Dev Builds » 20120827-1717

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 08:45:52 1976465 3668 67 2487 1114 -275.3 +/- 10.18
ncm-et-4 08:49:19 1978756 3690 57 2547 1086 -284.72 +/- 10.31
ncm-et-9 08:46:36 1979030 3659 66 2498 1095 -278.34 +/- 10.27
ncm-et-10 08:45:40 1965343 3642 65 2496 1081 -280.1 +/- 10.34
ncm-et-13 03:58:51 1989007 1670 36 1168 466 -286.67 +/- 15.79
ncm-et-15 08:48:44 1976461 3671 62 2508 1101 -279.36 +/- 10.24
20000 353 13704 5943 -280.14 +/- 4.41

Test Detail

ID Host Started (UTC) Duration Base NPS Games Wins Losses Draws Elo CLI PGN
118291 ncm-et-4 2020-03-10 10:34 01:12:00 1963380 500 7 339 154 -277.93 +/- 27.43
118290 ncm-et-9 2020-03-10 10:33 01:11:20 1971875 500 9 353 138 -293.29 +/- 29.11
118289 ncm-et-15 2020-03-10 10:33 01:13:06 1971566 500 10 320 170 -251.89 +/- 26.08
118288 ncm-et-3 2020-03-10 10:31 01:12:00 1963378 500 12 343 145 -276.69 +/- 28.41
118287 ncm-et-10 2020-03-10 10:31 01:11:35 1962936 500 9 323 168 -256.44 +/- 26.22
118286 ncm-et-9 2020-03-10 09:21 01:10:29 1970634 500 9 351 140 -290.66 +/- 28.9
118285 ncm-et-4 2020-03-10 09:21 01:12:54 1966459 500 6 326 168 -263.42 +/- 26.13
118284 ncm-et-15 2020-03-10 09:19 01:12:07 1969086 500 4 359 137 -308.24 +/- 29.1
118283 ncm-et-10 2020-03-10 09:19 01:11:12 1968626 500 11 339 150 -273.0 +/- 27.9
118282 ncm-et-3 2020-03-10 09:18 01:12:13 1962149 500 9 329 162 -263.42 +/- 26.74
118281 ncm-et-9 2020-03-10 08:07 01:12:48 1963995 500 12 343 145 -276.69 +/- 28.41
118280 ncm-et-15 2020-03-10 08:06 01:11:53 1972830 500 7 339 154 -277.93 +/- 27.43
118279 ncm-et-4 2020-03-10 08:06 01:13:07 1973747 500 10 368 122 -312.48 +/- 31.03
118278 ncm-et-10 2020-03-10 08:06 01:12:03 1960790 500 4 338 158 -280.42 +/- 26.94
118277 ncm-et-3 2020-03-10 08:05 01:11:35 1961534 500 10 334 156 -268.17 +/- 27.31
118276 ncm-et-3 2020-03-10 06:53 01:11:11 1961999 500 9 346 145 -284.22 +/- 28.37
118275 ncm-et-10 2020-03-10 06:53 01:11:59 1967098 500 9 357 134 -298.62 +/- 29.56
118274 ncm-et-9 2020-03-10 06:53 01:13:45 1970501 500 14 330 156 -258.75 +/- 27.36
118273 ncm-et-4 2020-03-10 06:53 01:12:40 1965537 500 8 340 152 -277.93 +/- 27.65
118272 ncm-et-15 2020-03-10 06:53 01:12:46 1972204 500 12 334 154 -265.78 +/- 27.53
45451 ncm-et-10 2018-07-11 22:51 00:20:23 1946940 142 3 99 40 -285.53 +/- 54.86
45450 ncm-et-9 2018-07-11 22:49 00:22:28 1990981 159 1 101 57 -256.98 +/- 44.85
45449 ncm-et-3 2018-07-11 22:47 00:23:47 1991455 168 3 121 44 -302.96 +/- 52.26
45448 ncm-et-15 2018-07-11 22:47 00:24:09 1981068 171 2 127 42 -323.41 +/- 53.54
45447 ncm-et-13 2018-07-11 22:46 00:24:49 1989717 170 1 125 44 -322.24 +/- 52.03
45446 ncm-et-4 2018-07-11 22:45 00:26:30 1990982 190 4 132 54 -284.01 +/- 46.96
45445 ncm-et-10 2018-07-11 21:37 01:12:19 1982010 500 7 348 145 -289.36 +/- 28.32
45444 ncm-et-9 2018-07-11 21:36 01:11:44 1988769 500 8 339 153 -276.69 +/- 27.55
45443 ncm-et-15 2018-07-11 21:35 01:11:24 1982792 500 7 345 148 -285.49 +/- 28.02
45442 ncm-et-13 2018-07-11 21:35 01:10:46 1989401 500 8 373 119 -322.67 +/- 31.43
45441 ncm-et-3 2018-07-11 21:34 01:12:20 1988296 500 10 337 153 -271.78 +/- 27.59
45440 ncm-et-4 2018-07-11 21:33 01:10:56 1989875 500 5 333 162 -273.0 +/- 26.62
45439 ncm-et-10 2018-07-11 20:26 01:10:43 1980914 500 11 343 146 -277.93 +/- 28.3
45438 ncm-et-3 2018-07-11 20:22 01:10:35 1991299 500 7 345 148 -285.49 +/- 28.02
45437 ncm-et-9 2018-07-11 20:22 01:12:24 1987193 500 4 333 163 -274.22 +/- 26.48
45436 ncm-et-15 2018-07-11 20:22 01:11:33 1979981 500 8 344 148 -282.94 +/- 28.04
45435 ncm-et-13 2018-07-11 20:22 01:11:48 1987193 500 12 338 150 -270.57 +/- 27.91
45434 ncm-et-4 2018-07-11 20:20 01:11:03 1989719 500 10 358 132 -298.62 +/- 29.8
45433 ncm-et-9 2018-07-11 19:10 01:11:38 1988296 500 9 348 143 -286.77 +/- 28.58
45432 ncm-et-13 2018-07-11 19:09 01:11:28 1989717 500 15 332 153 -259.91 +/- 27.65
45431 ncm-et-10 2018-07-11 19:09 01:15:26 1953431 500 11 349 140 -285.49 +/- 28.92
45430 ncm-et-15 2018-07-11 19:09 01:11:46 1982164 500 12 340 148 -273.0 +/- 28.11
45429 ncm-et-3 2018-07-11 19:09 01:12:11 1991614 500 7 332 161 -269.37 +/- 26.77
45428 ncm-et-4 2018-07-11 19:09 01:10:09 1990349 500 7 351 142 -293.29 +/- 28.64

Commit

Commit ID 3df2c01b5769c7ae996fb5b992c06e4a5428ad35
Author Marco Costalba
Date 2012-08-27 17:17:02 UTC
Correctly handle handover of setup states Before the search we setup the starting position doing all the moves (sent by GUI) from start position to the position just before to start searching. To do this we use a set of StateInfo records used by each do_move() call. These records shall be kept valid during all the search because repetition draw detection uses them to back track all the earlier positions keys. The problem is that, while searching, the GUI could send another 'position' command, this calls set_position() that clears the states! Of course a crash follows shortly. Before searching all the relevant parameters are copied in start_searching() just for this reason: to fully detach data accessed during the search from the UCI protocol handling. So the natural solution would be to copy also the setup states. Unfortunatly this approach does not work because StateInfo contains a pointer to the previous record, so naively copying and then freeing the original memory leads to a crash. That's why we use two std::auto_ptr (one belonging to UCI and another to Search) to safely transfer ownership of the StateInfo records to the search, after we have setup the root position. As a nice side-effect all the possible memory leaks are magically sorted out for us by std::auto_ptr semantic. No functional change. Signed-off-by: Marco Costalba <mcostalba@gmail.com>
Copyright 2011–2024 Next Chess Move LLC