Dev Builds » 20140212-1347

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 06:36:45 1984565 2684 158 1126 1400 -131.21 ± 8.92
ncm-et-4 06:36:58 1987408 2644 169 1098 1377 -127.51 ± 9.01
ncm-et-5 01:45:33 2023446 722 37 285 400 -124.4 ± 16.47
ncm-et-6 01:45:33 2025653 724 41 323 360 -142.87 ± 17.7
ncm-et-7 01:45:45 1960369 684 48 283 353 -124.43 ± 17.88
ncm-et-8 01:45:42 2023201 721 36 301 384 -133.96 ± 16.92
ncm-et-9 02:23:18 2010284 980 47 437 496 -146.35 ± 14.98
ncm-et-10 06:36:23 1983243 2690 137 1134 1419 -135.21 ± 8.81
ncm-et-11 01:45:49 2011032 698 46 271 381 -116.14 ± 17.04
ncm-et-12 01:45:25 2004144 722 47 282 393 -117.35 ± 16.78
ncm-et-13 06:36:47 1987806 2662 146 1108 1408 -131.5 ± 8.86
ncm-et-14 01:45:23 2017826 712 49 307 356 -131.89 ± 17.86
ncm-et-15 06:36:41 1988963 2665 131 1070 1464 -127.9 ± 8.61
ncm-et-16 01:45:31 1982223 692 37 280 375 -127.43 ± 17.1
20000 1129 8305 10566 -130.46 ± 3.24

Test Detail

ID Host Started (UTC) Duration Base NPS Games Wins Losses Draws Elo CLI PGN
95989 ncm-et-13 2019-09-13 21:23 00:37:28 1973148 248 18 104 126 -125.69 ± 30.07
95988 ncm-et-4 2019-09-13 21:23 00:37:30 1972037 252 19 104 129 -121.97 ± 29.71
95987 ncm-et-10 2019-09-13 21:23 00:37:29 1965538 255 9 116 130 -155.38 ± 29.11
95986 ncm-et-9 2019-09-13 21:23 00:37:32 1982485 253 12 105 136 -133.98 ± 28.41
95985 ncm-et-15 2019-09-13 21:23 00:37:21 1968936 251 12 96 143 -120.93 ± 27.38
95984 ncm-et-3 2019-09-13 21:23 00:37:14 1971104 252 13 108 131 -137.77 ± 29.17
95314 ncm-et-4 2019-09-07 20:21 00:26:14 1974073 180 15 79 86 -129.17 ± 36.74
95313 ncm-et-13 2019-09-07 20:18 00:28:35 1976851 197 8 82 107 -137.23 ± 31.87
95312 ncm-et-3 2019-09-07 20:18 00:28:37 1975293 190 13 77 100 -121.79 ± 33.58
95311 ncm-et-15 2019-09-07 20:16 00:31:15 1979037 206 13 71 122 -100.54 ± 29.62
95310 ncm-et-10 2019-09-07 20:14 00:32:50 1977449 216 10 87 119 -129.54 ± 30.22
95309 ncm-et-4 2019-09-07 19:04 01:15:42 1984598 500 33 200 267 -120.67 ± 20.44
95308 ncm-et-13 2019-09-07 19:02 01:15:06 1968008 500 21 213 266 -140.62 ± 20.26
95307 ncm-et-3 2019-09-07 19:00 01:17:35 1935232 500 28 218 254 -138.99 ± 21.02
95306 ncm-et-15 2019-09-07 19:00 01:14:55 1971876 500 30 191 279 -116.0 ± 19.79
95305 ncm-et-10 2019-09-07 18:57 01:15:25 1957399 500 22 211 267 -138.18 ± 20.23
95304 ncm-et-4 2019-09-07 17:47 01:15:31 1968986 500 36 227 237 -139.81 ± 22.02
95303 ncm-et-13 2019-09-07 17:46 01:15:24 1968313 500 29 209 262 -130.94 ± 20.63
95302 ncm-et-3 2019-09-07 17:45 01:13:13 1979974 500 30 216 254 -135.76 ± 21.05
95301 ncm-et-10 2019-09-07 17:44 01:11:44 1978147 500 29 218 253 -138.18 ± 21.09
95300 ncm-et-15 2019-09-07 17:44 01:14:53 1986756 500 26 214 260 -137.37 ± 20.67
95299 ncm-et-15 2019-09-07 16:30 01:12:27 1979347 500 23 215 262 -140.62 ± 20.51
95298 ncm-et-13 2019-09-07 16:30 01:14:41 1976524 500 24 208 268 -134.15 ± 20.22
95297 ncm-et-10 2019-09-07 16:30 01:13:26 1964615 500 29 199 272 -123.02 ± 20.12
95296 ncm-et-3 2019-09-07 16:30 01:14:23 1978881 500 39 185 276 -104.49 ± 20.08
95295 ncm-et-4 2019-09-07 16:30 01:16:21 1982986 500 25 205 270 -130.94 ± 20.14
27024 ncm-et-7 2018-05-06 21:32 00:28:46 1964117 184 16 76 92 -117.59 ± 35.41
27023 ncm-et-16 2018-05-06 21:31 00:28:50 2021488 192 4 84 104 -154.14 ± 31.87
27022 ncm-et-11 2018-05-06 21:30 00:29:59 2004482 198 16 69 113 -95.32 ± 31.28
27021 ncm-et-14 2018-05-06 21:29 00:30:51 2018233 212 12 94 106 -141.76 ± 32.69
27020 ncm-et-15 2018-05-06 21:29 00:31:30 2019045 208 6 78 124 -125.45 ± 28.68
27019 ncm-et-13 2018-05-06 21:29 00:31:33 2027867 217 14 81 122 -110.89 ± 30.0
27018 ncm-et-4 2018-05-06 21:28 00:31:50 2005405 212 9 90 113 -139.84 ± 31.15
27017 ncm-et-6 2018-05-06 21:28 00:32:13 2025572 224 14 106 104 -151.65 ± 33.31
27016 ncm-et-10 2018-05-06 21:28 00:32:04 2019536 219 14 95 110 -134.9 ± 32.15
27015 ncm-et-5 2018-05-06 21:28 00:32:33 2023773 222 15 85 122 -113.42 ± 30.17
27014 ncm-et-8 2018-05-06 21:28 00:32:54 2024099 221 15 101 105 -142.72 ± 33.13
27013 ncm-et-12 2018-05-06 21:28 00:32:51 1987615 222 15 95 112 -131.09 ± 31.88
27012 ncm-et-9 2018-05-06 21:27 00:33:35 2025900 227 6 101 120 -154.91 ± 29.93
27011 ncm-et-3 2018-05-06 21:26 00:34:47 2026063 242 5 112 125 -165.0 ± 29.31
27010 ncm-et-12 2018-05-06 20:14 01:12:34 2020673 500 32 187 281 -111.37 ± 19.73
27009 ncm-et-6 2018-05-06 20:14 01:13:20 2025735 500 27 217 256 -138.99 ± 20.9
27008 ncm-et-16 2018-05-06 20:14 01:16:41 1942959 500 33 196 271 -117.55 ± 20.24
27007 ncm-et-8 2018-05-06 20:14 01:12:48 2022303 500 21 200 279 -130.15 ± 19.6
27006 ncm-et-14 2018-05-06 20:14 01:14:32 2017420 500 37 213 250 -127.76 ± 21.35
27005 ncm-et-10 2018-05-06 20:14 01:13:25 2020021 500 24 208 268 -134.15 ± 20.22
27004 ncm-et-7 2018-05-06 20:14 01:16:59 1956622 500 32 207 261 -126.97 ± 20.73
27003 ncm-et-3 2018-05-06 20:14 01:10:56 2025408 500 30 210 260 -130.94 ± 20.75
27002 ncm-et-15 2018-05-06 20:14 01:14:20 2017749 500 21 205 274 -134.15 ± 19.85
27001 ncm-et-9 2018-05-06 20:14 01:12:11 2022467 500 29 231 240 -148.85 ± 21.78
27000 ncm-et-4 2018-05-06 20:14 01:13:50 2023773 500 32 193 275 -116.0 ± 20.02
26999 ncm-et-11 2018-05-06 20:14 01:15:50 2017583 500 30 202 268 -124.6 ± 20.34
26998 ncm-et-13 2018-05-06 20:13 01:14:00 2023937 500 32 211 257 -130.15 ± 20.93
26997 ncm-et-5 2018-05-06 20:13 01:13:00 2023119 500 22 200 278 -129.35 ± 19.67

Commit

Commit ID e6523e56b8cc110aa38d698a993cb04b430e67b0
Author Marco Costalba
Date 2014-02-12 13:47:36 UTC
Move magic random to RKISS When initializing the magic numbers used to compute sliding attacks, we endless generate a random and test it as a possible magic. In the general case this takes a lot of iterations, but here, insteaad of picking a casual random, we rotate it a couple of times and generate a number that we know has a good probability to be a magic candidate. This is becuase the quantities by which we rotate the number are known in advance to produce quickly a good canidate. The patch, inspired by DON, just moves the shuffle to RKISS changing the boosters to take in account a left rotation instead of a right rotation as in the original. No functional change.
Copyright 2011–2024 Next Chess Move LLC