Dev Builds » 20190711-0930

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 08:34:16 1970685 3344 1965 62 1317 +224.5 ± 9.26
ncm-et-4 08:35:36 1951590 3306 1943 63 1300 +224.29 ± 9.32
ncm-et-9 08:34:25 1969684 3346 1970 78 1298 +222.64 ± 9.36
ncm-et-10 08:35:38 1964940 3336 1996 57 1283 +230.81 ± 9.39
ncm-et-13 08:37:15 1973060 3328 1913 72 1343 +216.44 ± 9.17
ncm-et-15 08:34:05 1975320 3340 1950 78 1312 +220.11 ± 9.3
20000 11737 410 7853 +223.1 ± 3.8

Test Detail

ID Host Started (UTC) Duration Base NPS Games Wins Losses Draws Elo CLI PGN
89216 ncm-et-4 2019-07-16 21:55 00:46:38 1967241 306 190 8 108 +238.0 ± 32.82
89215 ncm-et-13 2019-07-16 21:51 00:51:10 1978097 328 187 8 133 +212.73 ± 29.3
89214 ncm-et-3 2019-07-16 21:49 00:52:28 1961381 344 214 6 124 +243.36 ± 30.45
89213 ncm-et-10 2019-07-16 21:49 00:52:27 1962002 336 208 7 121 +239.86 ± 30.89
89212 ncm-et-15 2019-07-16 21:49 00:52:43 1967402 340 214 8 118 +244.04 ± 31.38
89211 ncm-et-9 2019-07-16 21:48 00:53:24 1966936 346 215 6 125 +243.03 ± 30.32
89210 ncm-et-4 2019-07-16 20:34 01:20:06 1940869 500 293 6 201 +227.04 ± 23.61
89209 ncm-et-15 2019-07-16 20:32 01:15:51 1971601 500 283 8 209 +214.85 ± 23.15
89208 ncm-et-10 2019-07-16 20:32 01:16:33 1961998 500 308 8 184 +240.82 ± 24.91
89207 ncm-et-13 2019-07-16 20:32 01:18:07 1972652 500 304 11 185 +233.32 ± 24.92
89206 ncm-et-3 2019-07-16 20:31 01:17:39 1978178 500 300 8 192 +232.26 ± 24.31
89205 ncm-et-9 2019-07-16 20:30 01:17:23 1963998 500 289 9 202 +219.87 ± 23.65
89204 ncm-et-4 2019-07-16 19:16 01:17:08 1961074 500 282 11 207 +210.89 ± 23.38
89203 ncm-et-13 2019-07-16 19:15 01:15:54 1975760 500 280 11 209 +208.93 ± 23.25
89202 ncm-et-15 2019-07-16 19:14 01:17:01 1973893 500 290 14 196 +215.85 ± 24.2
89201 ncm-et-10 2019-07-16 19:13 01:17:15 1972894 500 282 9 209 +212.86 ± 23.18
89200 ncm-et-3 2019-07-16 19:12 01:17:06 1981039 500 282 9 209 +212.86 ± 23.18
89199 ncm-et-9 2019-07-16 19:12 01:16:23 1974446 500 296 15 189 +220.88 ± 24.71
89198 ncm-et-4 2019-07-16 17:55 01:19:17 1923441 500 301 12 187 +229.12 ± 24.79
89197 ncm-et-3 2019-07-16 17:55 01:16:12 1982052 500 300 11 189 +229.12 ± 24.62
89196 ncm-et-10 2019-07-16 17:55 01:17:06 1961549 500 296 5 199 +231.21 ± 23.7
89195 ncm-et-15 2019-07-16 17:55 01:17:51 1972659 500 282 13 205 +208.93 ± 23.56
89194 ncm-et-9 2019-07-16 17:55 01:15:59 1979425 500 298 14 188 +223.94 ± 24.76
89193 ncm-et-13 2019-07-16 17:55 01:18:23 1969862 500 275 11 214 +204.07 ± 22.93
88940 ncm-et-4 2019-07-14 12:29 01:16:33 1975571 500 295 7 198 +228.08 ± 23.85
88939 ncm-et-13 2019-07-14 12:29 01:18:14 1964918 500 280 13 207 +206.98 ± 23.43
88938 ncm-et-10 2019-07-14 12:28 01:17:42 1959110 500 324 10 166 +256.44 ± 26.41
88937 ncm-et-15 2019-07-14 12:27 01:16:40 1976747 500 296 7 197 +229.12 ± 23.92
88936 ncm-et-3 2019-07-14 12:26 01:17:54 1965225 500 305 8 187 +237.58 ± 24.68
88935 ncm-et-9 2019-07-14 12:26 01:18:18 1966641 500 292 11 197 +220.88 ± 24.05
88934 ncm-et-13 2019-07-14 11:10 01:17:20 1971880 500 295 8 197 +227.04 ± 23.96
88933 ncm-et-3 2019-07-14 11:10 01:15:27 1966306 500 281 9 210 +211.87 ± 23.12
88932 ncm-et-4 2019-07-14 11:09 01:19:03 1929553 500 281 11 208 +209.91 ± 23.31
88931 ncm-et-15 2019-07-14 11:09 01:17:06 1979692 500 293 14 193 +218.86 ± 24.41
88930 ncm-et-10 2019-07-14 11:09 01:17:56 1954652 500 288 10 202 +217.85 ± 23.68
88929 ncm-et-9 2019-07-14 11:09 01:16:23 1968334 500 286 7 207 +218.86 ± 23.24
88928 ncm-et-3 2019-07-14 09:51 01:17:30 1960615 500 283 11 206 +211.87 ± 23.44
88927 ncm-et-4 2019-07-14 09:51 01:16:51 1963381 500 301 8 191 +233.32 ± 24.39
88926 ncm-et-10 2019-07-14 09:51 01:16:39 1982381 500 290 8 202 +221.9 ± 23.61
88925 ncm-et-9 2019-07-14 09:51 01:16:35 1968014 500 294 16 190 +217.85 ± 24.66
88924 ncm-et-15 2019-07-14 09:51 01:16:53 1985251 500 292 14 194 +217.85 ± 24.34
88923 ncm-et-13 2019-07-14 09:51 01:18:07 1978257 500 292 10 198 +221.9 ± 23.95

Commit

Commit ID fa1a2a0667ff81f04e94cab078c50c2d51f4cb5e
Author Daniel Axtens
Date 2019-07-11 09:30:09 UTC
Enable popcount and prefetch for ppc-64 PowerPC has had popcount instructions for a long time, at least as far back as POWER5 (released 2004). Enable them via a gcc builtin. Using a gcc builtin has the added bonus that if compiled for a processor that lacks a hardware instruction, gcc will include a software popcount implementation that does not use the instruction. It might be slower than the table lookups (or it might be faster) but it will certainly work. So this isn't going to break anything. On my POWER8 VM, this leads to a ~4.27% speedup. Fir prefetch, the gcc builtin generates a 'dcbt' instruction, which is supported at least as far back as the G5 (2002) and POWER4 (2001). This leads to a ~5% speedup on my POWER8 VM. No functional change
Copyright 2011–2024 Next Chess Move LLC