| Windows x64 for Haswell CPUs Windows x64 for modern computers + AVX2 Windows x64 for modern computers Windows x64 + SSSE3 Windows x64 Linux x64 for Haswell CPUs Linux x64 for modern computers + AVX2 Linux x64 for modern computers Linux x64 + SSSE3 Linux x64 | Author: Michael Chaly
Date: Wed May 31 09:06:30 2023 +0200 Timestamp: 1685516790 Small simplification for probcut in check Remove depth condition from there as not longer needed. Passed STC: https://tests.stockfishchess.org/tests/view/647367cad29264e4cfa753e6 LLR: 2.96 (-2.94,2.94) <-1.75,0.25> Total: 254336 W: 67830 L: 67847 D: 118659 Elo -0.02 Ptnml(0-2): 580, 28181, 69697, 28096, 614 Passed LTC: https://tests.stockfishchess.org/tests/view/647576184a36543c4c9f3af7 LLR: 2.95 (-2.94,2.94) <-1.75,0.25> Total: 80706 W: 22048 L: 21898 D: 36760 Elo +0.65 Ptnml(0-2): 28, 7721, 24712, 7857, 35 closes https://github.com/official-stockfish/Stockfish/pull/4594 bench 2381945 see source |
| Windows x64 for Haswell CPUs Windows x64 for modern computers + AVX2 Windows x64 for modern computers Windows x64 + SSSE3 Windows x64 Linux x64 for Haswell CPUs Linux x64 for modern computers + AVX2 Linux x64 for modern computers Linux x64 + SSSE3 Linux x64 | Author: Linmiao Xu
Date: Wed May 31 08:54:38 2023 +0200 Timestamp: 1685516078 Simplify nnue eval complexity calculation Remove a multiplier when blending nnue complexity with semi-classical complexity. Passed non-regression STC: https://tests.stockfishchess.org/tests/view/6473a71dd29264e4cfa75839 LLR: 2.95 (-2.94,2.94) <-1.75,0.25> Total: 124768 W: 33180 L: 33060 D: 58528 Elo +0.33 Ptnml(0-2): 314, 13797, 34030, 13941, 302 Passed non-regression LTC: https://tests.stockfishchess.org/tests/view/6474af3dd29264e4cfa768f4 LLR: 2.96 (-2.94,2.94) <-1.75,0.25> Total: 108180 W: 29008 L: 28884 D: 50288 Elo +0.40 Ptnml(0-2): 29, 10420, 33075, 10530, 36 closes https://github.com/official-stockfish/Stockfish/pull/4592 bench 2316827 see source |
| Windows x64 for Haswell CPUs Windows x64 for modern computers + AVX2 Windows x64 for modern computers Windows x64 + SSSE3 Windows x64 Linux x64 for Haswell CPUs Linux x64 for modern computers + AVX2 Linux x64 for modern computers Linux x64 + SSSE3 Linux x64 | Author: Linmiao Xu
Date: Wed May 31 08:51:22 2023 +0200 Timestamp: 1685515882 Update NNUE architecture to SFNNv6 with larger L1 size of 1536 Created by training a new net from scratch with L1 size increased from 1024 to 1536. Thanks to Vizvezdenec for the idea of exploring larger net sizes after recent training data improvements. A new net was first trained with lambda 1.0 and constant LR 8.75e-4. Then a strong net from a later epoch in the training run was chosen for retraining with start-lambda 1.0 and initial LR 4.375e-4 decaying with gamma 0.995. Retraining was performed a total of 3 times, for this 4-step process: 1. 400 epochs, lambda 1.0 on filtered T77+T79 v6 deduplicated data 2. 800 epochs, end-lambda 0.75 on T60T70wIsRightFarseerT60T74T75T76.binpack 3. 800 epochs, end-lambda 0.75 and early-fen-skipping 28 on the master dataset 4. 800 epochs, end-lambda 0.7 and early-fen-skipping 28 on the master dataset In the training sequence that reached the new nn-8d69132723e2.nnue net, the epochs used for the 3x retraining runs were: 1. epoch 379 trained on T77T79-filter-v6-dd.min.binpack 2. epoch 679 trained on T60T70wIsRightFarseerT60T74T75T76.binpack 3. epoch 799 trained on the master dataset For training from scratch: python3 easy_train.py \ --experiment-name new-L1-1536-T77T79-filter-v6dd \ --training-dataset /data/T77T79-filter-v6-dd.min.binpack \ --max_epoch 400 \ --lambda 1.0 \ --start-from-engine-test-net False \ --engine-test-branch linrock/Stockfish/L1-1536 \ --nnue-pytorch-branch linrock/Stockfish/misc-fixes-L1-1536 \ --tui False \ --gpus "0," \ --seed $RANDOM Retraining commands were similar to each other. For the 3rd retraining run: python3 easy_train.py \ --experiment-name L1-1536-T77T79-v6dd-Re1-LeelaFarseer-Re2-masterDataset-Re3-sameData \ --training-dataset /data/leela96-dfrc99-v2-T60novdecT80juntonovjanfebT79aprmayT78jantosepT77dec-v6dd.binpack \ --early-fen-skipping 28 \ --max_epoch 800 \ --start-lambda 1.0 \ --end-lambda 0.7 \ --lr 4.375e-4 \ --gamma 0.995 \ --start-from-engine-test-net False \ --start-from-model /data/L1-1536-T77T79-v6dd-Re1-LeelaFarseer-Re2-masterDataset-nn-epoch799.nnue \ --engine-test-branch linrock/Stockfish/L1-1536 \ --nnue-pytorch-branch linrock/nnue-pytorch/misc-fixes-L1-1536 \ --tui False \ --gpus "0," \ --seed $RANDOM The T77+T79 data used is a subset of the master dataset available at: https://robotmoon.com/nnue-training-data/ T60T70wIsRightFarseerT60T74T75T76.binpack is available at: https://drive.google.com/drive/folders/1S9-ZiQa_3ApmjBtl2e8SyHxj4zG4V8gG Local elo at 25k nodes per move vs. nn-e1fb1ade4432.nnue (L1 size 1024): nn-epoch759.nnue : 26.9 +/- 1.6 Failed STC https://tests.stockfishchess.org/tests/view/64742485d29264e4cfa75f97 LLR: -2.94 (-2.94,2.94) <0.00,2.00> Total: 13728 W: 3588 L: 3829 D: 6311 Elo -6.10 Ptnml(0-2): 71, 1661, 3610, 1482, 40 Failing LTC https://tests.stockfishchess.org/tests/view/64752d7c4a36543c4c9f3618 LLR: -1.91 (-2.94,2.94) <0.50,2.50> Total: 35424 W: 9522 L: 9603 D: 16299 Elo -0.79 Ptnml(0-2): 24, 3579, 10585, 3502, 22 Passed VLTC 180+1.8 https://tests.stockfishchess.org/tests/view/64752df04a36543c4c9f3638 LLR: 2.95 (-2.94,2.94) <0.50,2.50> Total: 47616 W: 13174 L: 12863 D: 21579 Elo +2.27 Ptnml(0-2): 13, 4261, 14952, 4566, 16 Passed VLTC SMP 60+0.6 th 8 https://tests.stockfishchess.org/tests/view/647446ced29264e4cfa761e5 LLR: 2.94 (-2.94,2.94) <0.50,2.50> Total: 19942 W: 5694 L: 5451 D: 8797 Elo +4.23 Ptnml(0-2): 6, 1504, 6707, 1749, 5 closes https://github.com/official-stockfish/Stockfish/pull/4593 bench 2222567 see source |
| Windows x64 for Haswell CPUs Windows x64 for modern computers + AVX2 Windows x64 for modern computers Windows x64 + SSSE3 Windows x64 Linux x64 for Haswell CPUs Linux x64 for modern computers + AVX2 Linux x64 for modern computers Linux x64 + SSSE3 Linux x64 | Author: windfishballad
Date: Sun May 28 20:05:50 2023 +0200 Timestamp: 1685297150 Removed quadratic term in optimism Remove term which is quadratic in optimism in the eval. Simplifies and should also remove the bias towards side to move making the eval better for analysis. STC: https://tests.stockfishchess.org/tests/view/6470a9d8c29e0d4352b0bca5 LLR: 2.94 (-2.94,2.94) <-1.75,0.25> Total: 154432 W: 41127 L: 41040 D: 72265 Elo +0.20 Ptnml(0-2): 380, 17094, 42190, 17163, 389 LTC: https://tests.stockfishchess.org/tests/view/6471e9b3e549d9cf2fb219ef LLR: 2.94 (-2.94,2.94) <-1.75,0.25> Total: 127926 W: 34474 L: 34369 D: 59083 Elo +0.29 Ptnml(0-2): 43, 12505, 38776, 12582, 57 closes https://github.com/official-stockfish/Stockfish/pull/4590 Bench: 2541211 see source |
| Windows x64 for Haswell CPUs Windows x64 for modern computers + AVX2 Windows x64 for modern computers Windows x64 + SSSE3 Windows x64 Linux x64 for Haswell CPUs Linux x64 for modern computers + AVX2 Linux x64 for modern computers Linux x64 + SSSE3 Linux x64 | Author: Joost VandeVondele
Date: Sun May 28 20:03:33 2023 +0200 Timestamp: 1685297013 Update CPU contributors list update CPU contributors list, the previous update was a couple of months ago, and unfortunately, was not quite accurate for the number of games played. This version is based clean calculation from the DB and an updated script that tracks things (see https://github.com/glinscott/fishtest/pull/1702). closes https://github.com/official-stockfish/Stockfish/pull/4589 No functional change see source |
| Windows x64 for Haswell CPUs Windows x64 for modern computers + AVX2 Windows x64 for modern computers Windows x64 + SSSE3 Windows x64 Linux x64 for Haswell CPUs Linux x64 for modern computers + AVX2 Linux x64 for modern computers Linux x64 + SSSE3 Linux x64 | Author: cj5716
Date: Thu May 25 20:49:11 2023 +0200 Timestamp: 1685040551 Remove ss->ttHit condition where ttValue != VALUE_NONE Simplification is done at 3 separate places in the code. Thanks to peregrineshahin for helping me find 2 of such places. (See original PR #4584) Passed non-regression test LLR: 2.93 (-2.94,2.94) <-1.75,0.25> Total: 120256 W: 32204 L: 32085 D: 55967 Elo +0.34 Ptnml(0-2): 292, 12473, 34483, 12584, 296 https://tests.stockfishchess.org/tests/view/646f045968661bfd984325e3 closes https://github.com/official-stockfish/Stockfish/pull/4587 No functional change see source |
| Windows x64 for Haswell CPUs Windows x64 for modern computers + AVX2 Windows x64 for modern computers Windows x64 + SSSE3 Windows x64 Linux x64 for Haswell CPUs Linux x64 for modern computers + AVX2 Linux x64 for modern computers Linux x64 + SSSE3 Linux x64 | Author: FauziAkram
Date: Thu May 25 20:42:43 2023 +0200 Timestamp: 1685040163 Simplify Futility pruning for captures Passed STC: LLR: 2.95 (-2.94,2.94) <-1.75,0.25> Total: 412928 W: 109433 L: 109620 D: 193875 Elo -0.16 Ptnml(0-2): 1071, 45929, 112650, 45744, 1070 https://tests.stockfishchess.org/tests/view/6468eac40db5177f2b76ef4d Passed LTC: LLR: 2.95 (-2.94,2.94) <-1.75,0.25> Total: 190200 W: 51465 L: 51420 D: 87315 Elo +0.08 Ptnml(0-2): 58, 18585, 57788, 18592, 77 https://tests.stockfishchess.org/tests/view/646b66520db5177f2b772a84 closes https://github.com/official-stockfish/Stockfish/pull/4583 bench: 2486604 see source |
| Windows x64 for Haswell CPUs Windows x64 for modern computers + AVX2 Windows x64 for modern computers Windows x64 + SSSE3 Windows x64 Linux x64 for Haswell CPUs Linux x64 for modern computers + AVX2 Linux x64 for modern computers Linux x64 + SSSE3 Linux x64 | Author: FauziAkram
Date: Thu May 25 20:40:09 2023 +0200 Timestamp: 1685040009 Simplify Prune moves with negative SEE Passed STC: LLR: 2.95 (-2.94,2.94) <-1.75,0.25> Total: 57760 W: 15472 L: 15286 D: 27002 Elo +1.12 Ptnml(0-2): 123, 6025, 16430, 6147, 155 https://tests.stockfishchess.org/tests/view/6468eb6b0db5177f2b76ef62 Passed LTC: LLR: 2.95 (-2.94,2.94) <-1.75,0.25> Total: 93966 W: 25274 L: 25141 D: 43551 Elo +0.49 Ptnml(0-2): 33, 8498, 29792, 8623, 37 https://tests.stockfishchess.org/tests/view/6469570b0db5177f2b76f81b closes: https://github.com/official-stockfish/Stockfish/pull/4579 Bench: 2304063 see source |
| Windows x64 for Haswell CPUs Windows x64 for modern computers + AVX2 Windows x64 for modern computers Windows x64 + SSSE3 Windows x64 Linux x64 for Haswell CPUs Linux x64 for modern computers + AVX2 Linux x64 for modern computers Linux x64 + SSSE3 Linux x64 | Author: Muzhen Gaming
Date: Thu May 25 20:37:53 2023 +0200 Timestamp: 1685039873 Simplify delta calculation in aspiration window Simplification STC: https://tests.stockfishchess.org/tests/view/6468cb200db5177f2b76ecbb LLR: 2.93 (-2.94,2.94) <-1.75,0.25> Total: 220416 W: 58503 L: 58487 D: 103426 Elo +0.03 Ptnml(0-2): 596, 24384, 60188, 24488, 552 Simplification LTC: https://tests.stockfishchess.org/tests/view/646a15840db5177f2b770704 LLR: 2.94 (-2.94,2.94) <-1.75,0.25> Total: 177756 W: 47882 L: 47825 D: 82049 Elo +0.11 Ptnml(0-2): 55, 17430, 53858, 17473, 62 closes https://github.com/official-stockfish/Stockfish/pull/4581 Bench: 2304063 see source |
| Windows x64 for Haswell CPUs Windows x64 for modern computers + AVX2 Windows x64 for modern computers Windows x64 + SSSE3 Windows x64 Linux x64 for Haswell CPUs Linux x64 for modern computers + AVX2 Linux x64 for modern computers Linux x64 + SSSE3 Linux x64 | Author: candirufish
Date: Thu May 25 20:35:13 2023 +0200 Timestamp: 1685039713 More Depth Reduction Reduce more for depth > 3 and depth < 12 LTC: https://tests.stockfishchess.org/tests/view/646c5abbd1f14fd69a6f2fab LLR: 2.95 (-2.94,2.94) <0.50,2.50> Total: 197280 W: 53405 L: 52797 D: 91078 Elo +1.07 Ptnml(0-2): 62, 19025, 59886, 19577, 90 STC: https://tests.stockfishchess.org/tests/view/646bee71d1f14fd69a6f259d LLR: 2.95 (-2.94,2.94) <0.00,2.00> Total: 100832 W: 26861 L: 26466 D: 47505 Elo +1.36 Ptnml(0-2): 240, 10985, 27622, 11278, 291 https://github.com/official-stockfish/Stockfish/pull/4585 bench: 2276617 see source |
| Windows x64 for Haswell CPUs Windows x64 for modern computers + AVX2 Windows x64 for modern computers Windows x64 + SSSE3 Windows x64 Linux x64 for Haswell CPUs Linux x64 for modern computers + AVX2 Linux x64 for modern computers Linux x64 + SSSE3 Linux x64 | Author: Michael Chaly
Date: Sat May 20 10:19:03 2023 +0200 Timestamp: 1684570743 Small simplification in low depth pruning Uncap low depth pruning lmr depth. It's anyway capped for most cases apart from futility pruning for captures - removes one std::min call. Passed STC: https://tests.stockfishchess.org/tests/view/645e8fa6d55cccb2e64225a1 LLR: 2.95 (-2.94,2.94) <-1.75,0.25> Total: 184064 W: 49039 L: 48982 D: 86043 Elo +0.11 Ptnml(0-2): 462, 20353, 50349, 20402, 466 Passed LTC: https://tests.stockfishchess.org/tests/view/645f4d48d55cccb2e6423335 LLR: 2.96 (-2.94,2.94) <-1.75,0.25> Total: 83886 W: 22613 L: 22465 D: 38808 Elo +0.61 Ptnml(0-2): 31, 8090, 25546, 8252, 24 closes https://github.com/official-stockfish/Stockfish/pull/4566 bench 3201883 see source |
| Windows x64 for Haswell CPUs Windows x64 for modern computers + AVX2 Windows x64 for modern computers Windows x64 + SSSE3 Windows x64 Linux x64 for Haswell CPUs Linux x64 for modern computers + AVX2 Linux x64 for modern computers Linux x64 + SSSE3 Linux x64 | Author: pb00067
Date: Sat May 20 10:16:04 2023 +0200 Timestamp: 1684570564 Simplify SEE verfication logic Passed STC https://tests.stockfishchess.org/tests/view/6461d51887f6567dd4df27d0 LLR: 2.94 (-2.94,2.94) <-1.75,0.25> Total: 177056 W: 47181 L: 47118 D: 82757 Elo +0.12 Ptnml(0-2): 456, 19381, 48792, 19442, 457 Passed LTC https://tests.stockfishchess.org/tests/view/64631a9287f6567dd4df4502 2.94 (-2.94,2.94) <-1.75,0.25> Total: 104346 W: 28062 L: 27935 D: 48349 Elo +0.42 Ptnml(0-2): 25, 10190, 31631, 10287, 40 closes https://github.com/official-stockfish/Stockfish/pull/4578 bench: 2903251 see source |
| Windows x64 for Haswell CPUs Windows x64 for modern computers + AVX2 Windows x64 for modern computers Windows x64 + SSSE3 Windows x64 Linux x64 for Haswell CPUs Linux x64 for modern computers + AVX2 Linux x64 for modern computers Linux x64 + SSSE3 Linux x64 | Author: xoto10
Date: Sat May 20 10:05:19 2023 +0200 Timestamp: 1684569919 Simplify optimism calculation This change removes one of the constants in the calculation of optimism. It also changes the 2 constants used with the scale value so that they are independent, instead of applying a constant to the scale and then adjusting it again when it is applied to the optimism. This might make the tuning of these constants cleaner and more reliable in the future. STC 10+0.1 (accidentally run as an Elo gainer: LLR: 2.93 (-2.94,2.94) <0.00,2.00> Total: 154080 W: 41119 L: 40651 D: 72310 Elo +1.06 Ptnml(0-2): 375, 16840, 42190, 17212, 423 https://tests.stockfishchess.org/tests/live_elo/64653eabf3b1a4e86c317f77 LTC 60+0.6: LLR: 2.95 (-2.94,2.94) <-1.75,0.25> Total: 217434 W: 58382 L: 58363 D: 100689 Elo +0.03 Ptnml(0-2): 66, 21075, 66419, 21088, 69 https://tests.stockfishchess.org/tests/live_elo/6465d077f3b1a4e86c318d6c closes https://github.com/official-stockfish/Stockfish/pull/4576 bench: 3190961 see source |
| Windows x64 for Haswell CPUs Windows x64 for modern computers + AVX2 Windows x64 for modern computers Windows x64 + SSSE3 Windows x64 Linux x64 for Haswell CPUs Linux x64 for modern computers + AVX2 Linux x64 for modern computers Linux x64 + SSSE3 Linux x64 | Author: Stefan Geschwentner
Date: Sat May 20 09:58:51 2023 +0200 Timestamp: 1684569531 Small simplification in history pruning. Remove the constant term of the history threshold which lowers the chance that pruning occurs. As compensation allow pruning at a slightly higher depth. Passed STC: https://tests.stockfishchess.org/tests/view/64634c9a87f6567dd4df4901 LLR: 2.95 (-2.94,2.94) <-1.75,0.25> Total: 101536 W: 27156 L: 27012 D: 47368 Elo +0.49 Ptnml(0-2): 266, 11165, 27772, 11289, 276 Passed LTC: https://tests.stockfishchess.org/tests/view/6463d68b17982fde89d2bc2b LLR: 2.94 (-2.94,2.94) <-1.75,0.25> Total: 32154 W: 8741 L: 8543 D: 14870 Elo +2.14 Ptnml(0-2): 8, 3093, 9687, 3271, 18 Passed LTC: retest on top of VLTC tuning PR 4571 because this changes the history depth factor (use this new factor here) https://tests.stockfishchess.org/tests/view/6467300e165c4b29ec0afd3f LLR: 2.94 (-2.94,2.94) <-1.75,0.25> Total: 99270 W: 26840 L: 26707 D: 45723 Elo +0.47 Ptnml(0-2): 36, 9753, 29928, 9878, 40 closes https://github.com/official-stockfish/Stockfish/pull/4578 Bench: 2984341 see source |
| Windows x64 for Haswell CPUs Windows x64 for modern computers + AVX2 Windows x64 for modern computers Windows x64 + SSSE3 Windows x64 Linux x64 for Haswell CPUs Linux x64 for modern computers + AVX2 Linux x64 for modern computers Linux x64 + SSSE3 Linux x64 | Author: Muzhen Gaming
Date: Sat May 20 09:21:46 2023 +0200 Timestamp: 1684567306 Search tuning at very long time control Many search parameter changes, tuned (https://tests.stockfishchess.org/tests/view/645e4c67d55cccb2e64220ff) at ~300k games @ VLTC (120+1.2). Failed STC: https://tests.stockfishchess.org/tests/view/6465fcd77968ca827c1410c2 LLR: -2.95 (-2.94,2.94) <0.00,2.00> Total: 33824 W: 8863 L: 9067 D: 15894 Elo -2.10 Ptnml(0-2): 89, 3833, 9266, 3641, 83 Neutral LTC: https://tests.stockfishchess.org/tests/view/646385ce87f6567dd4df4e37 Elo: -0.48 +-1.2 (95%) LOS: 22.2% Total: 60000 W: 16235 L: 16318 D: 27447 Elo -0.48 Ptnml(0-2): 27, 5831, 18366, 5750, 26 nElo: -1.08 +-2.8 (95%) PairsRatio: 0.99 Passed VLTC 180+1.8: https://tests.stockfishchess.org/tests/view/646385f787f6567dd4df4e3e LLR: 2.95 (-2.94,2.94) <0.50,2.50> Total: 126448 W: 34704 L: 34258 D: 57486 Elo +1.23 Ptnml(0-2): 9, 10970, 40825, 11406, 14 Passed VLTC SMP 60+0.6 8thread: https://tests.stockfishchess.org/tests/view/646628de884ce93b65df2ac9 LLR: 2.95 (-2.94,2.94) <0.50,2.50> Total: 59456 W: 16791 L: 16487 D: 26178 Elo +1.78 Ptnml(0-2): 5, 4473, 20467, 4779, 4 closes https://github.com/official-stockfish/Stockfish/pull/4574 Bench: 3347573 see source |
| Windows x64 for Haswell CPUs Windows x64 for modern computers + AVX2 Windows x64 for modern computers Windows x64 + SSSE3 Windows x64 Linux x64 for Haswell CPUs Linux x64 for modern computers + AVX2 Linux x64 for modern computers Linux x64 + SSSE3 Linux x64 | Author: Joost VandeVondele
Date: Sat May 20 09:16:46 2023 +0200 Timestamp: 1684567006 Update WLD model using data of May, recalibrate the WLD model. closes https://github.com/official-stockfish/Stockfish/pull/4577 No functional change see source |
| Windows x64 for Haswell CPUs Windows x64 for modern computers + AVX2 Windows x64 for modern computers Windows x64 + SSSE3 Windows x64 Linux x64 for Haswell CPUs Linux x64 for modern computers + AVX2 Linux x64 for modern computers Linux x64 + SSSE3 Linux x64 | Author: Michael Chaly
Date: Sun May 7 22:36:57 2023 +0200 Timestamp: 1683491817 Refine deeper post-lmr searches This patch improves logic conditions for performing deeper searches after passed LMR. Instead of exceeding alpha by some margin now it requires to exceed the current best value - which may be lower than alpha (but never bigger since we update alpha with bestvalue if it exceeds alpha). Passed STC: https://tests.stockfishchess.org/tests/view/6455f78008858de8313775b6 LLR: 2.94 (-2.94,2.94) <0.00,2.00> Total: 209344 W: 55993 L: 55448 D: 97903 Elo +0.90 Ptnml(0-2): 507, 22798, 57526, 23325, 516 Passed LTC: https://tests.stockfishchess.org/tests/view/64572d46eb75932ccfebff97 LLR: 2.94 (-2.94,2.94) <0.50,2.50> Total: 66288 W: 17867 L: 17514 D: 30907 Elo +1.85 Ptnml(0-2): 21, 6240, 20269, 6593, 21 closes https://github.com/official-stockfish/Stockfish/pull/4559 bench 3808503 see source |
| Windows x64 for Haswell CPUs Windows x64 for modern computers + AVX2 Windows x64 for modern computers Windows x64 + SSSE3 Windows x64 Linux x64 for Haswell CPUs Linux x64 for modern computers + AVX2 Linux x64 for modern computers Linux x64 + SSSE3 Linux x64 | Author: Michael Chaly
Date: Sun May 7 22:32:21 2023 +0200 Timestamp: 1683491541 Small cleanup In search remove one condition check and reorder conditions. Removes some code. Passed non-regression test: https://tests.stockfishchess.org/tests/view/64548fa06206ee34ebf853ad LLR: 2.95 (-2.94,2.94) <-1.75,0.25> Total: 282976 W: 75327 L: 75374 D: 132275 Elo -0.06 Ptnml(0-2): 604, 29673, 80995, 29598, 618 closes https://github.com/official-stockfish/Stockfish/pull/4557 No functional change see source |
| Windows x64 for Haswell CPUs Windows x64 for modern computers + AVX2 Windows x64 for modern computers Windows x64 + SSSE3 Windows x64 Linux x64 for Haswell CPUs Linux x64 for modern computers + AVX2 Linux x64 for modern computers Linux x64 + SSSE3 Linux x64 | Author: peregrineshahin
Date: Sun May 7 22:31:03 2023 +0200 Timestamp: 1683491463 Clean up after "Simplify away complexity in evaluation" closes https://github.com/official-stockfish/Stockfish/pull/4555 No functional change. see source |
| Windows x64 for Haswell CPUs Windows x64 for modern computers + AVX2 Windows x64 for modern computers Windows x64 + SSSE3 Windows x64 Linux x64 for Haswell CPUs Linux x64 for modern computers + AVX2 Linux x64 for modern computers Linux x64 + SSSE3 Linux x64 | Author: Michael Chaly
Date: Fri May 5 07:12:47 2023 +0200 Timestamp: 1683263567 Reduce more if current node has a lot of refuted moves. This patch refines idea of cutoff count - in master we reduce more if current node has at least 4 moves that are refuted by search, this patch increases this count by 1 if refutation happened without having a tt move. Passed STC: https://tests.stockfishchess.org/tests/view/645363c36206ee34ebf8191d LLR: 2.95 (-2.94,2.94) <0.00,2.00> Total: 67616 W: 18220 L: 17874 D: 31522 Elo +1.78 Ptnml(0-2): 142, 7346, 18504, 7656, 160 Passed LTC: https://tests.stockfishchess.org/tests/view/6453a0ea6206ee34ebf82796 LLR: 2.94 (-2.94,2.94) <0.50,2.50> Total: 195228 W: 52741 L: 52140 D: 90347 Elo +1.07 Ptnml(0-2): 53, 18718, 59482, 19297, 64 closes https://github.com/official-stockfish/Stockfish/pull/4556 bench 3448916 see source |
| Windows x64 for Haswell CPUs Windows x64 for modern computers + AVX2 Windows x64 for modern computers Windows x64 + SSSE3 Windows x64 Linux x64 for Haswell CPUs Linux x64 for modern computers + AVX2 Linux x64 for modern computers Linux x64 + SSSE3 Linux x64 | Author: candirufish
Date: Wed May 3 20:41:33 2023 +0200 Timestamp: 1683139293 Adjust reductions Decrease further on cutNodes with tte->depth() >= depth + 3 condition. LTC: https://tests.stockfishchess.org/tests/view/644dc84bccf5e93df5e50c13 LLR: 2.94 (-2.94,2.94) <0.50,2.50> Total: 155346 W: 42184 L: 41660 D: 71502 Elo +1.17 Ptnml(0-2): 59, 14765, 47504, 15283, 62 STC: https://tests.stockfishchess.org/tests/view/644d05de68e01d8194cd9bbb LLR: 2.93 (-2.94,2.94) <0.00,2.00> Total: 121888 W: 32868 L: 32444 D: 56576 Elo +1.21 Ptnml(0-2): 332, 13273, 33343, 13631, 365 closes https://github.com/official-stockfish/Stockfish/pull/4552 bench: 3739675 see source |
| Windows x64 for Haswell CPUs Windows x64 for modern computers + AVX2 Windows x64 for modern computers Windows x64 + SSSE3 Windows x64 Linux x64 for Haswell CPUs Linux x64 for modern computers + AVX2 Linux x64 for modern computers Linux x64 + SSSE3 Linux x64 | Author: Linmiao Xu
Date: Wed May 3 20:37:57 2023 +0200 Timestamp: 1683139077 Update 7 eval and optimism params Params found using spsa at 30+0.3 with this tuning config: ``` // evaluate.cpp int nnueOptScaleBase = 1001; int nnueComplexityMult = 406; int nnueComplexityOptOffset = 424; int evalOptComplexityOffset = 272; int evalOptScaleOffset = 748; TUNE(SetRange(801, 1201), nnueOptScaleBase); TUNE(SetRange(306, 506), nnueComplexityMult); TUNE(SetRange(324, 524), nnueComplexityOptOffset); TUNE(SetRange(172, 372), evalOptComplexityOffset); TUNE(SetRange(648, 848), evalOptScaleOffset); // search.cpp int searchOptBase = 120; int searchOptDenom = 161; TUNE(SetRange(20, 220), searchOptBase); TUNE(SetRange(111, 211), searchOptDenom); ``` Passed STC: https://tests.stockfishchess.org/tests/view/644dda8accf5e93df5e50cbe LLR: 2.93 (-2.94,2.94) <0.00,2.00> Total: 136800 W: 36682 L: 36237 D: 63881 Elo +1.13 Ptnml(0-2): 353, 14910, 37492, 15229, 416 Passed LTC: https://tests.stockfishchess.org/tests/view/644eaedb3f31c3bbe4a3d345 LLR: 2.94 (-2.94,2.94) <0.50,2.50> Total: 64548 W: 17624 L: 17272 D: 29652 Elo +1.89 Ptnml(0-2): 33, 6112, 19631, 6466, 32 closes https://github.com/official-stockfish/Stockfish/pull/4550 bench 3670343 see source |
| Windows x64 for Haswell CPUs Windows x64 for modern computers + AVX2 Windows x64 for modern computers Windows x64 + SSSE3 Windows x64 Linux x64 for Haswell CPUs Linux x64 for modern computers + AVX2 Linux x64 for modern computers Linux x64 + SSSE3 Linux x64 | Author: Linmiao Xu
Date: Tue Apr 25 08:19:00 2023 +0200 Timestamp: 1682403540 Update default net to nn-e1fb1ade4432.nnue Created by retraining nn-dabb1ed23026.nnue with a dataset composed of: * The previous best dataset (nn-1ceb1a57d117.nnue dataset) * Adding de-duplicated T80 data from feb2023 and the last 10 days of jan2023, filtered with v6-dd Initially trained with the same options as the recent master net (nn-1ceb1a57d117.nnue). Around epoch 890, training was manually stopped and max epoch increased to 1000. ``` python3 easy_train.py \ --experiment-name leela96-dfrc99-T60novdec-v2-T80augsep-v6-T80junjuloctnovjanfebT79aprmayT78jantosepT77dec-v6dd \ --training-dataset /data/leela96-dfrc99-T60novdec-v2-T80augsep-v6-T80junjuloctnovjanfebT79aprmayT78jantosepT77dec-v6dd.binpack \ --nnue-pytorch-branch linrock/nnue-pytorch/misc-fixes \ --start-from-engine-test-net True \ --early-fen-skipping 30 \ --start-lambda 1.0 \ --end-lambda 0.7 \ --max_epoch 900 \ --lr 4.375e-4 \ --gamma 0.995 \ --tui False \ --gpus "0," \ --seed $RANDOM ``` The same v6-dd filtering and binpack minimizer was used for preparing the recent nn-1ceb1a57d117.nnue dataset. ``` python3 interleave_binpacks.py \ leela96-filt-v2.binpack \ dfrc99-filt-v2.binpack \ T60-nov2021-12tb7p-eval-filt-v2.binpack \ T60-dec2021-12tb7p-eval-filt-v2.binpack \ filt-v6/test80-aug2022-16tb7p-filter-v6.min-mar2023.binpack \ filt-v6/test80-sep2022-16tb7p-filter-v6.min-mar2023.binpack \ filt-v6-dd/test80-jun2022-16tb7p-filter-v6-dd.min-mar2023.binpack \ filt-v6-dd/test80-jul2022-16tb7p-filter-v6-dd.binpack \ filt-v6-dd/test80-oct2022-16tb7p-filter-v6-dd.binpack \ filt-v6-dd/test80-nov2022-16tb7p-filter-v6-dd.binpack \ filt-v6-dd/test80-jan2022-3of3-16tb7p-filter-v6-dd.min-mar2023.binpack \ filt-v6-dd/test80-feb2023-16tb7p-filter-v6-dd.min-mar2023.binpack \ filt-v6-dd/test79-apr2022-16tb7p-filter-v6-dd.binpack \ filt-v6-dd/test79-may2022-16tb7p-filter-v6-dd.binpack \ filt-v6-dd/test78-jantomay2022-16tb7p-filter-v6-dd.min-mar2023.binpack \ filt-v6-dd/test78-juntosep2022-16tb7p-filter-v6-dd.binpack \ filt-v6-dd/test77-dec2021-16tb7p-filter-v6-dd.binpack \ /data/leela96-dfrc99-T60novdec-v2-T80augsep-v6-T80junjuloctnovjanfebT79aprmayT78jantosepT77dec-v6dd.binpack ``` Links for downloading the training data components can be found at: https://robotmoon.com/nnue-training-data/ Local elo at 25k nodes per move: nn-epoch919.nnue : 2.6 +/- 2.8 Passed STC vs. nn-dabb1ed23026.nnue https://tests.stockfishchess.org/tests/view/644420df94ff3db5625f2af5 LLR: 2.94 (-2.94,2.94) <0.00,2.00> Total: 125960 W: 33898 L: 33464 D: 58598 Elo +1.20 Ptnml(0-2): 351, 13920, 34021, 14320, 368 Passed LTC vs. nn-1ceb1a57d117.nnue https://tests.stockfishchess.org/tests/view/64469f128d30316529b3dc46 LLR: 2.95 (-2.94,2.94) <0.50,2.50> Total: 24544 W: 6817 L: 6542 D: 11185 Elo +3.89 Ptnml(0-2): 8, 2252, 7488, 2505, 19 closes https://github.com/official-stockfish/Stockfish/pull/4546 bench 3714847 see source |
| Windows x64 for Haswell CPUs Windows x64 for modern computers + AVX2 Windows x64 for modern computers Windows x64 + SSSE3 Windows x64 Linux x64 for Haswell CPUs Linux x64 for modern computers + AVX2 Linux x64 for modern computers Linux x64 + SSSE3 Linux x64 | Author: Linmiao Xu
Date: Tue Apr 25 08:17:22 2023 +0200 Timestamp: 1682403442 Created by retraining the master net with these changes to the dataset: * Extending v6 filtering to data from T77 dec2021, T79 may2022, and T80 nov2022 * Reducing the number of duplicate positions, prioritizing position scores seen later in time * Using a binpack minimizer to reduce the overall data size Trained the same way as the previous master net, aside from the dataset changes: ``` python3 easy_train.py \ --experiment-name leela96-dfrc99-T60novdec-v2-T80augsep-v6-T80junjuloctnovT79aprmayT78jantosepT77dec-v6dd \ --training-dataset /data/leela96-dfrc99-T60novdec-v2-T80augsep-v6-T80junjuloctnovT79aprmayT78jantosepT77dec-v6dd.binpack \ --nnue-pytorch-branch linrock/nnue-pytorch/misc-fixes \ --start-from-engine-test-net True \ --early-fen-skipping 30 \ --start-lambda 1.0 \ --end-lambda 0.7 \ --max_epoch 900 \ --lr 4.375e-4 \ --gamma 0.995 \ --tui False \ --gpus "0," \ --seed $RANDOM ``` The new v6-dd filtering reduces duplicate positions by iterating over hourly data files within leela test runs, starting with the most recent, then keeping positions the first time they're seen and ignoring positions that are seen again. This ordering was done with the assumption that position scores seen later in time are generally more accurate than scores seen earlier in the test run. Positions are de-duplicated based on piece orientations, the first token in fen strings. The binpack minimizer was run with default settings after first merging monthly data into single binpacks. ``` python3 interleave_binpacks.py \ leela96-filt-v2.binpack \ dfrc99-filt-v2.binpack \ T60-nov2021-12tb7p-eval-filt-v2.binpack \ T60-dec2021-12tb7p-eval-filt-v2.binpack \ filt-v6/test80-aug2022-16tb7p-filter-v6.min-mar2023.binpack \ filt-v6/test80-sep2022-16tb7p-filter-v6.min-mar2023.binpack \ filt-v6-dd/test80-jun2022-16tb7p-filter-v6-dd.min-mar2023.binpack \ filt-v6-dd/test80-jul2022-16tb7p-filter-v6-dd.binpack \ filt-v6-dd/test80-oct2022-16tb7p-filter-v6-dd.binpack \ filt-v6-dd/test80-nov2022-16tb7p-filter-v6-dd.binpack \ filt-v6-dd/test79-apr2022-16tb7p-filter-v6-dd.binpack \ filt-v6-dd/test79-may2022-16tb7p-filter-v6-dd.binpack \ filt-v6-dd/test78-jantomay2022-16tb7p-filter-v6-dd.min-mar2023.binpack \ filt-v6-dd/test78-juntosep2022-16tb7p-filter-v6-dd.binpack \ filt-v6-dd/test77-dec2021-16tb7p-filter-v6-dd.binpack \ /data/leela96-dfrc99-T60novdec-v2-T80augsep-v6-T80junjuloctnovT79aprmayT78jantosepT77dec-v6dd.binpack ``` The code for v6-dd filtering is available along with training data preparation scripts at: https://github.com/linrock/nnue-data Links for downloading the training data components: https://robotmoon.com/nnue-training-data/ The binpack minimizer is from: #4447 Local elo at 25k nodes per move: nn-epoch859.nnue : 1.2 +/- 2.6 Passed STC: https://tests.stockfishchess.org/tests/view/643aad7db08900ff1bc5a832 LLR: 2.93 (-2.94,2.94) <0.00,2.00> Total: 565040 W: 150225 L: 149162 D: 265653 Elo +0.65 Ptnml(0-2): 1875, 62137, 153229, 63608, 1671 Passed LTC: https://tests.stockfishchess.org/tests/view/643ecf2fa43cf30e719d2042 LLR: 2.94 (-2.94,2.94) <0.50,2.50> Total: 1014840 W: 274645 L: 272456 D: 467739 Elo +0.75 Ptnml(0-2): 515, 98565, 306970, 100956, 414 closes https://github.com/official-stockfish/Stockfish/pull/4545 bench 3476305 see source |
| Windows x64 for Haswell CPUs Windows x64 for modern computers + AVX2 Windows x64 for modern computers Windows x64 + SSSE3 Windows x64 Linux x64 for Haswell CPUs Linux x64 for modern computers + AVX2 Linux x64 for modern computers Linux x64 + SSSE3 Linux x64 | Author: Boštjan Mejak
Date: Tue Apr 25 08:08:48 2023 +0200 Timestamp: 1682402928 Update AUTHORS Improved some comments in the AUTHORS file, sort contributors closes https://github.com/official-stockfish/Stockfish/pull/4520 No functional change see source |