| Windows x64 for Haswell CPUs Windows x64 for modern computers Windows x64 Windows 32 Linux x64 for Haswell CPUs Linux x64 for modern computers Linux x64 | Author: mstembera
Date: Sun Oct 21 08:15:04 2018 +0200 Timestamp: 1540102504 Small simplification in castling rights There is no need for a special struct with a static member to generate castling rights. No functional change. see source |
| Windows x64 for Haswell CPUs Windows x64 for modern computers Windows x64 Windows 32 Linux x64 for Haswell CPUs Linux x64 for modern computers Linux x64 | Author: ElbertoOne
Date: Sun Oct 14 20:40:57 2018 +0200 Timestamp: 1539542457 Simplify check extensions Remove the !moveCountPruning condition for check extensions, which seems not necessary. STC: LLR: 2.95 (-2.94,2.94) [-3.00,1.00] Total: 22238 W: 4835 L: 4715 D: 12688 Elo +1.87 http://tests.stockfishchess.org/tests/view/5bb3241a0ebc592439f6d2ac LTC: LLR: 2.95 (-2.94,2.94) [-3.00,1.00] Total: 36593 W: 5898 L: 5802 D: 24893 Elo +0.91 http://tests.stockfishchess.org/tests/view/5bb34c220ebc592439f6d5dc Bench: 4274207 see source |
| Windows x64 for Haswell CPUs Windows x64 for modern computers Windows x64 Windows 32 Linux x64 for Haswell CPUs Linux x64 for modern computers Linux x64 | Author: Joost VandeVondele
Date: Sun Oct 14 20:33:52 2018 +0200 Timestamp: 1539542032 Randomize draw eval The patch adds a small random component (+-1) to VALUE_DRAW for the evaluation of draw positions (mostly 3folds). This random component is not static, but potentially different for each visit of the node (hence derived from the node counter). The effect is that in positions with many 3fold draw lines, different lines are followed at each iteration. This keeps the search much more dynamic, as opposed to being locked to one particular 3fold. An example of a position where master suffers from 3fold-blindness and this patch solves quickly is the famous TCEC game 53: FEN: 3r2k1/pr6/1p3q1p/5R2/3P3p/8/5RP1/3Q2K1 b - - 0 51 master doesn't see that this is a lost position (draw eval up to depth 50) as Qf6-e6 d4-d5 (found by patch at depth 23) leads to a loss. The 3fold-blindness is more important at longer TC, the patch was yellow STC and LTC, but passed VLTC: STC LLR: -2.95 (-2.94,2.94) [0.00,5.00] Total: 46328 W: 10048 L: 9953 D: 26327 Elo +0.71 http://tests.stockfishchess.org/tests/view/5b9c0ca20ebc592cf275f7c7 LTC LLR: -2.95 (-2.94,2.94) [0.00,5.00] Total: 54663 W: 8938 L: 8846 D: 36879 Elo +0.58 http://tests.stockfishchess.org/tests/view/5b9ca1610ebc592cf27601d3 VLTC LLR: 2.95 (-2.94,2.94) [0.00,5.00] Total: 31789 W: 4512 L: 4284 D: 22993 Elo +2.49 http://tests.stockfishchess.org/tests/view/5b9d1a670ebc592cf276076d Credit to @crossbr for pointing to this problem repeatedly, and giving the hint that many draw lines are typical in those situations. Bench: 4756639 see source |
| Windows x64 for Haswell CPUs Windows x64 for modern computers Windows x64 Windows 32 Linux x64 for Haswell CPUs Linux x64 for modern computers Linux x64 | Author: Guenther Demetz
Date: Sun Oct 14 20:19:46 2018 +0200 Timestamp: 1539541186 Correctly track down pv even in fail-high case Currently we update (track up) the pv even in the fail high case. However most times in such cases the pv in the ply below remains unset because there we have value == alpha and so finally we see truncated pv's (=just one move) in fail high cases. Of course tracking down these pv's (+sending them to the gui) comes at a certian cost, but no-regression tests passed: STC: LLR: 2.96 (-2.94,2.94) [-3.00,1.00] Total: 16300 W: 3556 L: 3424 D: 9320 Elo +2.81 http://tests.stockfishchess.org/tests/view/5b9b73500ebc592cf275ea92 LTC: LLR: 2.96 (-2.94,2.94) [-3.00,1.00] Total: 202411 W: 32734 L: 32897 D: 136780 Elo -0.28 http://tests.stockfishchess.org/tests/view/5b9baed10ebc592cf275ef6d N.B.: Digging also into qsearch was tried in another version but seemed not to pass the tests. This means that we don't always will get a pv until the very tips. No functional change see source |
| Windows x64 for Haswell CPUs Windows x64 for modern computers Windows x64 Windows 32 Linux x64 for Haswell CPUs Linux x64 for modern computers Linux x64 | Author: Miguel Lahoz
Date: Sun Oct 14 20:15:16 2018 +0200 Timestamp: 1539540916 Simplify evaluation of blockers_for_king Currently, we have two evaluation terms which account for pinned pieces. One is for all pinned pieces in kingDanger computation and another for just pinned pawns in ThreatByRank. We can increase the relevant bonus for kingDanger calculation and do away with the ThreatByRank, which seems to just add more complexity. STC: LLR: 2.95 (-2.94,2.94) [-3.00,1.00] Total: 113353 W: 24299 L: 24356 D: 64698 Elo -0.17 http://tests.stockfishchess.org/tests/view/5ba348c20ebc592cf2766e61 LTC: LLR: 2.95 (-2.94,2.94) [-3.00,1.00] Total: 96458 W: 15514 L: 15511 D: 65433 Elo +0.01 http://tests.stockfishchess.org/tests/view/5ba398830ebc592cf2767563 At 100k games, I thought it struggles a bit, but some related [0,4] tests attempting individual tweaks seem to fail: I tried directly tweaking ThreatByRank: http://tests.stockfishchess.org/tests/view/5ba3c6300ebc592cf276791c http://tests.stockfishchess.org/tests/view/5ba3c6190ebc592cf2767917 @Vizveznedec was also recently trying to tweak the same coeffecients for kingDanger calculation: http://tests.stockfishchess.org/tests/view/5ba2c7320ebc592cf27664b2 http://tests.stockfishchess.org/tests/view/5ba2c8220ebc592cf27664b8 http://tests.stockfishchess.org/tests/view/5ba2c7880ebc592cf27664b4 http://tests.stockfishchess.org/tests/view/5ba2c7ce0ebc592cf27664b6 Bench: 4648095 see source |
| Windows x64 for Haswell CPUs Windows x64 for modern computers Windows x64 Windows 32 Linux x64 for Haswell CPUs Linux x64 for modern computers Linux x64 | Author: Joost VandeVondele
Date: Sun Oct 14 20:10:47 2018 +0200 Timestamp: 1539540647 small ttCapture simplification. ttCapture can be assigned to only once outside of the main loop. The patch seems functional at higher depths (seems possible in the case of non-legal TTmoves that are captures). passed STC LLR: 2.94 (-2.94,2.94) [-3.00,1.00] Total: 23189 W: 5098 L: 4980 D: 13111 Elo +1.77 http://tests.stockfishchess.org/tests/view/5bb3822c0ebc592439f6d966 passed LTC LLR: 2.95 (-2.94,2.94) [-3.00,1.00] Total: 10336 W: 1665 L: 1529 D: 7142 Elo +4.57 http://tests.stockfishchess.org/tests/view/5bb39a190ebc592439f6db8a unchanged bench: 4312846 see source |
| Windows x64 for Haswell CPUs Windows x64 for modern computers Windows x64 Windows 32 Linux x64 for Haswell CPUs Linux x64 for modern computers Linux x64 | Author: 31m059
Date: Sun Oct 14 20:02:31 2018 +0200 Timestamp: 1539540151 Combo This PR is a combination of two unrelated [0, 4] patches that appeared promising but not quite strong enough to pass on their own. The combination initially failed STC with a positive score after a long run, and the subsequent speculative LTC test passed. * tweak_threatOnQueen4 : Increase the middlegame components of ThreatByMinor[QUEEN] and ThreatByRook[QUEEN] by 15 each. Bryan's (@crossbr) analysis of CCC Bonus Game 10 inspired several tests on penalizing a queen with limited safe mobility. While attempting to implement this idea, I noticed that when I did not include the queen's current square in the calculations, the Elo gains seemed to vanish--and only then did I have the idea to revisit ThreatByMinor[QUEEN] and ThreatByRook[QUEEN], adding a corresponding value to each. Without Bryan's work, this test would never have been submitted. I would also like to recognize the efforts and contributions of @SFisGOD, who also vigorously worked on this idea. * Use pure static eval for null move pruning : This idea was directly re-purposed from a promising test by Jerry Donald Watson (@jerrydonaldwatson) in August. It was also independently developed and tested by Stefan Geschwentner (@locutus2) previously. Thank you all! STC (failed yellow): LLR: -2.96 (-2.94,2.94) [0.00,4.00] Total: 83913 W: 17986 L: 17825 D: 48102 Elo +0.67 http://tests.stockfishchess.org/tests/view/5bbc59300ebc592439f76aa5 LTC: LLR: 2.95 (-2.94,2.94) [0.00,4.00] Total: 137198 W: 22351 L: 21772 D: 93075 Elo +1.47 http://tests.stockfishchess.org/tests/view/5bbce35f0ebc592439f77639 Bench: 4312846 see source |
| Windows x64 for Haswell CPUs Windows x64 for modern computers Windows x64 Windows 32 Linux x64 for Haswell CPUs Linux x64 for modern computers Linux x64 | Author: Eduardo Caceres
Date: Thu Sep 27 21:39:36 2018 +0200 Timestamp: 1538077176 Fix two typos in comments Note by snicolet: I use this non-functional change patch as a pretext to correct the wrong bench number I introduced in the message of the previous commit. Bench: 4059356 see source |
| Windows x64 for Haswell CPUs Windows x64 for modern computers Windows x64 Windows 32 Linux x64 for Haswell CPUs Linux x64 for modern computers Linux x64 | Author: Joost VandeVondele
Date: Thu Sep 27 21:28:38 2018 +0200 Timestamp: 1538076518 Remove essentially unused code this was added recently as part of a larger commit, but only changes eval of positions at MAX_PLY depth a little. Can be safely removed: passed STC: LLR: 2.95 (-2.94,2.94) [-3.00,1.00] Total: 7424 W: 1640 L: 1492 D: 4292 Elo +6.93 http://tests.stockfishchess.org/html/live_elo.html?5ba3bcbe0ebc592cf27677ff passed LTC: LLR: 2.96 (-2.94,2.94) [-3.00,1.00] Total: 73554 W: 12028 L: 11990 D: 49536 Elo +0.18 http://tests.stockfishchess.org/html/live_elo.html?5ba397ee0ebc592cf2767556 unchanged Bench: 4248710 see source |
| Windows x64 for Haswell CPUs Windows x64 for modern computers Windows x64 Windows 32 Linux x64 for Haswell CPUs Linux x64 for modern computers Linux x64 | Author: protonspring
Date: Thu Sep 27 21:18:18 2018 +0200 Timestamp: 1538075898 Two simplifications in passed pawns evaluation These two simplifications appear to be affecting and/or offsetting each other. Neither can be removed independently, but in combination they pass -3,1. STC LLR: 2.96 (-2.94,2.94) [-3.00,1.00] Total: 36391 W: 7888 L: 7795 D: 20708 Elo +0.89 http://tests.stockfishchess.org/tests/view/5b9bce410ebc592cf275f1b2 LTC LLR: 2.96 (-2.94,2.94) [-3.00,1.00] Total: 19513 W: 3237 L: 3114 D: 13162 Elo +2.19 http://tests.stockfishchess.org/tests/view/5b9c0edf0ebc592cf275f80e Closes https://github.com/official-stockfish/Stockfish/pull/1769 bench 4059356 see source |
| Windows x64 for Haswell CPUs Windows x64 for modern computers Windows x64 Windows 32 Linux x64 for Haswell CPUs Linux x64 for modern computers Linux x64 | Author: Rocky640
Date: Thu Sep 27 20:58:40 2018 +0200 Timestamp: 1538074720 Pawn PSQT Tuned Tested against master "Tweak opposite color bishops endgame scaling" using values from a 100K SPSA with ck=10 Passed STC http://tests.stockfishchess.org/tests/view/5ba7fe7a0ebc592cf276b971 LLR: 2.95 (-2.94,2.94) [0.00,4.00] Total: 27717 W: 6052 L: 5782 D: 15883 Elo +3.38 Passed LTC http://tests.stockfishchess.org/tests/view/5ba815790ebc592cf276bb6b LLR: 2.95 (-2.94,2.94) [0.00,4.00] Total: 17486 W: 2919 L: 2712 D: 11855 Elo +4.11 bench: 4441247 see source |
| Windows x64 for Haswell CPUs Windows x64 for modern computers Windows x64 Windows 32 Linux x64 for Haswell CPUs Linux x64 for modern computers Linux x64 | Author: Joost VandeVondele
Date: Thu Sep 27 20:48:11 2018 +0200 Timestamp: 1538074091 Remove unneeded branch Storing unconditionally the current generation and bound is equivalent to master. Part of the condition was added as a speed optimization in #429. Here the branch is fully eliminated. passed STC single-threaded: LLR: 2.96 (-2.94,2.94) [-3.00,1.00] Total: 73515 W: 16378 L: 16359 D: 40778 Elo +0.09 http://tests.stockfishchess.org/tests/view/5b2fc38c0ebc5902b2e57fd5 passed STC multi-threaded: LLR: 2.95 (-2.94,2.94) [-3.00,1.00] Total: 63725 W: 12916 L: 12874 D: 37935 Elo +0.23 http://tests.stockfishchess.org/tests/view/5b307b8f0ebc5902b2e5895f The multithreaded test was run after a plausible suggestion by @mstembera that the effect of this could be larger with many cores. The result seems to indicate this doesn't really matter on the 8core architecture abundantly available on fishtest. No functional change see source |
| Windows x64 for Haswell CPUs Windows x64 for modern computers Windows x64 Windows 32 Linux x64 for Haswell CPUs Linux x64 for modern computers Linux x64 | Author: Vizvezdenec
Date: Mon Sep 10 12:22:44 2018 +0200 Timestamp: 1536574964 Tweak opposite colord bishops endgame scaling. Make scale factor dependant on asymmetry of pawn structure. STC http://tests.stockfishchess.org/tests/view/5b92a2a80ebc592cf2753dd4 LLR: 2.96 (-2.94,2.94) [0.00,5.00] Total: 31490 W: 6870 L: 6587 D: 18033 Elo +3.12 LTC http://tests.stockfishchess.org/tests/view/5b92f8170ebc592cf2754438 LLR: 2.95 (-2.94,2.94) [0.00,5.00] Total: 54928 W: 8988 L: 8653 D: 37287 Elo +2.12 This patch shows that SF can use some more complicated endgame heuristics to evaluate endgames better from the distance. Closes https://github.com/official-stockfish/Stockfish/pull/1767 Bench: 4248710 see source |
| Windows x64 for Haswell CPUs Windows x64 for modern computers Windows x64 Windows 32 Linux x64 for Haswell CPUs Linux x64 for modern computers Linux x64 | Author: ElbertoOne
Date: Tue Sep 4 10:43:02 2018 +0200 Timestamp: 1536050582 Parameter tweaks in PSQT and NMP This patch is a combinaison of two parameters tweaks patches which have failed as strong yellows at LTC recently, by Alain Savard (Rocky640) and Fabian Fichter (ianfab): http://tests.stockfishchess.org/tests/view/5b8a71e60ebc592cf2749b1d http://tests.stockfishchess.org/tests/view/5b81ce3b0ebc5902bdbb6585 Passed STC: LLR: 2.95 (-2.94,2.94) [0.00,4.00] Total: 57200 W: 12392 L: 12008 D: 32800 Elo +2.33 http://tests.stockfishchess.org/tests/view/5b8d0a5a0ebc592cf274c48f And LTC: LLR: 2.96 (-2.94,2.94) [0.00,4.00] Total: 37215 W: 6233 L: 5962 D: 25020 Elo +2.53 http://tests.stockfishchess.org/tests/view/5b8d56090ebc592cf274cb53 Closes https://github.com/official-stockfish/Stockfish/pull/1764 Bench: 4136116 --------------- How to continue from there? The null move reduction formula in line 769 of search.cpp is quite convoluted and full of mysterious magic constants at the moment, it would certainly be nice to simplify it and/or gain more Elo from it: ``` Depth R = ( (823 + 67 * depth / ONE_PLY) / 256 + std::min(int(eval - beta) / 200, 3)) * ONE_PLY; ``` see source |
| Windows x64 for Haswell CPUs Windows x64 for modern computers Windows x64 Windows 32 Linux x64 for Haswell CPUs Linux x64 for modern computers Linux x64 | Author: Stéphane Nicolet
Date: Mon Sep 3 22:11:30 2018 +0200 Timestamp: 1536005490 Update list of authors And also fix some spaces and formatting oddities in the code. No functional change see source |
| Windows x64 for Haswell CPUs Windows x64 for modern computers Windows x64 Windows 32 Linux x64 for Haswell CPUs Linux x64 for modern computers Linux x64 | Author: Stéphane Nicolet
Date: Sat Sep 1 11:30:38 2018 +0200 Timestamp: 1535794238 Re-introduce "keep pawns on both flanks" Re-introduce the "keep pawns on both flanks" idea. STC yellow: LLR: -2.95 (-2.94,2.94) [0.00,5.00] Total: 93279 W: 20175 L: 19853 D: 53251 Elo +1.20 http://tests.stockfishchess.org/tests/view/5b8a00370ebc592cf274916a LTC: LLR: 2.96 (-2.94,2.94) [0.00,5.00] Total: 11440 W: 1960 L: 1792 D: 7688 Elo +5.10 http://tests.stockfishchess.org/tests/view/5b8a329f0ebc592cf2749615 Closes https://github.com/official-stockfish/Stockfish/pull/1761 Bench: 4609645 see source |
| Windows x64 for Haswell CPUs Windows x64 for modern computers Windows x64 Windows 32 Linux x64 for Haswell CPUs Linux x64 for modern computers Linux x64 | Author: Rocky640
Date: Sat Sep 1 04:33:17 2018 +0200 Timestamp: 1535769197 Long Diagonal Tweaks a) Reduce PSQT values along the long diagonals on non-central squares and increase the LongDiagonal bonus accordingly. The effect is to penalise bishops on the long diagonal which can not "see" the 2 central squares. The "good" bishops still have more or less the same bonus as current master. b) For a bishop on a central square, because of the "| s" term in the code, the LongDiagonalBonus was always given. So while being there, remove the "| s" and compensate the central Bishop PSQT accordingly. Passed STC LLR: 2.95 (-2.94,2.94) [0.00,4.00] Total: 44498 W: 9658 L: 9323 D: 25517 Elo +2.62 http://tests.stockfishchess.org/tests/view/5b8992770ebc592cf2748942 Passed LTC LLR: 2.95 (-2.94,2.94) [0.00,4.00] Total: 63092 W: 10324 L: 9975 D: 42793 Elo +1.92 http://tests.stockfishchess.org/tests/view/5b89a17a0ebc592cf2748b59 Closes https://github.com/official-stockfish/Stockfish/pull/1760 bench: 4693901 see source |
| Windows x64 for Haswell CPUs Windows x64 for modern computers Windows x64 Windows 32 Linux x64 for Haswell CPUs Linux x64 for modern computers Linux x64 | Author: protonspring
Date: Wed Aug 29 02:49:10 2018 +0200 Timestamp: 1535503750 Remove PawnsOnBothFlanks It looks like PawnsOnBothFlanks can be removed from initiative(). A barrage of tests seem to confirm that the adjustment to -110 does not gain elo to offset any potential loss by removing PawnsOnBothFlanks. STC LLR: 2.96 (-2.94,2.94) [-3.00,1.00] Total: 22014 W: 4760 L: 4639 D: 12615 Elo +1.91 http://tests.stockfishchess.org/tests/view/5b7f50cc0ebc5902bdbb3a3e LTC LLR: 2.96 (-2.94,2.94) [-3.00,1.00] Total: 40561 W: 6667 L: 6577 D: 27317 Elo +0.77 http://tests.stockfishchess.org/tests/view/5b801f9f0ebc5902bdbb4467 The barrage of 0,4 tests on the -136 value are in my ps_tunetests branch. http://tests.stockfishchess.org/tests/user/protonspring Closes https://github.com/official-stockfish/Stockfish/pull/1751 Bench: 4413173 ------------- How to continue from there? The fact that endgames with all the pawns on only one flank are drawish is a well-known chess idea, so it seems quite strange that this can be removed so easily without losing Elo. In the past there had been attempts to improve on PawnsOnBothFlanks with similar concepts (for instance using the pawn span value), but the tests were at best neutral. Maybe Stockfish is now mature enough that these refined ideas would work to replace PawnsOnBothFlanks? see source |
| Windows x64 for Haswell CPUs Windows x64 for modern computers Windows x64 Windows 32 Linux x64 for Haswell CPUs Linux x64 for modern computers Linux x64 | Author: MJZ1977
Date: Wed Aug 29 02:28:09 2018 +0200 Timestamp: 1535502489 Fix bug with "excludedMove" for probcut Bugfix: "excludedMove" has to be skipped in the probcut loop too. If it is not skipped, the probcut can exit quickly with a wrong return value corresponding to the excluded move. See the following forum thread for a discussion: https://groups.google.com/forum/?fromgroups=#!topic/fishcooking/GGithf_VwSU STC : LLR: 2.95 (-2.94,2.94) [-3.00,1.00] Total: 17130 W: 3747 L: 3617 D: 9766 Elo +2.64 http://tests.stockfishchess.org/tests/view/5b8460c40ebc5902bdbb999a LTC : LLR: 2.96 (-2.94,2.94) [-3.00,1.00] Total: 12387 W: 2064 L: 1930 D: 8393 Elo +3.76 http://tests.stockfishchess.org/tests/view/5b8466f90ebc5902bdbb9a21 To go further : it can be perhaps useful to tune the singular extension search parameters. Closes https://github.com/official-stockfish/Stockfish/pull/1754 Bench: 4308541 see source |
| Windows x64 for Haswell CPUs Windows x64 for modern computers Windows x64 Windows 32 Linux x64 for Haswell CPUs Linux x64 for modern computers Linux x64 | Author: Steinar H. Gunderson
Date: Wed Aug 29 02:00:20 2018 +0200 Timestamp: 1535500820 Shrink the hash table of tablebases back to 4096 entries There is no need to make this as large as 65536 just for the sake of the single 7-man tablebase that happens to have the key 0xf9247fff. Idea for the fix by Ronald de Man, who suggested simply to allow more buckets past the end. We also implement Robin Hood hashing for the hash table, which takes the worst -case search for full 7-man tablebases down from 68 to 11 probes (Also takes the average probe length from 2.06 to 2.05). For a table with 8K entries, the corresponding numbers would be worst-case from 9 to 4, with average from 1.30 to 1.29. https://github.com/official-stockfish/Stockfish/pull/1747 No functional change see source |
| Windows x64 for Haswell CPUs Windows x64 for modern computers Windows x64 Windows 32 Linux x64 for Haswell CPUs Linux x64 for modern computers Linux x64 | Author: Ondrej Mosnacek
Date: Wed Aug 29 01:24:45 2018 +0200 Timestamp: 1535498685 Refactor pure static eval code This commit tries to make the new pure static eval code more readable by splitting up the nested assignments into separate lines and making a few more cosmetic tweaks. No functional change. see source |
| Windows x64 for Haswell CPUs Windows x64 for modern computers Windows x64 Windows 32 Linux x64 for Haswell CPUs Linux x64 for modern computers Linux x64 | Author: protonspring
Date: Wed Aug 29 01:07:38 2018 +0200 Timestamp: 1535497658 make DistanceRing more consistent This is a non-functional change. By pre-incrementing minKingPawnDistance instead of post-incrementing, we can remove this -1. This also makes DistanceRing more consistent with the rest of stockfish since it now holds an actual "distance" instead of a less natural distance-1. In current master, PseudoAttacks[KING][ksq] == DistanceRingBB[ksq][0] With this patch, it will be PseudoAttacks[KING][ksq] == DistanceRingBB[ksq][1] ie squares at distance 1 from the king. This is more natural use of distance. The current array size DistanceRingBB[SQUARE_NB][8] is still OK with the new definition, because maximum distance between two squares on a chess board is seven (for example Kh1 and a8). No functional change. see source |
| Windows x64 for Haswell CPUs Windows x64 for modern computers Windows x64 Windows 32 Linux x64 for Haswell CPUs Linux x64 for modern computers Linux x64 | Author: Vizvezdenec
Date: Wed Aug 29 00:53:31 2018 +0200 Timestamp: 1535496811 Tweak stat bonus formula Tweak stat bonus formula on top of latest elo gain by @snicolet STC http://tests.stockfishchess.org/tests/view/5b830a810ebc5902bdbb7e9c LLR: 2.95 (-2.94,2.94) [0.00,4.00] Total: 27797 W: 6113 L: 5842 D: 15842 Elo +3.39 LTC http://tests.stockfishchess.org/tests/view/5b831f2c0ebc5902bdbb8038 LLR: 2.95 (-2.94,2.94) [0.00,4.00] Total: 13655 W: 2294 L: 2099 D: 9262 Elo +4.96 I think that more elo can be found in tweaks of this parameters so I plan to further try some "hand-tuning", including increasing/decreasing ratio of two constants and making bonus assimetric to 0. Thx to @AndyGrant for helping with github and @jerrydonaldwatson for original idea. Closes https://github.com/official-stockfish/Stockfish/pull/1748 Bench: 4172767 see source |
| Windows x64 for Haswell CPUs Windows x64 for modern computers Windows x64 Windows 32 Linux x64 for Haswell CPUs Linux x64 for modern computers Linux x64 | Author: VoyagerOne
Date: Wed Aug 29 00:41:53 2018 +0200 Timestamp: 1535496113 Don't modify Eval with search stats at ttHits STC: LLR: 2.95 (-2.94,2.94) [-3.00,1.00] Total: 28344 W: 6148 L: 6040 D: 16156 Elo +1.32 http://tests.stockfishchess.org/tests/view/5b7d6b4e0ebc5902bdbb1914 LTC: LLR: 2.96 (-2.94,2.94) [-3.00,1.00] Total: 41084 W: 6769 L: 6680 D: 27635 Elo +0.75 http://tests.stockfishchess.org/tests/view/5b7d7f5b0ebc5902bdbb1b85 Bench: 4457440 see source |
| Windows x64 for Haswell CPUs Windows x64 for modern computers Windows x64 Windows 32 Linux x64 for Haswell CPUs Linux x64 for modern computers Linux x64 | Author: Stefan Geschwentner
Date: Mon Aug 20 21:52:29 2018 +0200 Timestamp: 1534794749 Store only unchanged static evaluations in TT A recent commit introduced a decrease of the static evaluation of an inner node dependent on the previous stat score, which finally was also stored in the transposition table. Now only the unchanged static evaluation are stored there. Remark: For the case that a static evaluation can be retrieved from the transposition table the value is now used unchanged. Another test which also applies the modification in this case failed: http://tests.stockfishchess.org/tests/view/5b7af6df0ebc5902bdbae2f6 STC: LLR: 2.95 (-2.94,2.94) [0.00,5.00] Total: 6707 W: 1547 L: 1383 D: 3777 Elo +8.50 http://tests.stockfishchess.org/tests/view/5b7a92df0ebc5902bdbadcf3 LTC: LLR: 2.95 (-2.94,2.94) [0.00,5.00] Total: 36203 W: 6046 L: 5781 D: 24376 Elo +2.54 http://tests.stockfishchess.org/tests/view/5b7abaa10ebc5902bdbadfa9 Closes https://github.com/official-stockfish/Stockfish/pull/1742 Bench: 4457440 see source |