Dev Builds » 20210525-1940

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 08:16:14 1964644 3340 2697 17 626 +384.02 +/- 13.61
ncm-et-4 08:15:47 1958859 3345 2717 12 616 +390.23 +/- 13.71
ncm-et-9 08:16:26 1961533 3332 2710 19 603 +389.18 +/- 13.88
ncm-et-10 08:16:28 1951560 3311 2723 8 580 +401.91 +/- 14.13
ncm-et-13 08:15:49 1961208 3342 2683 11 648 +381.24 +/- 13.36
ncm-et-15 08:16:35 1962285 3330 2742 13 575 +401.41 +/- 14.2
20000 16272 80 3648 +391.17 +/- 5.63

Test Detail

ID Host Started (UTC) Duration Base NPS Games Wins Losses Draws Elo CLI PGN
147600 ncm-et-15 2021-05-27 11:46 00:50:07 1963732 339 280 1 58 +405.13 +/- 45.39
147599 ncm-et-10 2021-05-27 11:46 00:50:24 1945554 337 278 0 59 +407.2 +/- 44.87
147598 ncm-et-9 2021-05-27 11:45 00:50:46 1963234 336 271 1 64 +385.17 +/- 43.08
147597 ncm-et-4 2021-05-27 11:45 00:50:51 1968623 344 287 1 56 +414.37 +/- 46.26
147596 ncm-et-3 2021-05-27 11:45 00:51:25 1969132 354 292 0 62 +407.13 +/- 43.73
147595 ncm-et-13 2021-05-27 11:44 00:52:06 1949194 340 266 2 72 +360.09 +/- 40.55
147594 ncm-et-15 2021-05-27 10:30 01:15:21 1972207 500 410 2 88 +397.72 +/- 36.65
147593 ncm-et-13 2021-05-27 10:30 01:13:14 1962613 500 397 1 102 +374.11 +/- 33.87
147592 ncm-et-3 2021-05-27 10:30 01:13:56 1966924 500 415 1 84 +410.58 +/- 37.5
147591 ncm-et-9 2021-05-27 10:30 01:14:46 1954813 500 413 2 85 +404.05 +/- 37.32
147590 ncm-et-4 2021-05-27 10:30 01:14:36 1970633 500 411 4 85 +395.65 +/- 37.35
147589 ncm-et-10 2021-05-27 10:30 01:15:05 1956643 500 410 2 88 +397.72 +/- 36.65
147503 ncm-et-13 2021-05-26 17:46 00:01:13 1965542 2 2 0 0 +1199.83 +/- 312.71
147502 ncm-et-4 2021-05-26 17:46 00:01:09 1960310 1 1 0 0 +1129.65 +/- 376.02
147501 ncm-et-9 2021-05-26 16:35 01:11:27 1969564 496 408 2 86 +400.39 +/- 37.09
147500 ncm-et-10 2021-05-26 16:35 01:11:32 1949349 474 405 2 67 +436.7 +/- 42.24
147499 ncm-et-15 2021-05-26 16:35 01:12:19 1967188 491 412 2 77 +418.5 +/- 39.29
147498 ncm-et-3 2021-05-26 16:34 01:12:37 1965073 486 384 4 98 +364.88 +/- 34.69
147497 ncm-et-13 2021-05-26 16:31 01:13:37 1959694 500 387 1 112 +356.21 +/- 32.24
147496 ncm-et-4 2021-05-26 16:31 01:13:57 1965225 500 397 1 102 +374.11 +/- 33.87
147495 ncm-et-15 2021-05-26 15:20 01:14:17 1959839 500 411 1 88 +401.92 +/- 36.6
147494 ncm-et-10 2021-05-26 15:19 01:15:10 1950697 500 420 1 79 +421.93 +/- 38.73
147493 ncm-et-3 2021-05-26 15:19 01:14:18 1962926 500 403 3 94 +381.7 +/- 35.44
147492 ncm-et-9 2021-05-26 15:19 01:15:50 1958382 500 401 5 94 +374.11 +/- 35.47
147491 ncm-et-4 2021-05-26 15:17 01:12:59 1968167 500 413 1 86 +406.2 +/- 37.04
147490 ncm-et-13 2021-05-26 15:17 01:13:46 1959852 500 401 2 97 +379.78 +/- 34.83
147489 ncm-et-15 2021-05-26 14:06 01:12:41 1964459 500 408 2 90 +393.6 +/- 36.22
147488 ncm-et-9 2021-05-26 14:05 01:13:18 1969413 500 418 1 81 +417.32 +/- 38.22
147487 ncm-et-10 2021-05-26 14:05 01:14:15 1948729 500 405 1 94 +389.56 +/- 35.35
147486 ncm-et-3 2021-05-26 14:04 01:14:13 1955739 500 394 3 103 +364.98 +/- 33.79
147485 ncm-et-4 2021-05-26 14:03 01:13:15 1955720 500 401 3 96 +377.87 +/- 35.05
147484 ncm-et-13 2021-05-26 14:02 01:14:11 1961697 500 418 0 82 +419.61 +/- 37.91
147483 ncm-et-9 2021-05-26 12:49 01:14:51 1960055 500 404 2 94 +385.59 +/- 35.4
147482 ncm-et-15 2021-05-26 12:49 01:16:45 1952383 500 404 3 93 +383.64 +/- 35.64
147481 ncm-et-3 2021-05-26 12:49 01:15:03 1974839 500 402 3 95 +379.78 +/- 35.24
147480 ncm-et-4 2021-05-26 12:49 01:14:15 1941795 500 395 1 104 +370.41 +/- 33.52
147479 ncm-et-10 2021-05-26 12:48 01:15:40 1955438 500 399 1 100 +377.87 +/- 34.22
147478 ncm-et-13 2021-05-26 12:47 01:14:42 1966767 500 406 3 91 +387.57 +/- 36.04
147477 ncm-et-9 2021-05-26 11:33 01:15:28 1955276 500 395 6 99 +361.43 +/- 34.54
147476 ncm-et-3 2021-05-26 11:33 01:14:42 1957875 500 407 3 90 +389.56 +/- 36.25
147475 ncm-et-13 2021-05-26 11:33 01:13:00 1964312 500 406 2 92 +389.56 +/- 35.8
147474 ncm-et-10 2021-05-26 11:33 01:14:22 1954512 500 406 1 93 +391.57 +/- 35.55
147473 ncm-et-4 2021-05-26 11:33 01:14:45 1940401 500 412 1 87 +404.05 +/- 36.82
147472 ncm-et-15 2021-05-26 11:33 01:15:05 1956192 500 417 2 81 +412.8 +/- 38.27

Commit

Commit ID 9d53129075177cb11b63b43236556051ba60f7dd
Author Tomasz Sobczyk
Date 2021-05-25 19:40:51 UTC
Expose the lazy threshold for the feature transformer PSQT as a parameter. Definition of the lazy threshold moved to evaluate.cpp where all others are. Lazy threshold only used for real searches, not used for the "eval" call. This preserves the purity of NNUE evaluation, which is useful to verify consistency between the engine and the NNUE trainer. closes https://github.com/official-stockfish/Stockfish/pull/3499 No functional change
Copyright 2011–2024 Next Chess Move LLC