Dev Builds » 20210519-1734

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 06:47:54 582589 4002 359 1784 1859 -129.38 ± 5.06 84 1304 567 45 1 -282.1 ± 14.31
ncm-dbt-02 06:48:34 586477 3998 318 1802 1878 -135.43 ± 5.05 86 1361 504 47 1 -301.23 ± 15.19
ncm-dbt-03 06:47:27 587046 4000 335 1816 1849 -135.05 ± 5.06 102 1312 551 35 0 -294.28 ± 14.51
ncm-dbt-04 06:46:52 566993 4000 287 1784 1929 -136.67 ± 5.08 101 1335 524 40 0 -299.97 ± 14.89
ncm-dbt-05 06:47:29 582512 4000 350 1820 1830 -133.95 ± 5.12 104 1301 556 39 0 -290.01 ± 14.45
20000 1649 9006 9345 -134.09 ± 2.27 477 6613 2702 206 2 -293.42 ± 6.55

Test Detail

ID Host Base NPS Games WLD Standard Elo Ptnml(0-2) Gamepair Elo CLI PGN
445610 ncm-dbt-01 583740 2 0 1 1 -189.7 ± 55.98 0 1 0 0 0 -1129.65 ± 376.02
445609 ncm-dbt-02 585759 498 43 221 234 -129.92 ± 14.44 9 168 64 8 0 -287.23 ± 43.03
445608 ncm-dbt-05 581860 500 46 224 230 -129.35 ± 14.24 9 167 67 7 0 -285.49 ± 42.08
445607 ncm-dbt-03 584579 500 39 234 227 -143.07 ± 14.28 15 168 64 3 0 -315.35 ± 43.03
445606 ncm-dbt-04 568116 500 32 224 244 -140.62 ± 16.51 22 157 62 9 0 -288.06 ± 43.65
445605 ncm-dbt-01 583740 500 45 219 236 -126.17 ± 14.29 8 166 68 8 0 -277.93 ± 41.74
445604 ncm-dbt-02 587028 500 41 230 229 -138.18 ± 15.76 16 167 57 10 0 -295.94 ± 45.36
445603 ncm-dbt-03 586858 500 46 232 222 -135.76 ± 14.64 13 166 65 6 0 -295.94 ± 42.75
445602 ncm-dbt-01 583866 500 36 229 235 -141.44 ± 13.01 8 180 59 3 0 -330.23 ± 44.94
445601 ncm-dbt-05 581860 500 37 227 236 -138.99 ± 14.21 15 162 71 2 0 -301.33 ± 40.6
445600 ncm-dbt-04 565587 500 42 216 242 -126.17 ± 13.24 7 163 77 3 0 -280.42 ± 38.93
445599 ncm-dbt-03 585632 500 35 232 233 -144.71 ± 13.67 12 176 59 3 0 -330.23 ± 44.94
445598 ncm-dbt-02 585590 500 35 217 248 -132.54 ± 14.18 7 177 57 9 0 -301.33 ± 45.48
445597 ncm-dbt-04 567403 500 41 228 231 -136.56 ± 14.1 13 164 70 3 0 -298.62 ± 41.01
445596 ncm-dbt-05 582027 500 40 227 233 -136.56 ± 14.79 14 165 65 6 0 -295.94 ± 42.75
445595 ncm-dbt-01 577807 500 41 225 234 -134.15 ± 13.97 13 160 75 2 0 -290.66 ± 39.4
445594 ncm-dbt-03 589155 500 37 231 232 -142.25 ± 14.12 13 172 61 4 0 -318.25 ± 44.18
445593 ncm-dbt-02 586816 500 42 225 233 -133.34 ± 12.49 4 178 65 3 0 -312.48 ± 42.67
445592 ncm-dbt-04 569031 500 31 220 249 -138.18 ± 13.69 9 176 60 5 0 -315.35 ± 44.57
445591 ncm-dbt-05 577192 500 41 227 232 -135.76 ± 14.98 16 159 70 5 0 -288.06 ± 41.11
445590 ncm-dbt-01 584201 500 46 224 230 -129.35 ± 15.24 15 155 73 7 0 -270.57 ± 40.25
445589 ncm-dbt-03 587622 500 41 228 231 -136.56 ± 14.62 13 167 64 6 0 -298.62 ± 43.1
445588 ncm-dbt-02 586350 500 38 231 231 -141.44 ± 14.51 14 170 61 5 0 -312.48 ± 44.19
445587 ncm-dbt-01 581527 500 49 217 234 -121.45 ± 14.34 9 157 77 7 0 -261.07 ± 39.14
445586 ncm-dbt-05 583572 500 43 244 213 -148.02 ± 15.2 21 162 64 3 0 -315.35 ± 43.03
445585 ncm-dbt-04 566533 500 37 229 234 -140.62 ± 14.17 12 173 60 5 0 -315.35 ± 44.57
445584 ncm-dbt-01 582110 500 50 220 230 -123.02 ± 15.13 12 154 77 6 1 -261.07 ± 39.14
445583 ncm-dbt-02 587282 500 44 225 231 -131.74 ± 14.03 11 163 72 4 0 -288.06 ± 40.46
445582 ncm-dbt-05 586012 500 39 227 234 -137.37 ± 13.71 10 172 64 4 0 -309.64 ± 43.08
445581 ncm-dbt-03 587877 500 42 218 240 -127.76 ± 14.93 14 154 76 6 0 -268.17 ± 39.39
445580 ncm-dbt-04 566809 500 27 225 248 -145.54 ± 13.64 13 174 61 2 0 -330.23 ± 44.08
445579 ncm-dbt-01 584748 500 52 230 218 -129.35 ± 14.24 11 161 73 5 0 -280.42 ± 40.21
445578 ncm-dbt-05 582110 500 58 219 223 -116.0 ± 15.0 11 148 82 9 0 -240.82 ± 37.9
445577 ncm-dbt-02 586308 500 32 232 236 -147.19 ± 15.22 18 169 59 3 1 -324.17 ± 44.97
445576 ncm-dbt-03 588132 500 42 228 230 -135.76 ± 14.81 15 161 69 5 0 -290.66 ± 41.43
445575 ncm-dbt-04 569190 500 33 218 249 -134.95 ± 15.32 14 166 61 9 0 -290.66 ± 44.0
445574 ncm-dbt-02 586689 500 43 221 236 -129.35 ± 13.54 7 169 69 5 0 -290.66 ± 41.43
445573 ncm-dbt-01 581569 500 40 219 241 -130.14 ± 14.06 8 170 65 7 0 -290.66 ± 42.73
445572 ncm-dbt-03 586520 500 53 213 234 -115.22 ± 13.2 7 148 93 2 0 -247.41 ± 34.91
445571 ncm-dbt-04 563275 500 44 224 232 -130.94 ± 14.04 11 162 73 4 0 -285.49 ± 40.16
445570 ncm-dbt-05 585464 500 46 225 229 -130.14 ± 13.34 8 166 73 3 0 -290.66 ± 40.08

Commit

Commit ID d37de3cb1de63da5b2f8e6978c45c5b36973063b
Author Vizvezdenec
Date 2021-05-19 17:34:37 UTC
Do more continuation history based pruning This patch increases lmrDepth threshold for continuation history based pruning in search. This part of code for a long time was known to be really TC sensitive - decreasing this threshold easily passed lower time controls but failed badly at LTC, on the other hand it increase was part of a tuning that resulted in being negative at STC but was +12 elo at 180+1.8. After recent simplification of special conditions that sometimes increase it from 4 to 5 it was logical to overall test at longer time controls if 5 is better than 4 with deeper searches. reduces strenght on STC https://tests.stockfishchess.org/tests/view/60a3a8bbce8ea25a3ef03c74 ELO: -2.57 +-2.0 (95%) LOS: 0.6% Total: 20000 W: 1820 L: 1968 D: 16212 Ptnml(0-2): 68, 1582, 6836, 1458, 56 Passed LTC with STC bounds https://tests.stockfishchess.org/tests/view/60a027395085663412d090ce LLR: 2.93 (-2.94,2.94) <-0.50,2.50> Total: 175256 W: 6774 L: 6548 D: 161934 Ptnml(0-2): 91, 5808, 75604, 6034, 91 Passed VLTC with LTC bounds https://tests.stockfishchess.org/tests/view/60a2bccce229097940a037a7 LLR: 2.96 (-2.94,2.94) <0.50,3.50> Total: 65736 W: 1224 L: 1092 D: 63420 Ptnml(0-2): 5, 1012, 30706, 1136, 9 closes https://github.com/official-stockfish/Stockfish/pull/3473 bench 3689330
Copyright 2011–2025 Next Chess Move LLC