Dev Builds » 20141103-1635

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 06:00:39 1989374 2367 185 736 1446 -82.39 +/- 8.57
ncm-et-4 06:01:26 1979407 2391 168 702 1521 -78.93 +/- 8.21
ncm-et-5 01:47:54 2021653 699 47 238 414 -97.41 +/- 16.08
ncm-et-6 01:48:01 2026227 718 55 211 452 -76.71 +/- 15.18
ncm-et-7 01:47:52 2009676 705 38 211 456 -87.03 +/- 14.78
ncm-et-8 01:47:58 1883578 668 43 203 422 -84.87 +/- 15.6
ncm-et-9 06:01:26 1990345 2396 162 703 1531 -79.82 +/- 8.16
ncm-et-10 06:01:00 1984313 2369 155 770 1444 -92.31 +/- 8.53
ncm-et-11 01:47:48 2017909 712 50 224 438 -86.66 +/- 15.49
ncm-et-12 01:47:49 2018477 717 51 234 432 -90.68 +/- 15.7
ncm-et-13 06:01:28 1989600 2387 150 735 1502 -86.92 +/- 8.27
ncm-et-14 01:48:01 2018639 733 46 228 459 -88.11 +/- 14.99
ncm-et-15 06:01:18 1992802 2411 153 738 1520 -86.02 +/- 8.22
ncm-et-16 01:48:05 2020185 727 58 230 439 -83.79 +/- 15.62
20000 1361 6163 12476 -85.08 +/- 2.88

Test Detail

ID Host Started (UTC) Duration Base NPS Games Wins Losses Draws Elo CLI PGN
85190 ncm-et-10 2019-06-14 08:25 00:23:25 1960629 149 10 54 85 -105.75 +/- 35.92
85189 ncm-et-3 2019-06-14 08:24 00:23:56 1968003 152 17 55 80 -88.74 +/- 37.96
85188 ncm-et-13 2019-06-14 08:24 00:24:38 1975152 163 14 40 109 -55.9 +/- 30.35
85187 ncm-et-9 2019-06-14 08:22 00:26:37 1974202 171 13 58 100 -93.63 +/- 33.03
85186 ncm-et-4 2019-06-14 08:21 00:27:06 1966927 172 16 50 106 -69.6 +/- 31.82
85185 ncm-et-15 2019-06-14 08:20 00:28:31 1983294 193 8 60 125 -95.98 +/- 28.03
85184 ncm-et-10 2019-06-14 07:08 01:15:40 1971412 500 32 155 313 -87.26 +/- 18.17
85183 ncm-et-3 2019-06-14 07:07 01:15:52 1972701 500 36 133 331 -68.27 +/- 17.36
85182 ncm-et-13 2019-06-14 07:06 01:16:34 1968783 500 23 165 312 -101.46 +/- 18.02
85181 ncm-et-9 2019-06-14 07:06 01:14:43 1969565 500 33 141 326 -76.25 +/- 17.55
85180 ncm-et-4 2019-06-14 07:05 01:14:52 1961078 500 44 142 314 -68.99 +/- 18.31
85179 ncm-et-15 2019-06-14 07:04 01:14:40 1982839 500 38 144 318 -74.79 +/- 18.03
85178 ncm-et-10 2019-06-14 05:50 01:17:01 1964777 500 38 169 293 -93.2 +/- 19.24
85177 ncm-et-3 2019-06-14 05:49 01:16:42 1976171 500 39 161 300 -86.52 +/- 18.92
85176 ncm-et-4 2019-06-14 05:49 01:14:54 1964455 500 28 160 312 -93.95 +/- 18.14
85175 ncm-et-13 2019-06-14 05:49 01:16:22 1974370 500 36 153 311 -82.83 +/- 18.34
85174 ncm-et-15 2019-06-14 05:47 01:15:53 1976382 500 34 169 297 -96.19 +/- 18.98
85173 ncm-et-9 2019-06-14 05:47 01:17:51 1973270 500 29 144 327 -81.37 +/- 17.42
85172 ncm-et-15 2019-06-14 04:32 01:14:20 1979296 500 35 139 326 -73.34 +/- 17.59
85171 ncm-et-10 2019-06-14 04:32 01:17:11 1970483 500 32 163 305 -93.2 +/- 18.56
85170 ncm-et-3 2019-06-14 04:32 01:16:40 1967079 500 45 159 296 -80.63 +/- 19.19
85169 ncm-et-4 2019-06-14 04:32 01:16:36 1963701 500 34 148 318 -80.63 +/- 17.96
85168 ncm-et-13 2019-06-14 04:32 01:15:49 1971423 500 30 147 323 -82.83 +/- 17.64
85167 ncm-et-9 2019-06-14 04:32 01:14:14 1971108 500 34 135 331 -71.16 +/- 17.32
20024 ncm-et-8 2018-04-17 15:46 00:25:20 1977609 168 11 40 117 -60.58 +/- 28.4
20023 ncm-et-5 2018-04-17 15:41 00:30:33 2023120 199 16 66 117 -89.21 +/- 30.53
20022 ncm-et-7 2018-04-17 15:40 00:31:24 1996885 205 14 56 135 -72.2 +/- 27.23
20021 ncm-et-6 2018-04-17 15:39 00:32:21 2026228 218 15 63 140 -77.77 +/- 27.02
20020 ncm-et-11 2018-04-17 15:38 00:32:36 2019534 212 17 68 127 -85.25 +/- 29.16
20019 ncm-et-3 2018-04-17 15:38 00:32:10 2027046 215 8 66 141 -96.11 +/- 26.15
20018 ncm-et-10 2018-04-17 15:38 00:32:23 2018070 220 14 70 136 -90.43 +/- 27.71
20017 ncm-et-16 2018-04-17 15:38 00:32:48 2019861 227 14 67 146 -82.64 +/- 26.34
20016 ncm-et-12 2018-04-17 15:38 00:33:03 2018559 217 13 82 122 -114.44 +/- 29.94
20015 ncm-et-9 2018-04-17 15:38 00:33:36 2027046 225 20 67 138 -73.66 +/- 27.87
20014 ncm-et-4 2018-04-17 15:37 00:33:33 1996020 219 15 62 142 -75.74 +/- 26.72
20013 ncm-et-13 2018-04-17 15:37 00:33:48 2023937 224 18 72 134 -85.44 +/- 28.39
20012 ncm-et-15 2018-04-17 15:37 00:34:00 2018070 218 11 73 134 -101.61 +/- 27.76
20011 ncm-et-14 2018-04-17 15:37 00:34:22 2019371 233 18 64 151 -69.51 +/- 26.02
20010 ncm-et-11 2018-04-17 14:22 01:15:12 2016284 500 33 156 311 -87.26 +/- 18.28
20009 ncm-et-7 2018-04-17 14:22 01:16:28 2022467 500 24 155 321 -93.2 +/- 17.61
20008 ncm-et-15 2018-04-17 14:22 01:13:54 2016933 500 27 153 320 -89.48 +/- 17.73
20007 ncm-et-5 2018-04-17 14:22 01:17:21 2020186 500 31 172 297 -100.71 +/- 18.93
20006 ncm-et-12 2018-04-17 14:22 01:14:46 2018396 500 38 152 310 -80.63 +/- 18.42
20005 ncm-et-3 2018-04-17 14:22 01:15:19 2025245 500 40 162 298 -86.52 +/- 19.03
20004 ncm-et-8 2018-04-17 14:22 01:22:38 1789548 500 32 163 305 -93.2 +/- 18.56
20003 ncm-et-9 2018-04-17 14:22 01:14:25 2026882 500 33 158 309 -88.74 +/- 18.38
20002 ncm-et-6 2018-04-17 14:22 01:15:40 2026226 500 40 148 312 -76.25 +/- 18.35
20001 ncm-et-16 2018-04-17 14:22 01:15:17 2020510 500 44 163 293 -84.31 +/- 19.32
20000 ncm-et-13 2018-04-17 14:22 01:14:17 2023938 500 29 158 313 -91.71 +/- 18.11
19999 ncm-et-10 2018-04-17 14:22 01:15:20 2020510 500 29 159 312 -92.46 +/- 18.16
19998 ncm-et-14 2018-04-17 14:22 01:13:39 2017908 500 28 164 308 -96.94 +/- 18.33
19997 ncm-et-4 2018-04-17 14:22 01:14:25 2024263 500 31 140 329 -76.98 +/- 17.37

Commit

Commit ID d12378497cb24f40d3510cdcfaecd1335f523196
Author lucasart
Date 2014-11-03 16:35:02 UTC
Do not assume that enum are signed Clang 3.5 issues warning on constructs like: abs(f1 - f2). The thing is that f1 and f2 are enum types, and the range given (all positive) allows the compiler to choose an unsigned type (efficiency being one reason to prefer unsigned arithmetic). If f1 < f2 are unsigned, then f1 - f2 wraps around zero and the abs() becomes a no-op. It's the reinterpretation of the unsigned result (large value) as a signed int that happens to give the correct result, thanks to 2's complement. This is all tricky and dangerous! In the spirit of the standard, we assume nothing on the signedness of enums, and simply calculate the rank and file distances as: - rank_dist(r1, r2) = r1 < r2 ? r2 - r1 : r1 - r2 - file_dist(f1, f2) = f1 < f2 ? f2 - f1 : f1 - f2 this logic can in fact be applied to any enum we may use, so for better generality and to avoid code duplication, we use a template function diff() here. No functional change. Resolves #95
Copyright 2011–2024 Next Chess Move LLC