Dev Builds » 20261011-1259

Use this dev build

NCM plays each Stockfish dev build 20,000 times against Stockfish 15. This yields an approximate Elo difference and establishes confidence in the strength of the dev builds.

Summary

Host Duration Avg Base NPS Games WLD Standard Elo Ptnml(0-2) Gamepair Elo
ncm-dbt-01 06:41:36 583772 3998 1823 283 1892 +141.11 ± 4.25 0 16 460 1490 33 +341.14 ± 15.87
ncm-dbt-02 06:41:46 590545 4000 1829 282 1889 +141.74 ± 4.46 0 21 463 1464 52 +336.06 ± 15.83
ncm-dbt-03 06:42:20 584140 4008 1814 297 1897 +138.38 ± 4.31 0 14 497 1455 38 +328.72 ± 15.24
ncm-dbt-04 06:41:21 567904 3994 1838 315 1841 +139.53 ± 4.47 0 16 496 1431 54 +326.88 ± 15.26
ncm-dbt-05 06:42:04 607298 4000 1830 282 1888 +141.85 ± 4.31 1 14 461 1484 40 +341.64 ± 15.85
20000 9134 1459 9407 +140.52 ± 1.95 1 81 2377 7324 217 +334.81 ± 6.97

Test Detail

ID Host Base NPS Games WLD Standard Elo Ptnml(0-2) Gamepair Elo CLI PGN
486414 ncm-dbt-03 581527 8 4 0 4 +190.56 ± 13.88 0 0 0 4 0 +1370.41 ± 651.08 ↓
486413 ncm-dbt-04 567086 494 220 43 231 +130.26 ± 12.87 0 2 72 167 6 +296.15 ± 40.27 ↓
486412 ncm-dbt-01 583614 498 228 39 231 +138.8 ± 12.33 0 3 58 184 4 +332.52 ± 45.35 ↓
486411 ncm-dbt-05 602248 500 218 29 253 +138.18 ± 12.12 0 1 64 180 5 +327.18 ± 42.83 ↓
486410 ncm-dbt-02 592806 500 221 45 234 +127.76 ± 12.26 0 2 73 172 3 +295.94 ± 39.99 ↓
486409 ncm-dbt-03 582611 500 226 40 234 +135.76 ± 12.62 0 2 66 176 6 +315.35 ± 42.24 ↓
486408 ncm-dbt-04 569230 500 225 48 227 +128.55 ± 12.63 0 3 71 172 4 +295.94 ± 40.69 ↓
486407 ncm-dbt-02 593239 500 235 32 233 +149.68 ± 12.23 0 1 53 188 8 +363.2 ± 47.43 ↓
486406 ncm-dbt-01 584790 500 228 35 237 +141.44 ± 11.55 0 1 58 188 3 +346.12 ± 45.18 ↓
486405 ncm-dbt-05 601313 500 228 40 232 +137.37 ± 12.76 1 0 65 178 6 +324.17 ± 42.47 ↓
486404 ncm-dbt-03 584958 500 228 38 234 +138.99 ± 12.09 0 0 66 178 6 +327.17 ± 41.95 ↓
486403 ncm-dbt-04 568991 500 240 31 229 +154.7 ± 10.74 0 0 45 201 4 +401.91 ± 51.63 ↓
486402 ncm-dbt-02 587707 500 240 37 223 +149.68 ± 14.24 0 4 54 177 15 +339.63 ± 47.1 ↓
486401 ncm-dbt-01 582318 500 227 35 238 +140.62 ± 10.9 0 0 59 190 1 +349.43 ± 44.59 ↓
486400 ncm-dbt-05 604262 500 232 35 233 +144.71 ± 11.38 0 0 57 189 4 +356.21 ± 45.43 ↓
486399 ncm-dbt-03 585126 500 230 34 236 +143.89 ± 11.65 0 2 53 192 3 +356.21 ± 47.52 ↓
486398 ncm-dbt-04 567324 500 226 37 237 +138.18 ± 12.53 0 1 66 176 7 +321.19 ± 42.12 ↓
486397 ncm-dbt-02 590396 500 233 33 234 +147.19 ± 11.92 0 3 48 195 4 +366.78 ± 50.11 ↓
486396 ncm-dbt-01 583782 500 226 36 238 +138.99 ± 12.9 0 3 61 179 7 +324.17 ± 44.15 ↓
486395 ncm-dbt-05 604892 500 232 37 231 +143.07 ± 12.75 0 4 53 187 6 +342.85 ± 47.56 ↓
486394 ncm-dbt-03 585928 500 231 35 234 +143.89 ± 12.3 0 3 53 189 5 +349.43 ± 47.56 ↓
486393 ncm-dbt-04 568872 500 239 41 220 +145.54 ± 13.45 0 1 63 173 13 +330.23 ± 43.2 ↓
486392 ncm-dbt-02 585970 500 226 25 249 +148.02 ± 12.32 0 2 52 189 7 +359.68 ± 48.01 ↓
486391 ncm-dbt-01 583572 500 231 25 244 +152.18 ± 10.92 0 1 45 201 3 +393.6 ± 51.79 ↓
486390 ncm-dbt-05 602070 500 227 32 241 +143.07 ± 11.91 0 2 55 189 4 +349.43 ± 46.59 ↓
486389 ncm-dbt-03 584159 500 211 44 245 +120.67 ± 12.78 0 4 78 165 3 +273.0 ± 38.74 ↓
486388 ncm-dbt-04 567918 500 230 30 240 +147.19 ± 12.14 0 2 52 190 6 +359.68 ± 48.01 ↓
486387 ncm-dbt-01 583237 500 231 33 236 +145.54 ± 11.79 0 0 58 186 6 +352.79 ± 45.01 ↓
486386 ncm-dbt-05 624069 500 236 38 226 +145.54 ± 12.22 0 2 54 188 6 +352.8 ± 47.05 ↓
486385 ncm-dbt-02 593066 500 220 30 250 +138.99 ± 12.3 0 3 58 185 4 +333.32 ± 45.35 ↓
486384 ncm-dbt-03 582110 500 233 36 231 +144.71 ± 12.26 0 2 55 187 6 +349.43 ± 46.59 ↓
486383 ncm-dbt-04 567047 500 226 36 238 +138.99 ± 12.7 0 3 60 181 6 +327.18 ± 44.54 ↓
486382 ncm-dbt-01 584327 500 227 40 233 +136.56 ± 12.98 0 4 61 179 6 +318.25 ± 44.18 ↓
486381 ncm-dbt-02 590524 500 226 41 233 +134.95 ± 12.64 0 4 61 181 4 +318.25 ± 44.18 ↓
486380 ncm-dbt-05 602427 500 225 36 239 +138.18 ± 12.33 0 3 59 184 4 +330.23 ± 44.94 ↓
486379 ncm-dbt-03 585379 500 217 28 255 +138.18 ± 11.48 0 0 64 183 3 +333.32 ± 42.66 ↓
486378 ncm-dbt-04 566770 500 232 49 219 +133.34 ± 13.45 0 4 67 171 8 +301.33 ± 42.04 ↓
486377 ncm-dbt-01 584537 500 225 40 235 +134.95 ± 12.44 0 4 60 183 3 +321.19 ± 44.57 ↓
486376 ncm-dbt-02 590653 500 228 39 233 +138.18 ± 12.73 0 2 64 177 7 +321.19 ± 42.95 ↓
486375 ncm-dbt-03 585464 500 234 42 224 +140.62 ± 12.23 0 1 62 181 6 +333.32 ± 43.57 ↓
486374 ncm-dbt-05 617105 500 232 35 233 +144.71 ± 12.05 0 2 54 189 5 +352.8 ± 47.05 ↓

Commit

Commit ID 400c23f978897b34d4a5ed5c4c6c6f858d40f976
Author Carlos Esparza
Date 2026-10-11 12:59:52 UTC
Only generate legal moves in movegen The main changes here are by @rhirsch123. I have added some speedups and identified some more code that can be removed if we only ever deal with legal moves. The only(?) important implementation detail of legal move generation is that we now keep a bitboard of squares threatened by `~sideToMove` in `Position`, to help with king move generation. For speed and conciseness, we include x-ray threats through the `sideToMove`-colored king (this only matters when `sideToMove` is in check). As a bonus, these bitboards get re-used in quiet move scoring. Things that can be simplified once we have legal movegen: - We can merge the `legal()` and `pseudo_legal()` methods - Since we have threat bitboards, we no longer need `attackers_to_exist()` in `legal()`, so we can remove that method - We don't need a separate `generate<LEGAL>` method that performs legality checks - This also means the `NON_EVASIONS` movegen target becomes redundant The change to `ray_pass_bb()` is outdated but should still be beneficial for performance. passed gainer STC: https://tests.stockfishchess.org/tests/view/6ac2e35d5349156614d93a31 LLR: 2.93 (-2.94,2.94) <0.00,2.00> Total: 207232 W: 52791 L: 52245 D: 102196 Ptnml(0-2): 279, 23887, 54714, 24481, 255 passed non-reg LTC: https://tests.stockfishchess.org/tests/view/6ac2e3785349156614d93a33 LLR: 2.94 (-2.94,2.94) <-1.75,0.25> Total: 163212 W: 42028 L: 41955 D: 79229 Ptnml(0-2): 39, 17157, 47137, 17238, 35 `speedtest 1 256 60` by @Torom: ``` sf_base = 2282060 +/- 6930 (95%) sf_test = 2332255 +/- 5742 (95%) diff = 50195 +/- 10399 (95%) speedup = 2.19956% +/- 0.456% (95%) Probability candidate is faster than baseline by at least 0.00%: 100.00% 0.25%: 100.00% 0.50%: 100.00% 1.00%: 100.00% ``` A previous benchmark with shorter speedtests IIUC obtained `0.91296% +/- 0.108%`. The number of saved LOC is very modest, but the extra code is mostly helper functions, so I would call this a big conceptual simplification. closes https://github.com/official-stockfish/Stockfish/pull/7214 Bench: 1648623 Co-authored-by: Ryan Hirsch <hirsch28@purdue.edu>
Copyright 2011–2026 Next Chess Move LLC