Dev Builds » 20221016-0949

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 14. 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 10:00:17 1277106 3354 1330 389 1635 +100.16 ± 5.42 0 73 613 968 23 +213.56 ± 13.75
ncm-dbt-02 09:56:22 1299283 3314 1299 370 1645 +100.07 ± 5.69 3 83 582 960 29 +212.35 ± 14.13
ncm-dbt-03 09:59:00 1306676 3344 1350 397 1597 +101.83 ± 5.56 1 73 604 960 34 +214.97 ± 13.86
ncm-dbt-04 09:53:07 1303335 3314 1276 401 1637 +93.96 ± 5.6 2 87 624 922 22 +198.34 ± 13.64
ncm-dbt-05 09:59:56 1301565 3340 1298 366 1676 +99.59 ± 5.48 2 74 605 968 21 +213.21 ± 13.85
ncm-dbt-06 09:59:01 1303405 3334 1340 363 1631 +104.89 ± 5.57 2 74 564 999 28 +225.22 ± 14.35
20000 7893 2286 9821 +100.08 ± 2.27 10 464 3592 5777 157 +212.86 ± 5.68

Test Detail

ID Host Base NPS Games WLD Standard Elo Ptnml(0-2) Gamepair Elo CLI PGN
187081 ncm-dbt-04 1218354 152 55 19 78 +83.87 ± 25.59 0 4 33 38 1 +173.02 ± 59.92 ↓
187080 ncm-dbt-06 1214216 166 67 20 79 +101.13 ± 24.64 0 5 26 52 0 +223.06 ± 68.26 ↓
187079 ncm-dbt-02 1249593 164 67 17 80 +109.4 ± 25.76 0 4 26 50 2 +232.99 ± 68.49 ↓
187078 ncm-dbt-05 1217464 168 65 22 81 +90.94 ± 24.97 0 5 32 46 1 +190.85 ± 61.28 ↓
187077 ncm-dbt-03 1229128 174 68 18 88 +102.72 ± 24.66 0 5 28 53 1 +221.5 ± 65.74 ↓
187076 ncm-dbt-01 1206810 176 71 23 82 +97.21 ± 23.62 0 3 36 47 2 +201.54 ± 57.23 ↓
187075 ncm-dbt-04 1243173 500 188 54 258 +95.44 ± 14.4 0 15 88 145 2 +204.07 ± 36.53 ↓
187074 ncm-dbt-03 1229545 500 207 52 241 +111.37 ± 15.0 0 10 85 145 10 +230.16 ± 37.2 ↓
187073 ncm-dbt-05 1224153 500 196 51 253 +103.73 ± 14.03 0 13 80 156 1 +228.08 ± 38.34 ↓
187072 ncm-dbt-02 1244987 500 199 53 248 +104.49 ± 14.04 0 11 85 151 3 +226.0 ± 37.21 ↓
187071 ncm-dbt-06 1243757 500 190 56 254 +95.44 ± 14.7 1 14 87 146 2 +206.01 ± 36.74 ↓
187070 ncm-dbt-01 1132617 500 208 62 230 +104.49 ± 12.53 0 5 94 151 0 +232.26 ± 35.0 ↓
186817 ncm-dbt-02 1220862 500 203 57 240 +104.49 ± 15.96 2 15 74 153 6 +223.94 ± 39.6 ↓
186816 ncm-dbt-05 1220077 500 191 56 253 +96.19 ± 14.56 0 14 91 141 4 +202.15 ± 35.92 ↓
186815 ncm-dbt-03 1216882 500 204 57 239 +105.25 ± 14.51 0 10 90 143 7 +219.87 ± 36.1 ↓
186814 ncm-dbt-01 1203885 500 209 56 235 +109.83 ± 13.73 0 9 82 156 3 +240.82 ± 37.9 ↓
186813 ncm-dbt-06 1199541 500 205 54 241 +108.3 ± 14.52 0 11 83 150 6 +230.16 ± 37.67 ↓
186812 ncm-dbt-04 1238340 500 188 64 248 +88.0 ± 14.85 1 16 93 138 2 +187.16 ± 35.51 ↓
186811 ncm-dbt-02 1240353 500 198 54 248 +102.97 ± 14.18 0 11 88 147 4 +219.87 ± 36.54 ↓
186810 ncm-dbt-03 1232352 500 207 72 221 +96.19 ± 14.71 1 13 89 144 3 +206.01 ± 36.33 ↓
186809 ncm-dbt-05 1228359 500 199 60 241 +99.2 ± 15.19 0 15 88 140 7 +204.07 ± 36.53 ↓
186808 ncm-dbt-01 1180315 500 193 62 245 +93.2 ± 14.66 0 13 99 132 6 +190.85 ± 34.36 ↓
186807 ncm-dbt-06 1219714 500 207 59 234 +106.01 ± 14.04 1 9 83 155 2 +234.38 ± 37.67 ↓
186806 ncm-dbt-04 1224992 500 190 56 254 +95.44 ± 13.31 0 8 102 138 2 +204.07 ± 33.61 ↓
186805 ncm-dbt-03 1218013 500 194 64 242 +92.45 ± 14.94 0 16 93 136 5 +190.85 ± 35.52 ↓
186804 ncm-dbt-02 1230002 500 185 54 261 +93.2 ± 14.06 0 11 101 134 4 +194.57 ± 33.94 ↓
186803 ncm-dbt-01 1241491 500 191 56 253 +96.19 ± 13.8 0 11 95 142 2 +206.01 ± 35.09 ↓
186802 ncm-dbt-05 1223663 500 198 44 258 +110.6 ± 13.73 1 7 81 159 2 +247.41 ± 38.13 ↓
186801 ncm-dbt-06 1238707 500 202 56 242 +104.49 ± 14.35 0 11 87 147 5 +221.9 ± 36.76 ↓
186800 ncm-dbt-04 1200282 500 188 56 256 +93.95 ± 13.77 0 13 92 145 0 +204.07 ± 35.72 ↓
186799 ncm-dbt-03 1234326 500 204 55 241 +106.77 ± 13.89 0 8 90 147 5 +228.08 ± 36.04 ↓
186798 ncm-dbt-02 1228830 500 197 62 241 +96.19 ± 14.71 1 10 98 135 6 +200.24 ± 34.5 ↓
186797 ncm-dbt-01 1144689 500 196 47 257 +106.77 ± 14.67 0 14 77 155 4 +230.16 ± 39.03 ↓
186796 ncm-dbt-05 1239404 500 186 56 258 +92.45 ± 13.59 0 10 102 136 2 +196.45 ± 33.71 ↓
186795 ncm-dbt-04 1233802 500 191 59 250 +93.95 ± 14.96 1 13 94 137 5 +196.45 ± 35.33 ↓
186794 ncm-dbt-06 1227134 500 199 49 252 +107.54 ± 14.52 0 10 87 146 7 +226.0 ± 36.75 ↓
186793 ncm-dbt-03 1224680 500 197 57 246 +99.95 ± 13.68 0 9 95 143 3 +213.85 ± 35.03 ↓
186792 ncm-dbt-05 1210576 500 191 58 251 +94.7 ± 13.78 1 7 103 136 3 +202.15 ± 33.42 ↓
186791 ncm-dbt-02 1248992 500 199 57 244 +101.46 ± 14.17 0 11 90 145 4 +215.85 ± 36.11 ↓
186790 ncm-dbt-06 1215125 500 201 48 251 +109.83 ± 14.06 0 10 81 155 4 +238.66 ± 38.15 ↓
186789 ncm-dbt-04 1237091 500 206 72 222 +95.44 ± 15.41 0 17 89 137 7 +194.57 ± 36.29 ↓
186788 ncm-dbt-01 1188471 500 202 63 235 +99.2 ± 14.9 0 14 89 141 6 +206.01 ± 36.33 ↓
175479 ncm-dbt-04 1432142 12 4 1 7 +88.66 ± 75.75 0 0 3 3 0 +190.73 ± 238.47 ↓
175478 ncm-dbt-06 1451162 18 9 3 6 +120.35 ± 108.88 0 1 2 5 1 +217.63 ± 581.07 ↓
175477 ncm-dbt-05 1436602 22 8 3 11 +80.33 ± 54.5 0 0 6 5 0 +170.33 ± 140.48 ↓
175476 ncm-dbt-03 1447201 20 10 2 8 +147.12 ± 52.02 0 0 2 8 0 +381.52 ± 492.07 ↓
175474 ncm-dbt-01 1434298 28 11 2 15 +115.74 ± 49.12 0 0 5 9 0 +265.03 ± 178.38 ↓
175473 ncm-dbt-02 1440355 50 19 3 28 +115.22 ± 47.81 0 2 5 18 0 +263.42 ± 170.28 ↓
175472 ncm-dbt-04 1439342 50 24 5 21 +138.97 ± 34.19 0 0 6 19 0 +346.06 ± 166.42 ↓
175471 ncm-dbt-06 1439780 50 19 7 24 +85.04 ± 41.74 0 1 11 13 0 +181.7 ± 107.11 ↓
175470 ncm-dbt-05 1442247 50 19 7 24 +85.04 ± 51.0 0 3 7 15 0 +181.7 ± 134.48 ↓
175469 ncm-dbt-03 1448522 50 16 5 29 +77.7 ± 41.12 0 1 12 12 0 +164.07 ± 100.82 ↓
175468 ncm-dbt-01 1434158 50 16 5 29 +77.69 ± 35.69 0 0 14 11 0 +164.04 ± 86.61 ↓
175467 ncm-dbt-02 1442244 50 18 8 24 +70.43 ± 49.53 0 3 9 13 0 +147.19 ± 118.91 ↓
175466 ncm-dbt-04 1433722 50 23 7 20 +115.19 ± 47.81 0 0 11 12 2 +219.84 ± 104.3 ↓
175465 ncm-dbt-05 1440071 50 24 4 22 +147.15 ± 40.0 0 0 6 18 1 +346.06 ± 166.42 ↓
175464 ncm-dbt-03 1445158 50 22 8 20 +99.93 ± 36.88 0 0 11 14 0 +219.84 ± 104.3 ↓
175463 ncm-dbt-06 1437897 50 22 8 20 +99.93 ± 47.48 0 1 10 13 1 +200.24 ± 114.26 ↓
175462 ncm-dbt-01 1440646 50 15 4 31 +77.7 ± 41.12 0 1 12 12 0 +164.07 ± 100.82 ↓
175461 ncm-dbt-02 1446615 50 14 5 31 +63.22 ± 56.43 0 5 6 14 0 +130.94 ± 132.75 ↓
175460 ncm-dbt-04 1435449 50 19 8 23 +77.69 ± 45.93 0 1 13 10 1 +147.19 ± 95.14 ↓
175459 ncm-dbt-06 1450429 50 19 3 28 +115.22 ± 42.56 0 1 7 17 0 +263.42 ± 145.5 ↓
175458 ncm-dbt-01 1440790 50 18 9 23 +63.22 ± 48.63 0 3 10 12 0 +130.94 ± 112.59 ↓
175457 ncm-dbt-03 1447639 50 21 7 22 +99.94 ± 42.49 0 1 9 15 0 +219.87 ± 122.63 ↓
175456 ncm-dbt-05 1434601 50 21 5 24 +115.2 ± 36.62 0 0 9 16 0 +263.38 ± 120.78 ↓

Commit

Commit ID d6b6360ff5dcdffa141f50d0a81d0be7c6324c71
Author Stéphane Nicolet
Date 2022-10-16 09:49:07 UTC
Tweak the formula for NNUE complexity Joint work by Ofek Shochat and Stéphane Nicolet. passed STC: LLR: 2.95 (-2.94,2.94) <0.00,2.00> Total: 93288 W: 24996 L: 24601 D: 43691 Ptnml(0-2): 371, 10263, 24989, 10642, 379 https://tests.stockfishchess.org/tests/view/63448f4f4bc7650f07541987 passed LTC: LLR: 2.94 (-2.94,2.94) <0.50,2.50> Total: 84168 W: 22771 L: 22377 D: 39020 Ptnml(0-2): 47, 8181, 25234, 8575, 47 https://tests.stockfishchess.org/tests/view/6345186d4bc7650f07542fbd ================ It seems there are two effects with this patch: effect A : If Stockfish is winning at root, we have optimism > 0 for all leaves in the search tree where Stockfish is to move. There, if (psq - nnue) > 0 (ie if the advantage is more materialistic than positional), then the product D = optimism * (psq - nnue) will be positive, nnueComplexity will increase, and the eval will increase from SF point of view. So the effect A is that if Stockfish is winning at root, she will slightly favor in the search tree (in other words, search more) the positions where she can convert her advantage via materialist means. effect B : If Stockfish is losing at root, we have optimism > 0 for all leaves in the search tree where the opponent is to move. There, if (psq - nnue) < 0 (ie if the opponent advantage is more positional than materialistic), then the product D = optimism * (psq-nnue) will be negative, nnueComplexity will decrease, and the eval will decrease from the opponent point of view. So the effect B is that Stockfish will slightly favor in the search tree (search more) the branches where she can defend by slowly reducing the opponent positional advantage. ================= closes https://github.com/official-stockfish/Stockfish/pull/4195 bench: 4673898
Copyright 2011–2024 Next Chess Move LLC
Feedback & Support Terms of Service Privacy Policy Desktop Version