Dev Builds » 20210220-2149

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 06:42:00 583762 4000 298 1796 1906 -136.77 ± 4.79 79 1365 531 25 0 -307.89 ± 14.76
ncm-dbt-02 06:42:28 586284 4000 305 1824 1871 -138.89 ± 4.85 81 1390 496 33 0 -314.63 ± 15.31
ncm-dbt-03 06:43:12 584681 4000 306 1828 1866 -139.2 ± 5.1 110 1338 516 36 0 -305.45 ± 15.01
ncm-dbt-04 06:41:52 566773 4000 306 1823 1871 -138.69 ± 5.05 100 1355 508 36 1 -307.54 ± 15.13
ncm-dbt-05 06:41:46 581132 4000 317 1837 1846 -138.99 ± 4.91 92 1366 512 30 0 -311.06 ± 15.06
20000 1532 9108 9360 -138.51 ± 2.21 462 6814 2563 160 1 -309.29 ± 6.72

Test Detail

ID Host Base NPS Games WLD Standard Elo Ptnml(0-2) Gamepair Elo CLI PGN
456644 ncm-dbt-03 585421 500 33 215 252 -132.54 ± 14.53 10 170 62 8 0 -293.29 ± 43.72
456643 ncm-dbt-04 561873 500 25 232 243 -153.02 ± 14.3 16 179 51 4 0 -349.43 ± 48.52
456642 ncm-dbt-02 588217 500 45 232 223 -136.56 ± 13.73 12 165 71 2 0 -301.33 ± 40.6
456641 ncm-dbt-01 581693 500 36 223 241 -136.56 ± 13.73 12 165 71 2 0 -301.33 ± 40.6
456640 ncm-dbt-05 582611 500 25 226 249 -148.02 ± 13.15 10 184 53 3 0 -349.43 ± 47.56
456639 ncm-dbt-03 583447 500 31 219 250 -137.37 ± 14.43 14 164 68 4 0 -298.62 ± 41.71
456638 ncm-dbt-04 567442 500 35 231 234 -143.89 ± 14.07 11 180 53 6 0 -330.23 ± 47.44
456637 ncm-dbt-01 583196 500 39 230 231 -139.81 ± 14.01 12 171 63 4 0 -312.48 ± 43.44
456636 ncm-dbt-05 576210 500 44 222 234 -129.35 ± 14.07 8 169 66 7 0 -288.06 ± 42.4
456635 ncm-dbt-02 586350 500 25 221 254 -143.89 ± 12.3 6 186 56 2 0 -346.12 ± 46.15
456634 ncm-dbt-03 586604 500 43 229 228 -135.76 ± 12.62 6 176 66 2 0 -315.35 ± 42.24
456633 ncm-dbt-01 584453 500 45 227 228 -132.54 ± 13.65 11 162 75 2 0 -290.66 ± 39.4
456632 ncm-dbt-05 584159 500 46 233 221 -136.56 ± 15.29 19 153 74 4 0 -282.94 ± 39.86
456631 ncm-dbt-04 566178 500 45 225 230 -130.94 ± 15.54 13 164 64 8 1 -282.94 ± 42.98
456630 ncm-dbt-02 587749 500 36 229 235 -141.44 ± 12.61 7 181 60 2 0 -333.32 ± 44.47
456629 ncm-dbt-03 581277 500 40 235 225 -143.07 ± 14.46 15 169 62 4 0 -315.35 ± 43.81
456628 ncm-dbt-01 583908 500 34 229 237 -143.07 ± 14.1 13 173 60 4 0 -321.19 ± 44.57
456627 ncm-dbt-04 567561 500 49 225 226 -127.76 ± 13.75 10 159 78 3 0 -277.93 ± 38.66
456626 ncm-dbt-05 580821 500 36 236 228 -147.19 ± 14.15 16 170 62 2 0 -327.18 ± 43.69
456625 ncm-dbt-02 585337 500 32 222 246 -138.99 ± 13.1 8 177 62 3 0 -321.19 ± 43.77
456624 ncm-dbt-03 587113 500 36 226 238 -138.99 ± 14.39 13 169 63 5 0 -306.84 ± 43.45
456623 ncm-dbt-01 584327 500 33 221 246 -137.37 ± 14.08 10 174 60 6 0 -309.64 ± 44.55
456622 ncm-dbt-04 565666 500 30 224 246 -142.25 ± 15.01 16 168 60 6 0 -309.64 ± 44.55
456621 ncm-dbt-05 580904 500 55 237 208 -132.54 ± 13.1 7 171 69 3 0 -301.33 ± 41.33
456620 ncm-dbt-02 585042 500 30 231 239 -148.02 ± 14.49 15 176 54 5 0 -333.32 ± 47.06
456619 ncm-dbt-03 585337 500 44 237 219 -141.44 ± 15.7 19 162 62 7 0 -298.62 ± 43.77
456618 ncm-dbt-01 585337 500 38 226 236 -137.37 ± 12.96 6 180 60 4 0 -321.19 ± 44.57
456617 ncm-dbt-05 580987 500 32 241 227 -154.7 ± 14.04 17 177 54 2 0 -352.8 ± 47.05
456616 ncm-dbt-04 567918 500 43 222 235 -130.14 ± 12.59 5 171 72 2 0 -298.62 ± 40.29
456615 ncm-dbt-02 586901 500 42 227 231 -134.95 ± 13.41 9 170 68 3 0 -304.07 ± 41.65
456614 ncm-dbt-03 586308 500 44 231 225 -136.56 ± 14.1 14 161 73 2 0 -295.94 ± 39.99
456613 ncm-dbt-01 582903 500 41 219 240 -129.35 ± 13.54 10 160 78 2 0 -282.94 ± 38.55
456612 ncm-dbt-05 578959 500 37 216 247 -130.14 ± 13.34 6 172 67 5 0 -295.94 ± 42.08
456611 ncm-dbt-04 570148 500 40 227 233 -136.56 ± 13.92 11 169 66 4 0 -304.07 ± 42.38
456610 ncm-dbt-02 586266 500 43 231 226 -137.37 ± 14.6 14 165 66 5 0 -298.62 ± 42.41
456609 ncm-dbt-03 581943 500 35 236 229 -148.02 ± 15.03 19 167 60 4 0 -321.19 ± 44.57
456608 ncm-dbt-04 567403 500 39 237 224 -145.54 ± 14.75 18 165 64 3 0 -315.35 ± 43.03
456607 ncm-dbt-01 584285 500 32 221 247 -138.18 ± 12.12 5 180 64 1 0 -327.18 ± 42.83
456606 ncm-dbt-05 584411 500 42 226 232 -134.15 ± 13.61 9 170 67 4 0 -301.33 ± 42.04
456605 ncm-dbt-02 584411 500 52 231 217 -130.14 ± 15.06 10 170 59 11 0 -285.49 ± 44.52

Commit

Commit ID 7c30091a92abddb8265e53768b32751c49642040
Author Vizvezdenec
Date 2021-02-20 21:49:39 UTC
Introduce ProbCut for check evasions The idea of this patch can be described as follows: if we are in check and the transposition table move is a capture that returns a value far above beta, we can assume that the opponent just blundered a piece by giving check, and we return the transposition table value. This is similar to the usual probCut logic for quiet moves, but with a different threshold. Passed STC LLR: 2.94 (-2.94,2.94) {-0.25,1.25} Total: 33440 W: 3056 L: 2891 D: 27493 Ptnml(0-2): 110, 2338, 11672, 2477, 123 https://tests.stockfishchess.org/tests/view/602cd1087f517a561bc49bda Passed LTC LLR: 2.98 (-2.94,2.94) {0.25,1.25} Total: 10072 W: 401 L: 309 D: 9362 Ptnml(0-2): 2, 288, 4365, 378, 3 https://tests.stockfishchess.org/tests/view/602ceea57f517a561bc49bf0 The committed version has an additional fix to never return unproven wins in the tablebase range or the mate range. This fix passed tests for non- regression at STC and LTC: STC: LLR: 2.93 (-2.94,2.94) {-1.25,0.25} Total: 26240 W: 2354 L: 2280 D: 21606 Ptnml(0-2): 85, 1763, 9372, 1793, 107 https://tests.stockfishchess.org/tests/view/602d86a87f517a561bc49c7a LTC: LLR: 2.95 (-2.94,2.94) {-0.75,0.25} Total: 35304 W: 1299 L: 1256 D: 32749 Ptnml(0-2): 14, 1095, 15395, 1130, 18 https://tests.stockfishchess.org/tests/view/602d98d17f517a561bc49c83 Closes https://github.com/official-stockfish/Stockfish/pull/3362 Bench: 3830215
Copyright 2011–2026 Next Chess Move LLC