| 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: Mon Jun 3 23:12:13 2024 +0200 Timestamp: 1717449133 Simplify continuation histories Functional simplification. Simplify continuation histories, therefore increasing the effect of stats updates and movepicker bonuses for continuation history 3 plies deep. Passed STC: LLR: 2.93 (-2.94,2.94) <-1.75,0.25> Total: 229184 W: 59087 L: 59080 D: 111017 Elo +0.01 Ptnml(0-2): 554, 27248, 59002, 27213, 575 https://tests.stockfishchess.org/tests/view/665c7a09fd45fb0f907c223b Passed LTC: LLR: 2.95 (-2.94,2.94) <-1.75,0.25> Total: 44532 W: 11419 L: 11223 D: 21890 Elo +1.53 Ptnml(0-2): 18, 4787, 12457, 4989, 15 https://tests.stockfishchess.org/tests/view/665c8842fd45fb0f907c23ec closes https://github.com/official-stockfish/Stockfish/pull/5339 Bench: 1326444 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: Disservin
Date: Mon Jun 3 23:11:59 2024 +0200 Timestamp: 1717449119 Add helpers for managing aligned memory Previously, we had two type aliases, LargePagePtr and AlignedPtr, which required manually initializing the aligned memory for the pointer. The new helpers: - make_unique_aligned - make_unique_large_page are now available for allocating aligned memory (with large pages). They behave similarly to std::make_unique, ensuring objects allocated with these functions follow RAII. The old approach had issues with initializing non-trivial types or arrays of objects. The evaluation function of the network is now a unique pointer to an array instead of an array of unique pointers. Memory related functions have been moved into memory.h Passed High Hash Pressure Test Non-Regression STC: https://tests.stockfishchess.org/tests/view/665b2b36586058766677cfd2 LLR: 2.93 (-2.94,2.94) <-1.75,0.25> Total: 476992 W: 122426 L: 122677 D: 231889 Elo -0.18 Ptnml(0-2): 1145, 51027, 134419, 50744, 1161 Failed Normal Non-Regression STC: https://tests.stockfishchess.org/tests/view/665b2997586058766677cfc8 LLR: -2.94 (-2.94,2.94) <-1.75,0.25> Total: 877312 W: 225233 L: 226395 D: 425684 Elo -0.46 Ptnml(0-2): 2110, 94642, 246239, 93630, 2035 Probably a fluke since there shouldn't be a real slowndown and it has also passed the high hash pressure test. closes https://github.com/official-stockfish/Stockfish/pull/5332 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: Tomasz Sobczyk
Date: Mon Jun 3 08:54:24 2024 +0200 Timestamp: 1717397664 Fix GetProcessGroupAffinity call `GetProcessGroupAffinity` appears to require 4 byte alignment for `GroupArray` memory. See https://stackoverflow.com/q/78567676 for further information closes https://github.com/official-stockfish/Stockfish/pull/5340 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: Joost VandeVondele
Date: Sun Jun 2 09:23:28 2024 +0200 Timestamp: 1717313008 Revert "Adjust return bonus from tt cutoffs at fail highs" This reverts commit 783dfc2eb235236ff799618436d68d0c1a3f3807. could lead to a division by zero for: ttValue = (ttValue * tte->depth() + beta) / (tte->depth() + 1) as other threads can overwrite the tte with a QS depth of -1. closes https://github.com/official-stockfish/Stockfish/pull/5338 Bench: 1280020 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: Sat Jun 1 20:17:38 2024 +0200 Timestamp: 1717265858 Simplify statScore divisor into a constant Passed non-regression STC: https://tests.stockfishchess.org/tests/view/665b392ff4a1fd0c208ea864 LLR: 2.93 (-2.94,2.94) <-1.75,0.25> Total: 114752 W: 29628 L: 29495 D: 55629 Elo +0.40 Ptnml(0-2): 293, 13694, 29269, 13827, 293 Passed non-regression LTC: https://tests.stockfishchess.org/tests/view/665b588c11645bd3d3fac467 LLR: 2.95 (-2.94,2.94) <-1.75,0.25> Total: 65322 W: 16549 L: 16373 D: 32400 Elo +0.94 Ptnml(0-2): 30, 7146, 18133, 7322, 30 closes https://github.com/official-stockfish/Stockfish/pull/5337 bench 1241443 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 Jun 1 20:12:36 2024 +0200 Timestamp: 1717265556 Simplify tm, removing faster 1st move and 1.13 extraTime. Passed STC 10+0.1 : LLR: 2.93 (-2.94,2.94) <-1.75,0.25> Total: 349760 W: 90112 L: 90231 D: 169417 Elo -0.12 Ptnml(0-2): 784, 37970, 97496, 37841, 789 https://tests.stockfishchess.org/tests/view/665aeee00223e235f05b7d21 Passed LTC 60+0.6 : LLR: 2.95 (-2.94,2.94) <-1.75,0.25> Total: 140082 W: 35463 L: 35370 D: 69249 Elo +0.23 Ptnml(0-2): 59, 13492, 42851, 13575, 64 https://tests.stockfishchess.org/tests/view/665b15e78da109e362924e5a closes https://github.com/official-stockfish/Stockfish/pull/5334 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: MinetaS
Date: Sat Jun 1 20:09:47 2024 +0200 Timestamp: 1717265387 Simplify 50 move rule dampening Refactor the logic of 50 move rule dampening by removing a constant. Passed non-regression STC: LLR: 2.93 (-2.94,2.94) <-1.75,0.25> Total: 35232 W: 9214 L: 8992 D: 17026 Elo +2.19 Ptnml(0-2): 114, 4081, 8999, 4313, 109 https://tests.stockfishchess.org/tests/view/665a329013d08af3c1725610 Passed non-regression LTC: LLR: 2.94 (-2.94,2.94) <-1.75,0.25> Total: 38406 W: 9732 L: 9530 D: 19144 Elo +1.83 Ptnml(0-2): 14, 4132, 10708, 4336, 13 https://tests.stockfishchess.org/tests/view/665a370913d08af3c1725651 https://github.com/official-stockfish/Stockfish/pull/5327 Bench: 1059739 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: Sat Jun 1 20:08:08 2024 +0200 Timestamp: 1717265288 Simplify histories movepick formula Passed STC: LLR: 2.93 (-2.94,2.94) <-1.75,0.25> Total: 81440 W: 21100 L: 20929 D: 39411 Elo +0.73 Ptnml(0-2): 248, 9659, 20718, 9864, 231 https://tests.stockfishchess.org/tests/view/6659a8b7ea624d64ea5f3208 Passed LTC: LLR: 2.94 (-2.94,2.94) <-1.75,0.25> Total: 85758 W: 21763 L: 21607 D: 42388 Elo +0.63 Ptnml(0-2): 34, 9606, 23463, 9722, 54 https://tests.stockfishchess.org/tests/view/6659d7bff426908fcc6b692c closes https://github.com/official-stockfish/Stockfish/pull/5326 bench: 1280472 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 Jun 1 20:04:30 2024 +0200 Timestamp: 1717265070 Avoid changing bestvalue in the case the ttValue contains mate scores, do not return them as bestValue, since they are not proven. passed STC https://tests.stockfishchess.org/tests/view/665b1ea5586058766677cfa3 LLR: 2.93 (-2.94,2.94) <-1.75,0.25> Total: 58912 W: 15319 L: 15130 D: 28463 Elo +1.11 Ptnml(0-2): 141, 6562, 15854, 6765, 134 passed LTC: https://tests.stockfishchess.org/tests/view/665b2712586058766677cfc4 LLR: 2.94 (-2.94,2.94) <-1.75,0.25> Total: 141666 W: 35976 L: 35879 D: 69811 Elo +0.24 Ptnml(0-2): 61, 15513, 39584, 15618, 57 closes https://github.com/official-stockfish/Stockfish/pull/5335 Bench: 1336115 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 Jun 1 20:02:09 2024 +0200 Timestamp: 1717264929 Adjust return bonus from tt cutoffs at fail highs This is reintroduction of the recently simplified logic - if positive tt cutoff occurs return not a tt value but smth between it and beta. Difference is that instead of static linear combination there we use basically the same formula as we do in the main search - with the only difference being using tt depth instead of depth, which makes a lot of sense. Passed STC: https://tests.stockfishchess.org/tests/view/665b3a34f4a1fd0c208ea870 LLR: 2.95 (-2.94,2.94) <0.00,2.00> Total: 54944 W: 14239 L: 13896 D: 26809 Elo +2.17 Ptnml(0-2): 151, 6407, 14008, 6760, 146 Passed LTC: https://tests.stockfishchess.org/tests/view/665b520011645bd3d3fac341 LLR: 2.94 (-2.94,2.94) <0.50,2.50> Total: 90540 W: 23070 L: 22640 D: 44830 Elo +1.65 Ptnml(0-2): 39, 9903, 24965, 10315, 48 closes https://github.com/official-stockfish/Stockfish/pull/5336 bench 1381237 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: Sat Jun 1 19:59:07 2024 +0200 Timestamp: 1717264747 Update default smallnet to nn-37f18f62d772.nnue Created by training L1-128 from scratch with: - skipping based on simple eval in the trainer, for compatibility with regular binpacks without requiring pre-filtering all binpacks - minimum simple eval of 950, lower than 1000 previously - usage of some hse-v1 binpacks with minimum simple eval 1000 - addition of hse-v6 binpacks with minimum simple eval 500 - permuting the FT with 10k positions from fishpack32.binpack - torch.compile to speed up smallnet training Training is significantly slower when using non-pre-filtered binpacks due to the increased skipping required. This net was reached at epoch 339. ``` experiment-name: 128--S1-hse-1k-T80-v6-unfilt-less-sf--se-gt950-no-wld-skip training-dataset: /data/: - dfrc99-16tb7p.v2.min.binpack /data/hse-v1/: - leela96-filt-v2.min.high-simple-eval-1k.min-v2.binpack - test60-novdec2021-12tb7p-filter-v6-dd.min-mar2023.unmin.high-simple-eval-1k.min-v2.binpack - test77-nov2021-2tb7p.no-db.min.high-simple-eval-1k.min-v2.binpack - test77-dec2021-16tb7p.no-db.min.high-simple-eval-1k.min-v2.binpack - test77-jan2022-2tb7p.high-simple-eval-1k.min-v2.binpack - test78-jantomay2022-16tb7p-filter-v6-dd.min-mar2023.unmin.high-simple-eval-1k.min-v2.binpack - test78-juntosep2022-16tb7p-filter-v6-dd.min-mar2023.unmin.high-simple-eval-1k.min-v2.binpack - test79-apr2022-16tb7p.min.high-simple-eval-1k.min-v2.binpack - test79-may2022-16tb7p-filter-v6-dd.min-mar2023.unmin.high-simple-eval-1k.min-v2.binpack - test80-apr2022-16tb7p.min.high-simple-eval-1k.min-v2.binpack - test80-may2022-16tb7p.high-simple-eval-1k.min-v2.binpack - test80-jun2022-16tb7p-filter-v6-dd.min-mar2023.unmin.high-simple-eval-1k.min-v2.binpack - test80-jul2022-16tb7p.v6-dd.min.high-simple-eval-1k.min-v2.binpack - test80-sep2022-16tb7p-filter-v6-dd.min-mar2023.unmin.high-simple-eval-1k.min-v2.binpack - test80-nov2022-16tb7p-v6-dd.min.high-simple-eval-1k.min-v2.binpack /data/S11-mar2024/: - test80-2022-08-aug-16tb7p.v6-dd.min.binpack - test80-2022-10-oct-16tb7p.v6-dd.binpack - test80-2022-12-dec-16tb7p.min.binpack - test80-2023-01-jan-16tb7p.v6-sk20.min.binpack - test80-2023-02-feb-16tb7p.v6-sk20.min.binpack - test80-2023-03-mar-2tb7p.v6-sk16.min.binpack - test80-2023-04-apr-2tb7p.v6-sk16.min.binpack - test80-2023-05-may-2tb7p.v6.min.binpack - test80-2023-06-jun-2tb7p.binpack.min-v2.binpack - test80-2023-07-jul-2tb7p.binpack.min-v2.binpack - test80-2023-08-aug-2tb7p.v6.min.binpack - test80-2023-09-sep-2tb7p.binpack.hse-v6.binpack - test80-2023-10-oct-2tb7p.binpack.hse-v6.binpack - test80-2023-11-nov-2tb7p.binpack.hse-v6.binpack - test80-2023-12-dec-2tb7p.binpack.hse-v6.binpack - test80-2024-01-jan-2tb7p.binpack.hse-v6.binpack - test80-2024-02-feb-2tb7p.binpack.hse-v6.binpack - test80-2024-03-mar-2tb7p.binpack wld-fen-skipping: False nnue-pytorch-branch: linrock/nnue-pytorch/128-skipSimpleEval-lt950-torch-compile engine-test-branch: linrock/Stockfish/L1-128-nolazy engine-base-branch: linrock/Stockfish/L1-128 start-from-engine-test-net: False num-epochs: 500 start-lambda: 1.0 end-lambda: 1.0 ``` Training data can be found at: https://robotmoon.com/nnue-training-data/ Passed STC: https://tests.stockfishchess.org/tests/view/66549c16a86388d5e27daff5 LLR: 2.93 (-2.94,2.94) <0.00,2.00> Total: 196608 W: 51254 L: 50697 D: 94657 Elo +0.98 Ptnml(0-2): 722, 23244, 49796, 23839, 703 Passed LTC: https://tests.stockfishchess.org/tests/view/6658d1aa6b0e318cefa90122 LLR: 2.96 (-2.94,2.94) <0.50,2.50> Total: 122538 W: 31332 L: 30835 D: 60371 Elo +1.41 Ptnml(0-2): 69, 13407, 33811, 13922, 60 closes https://github.com/official-stockfish/Stockfish/pull/5333 bench 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: rn5f107s2
Date: Sat Jun 1 19:58:20 2024 +0200 Timestamp: 1717264700 MCP more after a bad singular search The idea is, that if we have the information that the singular search failed low and therefore produced an upperbound score, we can use the score from singularsearch as approximate upperbound as to what bestValue our non ttMoves will produce. If this value is well below alpha, we assume that all non-ttMoves will score below alpha and therfore can skip more moves. This patch also sets up variables for future patches wanting to use teh singular search result outside of singular extensions, in singularBound and singularValue, meaning further patches using this search result to affect various pruning techniques can be tried. Passed STC: https://tests.stockfishchess.org/tests/view/6658d13e6b0e318cefa90120 LLR: 2.94 (-2.94,2.94) <0.00,2.00> Total: 85632 W: 22112 L: 21725 D: 41795 Elo +1.57 Ptnml(0-2): 243, 10010, 21947, 10349, 267 Passed LTC: https://tests.stockfishchess.org/tests/view/6658dd356b0e318cefa9016a LLR: 2.94 (-2.94,2.94) <0.50,2.50> Total: 243978 W: 62014 L: 61272 D: 120692 Elo +1.06 Ptnml(0-2): 128, 26598, 67791, 27348, 124 closes https://github.com/official-stockfish/Stockfish/pull/5325 bench 1397172 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: Sat Jun 1 19:56:05 2024 +0200 Timestamp: 1717264565 Quadratic smallnet threshold with re-evaluation The threshold now decreases more quickly as pawn count decreases, using the smallnet more compared to before. Combo of two eval patches: https://tests.stockfishchess.org/tests/view/66576c5f6b0e318cefa8d26e https://tests.stockfishchess.org/tests/view/664ced40830eb9f886616a77 Passed STC: https://tests.stockfishchess.org/tests/view/66588c136b0e318cefa8ff21 LLR: 2.94 (-2.94,2.94) <0.00,2.00> Total: 112608 W: 29336 L: 28908 D: 54364 Elo +1.32 Ptnml(0-2): 344, 13223, 28718, 13699, 320 Passed LTC: https://tests.stockfishchess.org/tests/view/6658c8786b0e318cefa900f5 LLR: 2.96 (-2.94,2.94) <0.50,2.50> Total: 108288 W: 27493 L: 27026 D: 53769 Elo +1.50 Ptnml(0-2): 54, 11821, 29930, 12282, 57 closes https://github.com/official-stockfish/Stockfish/pull/5323 bench 1728074 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: Sat Jun 1 19:53:13 2024 +0200 Timestamp: 1717264393 Tweak first picked move (ttMove) reduction rule Tweak first picked move (ttMove) reduction rule: Instead of always resetting the reduction to 0, we now only do so if the current reduction is less than 2. If the current reduction is 2 or more, we decrease it by 2 instead. Passed STC: LLR: 2.93 (-2.94,2.94) <0.00,2.00> Total: 109504 W: 28340 L: 27919 D: 53245 Elo +1.34 Ptnml(0-2): 305, 12848, 28028, 13263, 308 https://tests.stockfishchess.org/tests/view/6658c2fa6b0e318cefa900c2 Passed LTC: LLR: 2.96 (-2.94,2.94) <0.50,2.50> Total: 130410 W: 33248 L: 32738 D: 64424 Elo +1.36 Ptnml(0-2): 53, 14139, 36328, 14615, 70 https://tests.stockfishchess.org/tests/view/6658dd8a6b0e318cefa90173 closes https://github.com/official-stockfish/Stockfish/pull/5321 bench: 1224588 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: Thu May 30 23:07:25 2024 +0200 Timestamp: 1717103245 Fix cross from Linux to Windows specifies Windows 7 required https://learn.microsoft.com/en-us/cpp/porting/modifying-winver-and-win32-winnt?view=msvc-170 closes https://github.com/official-stockfish/Stockfish/pull/5319 No functional change see source |
| Linux x64 for Haswell CPUs Linux x64 for modern computers + AVX2 Linux x64 for modern computers Linux x64 + SSSE3 Linux x64 | Author: Tomasz Sobczyk
Date: Thu May 30 23:05:25 2024 +0200 Timestamp: 1717103125 On linux use sysfs instead of lscpu Use sysfs (https://www.kernel.org/doc/Documentation/ABI/stable/sysfs-devices-node) to determine processor to NUMA node mapping. Avoids problems on some machines with high core count where lscpu was showing high cpu utilization. closes https://github.com/official-stockfish/Stockfish/pull/5315 No functional change see source |
| Linux x64 for Haswell CPUs Linux x64 for modern computers + AVX2 Linux x64 for modern computers Linux x64 + SSSE3 Linux x64 | Author: Tomasz Sobczyk
Date: Thu May 30 23:05:21 2024 +0200 Timestamp: 1717103121 Replace std::from_chars with std::stoull the former was not widely supported, requiring newer compiler versions. closes https://github.com/official-stockfish/Stockfish/pull/5313 No functional change see source |
| Linux x64 for Haswell CPUs Linux x64 for modern computers + AVX2 Linux x64 for modern computers Linux x64 + SSSE3 Linux x64 | Author: Tomasz Sobczyk
Date: Thu May 30 23:05:16 2024 +0200 Timestamp: 1717103116 Fix process' processor affinity determination on Windows. Specialize and privatize NumaConfig::get_process_affinity. Only enable NUMA capability for 64-bit Windows. Following #5307 and some more testing it was determined that the way affinity was being determined on Windows was incorrect, based on incorrect assumptions about GetNumaProcessorNodeEx. This patch fixes the issue by attempting to retrieve the actual process' processor affinity using Windows API. However one issue persists that is not addressable due to limitations of Windows, and will have to be considered a limitation. If affinities were set using SetThreadAffinityMask instead of SetThreadSelectedCpuSetMasks and GetProcessGroupAffinity returns more than 1 group it is NOT POSSIBLE to determine the affinity programmatically on Windows. In such case the implementation assumes no affinites are set and will consider all processors available for execution. closes https://github.com/official-stockfish/Stockfish/pull/5312 No functional change see source |
| Linux x64 for Haswell CPUs Linux x64 for modern computers + AVX2 Linux x64 for modern computers Linux x64 + SSSE3 Linux x64 | Author: Disservin
Date: Thu May 30 23:05:01 2024 +0200 Timestamp: 1717103101 NUMA: Fix concurrency counting for windows systems If there is more than 1 processor group, std::thread::hardware_concurrency should not be used. fixes #5307 closes https://github.com/official-stockfish/Stockfish/pull/5311 No functional change see source |
| 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 30 23:00:22 2024 +0200 Timestamp: 1717102822 Simplifying the malus for putting piece en prise formula Simplifying the malus for putting piece en prise formula by merging the minor pieces and pawns (removing the pawn exclusion from the formula). Passed STC: https://tests.stockfishchess.org/tests/view/66578d9c6b0e318cefa8d441 LLR: 2.99 (-2.94,2.94) <-1.75,0.25> Total: 314272 W: 80705 L: 80786 D: 152781 Elo -0.09 Ptnml(0-2): 873, 37577, 80366, 37398, 922 Passed LTC (before rebasing): https://tests.stockfishchess.org/tests/view/6657b5ee6b0e318cefa8d6ab LLR: 2.95 (-2.94,2.94) <-1.75,0.25> Total: 117000 W: 29447 L: 29324 D: 58229 Elo +0.37 Ptnml(0-2): 47, 12877, 32535, 12988, 53 Passed LTC (also after rebasing): https://tests.stockfishchess.org/tests/view/6658803d6b0e318cefa8fd99 LLR: 2.96 (-2.94,2.94) <-1.75,0.25> Total: 244992 W: 61807 L: 61814 D: 121371 Elo -0.01 Ptnml(0-2): 125, 27420, 67414, 27411, 126 closes https://github.com/official-stockfish/Stockfish/pull/5316 Bench: 1484840 see source |
| 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: Thu May 30 22:54:23 2024 +0200 Timestamp: 1717102463 Allow tt cutoffs for shallower depths in certain conditions Current master allows tt cutoffs only when depth from tt is strictly greater than current node depth. This patch also allows them when it's equal and if tt value is lower or equal to beta. Passed STC: https://tests.stockfishchess.org/tests/view/66578e2e6b0e318cefa8d447 LLR: 2.93 (-2.94,2.94) <0.00,2.00> Total: 26592 W: 6944 L: 6645 D: 13003 Elo +3.91 Ptnml(0-2): 67, 3039, 6795, 3318, 77 Passed LTC: https://tests.stockfishchess.org/tests/view/6657f46b6b0e318cefa8d7e9 LLR: 2.96 (-2.94,2.94) <0.50,2.50> Total: 142572 W: 36315 L: 35776 D: 70481 Elo +1.31 Ptnml(0-2): 70, 15666, 39288, 16179, 83 closes https://github.com/official-stockfish/Stockfish/pull/5314 Bench: 1368486 see source |
| 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 30 22:53:20 2024 +0200 Timestamp: 1717102400 Functional simplification in the transposition table Passed STC: LLR: 2.98 (-2.94,2.94) <-1.75,0.25> Total: 154848 W: 39838 L: 39750 D: 75260 Elo +0.20 Ptnml(0-2): 404, 16214, 44087, 16328, 391 https://tests.stockfishchess.org/tests/view/664892b088b8c6a2bbe430fc Passed LTC: LLR: 2.94 (-2.94,2.94) <-1.75,0.25> Total: 68172 W: 17296 L: 17137 D: 33739 Elo +0.81 Ptnml(0-2): 23, 6349, 21185, 6504, 25 https://tests.stockfishchess.org/tests/view/6648aabfa0781149e383e526 closes https://github.com/official-stockfish/Stockfish/pull/5263 Bench: 1623228 see source |
| Linux x64 for Haswell CPUs Linux x64 for modern computers + AVX2 Linux x64 for modern computers Linux x64 + SSSE3 Linux x64 | Author: Viren6
Date: Thu May 30 14:34:24 2024 +0200 Timestamp: 1717072464 Add extension condition to cutoffCnt Decrease cutoffCnt increment by 1 if extension is 2 or greater. Passed STC: https://tests.stockfishchess.org/tests/view/66577a696b0e318cefa8d34d LLR: 2.94 (-2.94,2.94) <0.00,2.00> Total: 99200 W: 25703 L: 25297 D: 48200 Elo +1.42 Ptnml(0-2): 253, 11660, 25390, 12022, 275 Passed LTC: https://tests.stockfishchess.org/tests/view/665787ab6b0e318cefa8d411 LLR: 2.94 (-2.94,2.94) <0.50,2.50> Total: 124530 W: 31659 L: 31161 D: 61710 Elo +1.39 Ptnml(0-2): 58, 13578, 34489, 14088, 52 closes https://github.com/official-stockfish/Stockfish/pull/5310 bench 1623228 see source |
| Linux x64 for Haswell CPUs Linux x64 for modern computers + AVX2 Linux x64 for modern computers Linux x64 + SSSE3 Linux x64 | Author: Robert Nurnberg @ elitebook
Date: Thu May 30 14:29:31 2024 +0200 Timestamp: 1717072171 Tweak and update the WDL model This PR updates the internal WDL model, using data from 2.5M games played by SF-dev (3c62ad7). Note that the normalizing constant has increased from 329 to 368. Changes to the fitting procedure: * the value for --materialMin was increased from 10 to 17: including data with less material leads to less accuracy for larger material count values * the data was filtered to only include single thread LTC games at 60+0.6 * the data was filtered to only include games from master against patches that are (approximatively) within 5 nElo of master For more information and plots of the model see PR#5309 closes https://github.com/official-stockfish/Stockfish/pull/5309 No functional change see source |
| 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: Thu May 30 14:28:07 2024 +0200 Timestamp: 1717072087 Update default main net to nn-ddcfb9224cdb.nnue Created by further tuning the spsa-tuned main net `nn-c721dfca8cd3.nnue` with the same methods described in https://github.com/official-stockfish/Stockfish/pull/5254 This net was reached at 61k / 120k spsa games at 70+0.7 th 7: https://tests.stockfishchess.org/tests/view/665639d0a86388d5e27dd259 Passed STC: https://tests.stockfishchess.org/tests/view/6657d44e6b0e318cefa8d771 LLR: 2.95 (-2.94,2.94) <0.00,2.00> Total: 114688 W: 29775 L: 29344 D: 55569 Elo +1.31 Ptnml(0-2): 274, 13633, 29149, 13964, 324 Passed LTC: https://tests.stockfishchess.org/tests/view/6657e1e46b0e318cefa8d7a6 LLR: 2.94 (-2.94,2.94) <0.50,2.50> Total: 88152 W: 22412 L: 21988 D: 43752 Elo +1.67 Ptnml(0-2): 56, 9560, 24409, 10006, 45 closes https://github.com/official-stockfish/Stockfish/pull/5308 Bench: 1434678 see source |