Dev Builds » 20241122-2234

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 06:50:24 583070 4000 1534 533 1933 +88.83 ± 4.9 2 93 822 1068 15 +188.08 ± 11.84
ncm-dbt-02 06:50:11 587729 4000 1554 551 1895 +89.02 ± 4.97 1 106 797 1081 15 +188.31 ± 12.05
ncm-dbt-03 06:53:21 587145 4002 1565 514 1923 +93.43 ± 4.87 1 84 797 1101 18 +198.69 ± 12.02
ncm-dbt-04 06:52:06 570412 4000 1566 537 1897 +91.43 ± 4.96 1 107 765 1116 11 +195.28 ± 12.31
ncm-dbt-05 06:53:55 584657 3998 1526 542 1930 +87.3 ± 4.92 2 105 807 1077 8 +185.9 ± 11.97
20000 7745 2677 9578 +90.0 ± 2.2 7 495 3988 5443 67 +191.22 ± 5.38

Test Detail

ID Host Base NPS Games WLD Standard Elo Ptnml(0-2) Gamepair Elo CLI PGN
387531 ncm-dbt-03 585674 2 0 0 2 0.0 ± 34.02 0 0 1 0 0 0.0 ± 30.47
387530 ncm-dbt-05 585717 498 201 73 224 +91.35 ± 14.07 0 13 97 137 2 +193.65 ± 34.74
387529 ncm-dbt-04 568474 500 201 52 247 +106.77 ± 13.56 0 10 82 157 1 +236.51 ± 37.9
387528 ncm-dbt-02 586435 500 189 75 236 +80.63 ± 13.79 0 15 106 129 0 +171.02 ± 33.15
387527 ncm-dbt-03 586985 500 198 64 238 +95.44 ± 14.1 0 11 98 137 4 +200.24 ± 34.5
387526 ncm-dbt-01 584622 500 193 61 246 +93.95 ± 13.61 0 10 100 138 2 +200.24 ± 34.09
387525 ncm-dbt-05 581693 500 191 65 244 +89.48 ± 13.69 0 12 101 136 1 +190.85 ± 33.97
387524 ncm-dbt-02 588558 500 185 63 252 +86.52 ± 15.09 0 20 91 136 3 +179.9 ± 35.84
387523 ncm-dbt-03 588899 500 189 64 247 +88.74 ± 13.83 0 12 103 133 2 +187.16 ± 33.6
387522 ncm-dbt-04 570709 500 190 76 234 +80.63 ± 14.52 0 19 99 131 1 +169.27 ± 34.4
387521 ncm-dbt-01 581236 500 186 77 237 +76.98 ± 14.83 1 18 104 125 2 +160.64 ± 33.54
387520 ncm-dbt-05 581902 500 190 61 249 +91.71 ± 14.19 1 13 92 144 0 +200.24 ± 35.72
387519 ncm-dbt-02 586181 500 195 58 247 +97.69 ± 13.66 1 8 95 145 1 +213.85 ± 35.03
387518 ncm-dbt-01 576169 500 186 66 248 +85.04 ± 13.9 0 14 103 132 1 +179.9 ± 33.65
387517 ncm-dbt-03 587792 500 212 61 227 +108.3 ± 13.23 0 9 81 160 0 +243.0 ± 38.14
387516 ncm-dbt-04 572356 500 185 60 255 +88.74 ± 13.83 0 14 97 139 0 +190.85 ± 34.75
387515 ncm-dbt-02 586858 500 195 80 225 +81.36 ± 14.11 0 13 113 120 4 +165.8 ± 31.92
387514 ncm-dbt-05 585421 500 178 69 253 +76.97 ± 13.53 0 13 116 120 1 +160.64 ± 31.43
387513 ncm-dbt-03 588387 500 193 73 234 +85.04 ± 13.44 0 11 109 129 1 +179.9 ± 32.5
387512 ncm-dbt-01 584706 500 187 60 253 +90.22 ± 13.55 0 11 102 136 1 +192.71 ± 33.75
387511 ncm-dbt-04 569749 500 202 78 220 +88.0 ± 14.12 0 14 100 134 2 +185.33 ± 34.19
387510 ncm-dbt-02 587579 500 195 79 226 +82.1 ± 13.83 0 15 104 131 0 +174.55 ± 33.5
387509 ncm-dbt-05 586350 500 184 85 231 +69.71 ± 14.31 0 21 109 120 0 +145.54 ± 32.74
387508 ncm-dbt-03 583447 500 189 66 245 +87.26 ± 13.95 0 14 100 135 1 +185.33 ± 34.19
387507 ncm-dbt-01 586223 500 202 73 225 +91.71 ± 13.09 1 6 106 137 0 +200.24 ± 32.81
387506 ncm-dbt-04 572638 500 194 63 243 +93.2 ± 13.12 0 8 104 137 1 +200.24 ± 33.24
387505 ncm-dbt-02 588728 500 203 59 238 +102.97 ± 13.38 0 8 92 148 2 +223.94 ± 35.61
387504 ncm-dbt-01 585253 500 197 69 234 +90.96 ± 14.02 0 12 101 134 3 +190.85 ± 33.97
387503 ncm-dbt-04 569589 500 198 71 231 +90.22 ± 14.16 0 12 103 131 4 +187.16 ± 33.6
387502 ncm-dbt-05 584076 500 189 60 251 +91.71 ± 14.49 1 14 91 143 1 +198.34 ± 35.92
387501 ncm-dbt-03 586985 500 196 67 237 +91.71 ± 13.57 0 9 106 132 3 +192.71 ± 32.93
387500 ncm-dbt-01 584453 500 202 67 231 +96.19 ± 14.26 0 11 98 136 5 +200.24 ± 34.5
387499 ncm-dbt-02 588600 500 197 61 242 +96.94 ± 14.12 0 13 90 145 2 +207.95 ± 36.13
387498 ncm-dbt-04 568912 500 196 66 238 +92.46 ± 14.79 1 15 89 143 2 +198.34 ± 36.31
387497 ncm-dbt-03 588856 500 193 57 250 +96.94 ± 13.81 0 9 100 137 4 +204.07 ± 34.04
387496 ncm-dbt-05 583070 500 197 61 242 +96.94 ± 13.65 0 10 96 142 2 +207.95 ± 34.86
387495 ncm-dbt-02 588899 500 195 76 229 +84.3 ± 14.18 0 14 106 127 3 +174.55 ± 33.13
387494 ncm-dbt-01 581902 500 181 60 259 +85.78 ± 13.46 0 11 108 130 1 +181.7 ± 32.68
387493 ncm-dbt-05 589027 500 196 68 236 +90.96 ± 13.24 0 9 105 135 1 +194.57 ± 33.11
387492 ncm-dbt-04 570869 500 200 71 229 +91.71 ± 14.04 0 15 91 144 0 +198.34 ± 35.92
387491 ncm-dbt-03 587282 500 195 62 243 +94.7 ± 14.09 1 9 99 138 3 +202.15 ± 34.28

Commit

Commit ID fb6be17ad40d321b5fff02395bc156568fce3091
Author Daniel Monroe
Date 2024-11-22 22:34:15 UTC
Simplify statscore at captures Simplify statscores for captures, setting them to 0 A recent tweak of Vizvezdenec finds substantial elo gain from giving captures a separate statscore, which is used mainly for reductions. The idea is that the old combination of quiet histories was inappropriate and that a value based on the capture history is more suitable. This simplification sets the statscore for captures to 0, suggesting that the elo gain came from rectifying the quiet history/capture mismatch. Passed STC (against a slightly older version of Viz's patch) https://tests.stockfishchess.org/tests/view/673ac6e286d5ee47d953f0ec LR: 2.96 (-2.94,2.94) <-1.75,0.25> Total: 57312 W: 14872 L: 14672 D: 27768 Ptnml(0-2): 152, 6761, 14649, 6923, 171 Passed LTC (against Viz's newest patch) https://tests.stockfishchess.org/tests/view/673cd00686d5ee47d953f2db LLR: 2.95 (-2.94,2.94) <-1.75,0.25> Total: 88236 W: 22510 L: 22358 D: 43368 Ptnml(0-2): 70, 9530, 24745, 9724, 49 closes https://github.com/official-stockfish/Stockfish/pull/5691 Bench: 959947
Copyright 2011–2024 Next Chess Move LLC