Dev Builds » 20210622-0951

Use this dev build

NCM plays each Stockfish dev build 20,000 times against Stockfish 7. This yields an approximate Elo difference and establishes confidence in the strength of the dev builds.

Summary

Host Duration Avg Base NPS Games Wins Losses Draws Elo
ncm-et-3 08:19:05 1964258 3344 2811 15 518 +419.76 +/- 14.98
ncm-et-4 08:19:30 1962068 3339 2750 9 580 +402.88 +/- 14.13
ncm-et-9 08:19:03 1962677 3355 2795 12 548 +412.25 +/- 14.55
ncm-et-10 08:19:18 1955230 3300 2774 6 520 +422.85 +/- 14.93
ncm-et-13 08:19:04 1963763 3331 2788 9 534 +417.64 +/- 14.74
ncm-et-15 08:19:06 1959926 3331 2757 11 563 +406.61 +/- 14.35
20000 16675 62 3263 +413.53 +/- 5.95

Test Detail

ID Host Started (UTC) Duration Base NPS Games Wins Losses Draws Elo CLI PGN
149366 ncm-et-10 2021-06-22 21:41 00:46:04 1953894 300 257 0 43 +444.95 +/- 53.05
149365 ncm-et-13 2021-06-22 21:39 00:48:17 1962161 331 286 1 44 +450.73 +/- 52.57
149364 ncm-et-15 2021-06-22 21:38 00:49:19 1961074 331 280 2 49 +424.14 +/- 49.67
149363 ncm-et-4 2021-06-22 21:37 00:51:00 1956179 339 273 3 63 +378.31 +/- 43.54
149362 ncm-et-3 2021-06-22 21:34 00:53:07 1958164 344 281 1 62 +395.6 +/- 43.82
149361 ncm-et-9 2021-06-22 21:34 00:53:34 1963550 355 295 1 59 +410.77 +/- 45.01
149360 ncm-et-10 2021-06-22 20:25 01:15:46 1955598 500 434 1 65 +457.52 +/- 42.89
149359 ncm-et-15 2021-06-22 20:24 01:13:06 1966166 500 415 2 83 +408.38 +/- 37.78
149358 ncm-et-13 2021-06-22 20:22 01:16:20 1957379 500 408 2 90 +393.6 +/- 36.22
149357 ncm-et-9 2021-06-22 20:20 01:13:17 1964765 500 414 2 84 +406.2 +/- 37.55
149356 ncm-et-4 2021-06-22 20:20 01:16:16 1959264 500 409 0 91 +399.81 +/- 35.89
149355 ncm-et-3 2021-06-22 20:19 01:14:36 1964462 500 427 1 72 +438.95 +/- 40.66
149354 ncm-et-15 2021-06-22 19:09 01:14:08 1966941 500 415 1 84 +410.58 +/- 37.5
149353 ncm-et-10 2021-06-22 19:09 01:15:30 1956491 500 422 0 78 +429.05 +/- 38.92
149352 ncm-et-13 2021-06-22 19:06 01:15:40 1967238 500 419 1 80 +419.61 +/- 38.47
149351 ncm-et-4 2021-06-22 19:04 01:14:43 1960207 500 416 0 84 +415.04 +/- 37.43
149350 ncm-et-9 2021-06-22 19:04 01:15:02 1960319 500 424 0 76 +433.94 +/- 39.45
149349 ncm-et-3 2021-06-22 19:03 01:15:39 1959084 500 423 4 73 +421.93 +/- 40.4
149348 ncm-et-15 2021-06-22 17:53 01:15:33 1954499 500 423 2 75 +426.65 +/- 39.83
149347 ncm-et-10 2021-06-22 17:51 01:17:06 1950699 500 415 1 84 +410.58 +/- 37.5
149346 ncm-et-4 2021-06-22 17:50 01:13:53 1965708 500 423 2 75 +426.65 +/- 39.83
149345 ncm-et-13 2021-06-22 17:50 01:15:18 1969089 500 422 1 77 +426.65 +/- 39.25
149344 ncm-et-9 2021-06-22 17:48 01:15:21 1952984 500 422 4 74 +419.61 +/- 40.12
149343 ncm-et-3 2021-06-22 17:47 01:14:38 1963395 500 412 2 86 +401.92 +/- 37.09
149342 ncm-et-15 2021-06-22 16:38 01:15:09 1951788 500 424 2 74 +429.05 +/- 40.11
149341 ncm-et-10 2021-06-22 16:35 01:15:48 1957867 500 415 2 83 +408.38 +/- 37.78
149340 ncm-et-4 2021-06-22 16:34 01:14:48 1959085 500 397 2 101 +372.25 +/- 34.1
149339 ncm-et-13 2021-06-22 16:34 01:15:03 1966000 500 426 2 72 +433.94 +/- 40.69
149338 ncm-et-9 2021-06-22 16:34 01:13:40 1961543 500 419 2 79 +417.32 +/- 38.77
149337 ncm-et-3 2021-06-22 16:33 01:13:38 1973430 500 416 2 82 +410.58 +/- 38.02
149336 ncm-et-15 2021-06-22 15:20 01:16:38 1954097 500 396 1 103 +372.25 +/- 33.69
149335 ncm-et-10 2021-06-22 15:20 01:13:47 1955419 500 420 1 79 +421.93 +/- 38.73
149334 ncm-et-13 2021-06-22 15:19 01:14:22 1960173 500 412 1 87 +404.05 +/- 36.82
149333 ncm-et-4 2021-06-22 15:19 01:14:45 1963541 500 418 0 82 +419.61 +/- 37.91
149332 ncm-et-3 2021-06-22 15:19 01:13:50 1960308 500 423 1 76 +429.05 +/- 39.52
149331 ncm-et-9 2021-06-22 15:18 01:15:08 1962150 500 404 3 93 +383.64 +/- 35.64
149330 ncm-et-15 2021-06-22 14:04 01:15:13 1964917 500 404 1 95 +387.57 +/- 35.16
149329 ncm-et-3 2021-06-22 14:04 01:13:37 1970967 500 429 4 67 +436.43 +/- 42.22
149328 ncm-et-13 2021-06-22 14:04 01:14:04 1964305 500 415 1 84 +410.58 +/- 37.5
149327 ncm-et-9 2021-06-22 14:04 01:13:01 1973432 500 417 0 83 +417.31 +/- 37.67
149326 ncm-et-4 2021-06-22 14:04 01:14:05 1970492 500 414 2 84 +406.2 +/- 37.55
149325 ncm-et-10 2021-06-22 14:04 01:15:17 1956647 500 411 1 88 +401.92 +/- 36.6

Commit

Commit ID 0470bcef0e1962b4f8da15108170b991d3f90d0e
Author Stéphane Nicolet
Date 2021-06-22 09:51:03 UTC
Detect fortresses a little bit quicker In the so-called "hybrid" method of evaluation of current master, we use the classical eval (because of its speed) instead of the NNUE eval when the classical material balance approximation hints that the position is "winning enough" to rely on the classical eval. This trade-off idea between speed and accuracy works well in general, but in some fortress positions the classical eval is just bad. So in shuffling branches of the search tree, we (slowly) increase the thresehold so that eventually we don't trust classical anymore and switch to NNUE evaluation. This patch increases that threshold faster, so that we switch to NNUE quicker in shuffling branches. Idea is to incite Stockfish to spend less time in fortresses lines in the search tree, and spend more time searching the critical lines. passed STC: LLR: 2.96 (-2.94,2.94) <-0.50,2.50> Total: 47872 W: 3908 L: 3720 D: 40244 Ptnml(0-2): 122, 3053, 17419, 3199, 143 https://tests.stockfishchess.org/tests/view/60cef34b457376eb8bcab79d passed LTC: LLR: 2.93 (-2.94,2.94) <0.50,3.50> Total: 73616 W: 2326 L: 2143 D: 69147 Ptnml(0-2): 21, 1940, 32705, 2119, 23 https://tests.stockfishchess.org/tests/view/60cf6d842114332881e73528 Retested at LTC against lastest master: LLR: 2.93 (-2.94,2.94) <0.50,3.50> Total: 18264 W: 642 L: 532 D: 17090 Ptnml(0-2): 6, 479, 8055, 583, 9 https://tests.stockfishchess.org/tests/view/60d18cd540925195e7a6c351 closes https://github.com/official-stockfish/Stockfish/pull/3578 Bench: 5139233
Copyright 2011–2024 Next Chess Move LLC