Günther Demetz [Tue, 23 Oct 2018 16:04:30 +0000]
Revert Pull Request #1771, see issue #1785 (#1786)
no functional change
bench: 4274207
mstembera [Sun, 14 Oct 2018 19:50:31 +0000]
Small simplification in castling rights
There is no need for a special struct with a static member
to generate castling rights.
No functional change.
Fabian Fichter [Sat, 20 Oct 2018 14:17:44 +0000]
Avoid pawn hash collissions with shogi pawns
shogi
ELO: 16.52 +-33.3 (95%) LOS: 83.5%
Total: 400 W: 200 L: 181 D: 19
euroshogi
ELO: 16.52 +-32.5 (95%) LOS: 84.1%
Total: 400 W: 191 L: 172 D: 37
minishogi
ELO: -13.90 +-26.2 (95%) LOS: 14.8%
Total: 400 W: 110 L: 126 D: 164
No functional change for non-shogi variants.
Fabian Fichter [Sat, 20 Oct 2018 12:36:20 +0000]
Tweak capture the flag bonus
king of the hill STC
LLR: 2.96 (-2.94,2.94) [0.00,10.00]
Total: 2713 W: 1223 L: 1095 D: 395
http://35.161.250.236:6543/tests/view/
5bc9a43a6e23db7639060c0f
racing kings STC
LLR: 2.95 (-2.94,2.94) [0.00,10.00]
Total: 1961 W: 955 L: 833 D: 173
http://35.161.250.236:6543/tests/view/
5bc9a44b6e23db7639060c12
Fabian Fichter [Thu, 18 Oct 2018 21:49:33 +0000]
Smoothen PSQTs for large boards
shogi
ELO: 36.62 +-47.5 (95%) LOS: 93.7%
Total: 200 W: 106 L: 85 D: 9
No functional change for board sizes <= 8x8.
ElbertoOne [Tue, 2 Oct 2018 07:07:29 +0000]
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
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
http://tests.stockfishchess.org/tests/view/
5bb34c220ebc592439f6d5dc
Bench: 4274207
Joost VandeVondele [Sun, 14 Oct 2018 18:33:24 +0000]
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
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
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
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
Guenther Demetz [Mon, 17 Sep 2018 06:42:58 +0000]
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
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
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
Miguel Lahoz [Thu, 20 Sep 2018 07:12:12 +0000]
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
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
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
Joost VandeVondele [Tue, 2 Oct 2018 14:34:40 +0000]
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
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
http://tests.stockfishchess.org/tests/view/
5bb39a190ebc592439f6db8a
unchanged bench: 4312846
31m059 [Mon, 8 Oct 2018 18:43:45 +0000]
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
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
http://tests.stockfishchess.org/tests/view/
5bbce35f0ebc592439f77639
Bench: 4312846
Fabian Fichter [Sat, 13 Oct 2018 10:48:01 +0000]
Add variants with large boards to readme
Fabian Fichter [Sat, 6 Oct 2018 17:34:25 +0000]
Support Courier chess
Introduce piece type wazir to support courier chess.
bench: 4648488
Fabian Fichter [Sat, 6 Oct 2018 16:43:11 +0000]
Support Jeson Mor
https://en.wikipedia.org/wiki/Jeson_Mor
Positions are considered to be winning if a player has the move
and one of his knights is on e5.
No functional change for other variants.
Fabian Fichter [Sat, 6 Oct 2018 15:48:45 +0000]
Support Janus and Embassy chess
No functional change for other variants.
Fabian Fichter [Sat, 29 Sep 2018 21:27:55 +0000]
Support capablanca chess (#6)
Generalize castling and add archbishop to support capablanca chess.
bench: 4636755
Fabian Fichter [Thu, 27 Sep 2018 20:18:34 +0000]
Add support for large-board variants (#6)
Add compile-time option for enabling a 12x10 board representation with
128 bit bitboards to enable support for chess variants with large boards.
Support shogi as first variant with board size >8x8.
Open issues for large board version:
- Evaluation needs to be consolidated (esp. PSQT)
- Endgame evaluation is disabled and requires to be fixed
before reenabling it.
- Memory consumption of history tables needs to be reduced.
No functional change for normal version.
Fabian Fichter [Sun, 7 Oct 2018 13:59:18 +0000]
Avoid overflow in bit-shift operation
Fixes error in Travis build.
Fabian Fichter [Sun, 30 Sep 2018 09:19:42 +0000]
Simplify move encoding
Using 32 bits for move encoding, the encoding of promotions
can be simplified, which results in a 5-10% speed-up.
Tested in combination with previous commit.
STC
ELO: 21.39 +-10.0 (95%) LOS: 100.0%
Total: 2000 W: 491 L: 368 D: 1141
http://35.161.250.236:6543/tests/view/
5bb096486e23db0fbab0ddf5
LTC
ELO: 14.95 +-9.3 (95%) LOS: 99.9%
Total: 2000 W: 420 L: 334 D: 1246
http://35.161.250.236:6543/tests/view/
5bb112c56e23db0fbab0de0a
No functional change.
Fabian Fichter [Sat, 29 Sep 2018 12:36:53 +0000]
Reserve 32 bit for moves in TT entries
Increases size of TT entries by 20%.
STC
ELO: 1.22 +-10.0 (95%) LOS: 59.4%
Total: 2000 W: 439 L: 432 D: 1129
http://35.161.250.236:6543/tests/view/
5baf7ef76e23db0fbab0ddee
LTC
ELO: -6.43 +-9.5 (95%) LOS: 9.2%
Total: 2000 W: 371 L: 408 D: 1221
http://35.161.250.236:6543/tests/view/
5baffcbf6e23db0fbab0ddf2
bench: 4985837
Fabian Fichter [Sat, 6 Oct 2018 14:36:55 +0000]
Reenable appveyor builds
Fabian Fichter [Fri, 5 Oct 2018 08:27:30 +0000]
Disable appveyor for now
Fabian Fichter [Sat, 29 Sep 2018 22:33:18 +0000]
Skip razoring for antichess variants
STC giveaway
LLR: 2.95 (-2.94,2.94) [0.00,10.00]
Total: 6849 W: 2738 L: 2561 D: 1550
http://35.161.250.236:6543/tests/view/
5b9bedb66e23db0fbab0dd7a
STC losers
LLR: -2.98 (-2.94,2.94) [-10.00,5.00]
Total: 1882 W: 805 L: 881 D: 196
http://35.161.250.236:6543/tests/view/
5babd6496e23db0fbab0dddf
LTC losers
LLR: 2.98 (-2.94,2.94) [-10.00,5.00]
Total: 1095 W: 510 L: 456 D: 129
http://35.161.250.236:6543/tests/view/
5bad0b6f6e23db0fbab0dde2
Fabian Fichter [Fri, 28 Sep 2018 08:54:08 +0000]
Update readme
Eduardo Caceres [Fri, 21 Sep 2018 21:18:46 +0000]
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
Joost VandeVondele [Thu, 20 Sep 2018 12:50:39 +0000]
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
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
http://tests.stockfishchess.org/html/live_elo.html?
5ba397ee0ebc592cf2767556
unchanged Bench: 4248710
protonspring [Fri, 14 Sep 2018 15:04:11 +0000]
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
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
http://tests.stockfishchess.org/tests/view/
5b9c0edf0ebc592cf275f80e
Closes https://github.com/official-stockfish/Stockfish/pull/1769
bench 4059356
Rocky640 [Sun, 23 Sep 2018 20:57:57 +0000]
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
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
bench: 4441247
Joost VandeVondele [Sun, 24 Jun 2018 16:14:48 +0000]
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
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
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
Fabian Fichter [Fri, 14 Sep 2018 17:18:21 +0000]
Skip razoring for capture the flag variants
STC king of the hill
LLR: 2.96 (-2.94,2.94) [0.00,10.00]
Total: 4999 W: 2157 L: 1999 D: 843
http://35.161.250.236:6543/tests/view/
5b9bede96e23db0fbab0dd7d
STC racing kings
LLR: 2.97 (-2.94,2.94) [0.00,10.00]
Total: 6888 W: 3098 L: 2909 D: 881
http://35.161.250.236:6543/tests/view/
5b9bee026e23db0fbab0dd81
ianfab [Tue, 7 Aug 2018 21:38:19 +0000]
Skip shallow depth pruning when there are forced captures
STC losers
LLR: 2.97 (-2.94,2.94) [0.00,10.00]
Total: 3558 W: 1779 L: 1629 D: 150
http://35.161.250.236:6543/tests/view/
5b6a11c66e23db0fbab0dc43
STC giveaway
LLR: 2.97 (-2.94,2.94) [0.00,10.00]
Total: 16582 W: 8319 L: 7979 D: 284
http://35.161.250.236:6543/tests/view/
5b6a11a86e23db0fbab0dc40
Fabian Fichter [Sun, 23 Sep 2018 09:55:20 +0000]
Disable endgame evaluation for shatranj
STC shatranj
LLR: 2.95 (-2.94,2.94) [0.00,10.00]
Total: 229 W: 76 L: 14 D: 139
http://35.161.250.236:6543/tests/view/
5ba763b06e23db0fbab0ddc7
Fabian Fichter [Sat, 22 Sep 2018 12:01:17 +0000]
Skip SEE pruning of captures for antichess variants
STC giveaway
LLR: 2.97 (-2.94,2.94) [0.00,10.00]
Total: 291 W: 147 L: 64 D: 80
http://35.161.250.236:6543/tests/view/
5ba630ff6e23db0fbab0ddc1
STC losers
LLR: 2.97 (-2.94,2.94) [0.00,10.00]
Total: 661 W: 331 L: 232 D: 98
http://35.161.250.236:6543/tests/view/
5ba631176e23db0fbab0ddc4
Fabian Fichter [Thu, 20 Sep 2018 20:08:15 +0000]
Tweak threat evaluation for antichess variants
STC giveaway
LLR: 2.95 (-2.94,2.94) [0.00,10.00]
Total: 462 W: 226 L: 136 D: 100
http://35.161.250.236:6543/tests/view/
5ba3fe7a6e23db0fbab0ddaa
STC losers
LLR: 2.97 (-2.94,2.94) [0.00,10.00]
Total: 334 W: 195 L: 99 D: 40
http://35.161.250.236:6543/tests/view/
5ba3fe8b6e23db0fbab0ddad
Fabian Fichter [Thu, 20 Sep 2018 19:42:11 +0000]
Skip SEE pruning of quiet moves for antichess variants
STC giveaway
LLR: 2.94 (-2.94,2.94) [0.00,10.00]
Total: 108 W: 86 L: 7 D: 15
http://35.161.250.236:6543/tests/view/
5ba3f8816e23db0fbab0dda0
STC losers
LLR: 2.95 (-2.94,2.94) [0.00,10.00]
Total: 285 W: 180 L: 82 D: 23
http://35.161.250.236:6543/tests/view/
5ba3fc4f6e23db0fbab0dda8
Fabian Fichter [Fri, 14 Sep 2018 17:26:33 +0000]
Skip futility pruning for extinction variants
STC extinction
LLR: 2.97 (-2.94,2.94) [0.00,10.00]
Total: 419 W: 251 L: 147 D: 21
http://35.161.250.236:6543/tests/view/
5b9bef726e23db0fbab0dd88
Fabian Fichter [Sat, 1 Sep 2018 09:22:40 +0000]
Tweak shelter strength for drop variants
STC crazyhouse
LLR: 2.95 (-2.94,2.94) [0.00,10.00]
Total: 7305 W: 3696 L: 3490 D: 119
http://35.161.250.236:6543/tests/view/
5b8a5c546e23db0fbab0dd0f
Fabian Fichter [Sun, 16 Sep 2018 21:52:52 +0000]
Fix #7: Checkmate evaluation for shatar and shogi
Evaluate illegal checkmates as losing.
Fabian Fichter [Sun, 16 Sep 2018 16:28:17 +0000]
Add king attack weights for fairy pieces
minishogi
ELO: 40.13 +-45.2 (95%) LOS: 96.1%
Total: 200 W: 98 L: 75 D: 27
euroshogi
ELO: 13.90 +-47.7 (95%) LOS: 71.7%
Total: 200 W: 101 L: 93 D: 6
makruk
ELO: -3.47 +-25.9 (95%) LOS: 39.6%
Total: 200 W: 28 L: 30 D: 142
shatranj
ELO: 12.17 +-28.3 (95%) LOS: 80.1%
Total: 200 W: 38 L: 31 D: 131
Fabian Fichter [Sun, 16 Sep 2018 16:26:44 +0000]
Consider shogi pawns in shelter evaluation
euroshogi
ELO: 41.89 +-47.2 (95%) LOS: 96.1%
Total: 200 W: 106 L: 82 D: 12
minishogi
ELO: 6.95 +-43.8 (95%) LOS: 62.3%
Total: 200 W: 84 L: 80 D: 36
Fabian Fichter [Sun, 16 Sep 2018 14:21:24 +0000]
Guard against overflow in piece types
No functional change.
Fabian Fichter [Sun, 2 Sep 2018 14:55:12 +0000]
Less early pruning for nCheck variants
STC 3check
LLR: 2.95 (-2.94,2.94) [0.00,10.00]
Total: 4562 W: 2177 L: 2017 D: 368
http://35.161.250.236:6543/tests/view/
5b8bfa486e23db0fbab0dd37
Fabian Fichter [Sat, 15 Sep 2018 14:19:47 +0000]
Add tic-tac-toe
Fabian Fichter [Sat, 8 Sep 2018 13:01:43 +0000]
No initiative bonus for extinction variants
STC giveaway
LLR: 2.96 (-2.94,2.94) [0.00,10.00]
Total: 1451 W: 646 L: 539 D: 266
http://35.161.250.236:6543/tests/view/
5b8bfc716e23db0fbab0dd3a
STC extinction
LLR: 2.95 (-2.94,2.94) [0.00,10.00]
Total: 2590 W: 1273 L: 1140 D: 177
http://35.161.250.236:6543/tests/view/
5b93c8af6e23db0fbab0dd66
ianfab [Sat, 25 Aug 2018 17:32:26 +0000]
Bonus for pieces in hand
STC crazyhouse
LLR: 2.96 (-2.94,2.94) [0.00,10.00]
Total: 5728 W: 2892 L: 2709 D: 127
http://35.161.250.236:6543/tests/view/
5b8192f96e23db0fbab0dcd9
Fabian Fichter [Mon, 10 Sep 2018 20:54:18 +0000]
Support connect4
bench: 4832716
Vizvezdenec [Mon, 10 Sep 2018 10:02:43 +0000]
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
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
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
Fabian Fichter [Sun, 2 Sep 2018 15:10:00 +0000]
Skip king safety evaluation if checking is illegal
STC racingkings
LLR: 2.97 (-2.94,2.94) [0.00,10.00]
Total: 120 W: 106 L: 10 D: 4
http://35.161.250.236:6543/tests/view/
5b8bfdd66e23db0fbab0dd3f
Fabian Fichter [Sat, 1 Sep 2018 09:24:37 +0000]
Increase king danger for drop variants
STC crazyhouse
LLR: 2.97 (-2.94,2.94) [0.00,10.00]
Total: 1516 W: 805 L: 682 D: 29
http://35.161.250.236:6543/tests/view/
5b8a5c6f6e23db0fbab0dd11
ElbertoOne [Mon, 3 Sep 2018 10:15:31 +0000]
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
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
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;
```
Stéphane Nicolet [Mon, 3 Sep 2018 10:46:05 +0000]
Update list of authors
And also fix some spaces and formatting oddities in the code.
No functional change
Fabian Fichter [Sat, 1 Sep 2018 09:21:05 +0000]
Tweak LMR pruning for mandatory captures
STC giveaway
LLR: 2.96 (-2.94,2.94) [0.00,10.00]
Total: 799 W: 359 L: 264 D: 176
http://35.161.250.236:6543/tests/view/
5b8a5c376e23db0fbab0dd0d
Fabian Fichter [Sat, 1 Sep 2018 09:25:45 +0000]
Tweak piece values for variants without checks
STC racingkings
LLR: 2.97 (-2.94,2.94) [0.00,10.00]
Total: 349 W: 211 L: 111 D: 27
http://35.161.250.236:6543/tests/view/
5b8a5c926e23db0fbab0dd13
ianfab [Sat, 25 Aug 2018 17:37:58 +0000]
Increase close enemies penalty
For drop and nCheck variants, increase close enemies penalty.
STC 3check
LLR: 2.96 (-2.94,2.94) [0.00,10.00]
Total: 1504 W: 762 L: 644 D: 98
http://35.161.250.236:6543/tests/view/
5b8194426e23db0fbab0dce0
STC crazyhouse
LLR: 2.95 (-2.94,2.94) [0.00,10.00]
Total: 3254 W: 1665 L: 1518 D: 71
http://35.161.250.236:6543/tests/view/
5b81942c6e23db0fbab0dcdd
ianfab [Sat, 25 Aug 2018 17:42:43 +0000]
Increase mobility bonus
Double mobility bonus for drop and antichess variants.
STC giveaway
LLR: 2.96 (-2.94,2.94) [0.00,10.00]
Total: 1946 W: 816 L: 705 D: 425
http://35.161.250.236:6543/tests/view/
5b81957f6e23db0fbab0dce8
STC crazyhouse
LLR: 2.95 (-2.94,2.94) [0.00,10.00]
Total: 3138 W: 1617 L: 1471 D: 50
http://35.161.250.236:6543/tests/view/
5b8195546e23db0fbab0dce2
Fabian Fichter [Sun, 2 Sep 2018 13:10:24 +0000]
Symmetrize evaluation for smaller board sizes
Piece-square tables remain asymmetrical.
Fabian Fichter [Sat, 1 Sep 2018 09:26:59 +0000]
Tweak tempo value for drop variants
STC crazyhouse
LLR: 2.95 (-2.94,2.94) [0.00,10.00]
Total: 773 W: 436 L: 324 D: 13
http://35.161.250.236:6543/tests/view/
5b8a5cb16e23db0fbab0dd15
Stéphane Nicolet [Sat, 1 Sep 2018 08:39:29 +0000]
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
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
http://tests.stockfishchess.org/tests/view/
5b8a329f0ebc592cf2749615
Closes https://github.com/official-stockfish/Stockfish/pull/1761
Bench: 4609645
Rocky640 [Fri, 31 Aug 2018 13:30:16 +0000]
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
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
http://tests.stockfishchess.org/tests/view/
5b89a17a0ebc592cf2748b59
Closes https://github.com/official-stockfish/Stockfish/pull/1760
bench: 4693901
protonspring [Wed, 29 Aug 2018 00:48:45 +0000]
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
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
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?
MJZ1977 [Wed, 29 Aug 2018 00:27:47 +0000]
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
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
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
Steinar H. Gunderson [Wed, 29 Aug 2018 00:00:14 +0000]
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
Ondrej Mosnacek [Tue, 28 Aug 2018 23:22:22 +0000]
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.
protonspring [Mon, 20 Aug 2018 16:29:46 +0000]
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.
Vizvezdenec [Tue, 28 Aug 2018 22:50:55 +0000]
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
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
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
VoyagerOne [Wed, 22 Aug 2018 19:51:20 +0000]
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
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
http://tests.stockfishchess.org/tests/view/
5b7d7f5b0ebc5902bdbb1b85
Bench: 4457440
Fabian Fichter [Tue, 28 Aug 2018 18:19:27 +0000]
Speed up attackers_to
Loop only over relevant piece types for a speed-up with no functional change.
ianfab [Sat, 25 Aug 2018 17:27:42 +0000]
Consider pieces in hand in king danger
STC crazyhouse
LLR: 2.96 (-2.94,2.94) [0.00,10.00]
Total: 583 W: 340 L: 231 D: 12
http://35.161.250.236:6543/tests/view/
5b8191f06e23db0fbab0dcd6
ianfab [Thu, 23 Aug 2018 20:26:20 +0000]
Remove lazy evaluation
STC racing kings
LLR: 2.96 (-2.94,2.94) [0.00,10.00]
Total: 160 W: 117 L: 26 D: 17
http://35.161.250.236:6543/tests/view/
5b7f9fca6e23db0fbab0dcc3
STC crazyhouse
ELO: 17.91 +-15.1 (95%) LOS: 99.0%
Total: 2000 W: 1037 L: 934 D: 29
http://35.161.250.236:6543/tests/view/
5b7f18a36e23db0fbab0dcb9
STC chess
ELO: -2.61 +-10.0 (95%) LOS: 30.5%
Total: 2000 W: 426 L: 441 D: 1133
http://35.161.250.236:6543/tests/view/
5b805f6b6e23db0fbab0dcca
bench: 5056893
Fabian Fichter [Mon, 27 Aug 2018 18:51:33 +0000]
Tweak probcut margin for nCheck chess
STC 3check
LLR: 2.97 (-2.94,2.94) [0.00,10.00]
Total: 5912 W: 2839 L: 2658 D: 415
http://35.161.250.236:6543/tests/view/
5b71df4f6e23db0fbab0dca6
ianfab [Wed, 22 Aug 2018 19:58:45 +0000]
More robust shogi move parsing
Allow optional equals sign at end of move string.
Stefan Geschwentner [Mon, 20 Aug 2018 10:06:52 +0000]
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
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
http://tests.stockfishchess.org/tests/view/
5b7abaa10ebc5902bdbadfa9
Closes https://github.com/official-stockfish/Stockfish/pull/1742
Bench: 4457440
Stéphane Nicolet [Fri, 17 Aug 2018 23:23:36 +0000]
Use an affine formula to mix stats and eval
Follow-up for the previous patch: we use an affine formula to mix stats
and evaluation in search. The idea is to give a bonus if the previous
move of the opponent was historically bad, and a malus if the previous
move of the opponent was historically good.
More precisely, if x is the stat score of the previous move by the opponent,
we implement the following formulas to tweak the evaluation at an internal
node of the tree for our pruning decisions at this node:
if x = 0, use v' = eval(P)
if x > 0, use v' = eval(P) - 5 - x/1024
if x < 0, use v' = eval(P) + 5 - x/1024
For reference, the previous master had this simpler rule:
if x > 0, use v' = eval(P) - 10
if x <= 0, use v' = eval(P)
STC:
LLR: 2.95 (-2.94,2.94) [0.00,5.00]
Total: 29322 W: 6359 L: 6088 D: 16875
http://tests.stockfishchess.org/tests/view/
5b76a5980ebc5902bdba957f
LTC:
LLR: 2.96 (-2.94,2.94) [0.00,5.00]
Total: 30893 W: 5154 L: 4910 D: 20829
http://tests.stockfishchess.org/tests/view/
5b76ca6d0ebc5902bdba9914
Closes https://github.com/official-stockfish/Stockfish/pull/1740
Bench: 4592766
Fabian Fichter [Fri, 17 Aug 2018 20:34:05 +0000]
Update readme with new variants
ianfab [Fri, 17 Aug 2018 19:24:03 +0000]
Support codrus chess
Variant of antichess with forced captures; losing the king wins.
http://www.chessvariants.com/diffobjective.dir/giveaway.html
http://www.binnewirtz.com/Schlagschach1.htm
ianfab [Fri, 17 Aug 2018 18:31:23 +0000]
Support breakthrough
https://en.wikipedia.org/wiki/Breakthrough_(board_game)
bench: 5016237
ianfab [Fri, 17 Aug 2018 18:06:40 +0000]
Insert missing value for clobber pieces
VoyagerOne [Thu, 16 Aug 2018 21:58:48 +0000]
Mix search stats with evaluation
Mix search stats with evaluation: if the opponent's move has a good historyStat,
then decrease the evaluation of the internal node a bit for the pruning decisions
during search.
STC;
LLR: 2.96 (-2.94,2.94) [0.00,5.00]
Total: 72083 W: 15683 L: 15203 D: 41197
http://tests.stockfishchess.org/tests/view/
5b74c3ea0ebc5902bdba7d41
LTC:
LLR: 2.95 (-2.94,2.94) [0.00,5.00]
Total: 29104 W: 4867 L: 4630 D: 19607
http://tests.stockfishchess.org/tests/view/
5b7565000ebc5902bdba851b
Closes https://github.com/official-stockfish/Stockfish/pull/1738
Bench: 4514101
-----------
How to continue from there?
• the use of the previous stat score can probably be simplified in lines 587 and 716
• we could try to use a continuous bonus based on the previous stat score, instead
of just a fixed offset of -10 when the opponent previous move was good.
----------
Comments by Stefan Geschwentner:
Interesting idea. Because only the eval in search is tweak this should only
influence the eval and static eval used at inner nodes, and not on the return
search value (which comes in the end from quiescence search), except through
saving in TT followed by a TT cutoff.
So essentialy this effects diverse pruning/reduction parts -- eval and static
eval are lowered for good opponent moves:
• tt cutoff (ttValue)
• improving (static eval)
• more razoring (eval)
• less futility pruning (eval)
• less null move pruning (eval + static eval) (but with little more depth)
• more probcut (static eval)
• more move futility pruning (static eval)
protonspring [Fri, 17 Aug 2018 08:19:32 +0000]
Simplify king file dependancy in evaluate_shelter()
Remove the special value we used for the file of the king in the
evaluate_shelter() function, and compensate by tweaking some of
the ShelterStrength[] array values.
STC
LLR: 2.94 (-2.94,2.94) [-3.00,1.00]
Total: 17069 W: 3782 L: 3652 D: 9635
http://tests.stockfishchess.org/tests/view/
5b75eb0d0ebc5902bdba8f3d
LTC
LLR: 2.95 (-2.94,2.94) [-3.00,1.00]
Total: 42639 W: 6973 L: 6887 D: 28779
http://tests.stockfishchess.org/tests/view/
5b75fd7f0ebc5902bdba906b
Closes https://github.com/official-stockfish/Stockfish/pull/1739
Bench: 4639508
ianfab [Thu, 16 Aug 2018 20:55:26 +0000]
Support clobber
https://en.wikipedia.org/wiki/Clobber
bench: 4737198
ianfab [Tue, 14 Aug 2018 20:31:48 +0000]
Support horde chess
- Add horde chess.
- Increase piece list size.
- Add configuration option for pawn double steps from first rank.
No functional change for existing variants.
ianfab [Mon, 13 Aug 2018 19:41:05 +0000]
Tweak probcut margin for extinction variants
extinction STC
LLR: 2.96 (-2.94,2.94) [0.00,10.00]
Total: 411 W: 238 L: 138 D: 35
http://35.161.250.236:6543/tests/view/
5b71df356e23db0fbab0dca3
ianfab [Mon, 13 Aug 2018 19:28:44 +0000]
Tweak bonus for racing kings
racing kings STC
LLR: 2.96 (-2.94,2.94) [0.00,10.00]
Total: 469 W: 262 L: 162 D: 45
http://35.161.250.236:6543/tests/view/
5b71dcbb6e23db0fbab0dca0
Stéphane Nicolet [Tue, 14 Aug 2018 07:24:26 +0000]
Double weight of capture history
We double in this patch the weight of the capture history table in the
local scoring of captures for move ordering.
The capture history table is indexed by the triplet (capturing piece,
capture square, captured piece) and gets information like "it seems to
have been historically good in that part of the search tree to capture
a pawn with a rook on g3, even if it seems to lose material", and affect
the normaly pure « Most Valuable Victim » ordering of captures.
Finished yellow at STC after 228842 games (posting a +1.36 Elo gain):
LLR: -2.95 (-2.94,2.94) [0.00,4.00]
Total: 228842 W: 50894 L: 50152 D: 127796
http://tests.stockfishchess.org/tests/view/
5b714bb00ebc5902bdba332d
Passed LTC:
LLR: 2.96 (-2.94,2.94) [0.00,4.00]
Total: 43251 W: 7425 L: 7131 D: 28695
http://tests.stockfishchess.org/tests/view/
5b71c7d40ebc5902bdba3e51
Thanks to user Vizvezdenec for running the LTC test.
Closes https://github.com/official-stockfish/Stockfish/pull/1736
Bench: 4272361
Alain SAVARD [Tue, 14 Aug 2018 06:31:56 +0000]
Remove pawncount array in imbalance
This is a natural follow up to last commit where values on the
QuadraticOurs diagonal and some piece value deltas were changed.
@Stefano80 tried to simplify the newly introduced pawncount array
using QuadraticOurs[1][1] =52 and a -30 adjustment on pawn values
His STC [-3,1] was green
http://tests.stockfishchess.org/tests/view/
5b707f5b0ebc5902bdba2745
but not his LTC[-3,1]
http://tests.stockfishchess.org/tests/view/
5b7095700ebc5902bdba2a49
So I started a 80000 30+0.3 SPSA on the QuadraticOurs diagonal and
on the piece values using @Stefano80 start values.
SPSA gave the new values QuadraticOurs[1][1] =38 and a -33 on pawn
values (the other changes on QuadraticOurs were kept, but were not
ignificant according to this test
http://tests.stockfishchess.org/tests/view/
5b710ccb0ebc5902bdba2f27)
STC
http://tests.stockfishchess.org/tests/view/
5b710b220ebc5902bdba2f19
LLR: 2.95 (-2.94,2.94) [-3.00,1.00]
Total: 50902 W: 11214 L: 11150 D: 28538
LTC
http://tests.stockfishchess.org/tests/view/
5b7124ef0ebc5902bdba3106
LLR: 2.96 (-2.94,2.94) [-3.00,1.00]
Total: 34271 W: 5852 L: 5753 D: 22666
Closes https://github.com/official-stockfish/Stockfish/pull/1735
bench: 4738555
GuardianRM [Sun, 12 Aug 2018 16:40:03 +0000]
Non-linear bonus for pawn count
This patch introduces a non-linear bonus for pawns, along with some
(linear) corrections for the other pieces types.
The original values were obtained by a massive non-linear tuning of both
pawns and other pieces by GuardianRM, while Alain Savard and Chris Cain
later simplified the patch by observing that, apart from the pawn case, the
tuned corrections were in fact almost affine and could be incorporated in
our current code base via the piece values in types.h (offset) and the diagonal
of the quadratic matrix (slope). See discussion in PR#1725 :
https://github.com/official-stockfish/Stockfish/pull/1725
STC:
LLR: 2.97 (-2.94,2.94) [0.00,5.00]
Total: 42948 W: 9662 L: 9317 D: 23969
http://tests.stockfishchess.org/tests/view/
5b6ff6e60ebc5902bdba1d87
LTC:
LLR: 2.97 (-2.94,2.94) [0.00,5.00]
Total: 19683 W: 3409 L: 3206 D: 13068
http://tests.stockfishchess.org/tests/view/
5b702dbd0ebc5902bdba216b
How to continue from there?
- Maybe the non-linearity for the pawn value could be somewhat tempered
again and a simpler linear correction for pawns would work?
Closes https://github.com/official-stockfish/Stockfish/pull/1734
Bench: 4681496
ianfab [Sun, 12 Aug 2018 09:49:53 +0000]
Fix #5: Set attack bitboards for all piece types
This fixes issues with uninitialized attack tables.
Bench results are deterministic now also for variants
that are missing some of the standard chess piece types.
No functional change for standard chess.
Fabian Fichter [Sun, 12 Aug 2018 08:44:59 +0000]
Reference wiki in readme
Stefano Cardanobile [Sun, 12 Aug 2018 08:02:51 +0000]
Combo of several promising parameter tweaks
Combo of several tuning patches which finished yellow at LTC.
[STC](http://tests.stockfishchess.org/tests/view/
5b6ead340ebc5902bdba14ce)
LR: 2.95 (-2.94,2.94) [0.00,4.00]
Total: 10668 W: 2445 L: 2239 D: 5984
Elo: 6.25 [1.76,10.69] (95%)
[LTC](http://tests.stockfishchess.org/tests/view/
5b6eb50e0ebc5902bdba151f)
LLR: 2.96 (-2.94,2.94) [0.00,4.00]
Total: 23761 W: 4155 L: 3923 D: 15683
Elo: 3.02 [0.29,5.67] (95%)
Original patches:
- [Piece values](http://tests.stockfishchess.org/tests/view/
5b6d2cc00ebc5902bdba02d5) by Stefano Cardanobile
- [Stat bonus](http://tests.stockfishchess.org/tests/view/
5b6adbc90ebc5902bdb9da73) by Stefan Geschwentner
- [Rook on pawn](http://tests.stockfishchess.org/tests/view/
5b62a95b0ebc5902bdb961c0) by Mark Tenzer
- [Hanging bonus](http://tests.stockfishchess.org/tests/view/
5b5d2fa00ebc5902bdb90855) by Ivan Ilvec
- [ss tweak](http://tests.stockfishchess.org/tests/view/
5b58b7240ebc5902bdb89025) by miguel-l
Bench: 4694813
ianfab [Fri, 10 Aug 2018 22:32:14 +0000]
Use negative piece values for antichess-like variants
giveaway STC
LLR: 7.17 (-2.94,2.94) [0.00,10.00]
Total: 234 W: 231 L: 2 D: 1
http://35.161.250.236:6543/tests/view/
5b6e12d56e23db0fbab0dc73
losers STC
LLR: 0.00 (-2.94,2.94) [0.00,10.00]
Total: 1342 W: 1071 L: 271 D: 0
http://35.161.250.236:6543/tests/view/
5b6e12f06e23db0fbab0dc76
ianfab [Thu, 9 Aug 2018 20:14:38 +0000]
Threat evaluation for variants with mandatory captures
giveaway STC
LLR: 2.96 (-2.94,2.94) [0.00,10.00]
Total: 704 W: 401 L: 290 D: 13
http://35.161.250.236:6543/tests/view/
5b6ca2256e23db0fbab0dc69
losers STC
LLR: 2.95 (-2.94,2.94) [0.00,10.00]
Total: 288 W: 187 L: 86 D: 15
http://35.161.250.236:6543/tests/view/
5b6ca24c6e23db0fbab0dc6c
Jerry Donald Watson [Sun, 12 Aug 2018 07:47:30 +0000]
Simple razoring: depth 1 only, no distinction between PV / NonPV
We simplify the razoring logic by applying it to all nodes at depth 1 only.
An added advantage is that only one razor margin is needed now, and we treat
PV and Non-PV nodes in the same manner.
How to continue?
- There may be some conditions in which depth 2 razoring is beneficial.
- We can see whether the razor margin can be tuned, perhaps even with a
different value for PV nodes.
- Perhaps we can unify the treatment of PV and Non-PV nodes in other parts
of the search as well.
STC:
LLR: 2.96 (-2.94,2.94) [-3.00,1.00]
Total: 5474 W: 1281 L: 1127 D: 3066
http://tests.stockfishchess.org/tests/view/
5b6de3b20ebc5902bdba0d1e
LTC:
LLR: 2.95 (-2.94,2.94) [-3.00,1.00]
Total: 62670 W: 10749 L: 10697 D: 41224
http://tests.stockfishchess.org/tests/view/
5b6dee340ebc5902bdba0eb0
In addition, we ran a fixed LTC test against a similar patch which also
passed SPRT [-3, 1]:
ELO: 0.23 +-2.1 (95%) LOS: 58.6%
Total: 36412 W: 6168 L: 6144 D: 24100
http://tests.stockfishchess.org/tests/view/
5b6e83940ebc5902bdba1485
We are opting for this patch as the more logical and simple of the two,
and it appears to be no less strong. Thanks in particular to @DU-jdto
for input into this patch.
Bench: 4476945
ianfab [Sat, 11 Aug 2018 20:05:48 +0000]
Fix Zobrist initialization for pawns in hand
No functional change for variants without drops.
ianfab [Sat, 11 Aug 2018 19:26:25 +0000]
Add benchmark support for variants
Use e.g. "bench crazyhouse" to run bench test for variants.
ianfab [Sat, 11 Aug 2018 13:36:00 +0000]
Add support for USI protocol
Specification: http://hgm.nubati.net/usi.html
- Switch protocol when receiving uci/usi command,
or when option "Protocol" is changed.
- Support parsing of SFEN, but internally use FEN.
- Use shogi coordinates (e.g., 3e3d) for all moves,
and notation with "*" for piece drops.
- Specify mate distance in ply instead of fullmoves.
- Todo: Adapt option names to be compliant with USI
- No whitespaces.
- Standard options should be prefixed by "USI_".
ianfab [Sat, 11 Aug 2018 13:54:48 +0000]
Prettify debug output for non-standard board sizes
Miguel Lahoz [Thu, 9 Aug 2018 14:04:36 +0000]
Remove Condition For Passed Pawns
Currently, we do not consider pawns passed if there is another pawn of
the same color in front of them. It appears that this condition is not
necessary. The idea is that the doubled pawns are likely to be weak and
one of them will be likely captured anyway. On the other hand, if we do
somehow manage to promote a pawn, then the pawn behind it becomes passed
as well. In any case, the end result is we end up with an extra
potentially passed pawn. The current evaluation for passed pawns already
handles this case by also scaling down this effect.
STC:
LLR: 2.96 (-2.94,2.94) [-3.00,1.00]
Total: 28291 W: 6287 L: 6178 D: 15826
http://tests.stockfishchess.org/tests/view/
5b6c4b960ebc5902bdb9f256
LTC:
LLR: 2.96 (-2.94,2.94) [-3.00,1.00]
Total: 30717 W: 5256 L: 5151 D: 20310
http://tests.stockfishchess.org/tests/view/
5b6c82980ebc5902bdb9f863
Bench: 4938285
Stefan Geschwentner [Wed, 8 Aug 2018 07:45:58 +0000]
LMR simplification
Unify the "quiet" and "non-quiet" reduction rules for use at any kind of moves.
The idea behind it was that both rules reduce at similiar cases in master:
one directly for late previous moves and the other indirectly by using a
bad stat score which is used for most move sorting and so approximates the
late move condition.
For captures/promotions the old rule was triggered in 25% but the new
rule only for 3% of all cases (so now more reductions are done, whereas
for quiet moves reductions keep the same level).
STC:
LLR: 2.95 (-2.94,2.94) [-3.00,1.00]
Total: 162327 W: 35976 L: 36134 D: 90217
http://tests.stockfishchess.org/tests/view/
5b6a9a430ebc5902bdb9d5c1
LTC:
LLR: 2.96 (-2.94,2.94) [-3.00,1.00]
Total: 29570 W: 5083 L: 4976 D: 19511
http://tests.stockfishchess.org/tests/view/
5b6bc5d00ebc5902bdb9e9d6
Bench: 4526980
ianfab [Wed, 8 Aug 2018 20:37:31 +0000]
Update name and author
ianfab [Tue, 7 Aug 2018 21:32:13 +0000]
Extend mandatory captures that are only moves
STC
LLR: 2.95 (-2.94,2.94) [0.00,10.00]
Total: 485 W: 290 L: 183 D: 12
LTC
LLR: 2.97 (-2.94,2.94) [0.00,10.00]
Total: 890 W: 483 L: 371 D: 36