64
195804 ncm-dbt-02 1230505 500 210 43 247 +120.67 ± 14.51 1 8 70 165 6 +268.17 ± 41.1
195803 ncm-dbt-05 1213031 500 223 37 240 +135.76 ± 13.57 0 3 68 169 10 +304.07 ± 41.65
195802 ncm-dbt-04 1217612 500 218 48 234 +123.02 ± 14.16 0 10 65 170 5 +275.45 ± 42.59
195801 ncm-dbt-01 1199811 500 204 44 252 +115.22 ± 13.2 0 5 84 157 4 +254.16 ± 37.27
195800 ncm-dbt-03 1242554 500 205 46 249 +114.45 ± 14.37 1 7 80 156 6 +249.64 ± 38.38
195799 ncm-dbt-06 1233297 500 213 48 239 +119.11 ± 14.19 0 8 76 159 7 +258.75 ± 39.42
195798 ncm-dbt-02 1230998 500 225 47 228 +129.35 ± 13.36 1 4 65 176 4 +301.33 ± 42.75
195797 ncm-dbt-05 1211464 500 212 29 259 +133.34 ± 13.26 0 6 60 179 5 +309.64 ± 44.55
195796 ncm-dbt-01 1147540 500 218 35 247 +133.34 ± 13.63 0 7 59 178 6 +306.84 ± 44.88
195795 ncm-dbt-03 1236682 500 220 43 237 +128.55 ± 13.73 0 5 71 166 8 +285.49 ± 40.81
195794 ncm-dbt-06 1228972 500 223 40 237 +133.34 ± 15.18 0 6 71 157 16 +280.42 ± 40.83
195793 ncm-dbt-04 1246968 500 207 40 253 +120.67 ± 13.67 0 6 77 161 6 +265.78 ± 39.12
195792 ncm-dbt-05 1253422 500 213 33 254 +130.94 ± 13.86 1 5 63 175 6 +301.33 ± 43.45
195791 ncm-dbt-02 1253210 500 208 42 250 +119.89 ± 14.83 1 10 67 166 6 +265.78 ± 41.9
195790 ncm-dbt-01 1201960 500 217 52 231 +119.11 ± 13.52 0 8 72 167 3 +268.17 ± 40.54
195789 ncm-dbt-03 1214268 500 215 43 242 +124.6 ± 14.47 0 8 71 162 9 +270.57 ± 40.83
195788 ncm-dbt-06 1232080 500 225 56 219 +122.24 ± 13.48 0 7 71 168 4 +275.45 ± 40.84
195787 ncm-dbt-04 1235559 500 209 41 250 +121.45 ± 13.31 0 4 80 160 6 +268.17 ± 38.21
195786 ncm-dbt-02 1250045 500 204 51 245 +109.83 ± 13.05 0 7 84 158 1 +245.2 ± 37.37
195785 ncm-dbt-05 1229751 500 197 42 261 +111.37 ± 13.56 0 6 88 151 5 +240.82 ± 36.39
195784 ncm-dbt-06 1232688 500 219 53 228 +119.89 ± 13.33 0 4 82 158 6 +263.42 ± 37.69
195783 ncm-dbt-03 1237891 500 206 39 255 +120.67 ± 13.84 0 7 75 162 6 +265.78 ± 39.69
195782 ncm-dbt-01 1208852 500 218 44 238 +126.17 ± 14.46 0 8 69 164 9 +275.45 ± 41.43
195781 ncm-dbt-04 1215941 500 202 49 249 +109.83 ± 13.73 0 7 88 150 5 +236.51 ± 36.44
195780 ncm-dbt-05 1225428 500 208 48 244 +115.22 ± 14.21 0 10 75 160 5 +251.89 ± 39.67
195779 ncm-dbt-02 1262433 500 207 34 259 +125.38 ± 13.61 0 6 71 167 6 +280.42 ± 40.83
195778 ncm-dbt-03 1252043 500 210 31 259 +130.14 ± 12.97 0 5 65 176 4 +301.33 ± 42.75
195777 ncm-dbt-06 1213105 500 205 35 260 +123.02 ± 13.11 0 5 74 167 4 +277.93 ± 39.92
195776 ncm-dbt-01 1205065 500 216 44 240 +124.6 ± 12.9 0 3 77 165 5 +280.42 ± 38.93
195775 ncm-dbt-04 1239988 500 206 42 252 +118.33 ± 14.03 0 6 82 154 8 +254.16 ± 37.82

Commit

Commit ID e699fee513ce26b3794ac43d08826c89106e10ea
Author Linmiao Xu
Date 2023-07-06 21:03:58 UTC
Update default net to nn-c38c3d8d3920.nnue This was a later epoch from the same experiment that led to the previous master net. After training, it was prepared the same way: 1. greedy permuting L1 weights with https://github.com/official-stockfish/Stockfish/pull/4620 2. leb128 compression with https://github.com/glinscott/nnue-pytorch/pull/251 3. greedy 2- and 3- cycle permuting with https://github.com/official-stockfish/Stockfish/pull/4640 Local elo at 25k nodes per move (vs. L1-1536 nn-fdc1d0fe6455.nnue): nn-epoch739.nnue : 20.2 +/- 1.7 Passed STC: https://tests.stockfishchess.org/tests/view/64a050b33ee09aa549c4e4c8 LLR: 2.95 (-2.94,2.94) <0.00,2.00> Total: 195552 W: 49977 L: 49430 D: 96145 Ptnml(0-2): 556, 22775, 50607, 23242, 596 Passed LTC: https://tests.stockfishchess.org/tests/view/64a127bd3ee09aa549c4f60c LLR: 2.94 (-2.94,2.94) <0.50,2.50> Total: 235452 W: 60327 L: 59609 D: 115516 Ptnml(0-2): 119, 25173, 66426, 25887, 121 closes https://github.com/official-stockfish/Stockfish/pull/4666 bench 2427629
Copyright 2011–2024 Next Chess Move LLC