Dev Builds » 20190701-1207

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 08:33:16 1967615 3331 2002 70 1259 +230.17 ± 9.51
ncm-et-4 08:33:13 1959783 3307 1960 67 1280 +226.22 ± 9.41
ncm-et-9 08:33:08 1969650 3326 1906 76 1344 +214.95 ± 9.17
ncm-et-10 08:33:26 1971202 3340 1972 80 1288 +223.16 ± 9.4
ncm-et-13 08:33:27 1973899 3358 1955 70 1333 +220.55 ± 9.21
ncm-et-15 08:33:19 1973948 3338 1920 59 1359 +218.62 ± 9.09
20000 11715 422 7863 +222.23 ± 3.79

Test Detail

ID Host Started (UTC) Duration Base NPS Games Wins Losses Draws Elo CLI PGN
87416 ncm-et-4 2019-07-02 07:23 00:46:30 1966461 307 192 6 109 +244.02 ± 32.57
87415 ncm-et-9 2019-07-02 07:21 00:49:24 1966315 326 198 9 119 +230.03 ± 31.22
87414 ncm-et-3 2019-07-02 07:19 00:51:00 1963840 331 202 4 125 +239.84 ± 30.13
87413 ncm-et-15 2019-07-02 07:18 00:51:53 1968339 338 205 8 125 +231.65 ± 30.39
87412 ncm-et-10 2019-07-02 07:18 00:52:12 1960332 340 200 10 130 +219.27 ± 29.82
87411 ncm-et-13 2019-07-02 07:15 00:55:15 1985684 358 200 6 152 +210.84 ± 27.15
87410 ncm-et-9 2019-07-02 06:03 01:16:31 1967704 500 275 12 213 +203.11 ± 23.02
87409 ncm-et-4 2019-07-02 06:03 01:19:39 1921278 500 304 11 185 +233.32 ± 24.92
87408 ncm-et-3 2019-07-02 06:02 01:15:32 1964764 500 301 11 188 +230.16 ± 24.69
87407 ncm-et-10 2019-07-02 05:59 01:17:32 1960782 500 294 9 197 +224.97 ± 23.99
87406 ncm-et-15 2019-07-02 05:59 01:18:10 1973282 500 287 6 207 +220.88 ± 23.2
87405 ncm-et-13 2019-07-02 05:58 01:15:53 1965688 500 303 10 187 +233.32 ± 24.74
87404 ncm-et-4 2019-07-02 04:45 01:17:03 1968540 500 296 11 193 +224.97 ± 24.33
87403 ncm-et-9 2019-07-02 04:44 01:17:59 1972348 500 298 12 190 +226.0 ± 24.57
87402 ncm-et-3 2019-07-02 04:43 01:18:00 1964763 500 316 15 169 +241.91 ± 26.25
87401 ncm-et-15 2019-07-02 04:42 01:15:54 1979644 500 290 6 204 +223.94 ± 23.4
87400 ncm-et-10 2019-07-02 04:40 01:18:16 1965240 500 298 12 190 +226.0 ± 24.57
87399 ncm-et-13 2019-07-02 04:40 01:16:54 1977269 500 295 8 197 +227.04 ± 23.96
87398 ncm-et-9 2019-07-02 03:27 01:16:22 1972499 500 294 16 190 +217.85 ± 24.66
87397 ncm-et-4 2019-07-02 03:26 01:17:10 1971276 500 283 9 208 +213.85 ± 23.25
87396 ncm-et-15 2019-07-02 03:25 01:15:33 1979940 500 278 11 211 +206.98 ± 23.12
87395 ncm-et-3 2019-07-02 03:24 01:18:27 1976002 500 269 10 221 +199.29 ± 22.46
87394 ncm-et-10 2019-07-02 03:23 01:16:21 1970788 500 289 14 197 +214.85 ± 24.13
87393 ncm-et-13 2019-07-02 03:22 01:16:32 1971105 500 286 9 205 +216.85 ± 23.45
87392 ncm-et-15 2019-07-02 02:08 01:16:29 1976538 500 290 11 199 +218.86 ± 23.91
87391 ncm-et-9 2019-07-02 02:08 01:18:04 1966008 500 268 6 226 +202.15 ± 22.01
87390 ncm-et-4 2019-07-02 02:06 01:18:53 1954875 500 302 7 191 +235.45 ± 24.35
87389 ncm-et-3 2019-07-02 02:06 01:16:33 1965074 500 300 9 191 +231.21 ± 24.42
87388 ncm-et-13 2019-07-02 02:05 01:16:06 1965688 500 289 16 195 +212.86 ± 24.31
87387 ncm-et-10 2019-07-02 02:05 01:16:38 1970479 500 289 9 202 +219.87 ± 23.65
87386 ncm-et-15 2019-07-02 00:49 01:17:49 1971571 500 280 9 211 +210.89 ± 23.05
87385 ncm-et-9 2019-07-02 00:48 01:18:07 1972352 500 285 14 201 +210.89 ± 23.85
87384 ncm-et-4 2019-07-02 00:48 01:17:12 1963226 500 298 11 191 +227.04 ± 24.48
87383 ncm-et-3 2019-07-02 00:48 01:17:12 1973794 500 310 11 179 +239.74 ± 25.37
87382 ncm-et-10 2019-07-02 00:48 01:16:00 1996849 500 306 17 177 +229.12 ± 25.63
87381 ncm-et-13 2019-07-02 00:47 01:17:00 1973293 500 283 8 209 +214.85 ± 23.15
87380 ncm-et-9 2019-07-01 23:31 01:16:41 1970330 500 288 7 205 +220.88 ± 23.38
87379 ncm-et-3 2019-07-01 23:30 01:16:32 1965073 500 304 10 186 +234.38 ± 24.81
87378 ncm-et-4 2019-07-01 23:30 01:16:46 1972826 500 285 12 203 +212.86 ± 23.67
87377 ncm-et-10 2019-07-01 23:30 01:16:27 1973948 500 296 9 195 +227.04 ± 24.13
87376 ncm-et-13 2019-07-01 23:30 01:15:47 1978566 500 299 13 188 +226.0 ± 24.74
87375 ncm-et-15 2019-07-01 23:30 01:17:31 1968325 500 290 8 202 +221.9 ± 23.61

Commit

Commit ID 217840a6a5a40b516cab59a450a9f36352997240
Author Joost VandeVondele
Date 2019-07-01 12:07:23 UTC
Introduce coordination between searching threads (#2204) this patch improves threading performance by introducing some coordination between threads. The observation is that threading is an area where a lot of Elo can potentially be gained: https://github.com/glinscott/fishtest/wiki/UsefulData#elo-from-threading At STC, 8 threads gain roughly 320 Elo, vs sequential at the same time, however, loses 66 Elo against a single thread with 8x more time. This 66 Elo should be partially recoverable with improved threading. To improve threading, this patch introduces some LMR at nodes that are already being searched by other threads. This requires some coordination between threads, avoiding however synchronisation. To do so, threads leave a trail of breadcrumbs to mark the nodes they are searching. These breadcrumbs are stored in a small hash table, which is only probed at low plies (currently ply < 8). A couple of variants of this patch passed both STC and LTC threaded tests. I picked the simpler, more robust version. I expect that further tests can find further improvements. STC (5+0.05 @ 8 threads): LLR: 2.95 (-2.94,2.94) [0.50,4.50] Total: 26209 W: 5359 L: 5079 D: 15771 http://tests.stockfishchess.org/tests/view/5d0a9b030ebc5925cf0a8e6f LTC (20+0.2 @ 8 threads): LLR: 2.96 (-2.94,2.94) [0.00,3.50] Total: 34832 W: 5650 L: 5382 D: 23800 http://tests.stockfishchess.org/tests/view/5d0c67a20ebc5925cf0aafa7 other passed/tested variants: http://tests.stockfishchess.org/tests/view/5d0a9b030ebc5925cf0a8e6f http://tests.stockfishchess.org/tests/view/5d0c67ca0ebc5925cf0aafa9 http://tests.stockfishchess.org/tests/view/5d0c67810ebc5925cf0aafa3 http://tests.stockfishchess.org/tests/view/5d0958ca0ebc5925cf0a74c6 For the sequential code there is no change in bench, and an earlier version of this patch passed a non-regression test. STC (10+0.1 @ 1 thread) LLR: 2.96 (-2.94,2.94) [-3.00,1.00] Total: 10471 W: 2364 L: 2220 D: 5887 http://tests.stockfishchess.org/tests/view/5d087ee20ebc5925cf0a6381 passed the additional non-regression tests at 2 and 4 threads 20+0.2 TC. The code was rebased on master prior to testing. 2 threads: LLR: 2.95 (-2.94,2.94) [-3.00,1.00] Total: 218863 W: 40927 L: 41153 D: 136783 http://tests.stockfishchess.org/tests/view/5d18c6c30ebc5925cf0b9566 4threads: LLR: 2.96 (-2.94,2.94) [-3.00,1.00] Total: 16839 W: 3017 L: 2889 D: 10933 http://tests.stockfishchess.org/tests/view/5d18c6ea0ebc5925cf0b9568 No functional change.
Copyright 2011–2024 Next Chess Move LLC