Dev Builds » 20140204-0724

You are viewing an old NCM Stockfish dev build test. You may find the most recent dev build tests using Stockfish 15 as the baseline here.

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 06:36:29 1989019 2651 147 1140 1364 -136.8 ± 9.04
ncm-et-4 06:37:13 1966694 2672 163 1122 1387 -130.51 ± 8.98
ncm-et-5 01:45:35 2024181 717 35 291 391 -129.76 ± 16.69
ncm-et-6 01:45:39 2025326 730 41 336 353 -148.89 ± 17.92
ncm-et-7 01:45:32 2010438 720 47 303 370 -129.17 ± 17.44
ncm-et-8 01:45:33 1944134 669 39 298 332 -141.91 ± 18.45
ncm-et-9 02:22:57 2003361 955 55 417 483 -138.61 ± 15.25
ncm-et-10 06:37:00 1980290 2679 160 1129 1390 -131.62 ± 8.96
ncm-et-11 01:45:37 2015323 718 35 314 369 -142.49 ± 17.35
ncm-et-12 01:45:30 2010868 704 33 285 386 -130.13 ± 16.76
ncm-et-13 06:37:17 1987168 2669 154 1118 1397 -131.42 ± 8.92
ncm-et-14 01:45:15 2019208 714 40 297 377 -130.92 ± 17.15
ncm-et-15 06:37:19 1982316 2677 155 1132 1390 -132.93 ± 8.95
ncm-et-16 01:45:37 2019376 725 46 290 389 -121.67 ± 16.9
20000 1150 8472 10378 -133.38 ± 3.28

Test Detail

ID Host Started (UTC) Duration Base NPS Games Wins Losses Draws Elo CLI PGN
96037 ncm-et-9 2019-09-14 03:11 00:37:16 1958929 240 14 99 127 -128.62 ± 29.64
96036 ncm-et-3 2019-09-14 03:11 00:37:28 1976694 249 21 109 119 -128.32 ± 31.19
96035 ncm-et-10 2019-09-14 03:11 00:37:24 1965704 253 14 105 134 -130.82 ± 28.81
96034 ncm-et-15 2019-09-14 03:11 00:37:32 1964455 254 12 112 130 -144.59 ± 29.29
96033 ncm-et-13 2019-09-14 03:11 00:37:33 1966003 245 17 101 127 -124.15 ± 29.85
96032 ncm-et-4 2019-09-14 03:11 00:37:32 1966613 249 18 103 128 -123.56 ± 29.79
95489 ncm-et-3 2019-09-09 11:01 00:29:35 1978101 192 7 90 95 -160.76 ± 34.27
95488 ncm-et-10 2019-09-09 11:01 00:29:43 1960921 200 11 82 107 -128.95 ± 32.2
95487 ncm-et-13 2019-09-09 11:00 00:30:34 1973273 204 10 89 105 -141.95 ± 32.62
95486 ncm-et-15 2019-09-09 11:00 00:30:56 1964147 205 9 83 113 -131.33 ± 30.98
95485 ncm-et-4 2019-09-09 10:59 00:31:25 1972043 209 15 87 107 -124.79 ± 32.61
95484 ncm-et-10 2019-09-09 09:45 01:14:46 1958638 500 31 195 274 -118.33 ± 20.06
95483 ncm-et-15 2019-09-09 09:45 01:14:01 1967088 500 43 210 247 -120.67 ± 21.56
95482 ncm-et-4 2019-09-09 09:45 01:13:37 1974050 500 30 201 269 -123.81 ± 20.29
95481 ncm-et-3 2019-09-09 09:44 01:15:51 1972032 500 29 206 265 -128.55 ± 20.47
95480 ncm-et-13 2019-09-09 09:44 01:15:30 1978200 500 30 225 245 -143.07 ± 21.53
95479 ncm-et-4 2019-09-09 08:31 01:12:51 1961891 500 24 221 255 -144.72 ± 20.89
95478 ncm-et-15 2019-09-09 08:30 01:13:52 1976179 500 28 204 268 -127.76 ± 20.3
95477 ncm-et-10 2019-09-09 08:29 01:15:01 1959699 500 23 213 264 -138.99 ± 20.41
95476 ncm-et-13 2019-09-09 08:29 01:14:00 1968967 500 25 196 279 -123.81 ± 19.69
95475 ncm-et-3 2019-09-09 08:28 01:14:57 1970796 500 21 206 273 -134.95 ± 19.9
95474 ncm-et-3 2019-09-09 07:14 01:13:27 1974527 500 27 218 255 -139.81 ± 20.95
95473 ncm-et-10 2019-09-09 07:14 01:14:32 1978331 500 38 216 246 -129.35 ± 21.57
95472 ncm-et-15 2019-09-09 07:14 01:15:18 1967716 500 24 211 265 -136.56 ± 20.38
95471 ncm-et-4 2019-09-09 07:14 01:16:08 1915337 500 31 203 266 -124.6 ± 20.46
95470 ncm-et-13 2019-09-09 07:14 01:14:04 1973898 500 22 216 262 -142.26 ± 20.49
27220 ncm-et-8 2018-05-07 10:07 00:26:59 1968985 169 10 67 92 -121.96 ± 34.71
27219 ncm-et-12 2018-05-07 10:03 00:30:25 2020023 204 10 82 112 -128.13 ± 31.23
27218 ncm-et-4 2018-05-07 10:03 00:30:52 1992894 214 4 89 121 -146.03 ± 29.18
27217 ncm-et-3 2018-05-07 10:03 00:31:05 2025903 210 13 78 119 -111.19 ± 30.29
27216 ncm-et-14 2018-05-07 10:02 00:31:41 2018395 214 11 85 118 -125.31 ± 30.43
27215 ncm-et-15 2018-05-07 10:02 00:31:57 2016285 218 10 81 127 -117.43 ± 28.88
27214 ncm-et-9 2018-05-07 10:02 00:32:02 2023282 215 14 88 113 -124.67 ± 31.55
27213 ncm-et-13 2018-05-07 10:02 00:32:01 2024427 220 14 97 109 -137.89 ± 32.35
27212 ncm-et-11 2018-05-07 10:01 00:32:22 2013225 218 11 103 104 -156.4 ± 33.11
27211 ncm-et-16 2018-05-07 10:01 00:32:27 2016611 225 13 81 131 -108.39 ± 28.63
27210 ncm-et-10 2018-05-07 10:01 00:32:28 2020348 226 13 98 115 -137.42 ± 31.32
27209 ncm-et-5 2018-05-07 10:01 00:32:33 2024917 217 11 84 122 -121.61 ± 29.8
27208 ncm-et-7 2018-05-07 10:01 00:32:51 2021663 220 19 93 108 -121.6 ± 32.7
27207 ncm-et-6 2018-05-07 10:00 00:33:31 2025408 230 13 112 105 -159.97 ± 33.14
27206 ncm-et-14 2018-05-07 08:47 01:13:34 2020022 500 29 212 259 -133.34 ± 20.78
27205 ncm-et-3 2018-05-07 08:47 01:14:06 2025081 500 29 233 238 -150.51 ± 21.89
27204 ncm-et-8 2018-05-07 08:47 01:18:34 1919283 500 29 231 240 -148.85 ± 21.78
27203 ncm-et-16 2018-05-07 08:47 01:13:10 2022142 500 33 209 258 -127.76 ± 20.89
27202 ncm-et-13 2018-05-07 08:47 01:13:35 2025408 500 36 194 270 -113.68 ± 20.33
27201 ncm-et-5 2018-05-07 08:47 01:13:02 2023445 500 24 207 269 -133.34 ± 20.17
27200 ncm-et-9 2018-05-07 08:47 01:13:39 2027872 500 27 230 243 -149.68 ± 21.59
27199 ncm-et-12 2018-05-07 08:47 01:15:05 2001714 500 23 203 274 -130.94 ± 19.9
27198 ncm-et-4 2018-05-07 08:47 01:14:48 1984034 500 41 218 241 -128.55 ± 21.86
27197 ncm-et-10 2018-05-07 08:47 01:13:06 2018395 500 30 220 250 -138.99 ± 21.26
27196 ncm-et-15 2018-05-07 08:47 01:13:43 2020347 500 29 231 240 -148.85 ± 21.78
27195 ncm-et-11 2018-05-07 08:47 01:13:15 2017422 500 24 211 265 -136.56 ± 20.38
27194 ncm-et-6 2018-05-07 08:47 01:12:08 2025244 500 28 224 248 -143.89 ± 21.34
27193 ncm-et-7 2018-05-07 08:47 01:12:41 1999214 500 28 210 262 -132.54 ± 20.61

Commit

Commit ID e5c3effdb1fee10a694e493773e62f433e642406
Author Lucas Braesch
Date 2014-02-04 07:24:46 UTC
Better document null search window Hopefully this patch makes the code more: * Self-documenting: Null search is always a zero window search, because it is testing for a fail high. It should never be done on a full window! The current code only works because we don't do it at PV nodes, and therefore (alpha, beta) = (beta-1, beta): that's the kind of "clever" trick we should avoid. * Idiot-proof: If we want to enable null search at PV nodes, all we need to do now is comment out the !PvNode condition. It's that simple! In theory, null search should not be done at PV nodes, because PV nodes should never fail high. But in practice, they DO fail high, because of aspiration windows, and search inconsistencies, for example. So it makes sense to keep that flexibility in the code. No functional change.
Copyright 2011–2024 Next Chess Move LLC