Dev Builds » 20230922-1725

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 09:54:44 1190840 3340 1498 261 1581 +135.1 ± 5.11 0 23 437 1160 50 +308.79 ± 16.29
ncm-dbt-02 09:51:40 1194342 3328 1461 256 1611 +131.77 ± 5.17 0 28 450 1139 47 +298.56 ± 16.06
ncm-dbt-03 09:55:14 1227577 3336 1460 280 1596 +128.44 ± 5.23 0 31 474 1115 48 +287.2 ± 15.64
ncm-dbt-04 09:54:52 1225572 3350 1445 276 1629 +126.55 ± 5.17 0 32 483 1119 41 +283.85 ± 15.49
ncm-dbt-05 09:50:38 1225051 3300 1447 269 1584 +129.74 ± 5.3 0 33 456 1111 50 +290.42 ± 15.96
ncm-dbt-06 09:55:07 1229096 3346 1463 270 1613 +129.57 ± 5.23 1 31 461 1134 46 +292.1 ± 15.87
20000 8774 1612 9614 +130.19 ± 2.12 1 178 2761 6778 282 +293.35 ± 6.47

Test Detail

ID Host Base NPS Games WLD Standard Elo Ptnml(0-2) Gamepair Elo CLI PGN
203990 ncm-dbt-05 1219288 300 139 23 138 +141.71 ± 19.33 0 3 38 99 10 +305.91 ± 56.45
203989 ncm-dbt-02 1192593 328 145 25 158 +133.28 ± 17.67 0 4 43 110 7 +293.96 ± 52.9
203988 ncm-dbt-01 1202183 340 155 28 157 +136.37 ± 16.82 0 0 53 107 10 +293.44 ± 46.63
203987 ncm-dbt-03 1219313 336 146 26 164 +129.8 ± 16.49 0 3 47 113 5 +291.0 ± 50.43
203986 ncm-dbt-06 1217595 346 147 32 167 +120.03 ± 17.24 0 5 54 108 6 +257.63 ± 46.92
203985 ncm-dbt-04 1208509 350 144 39 167 +107.54 ± 15.43 0 5 60 110 0 +240.82 ± 44.34
203984 ncm-dbt-05 1216027 500 218 37 245 +131.74 ± 13.3 0 3 71 168 8 +295.94 ± 40.69
203983 ncm-dbt-02 1196092 500 222 43 235 +130.14 ± 14.06 0 5 71 164 10 +285.49 ± 40.81
203982 ncm-dbt-03 1230893 500 213 47 240 +119.89 ± 14.19 0 10 69 166 5 +265.78 ± 41.36
203981 ncm-dbt-06 1213330 500 223 44 233 +130.15 ± 13.52 1 2 71 169 7 +295.94 ± 40.69
203980 ncm-dbt-04 1223613 500 217 37 246 +130.94 ± 13.87 0 4 72 164 10 +288.06 ± 40.46
203979 ncm-dbt-01 1205357 500 220 37 243 +133.34 ± 12.89 0 2 70 171 7 +304.07 ± 40.91
203978 ncm-dbt-05 1231824 500 210 49 241 +116.0 ± 13.37 0 7 78 162 3 +258.75 ± 38.88
203977 ncm-dbt-02 1193079 500 214 43 243 +123.81 ± 13.63 0 8 67 171 4 +280.42 ± 42.05
203976 ncm-dbt-01 1176535 500 230 48 222 +132.54 ± 13.83 0 7 61 175 7 +301.33 ± 44.13
203975 ncm-dbt-03 1224804 500 214 39 247 +126.97 ± 12.66 0 4 70 173 3 +293.29 ± 41.07
203974 ncm-dbt-04 1244256 500 219 46 235 +125.38 ± 13.43 0 6 70 169 5 +282.94 ± 41.13
203973 ncm-dbt-06 1225523 500 211 49 240 +116.77 ± 13.19 0 6 79 162 3 +261.07 ± 38.58
203972 ncm-dbt-05 1209263 500 218 28 254 +138.99 ± 12.7 0 4 57 184 5 +330.23 ± 45.79
203971 ncm-dbt-02 1193455 500 225 36 239 +138.18 ± 12.93 0 3 62 178 7 +321.19 ± 43.77
203970 ncm-dbt-01 1168957 500 237 36 227 +148.02 ± 12.1 0 2 51 191 6 +363.2 ± 48.51
203969 ncm-dbt-06 1231451 500 221 39 240 +132.54 ± 13.28 0 4 67 172 7 +301.33 ± 42.04
203968 ncm-dbt-03 1219088 500 227 39 234 +137.37 ± 13.71 0 4 64 172 10 +309.64 ± 43.08
203967 ncm-dbt-04 1221480 500 214 47 239 +120.67 ± 12.41 0 2 82 163 3 +273.0 ± 37.5
203966 ncm-dbt-05 1236137 500 227 45 228 +132.54 ± 13.47 0 0 80 158 12 +288.06 ± 37.72
203965 ncm-dbt-02 1195269 500 215 38 247 +128.55 ± 13.19 0 4 71 169 6 +290.66 ± 40.76
203964 ncm-dbt-06 1251810 500 215 36 249 +130.14 ± 13.52 0 5 68 170 7 +293.29 ± 41.75
203963 ncm-dbt-04 1231295 500 212 34 254 +129.35 ± 13.72 0 5 70 167 8 +288.06 ± 41.11
203962 ncm-dbt-01 1191869 500 214 40 246 +126.17 ± 13.06 0 3 76 165 6 +282.94 ± 39.21
203961 ncm-dbt-03 1220686 500 215 39 246 +127.76 ± 13.57 0 5 71 167 7 +285.49 ± 40.81
203960 ncm-dbt-05 1225891 500 214 41 245 +125.38 ± 13.79 0 8 66 171 5 +282.94 ± 42.37
203959 ncm-dbt-02 1193166 500 223 34 243 +138.18 ± 12.53 0 1 66 176 7 +321.19 ± 42.12
203958 ncm-dbt-04 1225869 500 223 39 238 +134.15 ± 13.61 0 5 64 173 8 +304.07 ± 43.1
203957 ncm-dbt-01 1211977 500 227 29 244 +145.54 ± 12.85 0 1 60 179 10 +339.63 ± 44.35
203956 ncm-dbt-06 1249505 500 228 35 237 +141.44 ± 13.59 0 3 62 174 11 +321.19 ± 43.77
203955 ncm-dbt-03 1244546 500 223 45 232 +129.35 ± 13.17 0 1 79 161 9 +285.49 ± 38.15
203954 ncm-dbt-05 1236928 500 221 46 233 +126.97 ± 14.11 0 8 66 169 7 +282.94 ± 42.37
203953 ncm-dbt-02 1196745 500 217 37 246 +130.94 ± 12.95 0 3 70 171 6 +298.62 ± 41.01
203952 ncm-dbt-04 1223983 500 216 34 250 +132.54 ± 13.47 0 5 65 173 7 +301.33 ± 42.75
203951 ncm-dbt-06 1214459 500 218 35 247 +133.34 ± 13.26 0 6 60 179 5 +309.64 ± 44.55
203950 ncm-dbt-03 1233714 500 222 45 233 +128.55 ± 13.73 0 4 74 163 9 +282.94 ± 39.86
203949 ncm-dbt-01 1179003 500 215 43 242 +124.6 ± 13.62 0 8 66 172 4 +282.94 ± 42.37

Commit

Commit ID 154b8d3ecb19d0b3fa9ec11cc3a1e666dfe0d2ce
Author Stefan Geschwentner
Date 2023-09-22 17:25:57 UTC
Remove VALUE_KNOWN_WIN. After removing classic evaluation VALUE_KNOWN_WIN is not anymore returned explicit evaluation. So remove and replace it with VALUE_TB_WIN_IN_MAX_PLY. Measurement on my big bench (bench 16 1 16 pos1000.fen) verifies that at least with current net the calculated evaluation lies always in the open interval (-VALUE_KNOWN_WIN, VALUE_KNOWN_WIN). So i consider this a non-functional change. But to be safe i tested this also at LTC as requested by Stephane Nicolet. STC: https://tests.stockfishchess.org/tests/view/64f9db40eaf01be8259a6ed5 LLR: 2.93 (-2.94,2.94) <-1.75,0.25> Total: 455296 W: 115981 L: 116217 D: 223098 Ptnml(0-2): 1415, 50835, 123420, 50527, 1451 LTC: https://tests.stockfishchess.org/tests/view/650bfd867ca0d3f7bbf25feb LLR: 2.95 (-2.94,2.94) <-1.75,0.25> Total: 35826 W: 9170 L: 8973 D: 17683 Ptnml(0-2): 12, 3523, 10645, 3722, 11 closes https://github.com/official-stockfish/Stockfish/pull/4792 Bench: 1603079
Copyright 2011–2024 Next Chess Move LLC