Dev Builds » 20240526-1832

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 11:35:59 1200823 4000 1800 307 1893 +136.26 +/- 4.72 0 34 500 1405 61 +312.48 +/- 15.25
ncm-dbt-02 11:33:50 1197387 3978 1757 298 1923 +133.65 +/- 4.82 1 46 488 1401 53 +306.42 +/- 15.44
ncm-dbt-03 11:37:42 1240168 4016 1779 280 1957 +136.26 +/- 4.66 1 33 494 1426 54 +315.43 +/- 15.34
ncm-dbt-05 11:33:03 1198807 3986 1781 297 1908 +135.88 +/- 4.73 1 33 499 1401 59 +312.13 +/- 15.26
ncm-dbt-06 11:36:58 1227132 4020 1779 296 1945 +134.51 +/- 4.74 1 35 514 1400 60 +307.16 +/- 15.04
20000 8896 1478 9626 +135.31 +/- 2.12 4 181 2495 7033 287 +310.7 +/- 6.82

Test Detail

ID Host Base NPS Games WLD Standard Elo Ptnml(0-2) Gamepair Elo CLI PGN
368822 ncm-dbt-03 1238327 16 6 2 8 +88.68 +/- 65.24 0 0 4 4 0 +190.76 +/- 189.19
368821 ncm-dbt-06 1246079 20 10 3 7 +126.91 +/- 57.06 0 0 3 7 0 +301.22 +/- 341.95
368820 ncm-dbt-02 1196694 478 218 42 218 +134.23 +/- 14.42 0 7 57 167 8 +303.33 +/- 45.67
368819 ncm-dbt-05 1197183 486 222 37 227 +139.26 +/- 13.93 0 4 60 169 10 +315.47 +/- 44.56
368818 ncm-dbt-01 1187556 500 225 29 246 +143.89 +/- 13.12 0 5 51 187 7 +342.85 +/- 48.46
368817 ncm-dbt-03 1246765 500 218 33 249 +134.95 +/- 13.78 1 4 61 177 7 +312.48 +/- 44.19
368816 ncm-dbt-06 1221617 500 214 31 255 +133.34 +/- 12.49 0 1 71 172 6 +306.84 +/- 40.47
368815 ncm-dbt-05 1197298 500 227 34 239 +141.44 +/- 13.21 0 3 60 178 9 +327.18 +/- 44.54
368814 ncm-dbt-02 1195858 500 212 28 260 +134.15 +/- 13.8 0 8 56 180 6 +309.64 +/- 45.98
368813 ncm-dbt-01 1180952 500 223 38 239 +134.95 +/- 14.13 0 6 63 171 10 +301.33 +/- 43.45
368812 ncm-dbt-03 1233325 500 212 32 256 +130.94 +/- 12.76 0 4 66 176 4 +304.07 +/- 42.38
368811 ncm-dbt-06 1234738 500 228 43 229 +134.95 +/- 13.03 0 2 69 171 8 +306.84 +/- 41.23
368810 ncm-dbt-05 1197294 500 226 35 239 +139.81 +/- 12.47 1 0 61 183 5 +336.46 +/- 43.96
368809 ncm-dbt-02 1194585 500 219 46 235 +125.38 +/- 14.13 0 6 74 161 9 +273.0 +/- 39.95
368808 ncm-dbt-01 1200613 500 216 51 233 +119.11 +/- 12.8 0 5 77 166 2 +270.57 +/- 39.07
368807 ncm-dbt-03 1251042 500 217 42 241 +126.97 +/- 12.85 0 2 77 165 6 +285.49 +/- 38.83
368806 ncm-dbt-06 1226625 500 219 38 243 +131.74 +/- 14.03 0 5 69 166 10 +290.66 +/- 41.43
368805 ncm-dbt-02 1195128 500 215 37 248 +129.35 +/- 12.61 0 6 61 182 1 +306.84 +/- 44.17
368804 ncm-dbt-05 1196644 500 223 43 234 +130.94 +/- 14.22 0 8 62 172 8 +293.29 +/- 43.72
368803 ncm-dbt-01 1235337 500 223 40 237 +133.34 +/- 13.45 0 3 70 168 9 +298.62 +/- 41.01
368802 ncm-dbt-03 1225720 500 232 36 232 +143.89 +/- 13.7 0 4 57 178 11 +330.23 +/- 45.79
368801 ncm-dbt-06 1225049 500 224 34 242 +138.99 +/- 12.9 0 4 58 182 6 +327.18 +/- 45.37
368800 ncm-dbt-02 1197411 500 218 37 245 +131.74 +/- 14.54 0 9 60 172 9 +293.29 +/- 44.36
368799 ncm-dbt-05 1208739 500 216 34 250 +132.54 +/- 13.1 0 4 66 174 6 +304.07 +/- 42.38
368798 ncm-dbt-01 1204299 500 225 36 239 +138.18 +/- 13.12 0 3 63 176 8 +318.25 +/- 43.39
368797 ncm-dbt-03 1240654 500 220 33 247 +136.56 +/- 11.97 0 3 59 186 2 +330.23 +/- 44.94
368796 ncm-dbt-06 1231300 500 216 39 245 +128.55 +/- 13.56 0 8 61 177 4 +295.94 +/- 44.08
368795 ncm-dbt-02 1198978 500 228 39 233 +138.18 +/- 12.33 0 3 59 184 4 +330.23 +/- 44.94
368794 ncm-dbt-05 1196275 500 225 35 240 +138.99 +/- 13.48 0 3 64 173 10 +315.35 +/- 43.03
368793 ncm-dbt-01 1199762 500 227 33 240 +142.25 +/- 12.98 0 2 61 178 9 +330.23 +/- 44.08
368792 ncm-dbt-03 1234302 500 235 35 230 +147.19 +/- 13.58 0 4 53 182 11 +342.85 +/- 47.56
368791 ncm-dbt-06 1224293 500 221 35 244 +135.76 +/- 14.98 1 7 58 173 11 +304.07 +/- 45.19
368790 ncm-dbt-02 1196290 500 220 31 249 +138.18 +/- 12.93 0 3 62 178 7 +321.19 +/- 43.77
368789 ncm-dbt-05 1198051 500 218 40 242 +129.35 +/- 13.54 0 7 63 175 5 +295.94 +/- 43.42
368788 ncm-dbt-01 1196294 500 226 48 226 +129.35 +/- 14.58 0 9 63 169 9 +285.49 +/- 43.31
368787 ncm-dbt-03 1253215 500 219 31 250 +137.37 +/- 13.15 0 6 55 184 5 +324.17 +/- 46.56
368786 ncm-dbt-06 1231033 500 231 42 227 +138.18 +/- 13.5 0 3 65 172 10 +312.48 +/- 42.67
368785 ncm-dbt-05 1198975 500 224 39 237 +134.95 +/- 13.03 0 4 63 177 6 +312.48 +/- 43.44
368784 ncm-dbt-02 1204155 500 227 38 235 +138.18 +/- 14.05 1 4 59 177 9 +318.25 +/- 44.96
368783 ncm-dbt-03 1238168 500 220 36 244 +134.15 +/- 13.8 0 6 62 174 8 +304.07 +/- 43.81
368782 ncm-dbt-06 1203457 500 216 31 253 +134.95 +/- 13.03 0 5 60 180 5 +315.35 +/- 44.57
368781 ncm-dbt-01 1201771 500 235 32 233 +149.68 +/- 12.01 0 1 52 190 7 +366.78 +/- 47.92

Commit

Commit ID 8bc3fd3871aaa2437105bdc141d5ac25a88ea885
Author Linmiao Xu
Date 2024-05-26 18:32:30 UTC
Lower smallnet threshold with tuned eval params The smallnet threshold is now below the training data range of the current smallnet (simple eval diff > 1k, nn-baff1edelf90.nnue) when no pawns are on the board. Params found with spsa at 93k / 120k games at 60+06: https://tests.stockfishchess.org/tests/view/664fa166a86388d5e27d7d6b Tuned on top of: https://github.com/official-stockfish/Stockfish/pull/5287 Passed STC: https://tests.stockfishchess.org/tests/view/664fc8b7a86388d5e27d8dac LLR: 2.96 (-2.94,2.94) <0.00,2.00> Total: 64672 W: 16731 L: 16371 D: 31570 Ptnml(0-2): 239, 7463, 16517, 7933, 184 Passed LTC: https://tests.stockfishchess.org/tests/view/664fd5f9a86388d5e27d8dfe LLR: 2.94 (-2.94,2.94) <0.50,2.50> Total: 210648 W: 53489 L: 52813 D: 104346 Ptnml(0-2): 102, 23129, 58164, 23849, 80 closes https://github.com/official-stockfish/Stockfish/pull/5288 Bench: 1717838
Copyright 2011–2024 Next Chess Move LLC