Dev Builds » 20140329-1013

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 05:58:46 1993604 2386 152 930 1304 -117.58 ± 9.19
ncm-et-4 05:58:49 1965805 2357 164 932 1261 -117.49 ± 9.4
ncm-et-5 01:46:14 2024182 725 48 300 377 -126.01 ± 17.26
ncm-et-6 01:46:14 2026226 727 57 304 366 -122.93 ± 17.65
ncm-et-7 01:46:18 1988439 693 43 295 355 -132.4 ± 17.8
ncm-et-8 01:46:16 1994947 698 49 282 367 -120.6 ± 17.49
ncm-et-9 05:59:02 1990109 2401 166 951 1284 -117.92 ± 9.32
ncm-et-10 05:58:13 1986261 2375 161 973 1241 -123.77 ± 9.5
ncm-et-11 01:46:02 2009269 723 53 296 374 -121.49 ± 17.38
ncm-et-12 01:46:31 1970794 708 38 282 388 -124.85 ± 16.78
ncm-et-13 05:59:06 1988301 2394 141 967 1286 -125.0 ± 9.27
ncm-et-14 01:46:21 2019452 724 36 309 379 -137.81 ± 17.08
ncm-et-15 05:59:03 1988269 2369 164 958 1247 -121.13 ± 9.48
ncm-et-16 01:46:29 2021325 720 33 321 366 -147.19 ± 17.42
20000 1305 8100 10595 -122.93 ± 3.24

Test Detail

ID Host Started (UTC) Duration Base NPS Games Wins Losses Draws Elo CLI PGN
93302 ncm-et-10 2019-08-21 00:57 00:22:54 1974754 157 12 61 84 -112.18 ± 36.68
93301 ncm-et-4 2019-08-21 00:57 00:23:32 1973116 156 9 63 84 -125.45 ± 36.39
93300 ncm-et-15 2019-08-21 00:56 00:24:10 1971260 159 13 65 81 -117.96 ± 37.65
93299 ncm-et-13 2019-08-21 00:54 00:25:57 1969733 172 4 74 94 -150.09 ± 33.54
93298 ncm-et-3 2019-08-21 00:53 00:26:50 1975152 168 10 59 99 -104.37 ± 32.89
93297 ncm-et-9 2019-08-21 00:52 00:28:33 1974985 188 15 70 103 -104.7 ± 33.01
93296 ncm-et-10 2019-08-20 23:39 01:16:45 1969454 500 38 207 255 -122.24 ± 21.11
93295 ncm-et-4 2019-08-20 23:39 01:16:35 1911949 500 38 203 259 -119.11 ± 20.91
93294 ncm-et-15 2019-08-20 23:39 01:16:03 1972515 500 36 207 257 -123.81 ± 20.98
93293 ncm-et-13 2019-08-20 23:38 01:15:05 1968781 500 40 212 248 -124.6 ± 21.49
93292 ncm-et-3 2019-08-20 23:36 01:15:53 1971426 500 32 177 291 -103.73 ± 19.24
93291 ncm-et-9 2019-08-20 23:36 01:14:21 1978799 500 37 196 267 -114.45 ± 20.5
93290 ncm-et-4 2019-08-20 22:23 01:14:48 1965730 500 38 201 261 -117.55 ± 20.81
93289 ncm-et-15 2019-08-20 22:22 01:16:11 1971131 500 34 209 257 -126.97 ± 20.96
93288 ncm-et-10 2019-08-20 22:21 01:17:19 1967695 500 29 209 262 -130.94 ± 20.63
93287 ncm-et-3 2019-08-20 22:21 01:14:31 1987538 500 36 194 270 -113.68 ± 20.33
93286 ncm-et-13 2019-08-20 22:21 01:16:24 1968636 500 27 200 273 -125.39 ± 20.03
93285 ncm-et-9 2019-08-20 22:20 01:15:12 1971273 500 31 190 279 -114.45 ± 19.81
93284 ncm-et-10 2019-08-20 21:05 01:14:57 1965132 500 31 190 279 -114.45 ± 19.81
93283 ncm-et-4 2019-08-20 21:05 01:17:29 1937508 500 30 202 268 -124.6 ± 20.34
93282 ncm-et-9 2019-08-20 21:05 01:14:27 1970175 500 35 198 267 -117.55 ± 20.47
93281 ncm-et-15 2019-08-20 21:05 01:16:10 1974989 500 45 197 258 -109.07 ± 21.03
93280 ncm-et-13 2019-08-20 21:04 01:15:17 1972656 500 32 187 281 -111.37 ± 19.73
93279 ncm-et-3 2019-08-20 21:04 01:15:19 1976694 500 23 208 269 -134.95 ± 20.15
25512 ncm-et-8 2018-05-02 19:32 00:29:29 2019701 198 13 83 102 -128.37 ± 33.32
25511 ncm-et-4 2018-05-02 19:32 00:30:01 2025081 201 13 75 113 -110.78 ± 31.18
25510 ncm-et-12 2018-05-02 19:31 00:30:28 2019861 208 9 81 118 -125.45 ± 30.11
25509 ncm-et-7 2018-05-02 19:31 00:30:28 1951628 193 13 80 100 -125.84 ± 33.65
25508 ncm-et-9 2018-05-02 19:30 00:31:39 2021651 213 17 96 100 -135.31 ± 34.07
25507 ncm-et-10 2018-05-02 19:30 00:31:56 2018558 218 12 101 105 -150.62 ± 32.97
25506 ncm-et-16 2018-05-02 19:29 00:32:28 2020836 220 9 102 109 -156.7 ± 32.05
25505 ncm-et-5 2018-05-02 19:29 00:32:19 2024590 225 16 89 120 -116.95 ± 30.61
25504 ncm-et-3 2018-05-02 19:29 00:32:37 2026390 218 20 88 110 -112.11 ± 32.36
25503 ncm-et-15 2018-05-02 19:29 00:32:48 2021326 210 15 83 112 -116.7 ± 31.69
25502 ncm-et-11 2018-05-02 19:29 00:32:38 2000305 223 19 89 115 -112.87 ± 31.53
25501 ncm-et-14 2018-05-02 19:29 00:33:01 2019534 224 13 101 110 -144.25 ± 32.17
25500 ncm-et-13 2018-05-02 19:29 00:33:05 2026063 222 13 90 119 -125.72 ± 30.56
25499 ncm-et-6 2018-05-02 19:28 00:33:54 2026063 227 12 97 118 -136.75 ± 30.76
25498 ncm-et-11 2018-05-02 18:14 01:13:24 2018233 500 34 207 259 -125.39 ± 20.86
25497 ncm-et-14 2018-05-02 18:14 01:13:20 2019371 500 23 208 269 -134.95 ± 20.15
25496 ncm-et-3 2018-05-02 18:14 01:13:36 2024428 500 31 204 265 -125.39 ± 20.51
25495 ncm-et-8 2018-05-02 18:14 01:16:47 1970194 500 36 199 265 -117.55 ± 20.58
25494 ncm-et-6 2018-05-02 18:14 01:12:20 2026390 500 45 207 248 -116.78 ± 21.53
25493 ncm-et-4 2018-05-02 18:14 01:16:24 1981448 500 36 188 276 -109.07 ± 20.04
25492 ncm-et-5 2018-05-02 18:14 01:13:55 2023774 500 32 211 257 -130.15 ± 20.93
25491 ncm-et-12 2018-05-02 18:14 01:16:03 1921728 500 29 201 270 -124.6 ± 20.22
25490 ncm-et-16 2018-05-02 18:14 01:14:01 2021815 500 24 219 257 -143.07 ± 20.79
25489 ncm-et-7 2018-05-02 18:14 01:15:50 2025251 500 30 215 255 -134.95 ± 21.0
25488 ncm-et-13 2018-05-02 18:14 01:13:18 2023937 500 25 204 271 -130.15 ± 20.09
25487 ncm-et-10 2018-05-02 18:14 01:14:22 2021977 500 39 205 256 -119.89 ± 21.07
25486 ncm-et-15 2018-05-02 18:14 01:13:41 2018395 500 21 197 282 -127.76 ± 19.45
25485 ncm-et-9 2018-05-02 18:14 01:14:50 2023773 500 31 201 268 -123.02 ± 20.36

Commit

Commit ID 9350d0dce5190fc88b4770ece79299c6d3edd2ed
Author Marco Costalba
Date 2014-03-29 10:13:42 UTC
Raise VALUE_INFINITE In some legal positions like this one: R6R/3Q4/1Q4Q1/4Q3/2Q4Q/Q4Q2/Np1Q4/kB1N1KB1 b -- 0 1 We can have a very high score, in this case 30177 and 29267 for midgame and endgame respectively, and because VALUE_INFINITE = 30001 we have an assert in interpolate() Midgame and endgame scores are stored in 16 bit signed integers so we can rise VALUE_INFINITE a little bit. This does not fix the possibility of overflow in general case, just makes the condition more difficult to trigger and anyhow better uses all the score width. Raising VALUE_INFINITE to 32000 seems to fix the problem for this particular case. No functional change.
Copyright 2011–2024 Next Chess Move LLC