Dev Builds » 20140805-0342

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 06:02:37 1988263 2391 160 853 1378 -103.67 +/- 8.86
ncm-et-4 06:02:01 1984626 2394 151 808 1435 -97.86 +/- 8.59
ncm-et-5 01:48:07 2023201 712 49 232 431 -91.35 +/- 15.68
ncm-et-6 01:48:22 2025900 729 44 267 418 -109.79 +/- 16.07
ncm-et-7 01:48:07 2008125 711 41 233 437 -96.21 +/- 15.41
ncm-et-8 01:47:54 1998961 702 38 230 434 -97.51 +/- 15.4
ncm-et-9 06:02:39 1990430 2378 163 858 1357 -104.59 +/- 8.96
ncm-et-10 06:02:28 1985100 2372 160 786 1426 -93.91 +/- 8.62
ncm-et-11 01:48:05 2017343 717 40 258 419 -109.08 +/- 15.94
ncm-et-12 01:48:13 1939373 682 44 250 388 -108.32 +/- 16.74
ncm-et-13 06:02:29 1992574 2392 142 818 1432 -100.94 +/- 8.58
ncm-et-14 01:48:17 2018723 715 45 260 410 -107.81 +/- 16.25
ncm-et-15 06:02:37 1992660 2399 161 831 1407 -99.68 +/- 8.74
ncm-et-16 01:48:17 2003220 706 57 244 405 -94.27 +/- 16.46
20000 1295 6928 11777 -100.57 +/- 3.01

Test Detail

ID Host Started (UTC) Duration Base NPS Games Wins Losses Draws Elo CLI PGN
87596 ncm-et-10 2019-07-03 15:50 00:24:17 1963080 151 11 51 89 -94.28 +/- 34.91
87595 ncm-et-9 2019-07-03 15:50 00:25:05 1967848 166 11 52 103 -87.62 +/- 31.87
87594 ncm-et-3 2019-07-03 15:49 00:25:38 1964609 163 11 65 87 -119.61 +/- 35.94
87593 ncm-et-13 2019-07-03 15:49 00:25:59 1981915 170 13 55 102 -87.65 +/- 32.5
87592 ncm-et-4 2019-07-03 15:48 00:26:48 1972231 176 12 57 107 -90.85 +/- 31.49
87591 ncm-et-15 2019-07-03 15:48 00:26:44 1969711 174 10 57 107 -96.24 +/- 31.21
87590 ncm-et-10 2019-07-03 14:32 01:16:59 1975214 500 44 169 287 -88.74 +/- 19.61
87589 ncm-et-15 2019-07-03 14:32 01:14:52 1989801 500 33 188 279 -111.37 +/- 19.84
87588 ncm-et-13 2019-07-03 14:32 01:15:56 1979200 500 29 176 295 -105.25 +/- 18.99
87587 ncm-et-9 2019-07-03 14:31 01:17:16 1969396 500 28 183 289 -111.37 +/- 19.26
87586 ncm-et-4 2019-07-03 14:31 01:15:37 1968383 500 29 170 301 -100.71 +/- 18.69
87585 ncm-et-3 2019-07-03 14:31 01:16:54 1964917 500 39 177 284 -98.44 +/- 19.69
87584 ncm-et-10 2019-07-03 13:16 01:15:18 1967707 500 27 171 302 -102.97 +/- 18.6
87583 ncm-et-13 2019-07-03 13:15 01:15:31 1971113 500 30 155 315 -88.74 +/- 18.03
87582 ncm-et-15 2019-07-03 13:14 01:16:51 1981886 500 34 183 283 -106.78 +/- 19.66
87581 ncm-et-9 2019-07-03 13:14 01:16:21 1984083 500 40 159 301 -84.31 +/- 18.88
87580 ncm-et-4 2019-07-03 13:13 01:16:39 1952309 500 32 168 300 -96.94 +/- 18.8
87579 ncm-et-3 2019-07-03 13:12 01:18:01 1973667 500 41 169 290 -90.97 +/- 19.43
87578 ncm-et-4 2019-07-03 11:57 01:15:00 1966468 500 31 163 306 -93.95 +/- 18.49
87577 ncm-et-3 2019-07-03 11:57 01:13:43 1975734 500 32 181 287 -106.78 +/- 19.43
87576 ncm-et-13 2019-07-03 11:57 01:16:46 1974365 500 29 170 301 -100.71 +/- 18.69
87575 ncm-et-10 2019-07-03 11:57 01:17:43 1962768 500 30 161 309 -93.2 +/- 18.32
87574 ncm-et-15 2019-07-03 11:57 01:15:52 1978256 500 31 160 309 -91.71 +/- 18.34
87573 ncm-et-9 2019-07-03 11:57 01:15:51 1968630 500 37 199 264 -116.78 +/- 20.65
21648 ncm-et-12 2018-04-22 02:09 00:27:51 2020836 182 17 66 99 -95.9 +/- 33.83
21647 ncm-et-8 2018-04-22 02:06 00:30:52 2019701 202 10 66 126 -98.91 +/- 28.48
21646 ncm-et-16 2018-04-22 02:05 00:31:37 2020512 206 13 83 110 -122.95 +/- 31.88
21645 ncm-et-9 2018-04-22 02:04 00:32:15 2028195 212 18 75 119 -95.77 +/- 30.61
21644 ncm-et-5 2018-04-22 02:04 00:32:39 2023772 212 13 71 128 -97.54 +/- 28.74
21643 ncm-et-11 2018-04-22 02:04 00:32:35 2014664 217 13 76 128 -103.85 +/- 28.9
21642 ncm-et-7 2018-04-22 02:04 00:32:41 1993948 211 17 76 118 -99.81 +/- 30.72
21641 ncm-et-10 2018-04-22 02:04 00:33:03 2020022 221 17 75 129 -93.37 +/- 29.08
21640 ncm-et-14 2018-04-22 02:04 00:33:02 2019538 215 19 78 118 -97.85 +/- 30.89
21639 ncm-et-4 2018-04-22 02:03 00:33:10 2023121 218 9 74 135 -106.84 +/- 27.42
21638 ncm-et-13 2018-04-22 02:03 00:33:34 2025081 222 16 72 134 -89.57 +/- 28.23
21637 ncm-et-6 2018-04-22 02:02 00:34:24 2027210 229 17 79 133 -96.47 +/- 28.64
21636 ncm-et-15 2018-04-22 02:02 00:34:31 2015638 225 19 77 129 -91.63 +/- 29.27
21635 ncm-et-3 2018-04-22 02:02 00:34:51 2023446 228 13 81 134 -106.87 +/- 28.23
21634 ncm-et-8 2018-04-22 00:48 01:17:02 1978221 500 28 164 308 -96.94 +/- 18.33
21633 ncm-et-5 2018-04-22 00:47 01:15:28 2022630 500 36 161 303 -88.74 +/- 18.72
21632 ncm-et-10 2018-04-22 00:47 01:15:08 2021814 500 31 159 310 -90.97 +/- 18.3
21631 ncm-et-4 2018-04-22 00:47 01:14:47 2025245 500 38 176 286 -98.44 +/- 19.58
21630 ncm-et-12 2018-04-22 00:47 01:20:22 1857910 500 27 184 289 -112.91 +/- 19.24
21629 ncm-et-16 2018-04-22 00:47 01:16:40 1985928 500 44 161 295 -82.83 +/- 19.23
21628 ncm-et-15 2018-04-22 00:47 01:13:47 2020673 500 34 166 300 -93.95 +/- 18.84
21627 ncm-et-14 2018-04-22 00:47 01:15:15 2017908 500 26 182 292 -112.14 +/- 19.07
21626 ncm-et-7 2018-04-22 00:47 01:15:26 2022303 500 24 157 319 -94.7 +/- 17.7
21625 ncm-et-13 2018-04-22 00:47 01:14:43 2023772 500 25 190 285 -119.11 +/- 19.39
21624 ncm-et-11 2018-04-22 00:47 01:15:30 2020022 500 27 182 291 -111.37 +/- 19.14
21623 ncm-et-3 2018-04-22 00:47 01:13:30 2027210 500 24 180 296 -112.14 +/- 18.83
21622 ncm-et-9 2018-04-22 00:47 01:15:51 2024428 500 29 190 281 -116.0 +/- 19.67
21621 ncm-et-6 2018-04-22 00:47 01:13:58 2024590 500 27 188 285 -116.0 +/- 19.44

Commit

Commit ID 8b88ca901779524ef17cb26c1278cb54ea3be6ae
Author lucasart
Date 2014-08-05 03:42:48 UTC
Reduce minimum memory requirement by 16MB Stockfish allocates the default hash (32MB) in main(), before entering UCI::loop(). If there is not enough memory, the program will crash even before UCI::loop() is entered and the GUI is given a change to specify a lower Hash value. This defective design could be resolved by doing a lazy allocation upon "isready" command, as the UCI protocol guarantees that "isready" will be sent at least once before any search. But it's a bit cumbersome when using Stockfish "manually" to have to remember to type "isready" everytime. So leave the current design, but reduce the default hash to 16MB instread of 32MB. In order to perform such quick searches (depth=13), there is no reason to use so much Hash anyway. Another benefit is to introduce a bit of hash pressure in bench, which increases chances to detect rare bugs related to TT replacement, for example. This is not a functional change, although it obviously changes the bench. bench 7461879
Copyright 2011–2024 Next Chess Move LLC