Dev Builds » 20140216-1051

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 05:57:11 1991021 2391 154 946 1291 -119.6 +/- 9.26
ncm-et-4 05:57:09 1980838 2353 150 954 1249 -123.69 +/- 9.44
ncm-et-5 01:46:13 2023774 716 51 328 337 -141.79 +/- 18.47
ncm-et-6 01:46:10 2027375 713 40 300 373 -132.81 +/- 17.27
ncm-et-7 01:46:01 2007086 695 36 312 347 -146.0 +/- 17.98
ncm-et-8 01:45:51 2023773 713 46 267 400 -111.35 +/- 16.54
ncm-et-9 05:57:34 1990075 2400 135 1002 1263 -131.44 +/- 9.37
ncm-et-10 05:57:18 1982502 2372 149 953 1270 -122.61 +/- 9.34
ncm-et-11 01:46:22 2004218 705 51 288 366 -121.52 +/- 17.56
ncm-et-12 01:46:11 2005331 722 41 318 363 -140.48 +/- 17.6
ncm-et-13 05:57:07 1986961 2389 158 928 1303 -116.12 +/- 9.21
ncm-et-14 01:46:13 2020347 719 44 291 384 -124.41 +/- 17.0
ncm-et-15 05:57:15 1988059 2402 128 1034 1240 -137.85 +/- 9.47
ncm-et-16 01:45:50 1995579 710 34 264 412 -116.75 +/- 16.03
20000 1217 8185 10598 -126.34 +/- 3.24

Test Detail

ID Host Started (UTC) Duration Base NPS Games Wins Losses Draws Elo CLI PGN
95046 ncm-et-4 2019-09-03 20:03 00:23:28 1964305 150 13 58 79 -107.54 +/- 38.02
95045 ncm-et-9 2019-09-03 20:02 00:25:04 1976232 168 7 60 101 -113.48 +/- 32.04
95044 ncm-et-10 2019-09-03 20:02 00:25:09 1961540 160 11 68 81 -129.45 +/- 37.55
95043 ncm-et-3 2019-09-03 20:01 00:25:30 1973586 173 9 67 97 -121.17 +/- 33.49
95042 ncm-et-13 2019-09-03 20:00 00:26:41 1965075 174 11 68 95 -118.17 +/- 34.2
95041 ncm-et-15 2019-09-03 19:59 00:27:34 1988286 175 4 70 101 -137.84 +/- 31.92
95040 ncm-et-4 2019-09-03 18:48 01:14:33 1971473 500 35 201 264 -119.89 +/- 20.62
95039 ncm-et-3 2019-09-03 18:45 01:15:24 1974996 500 27 182 291 -111.37 +/- 19.14
95038 ncm-et-10 2019-09-03 18:44 01:16:21 1968004 500 29 195 276 -119.89 +/- 19.92
95037 ncm-et-9 2019-09-03 18:44 01:16:49 1969861 500 30 208 262 -129.35 +/- 20.64
95036 ncm-et-13 2019-09-03 18:43 01:15:58 1969728 500 35 188 277 -109.83 +/- 19.97
95035 ncm-et-15 2019-09-03 18:43 01:15:23 1969256 500 32 206 262 -126.18 +/- 20.68
95034 ncm-et-4 2019-09-03 17:31 01:15:57 1963378 500 38 210 252 -124.6 +/- 21.26
95033 ncm-et-13 2019-09-03 17:29 01:13:27 1966613 500 35 184 281 -106.78 +/- 19.78
95032 ncm-et-10 2019-09-03 17:28 01:14:53 1960008 500 33 191 276 -113.68 +/- 19.99
95031 ncm-et-3 2019-09-03 17:28 01:15:19 1974670 500 41 192 267 -108.3 +/- 20.55
95030 ncm-et-9 2019-09-03 17:28 01:15:29 1963849 500 30 218 252 -137.37 +/- 21.16
95029 ncm-et-15 2019-09-03 17:27 01:15:07 1964457 500 23 233 244 -155.54 +/- 21.46
95028 ncm-et-10 2019-09-03 16:13 01:14:41 1964931 500 31 199 270 -121.46 +/- 20.26
95027 ncm-et-4 2019-09-03 16:13 01:17:06 1948568 500 28 207 265 -130.15 +/- 20.46
95026 ncm-et-9 2019-09-03 16:13 01:13:55 1980184 500 33 195 272 -116.78 +/- 20.19
95025 ncm-et-3 2019-09-03 16:13 01:14:40 1971569 500 32 210 258 -129.35 +/- 20.88
95024 ncm-et-15 2019-09-03 16:13 01:13:12 1969244 500 32 212 256 -130.94 +/- 20.98
95023 ncm-et-13 2019-09-03 16:12 01:15:12 1976396 500 33 197 270 -118.33 +/- 20.29
26828 ncm-et-4 2018-05-06 08:53 00:30:06 2026226 203 7 85 111 -140.72 +/- 31.12
26827 ncm-et-7 2018-05-06 08:52 00:30:36 1995922 195 9 87 99 -147.19 +/- 33.62
26826 ncm-et-11 2018-05-06 08:52 00:30:58 2018234 205 13 89 103 -135.25 +/- 33.23
26825 ncm-et-10 2018-05-06 08:52 00:31:09 2020185 212 14 93 105 -136.02 +/- 32.99
26824 ncm-et-8 2018-05-06 08:52 00:30:58 2023775 213 16 83 114 -113.12 +/- 31.43
26823 ncm-et-16 2018-05-06 08:51 00:31:55 2017102 210 6 77 127 -122.28 +/- 28.2
26822 ncm-et-14 2018-05-06 08:51 00:31:58 2019533 219 17 91 111 -122.2 +/- 32.1
26821 ncm-et-12 2018-05-06 08:50 00:32:21 2018557 222 11 99 112 -145.7 +/- 31.67
26820 ncm-et-6 2018-05-06 08:50 00:32:40 2028032 213 8 89 116 -139.11 +/- 30.52
26819 ncm-et-3 2018-05-06 08:50 00:32:48 2025735 218 12 101 105 -150.62 +/- 32.97
26818 ncm-et-5 2018-05-06 08:50 00:33:05 2023610 216 20 102 94 -138.84 +/- 35.29
26817 ncm-et-15 2018-05-06 08:49 00:33:20 2019533 227 17 96 114 -126.18 +/- 31.68
26816 ncm-et-9 2018-05-06 08:49 00:33:26 2024590 232 9 105 118 -152.93 +/- 30.64
26815 ncm-et-13 2018-05-06 08:49 00:33:19 2021489 215 15 85 115 -117.39 +/- 31.24
26814 ncm-et-16 2018-05-06 07:36 01:13:55 1974057 500 28 187 285 -114.45 +/- 19.46
26813 ncm-et-15 2018-05-06 07:36 01:12:39 2017583 500 20 217 263 -144.72 +/- 20.39
26812 ncm-et-13 2018-05-06 07:36 01:12:30 2022466 500 29 206 265 -128.55 +/- 20.47
26811 ncm-et-6 2018-05-06 07:36 01:13:30 2026718 500 32 211 257 -130.15 +/- 20.93
26810 ncm-et-8 2018-05-06 07:36 01:14:53 2023772 500 30 184 286 -110.6 +/- 19.45
26809 ncm-et-5 2018-05-06 07:36 01:13:08 2023939 500 31 226 243 -143.07 +/- 21.65
26808 ncm-et-7 2018-05-06 07:36 01:15:25 2018251 500 27 225 248 -145.54 +/- 21.32
26807 ncm-et-4 2018-05-06 07:36 01:15:59 2011078 500 29 193 278 -118.33 +/- 19.82
26806 ncm-et-12 2018-05-06 07:36 01:13:50 1992105 500 30 219 251 -138.18 +/- 21.21
26805 ncm-et-10 2018-05-06 07:36 01:15:05 2020347 500 31 207 262 -127.76 +/- 20.66
26804 ncm-et-14 2018-05-06 07:36 01:14:15 2021162 500 27 200 273 -125.39 +/- 20.03
26803 ncm-et-11 2018-05-06 07:36 01:15:24 1990202 500 38 199 263 -116.0 +/- 20.71
26802 ncm-et-3 2018-05-06 07:36 01:13:30 2025573 500 33 194 273 -116.0 +/- 20.14
26801 ncm-et-9 2018-05-06 07:35 01:12:51 2025736 500 26 216 258 -138.99 +/- 20.78

Commit

Commit ID d91079d4b02b410702dda082d5c489847f067a35
Author Marco Costalba
Date 2014-02-16 10:51:30 UTC
Fix material key for King Currently king has no material key associated because it can never happen to find a legal position without both kings, so there is no need to keep track of it. The consequence is that a position with only the two kings has material key set at zero and if the material hash table is empty any entry will match and this is wrong. Normally bug is hidden becuase the checking for a draw with pos.is_draw() is done earlier than evaluate() call, so that we never check in gameplay the material key of a position with two kings. Nevertheless the bug is there and can be reproduced setting at startup a position with only two kings and typing 'eval' from prompt. The fix is very simple: add a random key also for the king. Also fixed the condition in material.cpp to avoid asserting when a 'just 2 kings' postion is evaluated. No functional change.
Copyright 2011–2024 Next Chess Move LLC