Compare commits

...

448 Commits

Author SHA1 Message Date
Marco Costalba 4d120ee02e Stockfish 4
Stockfish bench signature is: 4132374
2013-08-20 09:01:25 +02:00
Tom Vijlbrief f45eee318b Fix crash when reaching max ply
Bug introduced in 1b7223a53c that
updated the ss base stack without increasing
the dimension.

No functional change.
2013-08-19 16:53:46 +02:00
Tom Vijlbrief 8c2fd2170a Remove useless condition in KXK endgame
Because eval is never called when in check.

No functional change.
2013-08-19 08:55:17 +02:00
Leonid Pechenik 91c2c44fb1 Further tweak movecount pruning
Passed both short TC
LLR: 2.95 (-2.94,2.94)
Total: 15140 W: 3125 L: 2976 D: 9039

And long TC
LLR: 2.95 (-2.94,2.94)
Total: 17118 W: 3165 L: 2974 D: 10979

bench: 4132374
2013-08-18 09:13:57 +02:00
Marco Costalba 27e9fc1067 Normalize "pawn in front of minor" patch
No functional change.
2013-08-17 11:05:55 +02:00
homoSapiensSapiens e005270fb6 Use constants arguments where possible
No functional changes.
2013-08-16 09:57:21 +02:00
Marco Costalba 4f55ed14d3 Revert using exceptions
Due to crashes. It will be reapplied once
we understand what's happening.

No functional change.
2013-08-15 09:36:26 +02:00
homoSapiensSapiens a6c0ba2100 Simplify DistanceRingsBB init
Verified by same benchmark and picking some random values.

No functional change.
2013-08-14 10:53:43 +02:00
homoSapiensSapiens fc316cbca9 Some renaming in TT store()
No functional change.
2013-08-14 09:38:35 +02:00
Marco Costalba 11b1a76f35 Use exceptions to stop the search
Instead of classical flags, throw an
exception when we want to immediately halt
the search. Currently only one type
is used for both UCI stop and threads
cut off.

No functional change.
2013-08-14 08:29:57 +02:00
Tom Vijlbrief bd8f463b7e Bonus for a pawn in front of knight/bishop
Idea originated from a post of Don Dailey
on talkchess and reported by Eelco.

This is the last succesful attempt of a long
series of trials (as usually happens, the
'idea' alone is not enough).

Passed both short 15secs TC
LLR: 2.97 (-2.94,2.94)
Total: 7629 W: 1645 L: 1515 D: 4469

And long 60secs TC
LLR: 2.96 (-2.94,2.94)
Total: 10218 W: 1932 L: 1775 D: 6511

bench: 4944581
2013-08-13 14:20:02 +02:00
Ryan Takker 4d14f97482 Remove Now Unneeded Help Text
With the new automatic setting of split depth
instead of a default, the user no longer needs
guidance on setting the split point.

Also threads now defaults to one.

No functional change.
2013-08-13 07:36:33 +02:00
Marco Costalba 15616ad199 Don't set Search::RootColor in Eval::trace
Search::RootColor is a global parameter set
before to start a search, it is not something
trace() should change.

This patch allows to add trace() calls, for
debugging, inside search itself without altering
the bench, and also ensures that the values
returned by trace() and evaluate() are fully
equivalent.

No functional change.
2013-08-11 07:02:50 +02:00
Marco Costalba 94a3608ab9 Fix GrainSize rounding error
The rounding formula is different between
positive and negative scores due to the
GrainSize/2 term that is asymmetric.

So use truncation instead of rounding. This
guarantees that evaluation is rounded to zero
in the same way for both positive and negative
scores.

Found with position's flip

bench: 4634244
2013-08-10 17:11:13 +02:00
Marco Costalba 5769509d72 Fix 'improving' condition
Because VALUE_NONE is 30002, it happens that
after a check the next move is never an improving
one.

After this patch bench signature is independent from
VALUE_NONE actual value.

bench: 4303194
2013-08-09 08:21:55 +02:00
Marco Costalba fff6b9f061 Increase LMR when not improving
Apply to LMR the same Eelco's idea
applied to move count pruning.

This is the result of a series of
attempts started by Thomas Kolarik.

Passed both short TC
LLR: 2.95 (-2.94, 2.94)
Total: 5675 W: 1241 L: 1117 D: 3317


And long TC:
LLR: 2.95 (-2.94, 2.94)
Total: 8748 W: 1689 L: 1539 D: 5520

bench: 4356801
2013-08-08 10:28:48 +02:00
Marco Costalba 56d2c3844a Further tweak Position::flip
No functional change.
2013-08-05 14:44:06 +02:00
Marco Costalba 23b6809f3d Rewrite flip() to use FEN string manipulation
Instead of dealing directly with internal parameters
just "flip" the FEN string and set the position from
that.

No functional change.
2013-08-05 12:58:14 +02:00
Marco Costalba f31847302d Streamline time computation
No functional change.
2013-08-03 18:30:43 +02:00
Marco Costalba b1a4a18d63 Update polyglot.ini with new "Min Split Depth" default
No functional change.
2013-08-03 16:41:18 +02:00
Dan Schmidt f7096ea7ce Refactor do_castle()
Not a real functional change, but bench changed due to different piecelist
reordering. To verify it a temporary my canonicalize_rooks function was
written as follows. It just ensures that the rook on the "smaller" square
is listed first.

void Position::canonicalize_rooks(Color c)
{
   if (pieceCount[c][ROOK] == 2)
   {
      Square s0 = pieceList[c][ROOK][0];
      Square s1 = pieceList[c][ROOK][1];
      if (s0 > s1)
      {
         pieceList[c][ROOK][0] = s1;
         pieceList[c][ROOK][1] = s0;
         index[s0] = 1;
         index[s1] = 0;
      }
   }
}

With this both bench and the test on Chess960 positions

./stockfish bench 128 1 8 Chess960.epd file > /dev/null

Gives same result.

bench: 4424151
2013-08-03 16:18:28 +02:00
Joona Kiiski a16ba5bbd1 Retire cpu_count()
Set threads number always to 1 at startup and let the
user explicitly to chose the number of threads.

Also preserve the useful behavior of automatically set
"Min Split Depth" according to the requested threads,
indeed this parameter is too technical for a casual user,
so, when left to zero, we set it on a sensible value.

No functional change
2013-08-02 16:48:25 +02:00
Marco Costalba 408e6ee9b6 Further factor out position update code
Along the lines of previous patch.

No functional change
2013-08-01 16:32:46 +02:00
Dan Schmidt 7b4f5c8f72 Factor out pieceList updating code
The new Position methods add_piece, move_piece, and remove_piece
now manage the member variables pieceList, pieceCount, and index,
and 9 blocks of code in Position that used to manipulate those
data structures by hand now call the new methods.

There is a slightly slowdown (< 1%) on Clang and on perft,
but the cleanup compensates the little speed loss.

No functional change.

Signed-off-by: Marco Costalba <mcostalba@gmail.com>
2013-08-01 15:50:19 +02:00
Marco Costalba 55948623e7 Rework Thread hierarchy
Introduce ThreadBase struct that is search
agnostic and just handles low level stuff,
and derive all the other specialized classes
form here.

In particular TimerThread does not hinerits
anymore all the search related stuff from Thread.

Also some renaming while there.

Suggested by Steven Edwards

No functional change.
2013-07-31 18:35:52 +02:00
Marco Costalba 4d46d29efe Fix a race at thread creation
At thread creation start_routine() is called
and from there the virtual function idle_loop()
because we do this inside Thread c'tor, where the
virtual mechanism is disabled, it could happen that
the base class idle_loop() is called instead.

The issue happens with TimerThread and MainThread
where, at launch, start_routine calls
Thread::idle_loop instead of the derived ones.

Normally this bug is hidden because c'tor finishes
before start_routine() is actually called in the
just created execution thread, but on some platforms
and in some cases this is not guaranteed and the
engine hangs.

Reported by Ted Wong on talkchess

No functional change.
2013-07-31 18:35:32 +02:00
Marco Costalba cc608a7aba Tidy up Position::pretty
No functional change.
2013-07-29 19:33:30 +02:00
Marco Costalba 1f40cd6d02 Small renaming
No functional change.
2013-07-29 19:32:59 +02:00
Marco Costalba 28bc8ed462 Speed up move generation
Pass the color as template parameter
to generate_all()

Speedup of 1,3% in perft and 2,5% in bench !

No functional change.
2013-07-29 19:01:50 +02:00
Eelco de Groot 5ee16a180a Increase pruning if evaluation is not improving
Add an additional set of margins to movecount pruning
to be used when static evaluation is getting worse
than previous move.

Here are the margins table with changing
depth (fm0 not improving, fm1 improving):

    d: 0, fm0: 3, fm1: 3
    d: 1, fm0: 4, fm1: 4
    d: 2, fm0: 6, fm1: 6
    d: 3, fm0: 7, fm1: 10
    d: 4, fm0: 11, fm1: 15
    d: 5, fm0: 15, fm1: 21
    d: 6, fm0: 21, fm1: 29
    d: 7, fm0: 27, fm1: 37
    d: 8, fm0: 35, fm1: 47
    d: 9, fm0: 42, fm1: 57
    d: 10, fm0: 51, fm1: 68
    d: 11, fm0: 60, fm1: 81
    d: 12, fm0: 70, fm1: 94
    d: 13, fm0: 81, fm1: 108
    d: 14, fm0: 92, fm1: 123
    d: 15, fm0: 104, fm1: 139

Good at both short TC

LLR: 2.97 (-2.94,2.94)
Total: 11502 W: 2503 L: 2361 D: 6638

And long TC

LLR: 2.98 (-2.94,2.94)
Total: 7189 W: 1421 L: 1277 D: 4491

bench: 4364793
2013-07-29 01:21:21 +02:00
Marco Costalba d30dfc084c Annotate an unlikely condition
No functional change.
2013-07-27 11:34:15 +02:00
Marco Costalba 6373e88b5b Fix an assert in KBK endgame
The endgame king + minor vs king is erroneusly
detected as king + minor vs king + minor

Here the fix is to detect king + minor earlier,
in particular to add these trivial cases to
endgame evaluation functions.

Spotted by Reuven Peleg

bench: 4727133
2013-07-27 08:25:45 +02:00
Tom Vijlbrief 7487eb0dca Rewrite pawn shield and storm code
Passes quickly both short TC:
LLR: 2.95 (-2.94,2.94)
Total: 5755 W: 1349 L: 1222 D: 3184

And long TC:
LLR: 2.95 (-2.94,2.94)
Total: 2744 W: 628 L: 505 D: 1611

bench: 4727133
2013-07-26 00:12:46 +02:00
Marco Costalba 2067a99c07 Fix a typo in bitboard.h
Introduced by previous patch.

Spotted by Joerg Oster

No functional change.
2013-07-25 07:44:27 +02:00
homoSapiensSapiens 002062ae93 Use #ifndef instead of #if !defined
And #ifdef instead of #if defined

This is more standard form (see for example iostream file).

No functional change.

Signed-off-by: Marco Costalba <mcostalba@gmail.com>
2013-07-24 19:49:17 +02:00
Marco Costalba 4064ee5406 Simplify captures ordering
A big simplification and removing of useless code.

Finished at 50% both at short TC (with SPRT) than
at long TC at fixed number of games:
ELO: -0.14 +-3.4 (95%) LOS: 46.8%
Total: 15206 W: 2836 L: 2842 D: 9528

bench: 5059948
2013-07-24 07:53:32 +02:00
Marco Costalba b0fd2b6b98 Revert "Halve king eval margin"
This reverts commit 4b3a0fdab0.

As Gary says: " It failed when I tried it at long TC previously, and only
barely passed this time.  Some anecdotal evidence is that it hurts vs other
engines as well (the Lightspeed rating list showed a 16 elo drop from previous
best version - still +- 5 error bars on both, but that's still significant)"

I also agree that if we have some doubts (like in this case) it is better to
be safe than sorry.

bench: 4615572
2013-07-24 07:46:25 +02:00
Ryan Schmitt d0bc951835 Tune pawn PSQT values
Reduces the influence of PSQT for entries such as
the extended center and the h-file.

Passed both short TC test:
LLR: 2.95 (-2.94,2.94)
Total: 23919 W: 5207 L: 5029 D: 13683

And long TC one:
LLR: 2.96 (-2.94,2.94)
Total: 5762 W: 1108 L: 974 D: 3680

Bench: 4617880

Signed-off-by: Marco Costalba <mcostalba@gmail.com>
2013-07-24 07:41:51 +02:00
Reuven Peleg 73131e7c78 Use arrow operator instead of * and .
No functional change.

Signed-off-by: Marco Costalba <mcostalba@gmail.com>
2013-07-22 20:22:40 +02:00
Reuven Peleg 1cc18d8a7a Better condition in is_pseudo_legal()
Simplify occupied destination condition.

No functional change.

Signed-off-by: Marco Costalba <mcostalba@gmail.com>
2013-07-22 20:02:31 +02:00
Marco Costalba c45c6d308f Small touches in move generation
No functional change.
2013-07-21 11:01:24 +02:00
Marco Costalba f73bb438aa Some renaming in MovePicker
No functional change.
2013-07-21 09:55:08 +02:00
Marco Costalba 71dd8a333f Rewrite and simplify SEE
This very speed critical code was full of clever (!)
tricks and subtle details.

So I have rewritten it in a more straithforward way
and, as very often happens, result is even faster
than original.

No functional change.
2013-07-21 01:04:29 +02:00
Marco Costalba 9207baed65 Revert "Fix critical SEE bug (take 2)"
This reverts commit 3e95800814

For some reason it fails the short TC test:
LLR: -2.96 (-2.94,2.94)
Total: 20033 W: 4214 L: 4265 D: 11554

bench: 4769737
2013-07-20 18:45:38 +02:00
Marco Costalba b191df5ebe Revert "Yet another attempt at signature-build"
Still broken on OS X

No functional change.
2013-07-20 15:15:31 +02:00
Marco Costalba 3e95800814 Fix critical SEE bug (take 2)
It is somewhat unbilievable but our SEE is broken !

    If the first SEE move is a king capture and square is
    defended then SEE continues instead of breaking.

    The bug shows only on normal SEE, not see_sign() so
    probing with a:

    dbg_hit_on_c(slIndex==1, captured == KING);

    reports just a tiny:

    Total 3465656 Hits 6646 hit rate (%) 0

    Bug was there since Retire seeValues[] and move PieceValue[] out of Position of 26/6/2011 (!)
    although for some reason didn't show immediately, indeed the
    bougous patch was a "No functional change" (!!)

    bench: 4699504
2013-07-20 14:24:47 +02:00
Marco Costalba 2ed56f4d5f Revert all the SEE stuff
The speed up seems to introduce some
functionality change.

Revert to original master for now.

bench: 4769737
2013-07-20 14:20:33 +02:00
Marco Costalba 7b0b463720 Yet another attempt at signature-build
This one should work on all flavours of sed

Suggested by by Louis Zulli

No functional change.
2013-07-20 14:05:04 +02:00
Marco Costalba a6c5b4c6fb Fix critical SEE bug
It is somewhat unbilievable but our SEE is broken !

If the first SEE move is a king capture and square is
defended then SEE continues instead of breaking.

The bug shows only on normal SEE, not see_sign() so
probing with a:

dbg_hit_on_c(slIndex==1, captured == KING);

reports just a tiny:

Total 3465656 Hits 6646 hit rate (%) 0

Bug was there since 351ef5c85b of 26/6/2011 (!)
although for some reason didn't show immediately, indeed the
bougous patch was a "No functional change" (!!)

bench: 4793754
2013-07-20 13:37:12 +02:00
Marco Costalba 0504a6975d Speedup see()
And rename next_attacker() SEE helper

This very simple patch is able to speed up
bench run of almost 2% !

No functional change.
2013-07-20 12:34:35 +02:00
Marco Costalba a5b5a91512 Fix signature-build under OSX
On OS X when you use -i an extension for the in-place
substitution a backup files is required.

http://stackoverflow.com/questions/4247068/sed-command-failing-on-mac-but-works-on-linux

So rewrite to make sed flushing sign.txt in one go and avoid
using -i option.

Reported by Louis Zulli

No functional change.
2013-07-20 02:38:14 +02:00
Reuven Peleg 1a8f63a896 Microptimize gives_check() for castling case
Without patch we have 333198 nps, with patch 334249.

A very small +0.3%, not a lot manily becuase this is a
side path that is taken very few times.

Anyhow idea is correct becuase first 'quick' condition
has an hit rate of about 95%.

No functional change.
2013-07-19 17:07:54 +02:00
Marco Costalba ee5514b8fd Small simplification in space eval scoring
No functional change.
2013-07-19 11:13:18 +02:00
Marco Costalba 99e547f4cb Rename MoveStack to ExtMove
Stack has no meaning here, while ExtMove (extended move),
better clarifies that we have a move + a score.

No functional change.
2013-07-19 10:27:58 +02:00
Marco Costalba 110644d918 Better document what we skip when in check
No functional change.
2013-07-19 09:37:31 +02:00
homoSapiensSapiens 4b3a0fdab0 Halve king eval margin
But still keep the same original
margin for score.

Passed both short TC test
LR: 2.95 (-2.94,2.94)
Total: 3710 W: 845 L: 726 D: 2139

And long TC
LLR: 2.95 (-2.94,2.94)
Total: 57859 W: 10939 L: 10532 D: 36388

bench: 4769737

Signed-off-by: Marco Costalba <mcostalba@gmail.com>
2013-07-19 08:16:39 +02:00
Marco Costalba 05e31c5e5f Drop grep and tr dependency in Makefile
Use only sed to get the bench signature.

No functional change.
2013-07-15 21:39:06 +02:00
Marco Costalba 46fdb14b2f Don't use __builtin_expect
Partially revert previous patch and use
unlikey() just as code annotation.

Actually it is better to rely on a profiler for branch prediction:

http://blog.man7.org/2012/10/how-much-do-builtinexpect-likely-and.html

"In fact, even when only one in ten thousand values is nonzero,
we're still at only roughly the break-even point"

No functional change,
2013-07-15 21:09:06 +02:00
Marco Costalba cbb1a8ed31 Better annotate unlikely conditions
And in case of gcc we win also a small
speed optimization due to better branch
prediction.

No functional change.
2013-07-15 21:01:02 +02:00
Reuven Peleg a6c5f60caa Simplify a condition in refutes()
No functional change.

Signed-off-by: Marco Costalba <mcostalba@gmail.com>
2013-07-15 20:40:49 +02:00
Marco Costalba 9518cc3254 Update 'make help'
No functional change.
2013-07-14 12:23:28 +02:00
Marco Costalba b0a177bc67 Add signature-profile-build make target
Extend patch 3f64a2af6a to profile builds.

here the make command is:

make signature-profile-build ARCH=xxx COMP=xxx

No functional change.
2013-07-14 11:57:06 +02:00
Marco Costalba 5b7b330616 Retire engine Tag
It is somewhat redundant and could make SF
name too long, so use just Version, in case
of a signature build Version will be set to
'sig-xxx' otherwise, if left empty, we fall
back on usual date stamp.

No functional change.
2013-07-14 11:13:13 +02:00
Marco Costalba 3f64a2af6a Add signature-build make target
When compiling with:

make signature-build ARCH=xxx COMP=xxx

After binary has been roduced, it will be run to
get the signature 'stockfish bench' and this
number will be used as Version, so that it
will be easy to track the original sources
from a binary.

No functinal change.
2013-07-14 10:59:25 +02:00
Marco Costalba 9749f1f14c Fix build on Intel compiler
Due to a strange issue (bug?) the ternary
operator does not return a BitCountType for
icc, so revert to the expression used
before bcbc9bfd1f

No functional change.
2013-07-13 23:13:30 +02:00
Marco Costalba 4ede49cd85 Fully qualify memset and memcpy
And other trivial touches.

Ispired by Lucas's DiscoCheck

No functional change.
2013-07-13 18:01:13 +02:00
Tom Vijlbrief 6960f41e03 Retire enoughMaterial + lower trapped rook threshold
Here speed up is the name of the game.

Speed up is gained:

- Removing the useless enoughMaterial code

- Limiting trapped rook evaluation to where it counts

Tested at long TC:
LLR: 2.97 (-2.94,2.94)
Total: 10061 W: 1948 L: 1790 D: 6323

bench: 4558173
2013-07-13 18:01:03 +02:00
Marco Costalba bf90499fc3 A useless assignment found by Clang’s static analyzer
Warning is: "Value stored to 'xxx' is never read" and
it is raised in SpNode case.

No functional change.
2013-07-13 16:57:03 +02:00
Marco Costalba 404c4122ce Fix build with MSVC 2013
Also add an assert hinted by MSVC code analysis tool.

No functional change.
2013-07-13 13:03:09 +02:00
Marco Costalba 2a0bbb9faa Fix printing of PV info: take 2
Now last PV line is printed twice, fix that.

No functional change.
2013-07-13 07:43:50 +02:00
Marco Costalba 128e097d03 Fix printing of PV info
It was erroneusly skipped after the
aspiration window rework.

Reported by Eelco.

No functional change.
2013-07-12 23:42:42 +02:00
Marco Costalba c1264e46d0 Rename some UCI options
Thanks to Don, Miguel, Louis and the other people
of talkchess forum for the suggestion:

http://www.talkchess.com/forum/viewtopic.php?t=48612

Also sync polyglot.ini with current UCI options

No functional change.
2013-07-11 16:07:10 +02:00
Marco Costalba 366f6b0dab Fix a crash with depth 1 perft
Bug recently introduced in e215a88cdd

No functional change.
2013-07-11 07:23:48 +02:00
Marco Costalba 58aee9a9ea Don't IID when in check also in PvNodes
This tiny functional change allows to
nicely simplify things.

Performed at 50% in short TC:
LLR: -0.43 (-2.94,2.94)
Total: 46406 W: 9681 L: 9565 D: 27160

And succesfully passed long TC reverse test:
LLR: -2.95 (-2.94,2.94)
Total: 4945 W: 858 L: 937 D: 3150

bench: 4507230
2013-07-09 19:03:11 +02:00
Marco Costalba 4b703b1429 Revert previous patch
Unfortunatly a reverse test at long TC failed:

master^ vs master
LLR: 1.37 (-2.94,2.94)
Total: 33682 W: 6294 L: 6071 D: 21317

So becuase short TC score is 50% there is a good
possibility patch is not scalable.

So revert it.

bench: 4507288
2013-07-09 08:04:12 +02:00
Marco Costalba 62d38f0196 Simplify "fail high upon reduction" in null search
Do not use threat move to detect the condition. This
let us to retire the big allows() function.

Test at short TC was within 50% score:
LLR: -2.95 (-2.94,2.94)
Total: 38272 W: 7941 L: 7940 D: 22391

To be verified with reverse long TC

bench: 4191565
2013-07-08 07:23:30 +02:00
Marco Costalba 7e575512ae Skip node-level cut-off tests when in check
No functional change.
2013-07-07 13:39:58 +02:00
Marco Costalba a55fb76dcc Simplify aspiration window code
Here the main difference is that now we center
aspiration window on last returned score. This allows
to simplify handling of mate scores.

We have done a reversed SPRT tests, where we wanted to
verify if master is stronger than this patch.

Long TC: master vs this patch (reverse test)
LLR: -2.95 (-2.94,2.94)
Total: 37992 W: 7012 L: 6920 D: 24060

bench: 4507288
2013-07-03 18:59:23 +02:00
Marco Costalba 838255ef91 Workaround github issue
Temporary revert aspiration window patch
so to be visible to everybody: it will be
re-applied with next patch

No functional change (together with next one)
2013-07-03 18:58:23 +02:00
Marco Costalba 6fbe027da0 Simplify aspiration window code
Here the main difference is that now we center
aspiration window on last returned score. This allows
to simplify handling of mate scores.

We have done a reversed SPRT tests, where we wanted to
verify if master is stronger than this patch.

Long TC: master vs this patch (reverse test)
LLR: -2.95 (-2.94,2.94)
Total: 37992 W: 7012 L: 6920 D: 24060

bench: 4507288
2013-07-03 09:06:03 +02:00
Marco Costalba ddcb572c41 Disable flto when debugging
Link-time optimization does not work well with
generation of debugging information:

http://gcc.gnu.org/onlinedocs/gcc/Optimize-Options.html

Reported by Louis Zulli

No functional change.
2013-07-03 08:21:21 +02:00
Marco Costalba 12c0dfc113 Revert "Remove confusing optimization"
This reverts commit e05c80a088.

we gain a speed up of 1.5% under gcc !

No functional change.
2013-07-02 20:12:57 +02:00
Marco Costalba db53883f06 Merge branch 'master' into aspiration
bench: 4507288
2013-07-02 07:27:08 +02:00
Marco Costalba 13ffb08136 Revert "Increase earlier aspiration window size"
This reverts commit b88bc7b766.
2013-07-02 07:25:39 +02:00
Marco Costalba 2d82db1d14 Entering a pawn endgame is no more dangerous
A simplification of the 'dangerous' definition.

Seems neutral at reverse test at long TC

master vs patch
LLR: -2.96 (-2.94,2.94)
Total: 16974 W: 3122 L: 3139 D: 10713

bench: 4689029
2013-07-02 07:24:17 +02:00
Marco Costalba b88bc7b766 Increase earlier aspiration window size
bench: 4377851
2013-07-01 19:29:38 +02:00
Marco Costalba e074a19f5c Merge branch 'master' into aspiration 2013-07-01 19:25:23 +02:00
Marco Costalba b8930d0c26 Fix a stale comment
No functional change.
2013-06-30 13:12:04 +02:00
Marco Costalba 4fc7734547 Simplify search results update
Also some rename while there.

No functional change.
2013-06-30 12:49:39 +02:00
Marco Costalba 92dcbfa658 Reorder conditions according to their frequency
This should minimize useless tests.

No functional change.
2013-06-30 11:35:53 +02:00
Marco Costalba b50eb6bea8 Center aspiration window on last returned score
bench: 4428212
2013-06-30 11:14:02 +02:00
Marco Costalba 6f079ae720 Simplify aspiration window loop
Don't open the window in case we find a mate score: this
will be takes care with next patch.

No functional change.
2013-06-30 10:54:09 +02:00
Marco Costalba 203fdc9ac1 Use calloc() in TranspositionTable::set_size()
Function calloc() already initializes memory to
zero, so avoid calling clear() afterwards.

Also some renaming while there (inspired by DiscoCheck).

No functional change.
2013-06-29 11:23:07 +02:00
Marco Costalba 17d41b3861 Fix some stale comments
No functional change.
2013-06-23 13:19:03 +02:00
Marco Costalba 8cff4862a6 Move SquareDistance[] to bitboard.cpp
No functional change.
2013-06-23 13:13:13 +02:00
Marco Costalba 908d98820b Don't explicitize enum values when not needed
Compiler will chose the correct values in sequential
order for you.

Also move file and rank bitboards definitions to
bitboard.h

No functional change.
2013-06-23 11:30:40 +02:00
Marco Costalba a4c11b71ac Retire in_front_bb(Color c, Square s) overload
Explciitly call rank_of() in the few places where
it is used.

No functional change.
2013-06-23 10:16:43 +02:00
Marco Costalba b2fadf32aa Retire ThisAndAdjacentFilesBB[]
It is unused. Also renamed attack_span_mask to
pawn_attack_span

No functional change.
2013-06-23 10:09:24 +02:00
Marco Costalba 378bcfe760 Simplify hidden_checkers()
De-templetize and pass color as function argument.
No speed change.

No functional change.
2013-06-23 09:03:40 +02:00
Marco Costalba fe2ed42661 Name functions along corresponding UCI commands
No functional change.
2013-06-22 12:46:03 +02:00
Marco Costalba e215a88cdd Micro-optimize perft
Avoid to call perft function when we just need to count
moves, at leaf nodes.

Speed up of almost 2%

No functional change.
2013-06-21 09:10:03 +02:00
Ryan Schmitt e95e69515a Include file attacks in 'major on pawn'
Passed both short TC:
LLR: 2.97 (-2.94,2.94)
Total: 57846 W: 12248 L: 11974 D: 33624

And long one:
LLR: 2.95 (-2.94,2.94)
Total: 9181 W: 1732 L: 1581 D: 5868

bench: 4609948
2013-06-19 07:22:10 +02:00
Reuven Peleg e05c80a088 Remove confusing optimization
Here we skip the call to pos.attacks_from<ROOK>(s) in the 98%
of cases, testing the first 2 members first. Unfortunatly
code is a bit triky and not clear. So we give up to the
speed optimization in exchange of more code clarity.

No functional change.

Signed-off-by: Marco Costalba <mcostalba@gmail.com>
2013-06-17 18:12:15 +02:00
Reuven Peleg 7b31e81d77 Merge some if statements in pos_is_ok()
No functional change.

Signed-off-by: Marco Costalba <mcostalba@gmail.com>
2013-06-17 18:02:59 +02:00
Marco Costalba c0964fc70f Remove redundant condition in probcut
When !ss->skipNullMove it is assured that excludedMove == MOVE_NONE

No functional change.
2013-06-17 09:30:59 +02:00
Marco Costalba cd782c11ec Rename piece_count and piece_list
No functional change.
2013-06-16 13:21:10 +02:00
Marco Costalba 5ea984ac35 Don't calculate pawnsOnSquares twice
And reformat some code while there.

No functional change.
2013-06-16 11:32:10 +02:00
Marco Costalba 1fd020a8ba Use move_pawns() in Pawns::probe
And rename some stuff.

No functional change.
2013-06-16 10:40:36 +02:00
Marco Costalba 02420d4670 Revert "Reduce more CUT nodes only if parent node is reduced"
This reverts commit d54e8a5955.

It was not proved with SPRT this tweak is stronger. So revert it
for now to follow fishtest guidelines.

bench: 5108393
2013-06-14 08:27:06 +02:00
Marco Costalba 2c7ab488a8 Fix description of TT entry
It was way outdated and wrong !

No functional change.
2013-06-14 08:21:02 +02:00
Marco Costalba d54e8a5955 Reduce more CUT nodes only if parent node is reduced
So when we are doing a LMR search at the parent ALL node.

This patch didn't prove stronger at 60" TC
LLR: -2.97 (-2.94,2.94)
Total: 22398 W: 4070 L: 4060 D: 14268

But, first, it scores at 50%, second (and most important for me) the opposite,
i.e. normal reduction when parent node is not reduced, seems very bad:
LLR: -2.95 (-2.94,2.94)
Total: 7036 W: 1446 L: 1534 D: 4056

According to Don, this idea of increased reduction of CUT nodes
works because if parent node is reduced, missing a cut-off due to
reduced depth search (meaning position is somehow tricky) forces
a full depth research at parent node, giving due insight in this
set of sensible positions.

IOW if we expect a node to fail-high at depth n, then we assume it
should fail-high also at depth n-1, if this doesn't happen it means
position is tricky enough to deserve a research at depth n+1.

bench: 4687419
2013-06-13 20:05:02 +02:00
Marco Costalba 4bebb15e94 Reduce more CUT nodes
We got a good result from this tweak, in line with
what was already found by Don Dailey.

At short TC:
LLR: 2.95 (-2.94,2.94)
Total: 13097 W: 2742 L: 2598 D: 7757

At long TC:
LLR: 2.97 (-2.94,2.94)
Total: 7281 W: 1408 L: 1265 D: 4608

bench: 5108393
2013-06-13 19:50:32 +02:00
Marco Costalba 3b8f66f8ac Introduce Cut/All node definitions
Follow Don Dailey definition of cut/all node:

"If the previous node was a cut node, we consider this an ALL node.
The only exception is for PV nodes which are a special case of ALL nodes.
In the PVS framework, the first zero width window searched from a PV
node is by our definition a CUT node and if you have to do a re-search
then it is suddenly promoted to a PV nodes (as per PVS search) and only
then can the cut and all nodes swap positions. In other words, these
internal search failures can force the status of every node in the subtree
to swap if it propagates back to the last PV nodes."

http://talkchess.com/forum/viewtopic.php?topic_view=threads&p=519741&t=47577

With this definition we have an hit rate higher than 90% on:

    if (!PvNode && depth > 4 * ONE_PLY)
        dbg_hit_on_c(cutNode, (bestValue >= beta));

And an hit rate of just 28% on:

    if (!PvNode && depth > 4 * ONE_PLY)
        dbg_hit_on_c(!cutNode, (bestValue >= beta));

No functional change.
2013-06-13 19:46:49 +02:00
Marco Costalba b6e9d901b0 Don't use std::vector::data()
It is a C++11 only function.

Reported by Eelco.

No functional change.
2013-06-13 07:42:43 +02:00
Marco Costalba 1b7223a53c Fix again early stop ss pointer
Fix was wrong becuase search starts from ss+1,
code is a bit tricky here, so rewrite in a way
to be more easy to read and understand.

Spotted by Eelco.

No functional change.
2013-06-09 23:36:46 +02:00
Marco Costalba de1dc4f2de Don't need to expose namespace Zobrist
It can be local to position.cpp

No functional change.
2013-06-09 23:27:07 +02:00
Marco Costalba a6e0f62a4f Zobrist::init() should be Position::init()
No functional change.
2013-06-09 13:54:38 +02:00
Marco Costalba 81e242a96d Convert pieceSquareTable to 3 dimensions
No functional change.
2013-06-09 13:10:21 +02:00
Marco Costalba dd1855eb2f More consistent 'piece' variable naming
No functional change.
2013-06-09 12:56:05 +02:00
Marco Costalba db4cd89cb8 Introduce operator~(Piece c)
Small syntactic sugar to reverse piece color.

No functional change.
2013-06-09 12:44:04 +02:00
Marco Costalba 7e95495b35 Retire psq_delta()
No functional change.
2013-06-09 12:32:16 +02:00
Marco Costalba 55eb7dd1e9 Use alpha instead of beta-1
It is more directly related to a fail-low.

No functional change.
2013-06-09 11:52:39 +02:00
Marco Costalba 902c0566a6 Fix incorrect 'ss' pointer in early stop check
The exclusion search used to verify one move is much
better than other shall be called with 'ss' and not
'ss+1'

No functional change.
2013-06-09 11:01:11 +02:00
Dariusz Orzechowski bc02cc0c8a Fix a typo
No functional change.
2013-06-08 11:01:28 +02:00
Marco Costalba 05c6f7a40b Fix search log when using skills
In case of we pick a sub-optimal move be
sure to print this, and not the best one
on seach log file.

Bug spotted by Guenther Demetz.

No functional change.
2013-06-08 10:56:20 +02:00
Marco Costalba 2a98042c21 Fix a crash when 'go' multiple times
Search is started after setting a position and
issuing UCI 'go' command. Then if we stop the search
and call 'go' again without setting a new position it
is assumed that the previous setup is preserved, but
this is not the case because what happens is that
SetupStates is reset to NULL, leading to a crash as
soon as RootPos.is_draw() is called because st->previous
is now stale.

UCI protocol is not very clear about requiring that a
position is setup always before launching a search,
so here we easy the life of GUI developers assuming
that the current state is preserved after returning
from a 'stop' command.

Bug reported by Gregor Cramer.

No functional change.
2013-06-01 16:19:42 +02:00
Marco Costalba 46409a7852 Assorted renaming in evaluation
And some reshuffle too.

No functional change.
2013-06-01 13:17:39 +02:00
jundery d8b266af8b Passed pawn tuning
A small number of tests with simulated
annealing at 15s indicated these values
may be better

And this is verified at long 60+0.05 TC
LLR: 2.95 (-2.94,2.94)
Total: 40658 W: 7821 L: 7501 D: 25336

bench: 4931544
2013-05-31 09:17:48 +02:00
Marco Costalba abb40777bf Shrink engine UCI name
Some GUI have problems with long names.

Reported by George Speight.

No functional change.
2013-05-27 17:43:38 +02:00
Marco Costalba 90abcac1c7 Add Pawn Structure also to polyglot.ini
No functional change.
2013-05-25 13:14:41 +02:00
Marco Costalba 7222f47350 Re-add "Pawn Structure" UCI option
And reshuffle the code to not special case
this parameter.

No functional change.
2013-05-25 12:38:14 +02:00
Marco Costalba eafb66e1aa More uniform tracing code
No functional change.
2013-05-25 12:01:50 +02:00
Uri Blass d4a02b135d Bunch of 3 small patches
This patch is the sum of:

- Grainsize of 4 instead of 8

- Removing "depth < DEPTH_ZERO"

- Change DEPTH_QS_RECAPTURES = -5 to -7

All the patches individually failed to pass SPRT but scored
around 50%.

Together they pass easily short TC:
LLR: 2.96 (-2.94,2.94)
Total: 4429 W: 964 L: 844 D: 2621

And with some difficult long TC of 60+0.05:
LLR: 2.95 (-2.94,2.94)
Total: 64133 W: 11968 L: 11532 D: 40633

bench: 4821467
2013-05-23 17:59:39 +02:00
Marco Costalba d3608c4e79 Microptimize MoveList loop
Add MOVE_NONE at the tail, this allows to loop
across MoveList checking for *it != MOVE_NONE,
and because *it is used imediately after compiler
is able to reuse it.

With this small patch perft speed increased of 3%

And it is also a semplification !

No functional change.
2013-05-19 22:00:49 +02:00
Marco Costalba 38cfbeeb50 Delay killers[] initialization
Most of the time we cut-off earlier, at captures, so this
results in useless work.

There is a small functionality change becuase 'ss' can change
from MovePicker c'tor to when killers are tried due, for
instance, to singular search.

bench: 4603795
2013-05-19 21:41:56 +02:00
Marco Costalba 77547a4ef1 Reduce countermoves less in LMR
Passed SPRT for both short TC 15+0.05:
LLR: 2.95 (-2.94,2.94)
Total: 17724 W: 3756 L: 3598 D: 10370

And long TC 60+0.05:
LLR: 2.95 (-2.94,2.94)
Total: 22672 W: 4232 L: 4011 D: 14429

bench: 4418832
2013-05-19 21:36:23 +02:00
Marco Costalba 8ceef92266 Mimic an iterator for looping across MoveList
Seems more conventional.

No functional change.
2013-05-19 13:28:25 +02:00
Joona Kiiski f7c013edd0 Use two counter moves instead of one
Very good at long 60"+0.05 TC
LLR: 2.95 (-2.94,2.94)
Total: 5954 W: 1151 L: 1016 D: 3787

[edit: slightly changed form original patch to avoid useless loop
 across killers when killer is MOVE_NONE]

bench: 4327405
2013-05-16 16:20:50 +02:00
Marco Costalba 148490f04c Rename Refutation to Countermove
Use proper naming according to:

http://chessprogramming.wikispaces.com/Countermove+Heuristic

The name of this idea is "Countermove Heuristic" and was
first introduced by Jos Uiterwijk in 1992

No functional change.
2013-05-15 20:59:56 +02:00
Uri Blass 7c6f346c90 Increased mobility array
Performed more or less well at short TC
LLR: 2.95 (-2.94,2.94)
Total: 50517 W: 9815 L: 9574 D: 31128

And a bit better at long TC
LLR: 2.96 (-2.94,2.94)
Total: 15564 W: 2805 L: 2624 D: 10135

bench: 4375253
2013-05-15 00:34:05 +02:00
Marco Costalba 1a34496761 Revert trapped rook bug fix
It seems that do  not limiting checking the
trapped rook only on rank 1 improves the
score.

At long TC
LLR: 2.97 (-2.94,2.94)
Total: 6581 W: 1346 L: 1204 D: 4031

bench: 4985012
2013-05-15 00:06:11 +02:00
Gary Linscott 049e5ca191 Minor bugfixes to refutation table
Don't update refutation table in case of
previous move is MOVE_NULL or MOVE_NONE
and don't try refutation if is already
a killer move.

Pass both short TC
LLR: 2.96 (-2.94,2.94)
Total: 4310 W: 953 L: 869 D: 2488

And long one
LLR: 2.95 (-2.94,2.94)
Total: 6707 W: 1254 L: 1184 D: 4269

bench: 4785954
2013-05-14 23:52:44 +02:00
Marco Costalba 19dd0de4ff Reformat previous patch
No functional change.
2013-05-13 20:42:44 +02:00
Joona Kiiski 8a61b030a6 Enable refuation table
Very good result both at short TC 15+0.05
LLR: 2.95 (-2.94,2.94)
Total: 2803 W: 596 L: 483 D: 1724

And at long TC 60+0.05
LLR: 2.95 (-2.94,2.94)
Total: 2862 W: 548 L: 431 D: 1883

bench: 4329221
2013-05-13 19:48:41 +02:00
Joona Kiiski c7e31d5aa8 Simple always overwrite Refutation table 2013-05-12 21:21:46 +01:00
Marco Costalba 818a3537a7 Use Them instead of ~Us
Unortunatly we have no guarantee that the call to
operator~(Color c) is resolved at compile time.

Perhaps the solution would be to use C++11 const_expr,
but for now simply use the good old-style ternary operator
that works as expected.

No functional change.
2013-05-11 11:49:44 +02:00
Marco Costalba bcbc9bfd1f Some code reformat in evaluate_pieces
No functional change.
2013-05-11 11:13:06 +02:00
Marco Costalba 7eda7335fd Simplify previous patch
No functional change.
2013-05-09 19:09:51 +02:00
Marco Costalba 02606a8c83 Merge 'passed_pawns' tweaks
Good at both short and long TC

15+0.05
LLR: 2.96 (-2.94,2.94)
Total: 28220 W: 5531 L: 5349 D: 17340

TC 60+0.05
LLR: 2.95 (-2.94,2.94)
Total: 12612 W: 2221 L: 2057 D: 8334

bench: 4857939
2013-05-08 23:04:11 +02:00
Reuven Peleg e7505324f6 Avoid explicit bitwise operators
No functional change.

Signed-off-by: Marco Costalba <mcostalba@gmail.com>
2013-05-05 22:51:18 +02:00
jundery 653c0527a7 Passed pawn eval
Use a stepped function to evaluate bonuses and add the bonus to the
middle game

bench: 4857939
2013-05-05 11:12:04 -06:00
Marco Costalba 7f4c7cd785 Merge increased 'movecount' pruning
Good at both short and long TC

15+0.05
LLR: 2.95 (-2.94,2.94)
Total: 13814 W: 2731 L: 2588 D: 8495

TC 60+0.05
LLR: 2.95 (-2.94,2.94)
Total: 18013 W: 3136 L: 2946 D: 11931

bench: 4306557
2013-05-05 13:46:26 +02:00
Marco Costalba 0958e5c6d3 Simplify previous condition
No functional change.
2013-05-05 12:31:32 +02:00
Marco Costalba 9fc77bc414 Fix trapped rook condition
A rook is trapped if on rank 1 as is the king.
Currently the condition aloows for the rook
to be also in front of the pawns as long
as king is on first rank.

Verified with short TC test:
LLR: -1.71 (-2.94,2.94)
Total: 23234 W: 4317 L: 4317 D: 14600

Here what it counts is that after 23K games
result is equal.

bench: 4696542
2013-05-05 12:27:11 +02:00
Marco Costalba 3b92159908 Further simplify previous patch
No functional change.
2013-05-04 12:15:31 +02:00
homoSapiensSapiens e00bb13e85 Merge some conditions
No functional change.

Signed-off-by: Marco Costalba <mcostalba@gmail.com>
2013-05-04 11:13:22 +02:00
Marco Costalba 3b41e62666 Drop some redundant defined(_WIN64)
When it is already defined(_WIN32).

According to Microsoft documentation:
http://msdn.microsoft.com/en-us/library/b0084kay.aspx

_WIN32 Defined for applications for Win32 and Win64. Always defined.

_WIN64 Defined for applications for Win64.

Patch suggested by Joona.

No functional change.
2013-05-03 15:24:54 +02:00
Marco Costalba 37c91aa94c Print time and node count before search ends
This info is normally printed together with
PV info in uci_pv() but when search is stopped,
for instance when max search time is reached,
uci_pv is not called and we miss this bits.

Suggested by gravy_train

No functional change.
2013-05-03 10:26:03 +02:00
Marco Costalba 43f67eab5f Merge mobility area tweak
A nice improvment.

Was good at 15+0.05
LLR: 2.96 (-2.94,2.94)
Total: 10731 W: 2176 L: 2040 D: 6515

And at 60"+0.05
LLR: 2.96 (-2.94,2.94)
Total: 10601 W: 1968 L: 1810 D: 6823

bench: 4676606
2013-05-03 10:12:31 +02:00
Gary Linscott 11d30b6298 Fix rounding issue 2013-05-02 14:37:55 -04:00
Gary Linscott 3edb15d183 More aggressive move count pruning 2013-05-02 09:47:34 -04:00
Marco Costalba d44ac0a485 Another take at TT alignment
This time revert to original version but using
uintptr_t instead of size_t

Suggested by Lucas.

No functional change.
2013-05-02 09:38:23 +02:00
Marco Costalba 481eda4ca0 Re-add "Cache line aligned TT"
But this time do not play with pointers, in
particular do not assume that size_t is an
unsigned type of the same width as pointers.

This code should be fully portable.

No functional change.
2013-05-01 23:42:16 +02:00
jhellis3 7323231786 Tweak Mobility Area
Only consider pawns and the king as restricting.
2013-05-01 02:37:50 -05:00
Marco Costalba e381951a24 Restore development version
No functional change.
2013-04-30 20:01:07 +02:00
Marco Costalba aa2368a687 Stockfish 3
Stockfish bench signature is: 4176431
2013-04-30 19:42:43 +02:00
Marco Costalba 293c44bc09 Revert "Cache line aligned TT"
This reverts commit 083fe58124

It seems to break Android build

No functional change.
2013-04-30 19:42:21 +02:00
Marco Costalba 06b9140e5c Temporary revert "Expose EvalInfo struct to search"
It is not needed for the release and introduces
a slowdown, although very small.

Probably it will be readded after the release.

No functional change.
2013-04-29 00:55:32 +02:00
Marco Costalba 156635749b Fix a 'value > VALUE_INFINITE' assert
This fixes an assert while testing with debug on.

Assert was due to static null pruning returning value

eval - futility_margin(depth, (ss-1)->futilityMoveCount)

That was sometimes higher than VALUE_INFINITE triggering
an assert at the caller site.

Because eval con be equal to ttValue and anyhow is read from
TT that can be corrupted in SMP case, we need to sanity
check it before to use.

bench: 4176431
2013-04-27 13:08:11 +02:00
Marco Costalba 083fe58124 Cache line aligned TT
Let TT clusters (16*4=64 bytes) to hold on a singe cache line.
This avoids the need for the double prefetch.

Original patches by Lucas and Jean-Francois that has also tested
on his AMD FX:

BIG HASHTABLE

./stockfish bench 1024 1 18 > /dev/null

Before:
1437642 nps
1426519 nps
1438493 nps

After:
1474482 nps
1476375 nps
1475877 nps

SMALL HASHTABLE

./stockfish bench 128 1 18 > /dev/null

Before:
1435207 nps
1435586 nps
1433741 nps

After:
1479143 nps
1471042 nps
1472286 nps

No functional change.
2013-04-26 19:38:11 +02:00
Marco Costalba e508494a99 Fix a crash introduced few days ago
Crash is due to uninitialized ss->futilityMoveCount that
when happens to be negative, yields to an out of range
access in futility_margin().

Bug is subtle because it shows itself only in SMP case.
Indeed in single thread mode we only use the

Stack ss[MAX_PLY_PLUS_2];

Allocated at the begin of id_loop() and due to pure
(bad) luck, it happens that for all the MAX_PLY_PLUS_2
elements, ss[i].futilityMoveCount >= 0

Note that the patch does not prevent futilityMoveCount
to be overwritten after, for instance singular search
or null verification, but to keep things readable and
because the effect is almost unmeasurable, we here
prefer a slightly incorrect but simpler patch.

bench: 4311634
2013-04-26 12:14:01 +02:00
Marco Costalba 2ef53ee368 Store Eval::Info in Search::Stack
Instead of a pointer. This should fix the issue of
remaining with a stale pointer when for instance calling
IID, but also null search verification, singular search
and razoring where we call search with the same ss
pointer. In this case ss->ei is overwritten in the
search() call and upon returning remains stale.

This patch could have a performance hit because Eval::Info
is big (176 bytes) and during splitting we copy 4 ss entries.

On the good side, this patch is a clean solution.

Proposed by Gary.

No functional change.
2013-04-25 21:52:26 +02:00
Marco Costalba d810441b35 Expose EvalInfo struct to search
Allow to use EvalInfo struct, populated by
evaluation(), in search.

In particular we allocate Eval::Info on the stack
and pass a pointer to this to evaluate().

Also add to Search::Stack a pointer to Eval::Info,
this allows to reference eval info of previous/next
nodes.

WARNING: Eval::Info is NOT initialized and is populated
by evaluate(), only if the latter is called, and this
does not happen in all the code paths, so care should be
taken when accessing this struct.

No functional change.
2013-04-25 12:57:37 +02:00
Ryan Schmitt 79bcb2ca54 Increase rook/queen on 7th bonus
Shows an increase at 15+0.05
LLR: 3.01 (-2.94,2.94)
Total: 20450 W: 4091 L: 3927 D: 12432

And at 60+0.05
LLR: 2.97 (-2.94,2.94)
Total: 61432 W: 10849 L: 10441 D: 40142

bench: 4493356

Signed-off-by: Marco Costalba <mcostalba@gmail.com>
2013-04-25 12:13:17 +02:00
Marco Costalba 289a767ab3 Merge Joona's increased static null pruning
The idea is to fail high more easily in static
null test if in the parent node we are already
very deep in the move list, so the propability
to fail high there is very low.

[edit: I have slightly changed the functionality
moving

ss->futilityMoveCount = moveCount;

At the end of the pruning code, this should not affect
ELO in anyway, but makes code more natural and logic]

Test with SPRT is good at 15+0.05
LLR: 2.96 (-2.94,2.94)
Total: 50653 W: 10024 L: 9780 D: 30849

And at 60+0.05
LLR: 2.97 (-2.94,2.94)
Total: 40799 W: 7227 L: 6921 D: 26651

bench: 4530093
2013-04-25 12:05:00 +02:00
Marco Costalba 4381ea23fe Fix cpu_count() on some platforms
When we use sysconf(_SC_NPROCESSORS_ONLN) to get number of
cores, we have to include sysconf library that is unistd.h

Sometimes it happens to work just becuase unistd.h indirectly
included by some other libraries, but not always.

Reported and fixed by Eyal BD

No functional change.
2013-04-25 10:56:56 +02:00
Joona Kiiski 817ca1820b Fix potential overflow 2013-04-23 07:26:36 +01:00
Joona Kiiski 8df17204f4 More aggressive post-futility pruning 2013-04-21 14:53:27 +01:00
Marco Costalba f84f04742a Skip a couple of popcount in previous patch
And some little tidy up

No functional change.
2013-04-19 10:31:18 +02:00
Marco Costalba cc40d1c46a Merge Joona's bishop+pawn tweak
The idea is to penalize a bishop in case of
its pawns are on the same colored squares.

Good at short 15"+0.05 TC
LLR: 2.95 (-2.94,2.94)
Total: 4252 W: 925 L: 806 D: 2521

And at longer 60"+0.05 TC
LLR: 2.95 (-2.94,2.94)
Total: 15006 W: 2743 L: 2564 D: 9699

bench: 5274705
2013-04-19 09:36:00 +02:00
Joona Kiiski 818e0b2d2b Try more aggressive version 2013-04-17 07:58:59 +01:00
Joona Kiiski 3e4dfb49a7 Give a small penalty for bishop for each pawn on the same colored square 2013-04-16 21:20:31 +01:00
Marco Costalba 87436e5544 Skip a redundant check
Spotted by Joona

No functional change.
2013-04-14 23:29:25 +02:00
Marco Costalba fe72c93141 De-templetize Position::is_draw()
Now that we always check for repetition we don't
need a template anymore.

No functional change.
2013-04-10 22:23:48 +02:00
Joona Kiiski 75221fcf5e Always check repetition
It seems stronger both at fast 15+0.05 TC with fixed game number test:
ELO: 2.74 +-2.7 (95%) LOS: 97.6%
Total: 24000 W: 4698 L: 4509 D: 14793

And also at long 60+0.05 TC with SPRT
LLR: 3.05 (-2.94,2.94)
Total: 38986 W: 6845 L: 6547 D: 25594

bench: 5157061
2013-04-10 22:20:40 +02:00
Marco Costalba a95cbca568 Simplify and speed up previous patch
Use an optinal argument instead of a template
parameter. Interestingly, not only is simpler,
but also faster, perhaps due to less L1 instruction
cache pressure because we don't duplicate the very
used SEE code path.

No functional change.
2013-04-09 23:32:06 +02:00
Joona Kiiski d23454854e Document asymmetric SEE pruning trick
Here are the tests:

sprt @ 60+0.05
ELO: 3.53 +-2.8 (95%) LOS: 99.3%
Total: 18794 W: 3098 L: 2907 D: 12789

16000 @ 60+0.05
ELO: 1.39 +-3.1 (95%) LOS: 81.0%
Total: 16000 W: 2689 L: 2625 D: 10686

16000 @ 15+0.05
ELO: 2.82 +-3.3 (95%) LOS: 95.1%
Total: 16000 W: 3148 L: 3018 D: 9834

No functional change

Signature: 4969307
2013-04-09 23:31:57 +02:00
Joona Kiiski c2902112e5 Don't treat king safety differently in AnalysisMode
Rationale:

- Current settings seem to make engine *significantly* weaker in analysis mode.
- In practice this setting only has effect when king safety scores are high.
- Even in analysis mode its far more important to know if one side is getting mated,
rather than get evaluation correct with 1cp accuracy.

No functional change
2013-04-09 23:29:58 +02:00
Marco Costalba 9498b2af82 Rescale UCI parameters to 100
And correspondingly modify internal ones
to compensate it.

No functional change.
2013-04-09 23:29:58 +02:00
Marco Costalba 2a5ae34bb2 Tweak some UCI parameters
According to Jean-Paul this setup should be stronger
than default.

And SPRT test seems to confirm it:

At fast TC 15"+0.05
ELO: 3.33 +-2.7 (95%) LOS: 99.2%
Total: 25866 W: 5461 L: 5213 D: 15192

At longer TC 60"+0.05
ELO: 7.27 +-5.0 (95%) LOS: 99.8%
Total: 6544 W: 1212 L: 1075 D: 4257

bench: 5473339

Signed-off-by: Marco Costalba <mcostalba@gmail.com>
2013-04-09 23:29:58 +02:00
Marco Costalba d3fe153fe6 Re-add 'Cowardice' and 'Aggressiveness' UCI options
I have lost my bet with Jean-Paul, so now I re-add
the two options...and I am glad of it :-)

No functional change.
2013-04-09 23:29:58 +02:00
Marco Costalba 889922041b Increase null verification threshold to 12 plies
Increasing depth limit to 10 plies seems stronger
after 16K games at 15"+0.05 (ELO: +3.56) and also
repeating the test at 60"+0.05 TC:

ELO: 2.08 +-3.1 (95%) LOS: 90.9%
Total: 16000 W: 2641 L: 2545 D: 10814

Moreover setting the limit to 12 is proved stronger
then limit set to 10 by direct SPRT test at 15"+0.05:

ELO: 2.56 +-2.0 (95%) LOS: 99.5%
Total: 46568 W: 9240 L: 8897 D: 28431

So we directly set the limit to 12, the strongest setup.

bench: 4361224
2013-04-09 23:29:58 +02:00
Joona Kiiski 7bad50773a Make use of asymmetric SEE 2013-04-05 19:05:32 +01:00
Joona Kiiski d3c3c4f8e7 Fix a silly bug 2013-04-05 19:05:32 +01:00
Joona Kiiski 2097cd1221 Introduce asymmetric SEE.
No functional change
2013-04-05 19:05:32 +01:00
Marco Costalba 0c1b40c5e2 Lower minimum allowed TT size to 1 MB
Setting a very low TT size could be used
for some specific testing.

No functional change.
2013-04-05 16:56:01 +02:00
Marco Costalba 8c10029df1 Revert "Double Impact of Gain tables"
This reverts commit 36c82b751c

Seems a regression against 2.3.1 tested with 20K games at 60"+0.05

With patch applied
ELO: 15.44 +-2.8 (95%) LOS: 100.0%
Total: 20000 W: 3928 L: 3040 D: 13032

Without patch applied
ELO: 18.76 +-2.8 (95%) LOS: 100.0%
Total: 20000 W: 3903 L: 2824 D: 13273

bench: 4781239
2013-04-05 08:59:38 +02:00
Hiraoka Takuya 6e2371a86b Don't early stop if we have a mated score
No functional change.
2013-04-04 21:39:48 +02:00
Joona Kiiski 36c82b751c Double Impact of Gain tables
Very unorthodox idea.

After 16000 games at 60"+0.05
ELO: 3.14 +-3.4 (95%) LOS: 96.6%
Total: 13407 W: 2278 L: 2157 D: 8972

bench: 4705335
2013-04-03 19:14:59 +02:00
Marco Costalba 7d42d02ec7 Set IID half way between d/2 and d-4
Master IID formula is depth / 2
Previous patch is depth - 4 * ONE_PLY

This one is the middle way:

(dept/2 + depth-4*ONE_PLY)/2  -> depth-2*ONE_PLY-depth/4

After 16000 games at 60+0.05 th 1
ELO: 4.08 +-3.1 (95%) LOS: 99.5%
Total: 16000 W: 2742 L: 2554 D: 10704

bench: 4781239
2013-03-30 22:26:30 +01:00
Marco Costalba c89274d8fb Merge branch 'master' into increase_iid 2013-03-29 22:50:04 +01:00
Marco Costalba f2638816bf Raise Min Split Depth
Raise the limit to 12 so to allow people to test
on many cores machines.

Suggested by Gary and Martin.

No functional change.
2013-03-25 20:04:49 +01:00
Marco Costalba 0b4ea54da9 Update bestValue when futility pruning (2)
Same idea of 5af8179647
in qsearch() but applied to search()

After 15500 games at 15+0.05
ELO: 4.48 +-3.4 (95%) LOS: 99.5%
Total: 15500 W: 3061 L: 2861 D: 9578

bench: 4985829
2013-03-24 23:32:21 +01:00
Joona Kiiski 09f1fdf52f Fix bogus mate scores in some positions
Always before pruning the move, it's important to check that:

bestValue > VALUE_MATED_IN_MAX_PLY

See example position:

8/2p1p3/P1NpP3/3k4/1P1BN3/2P1P3/2Q5/6K1 w - - 0 1
http://support.stockfishchess.org/discussions/problems/268-wrong-declaring-a-forced-mate-in-3-moves

This problem was present in 2.3.1, then it was fixed by my patch.

After 24000 games at 15+0.05
ELO: 2.40 +-4.4 (95%) LOS: 95.7%
Total: 24000 W: 4774 L: 4608 D: 14618

bench: 4465997
2013-03-23 21:28:51 +01:00
jundery d39b22927e Use ALL_PIECES value to reference attackedBy
No functional change
2013-03-23 12:17:12 +01:00
Marco Costalba 81e9cf043a Increase non-PV IID search depth
bench: 5146380
2013-03-22 20:57:31 +01:00
Marco Costalba 077e32efc9 Better document bitbase loop
Thanks to Lucas to spot the weak comment and
to Jundery to suggest a better one.

No functional change.
2013-03-19 19:08:41 +01:00
Gary Linscott 52f3e717fa Add KNPKB endgame
In a game vs Junior, SF had the option to trade into a winning
pawn endgame, and failed to do so. PGN at bottom.

This FEN was one key position: 8/2Nb1k2/6pp/4Pp2/5K1P/5PP1/8/8 w - - 5 62.

SF master chooses h5 here, with a fail high, which goes into the drawn KNPKB
ending. With the patch, SF correctly chooses Ke3, which maintains chances to win.

[Event "nTCEC - Stage 2a - Season 1"]
[Site "http://www.tcec-chess.net"]
[Date "2013.03.05"]
[Round "11.3"]
[White "Stockfish 210213"]
[Black "Junior 13.3"]
[Result "1/2-1/2"]
[Variant "normal"]

1. d4 f5 2. g3 Nf6 3. Bg2 e6 4. c4 d5 5. Nh3 c6 6. O-O Bd6 7. Bf4 Be7 8. Nd2 O-O
9. Qb3 a5 10. Rfd1 Ne4 11. Be3 Nd7 12. Nf4 Ndf6 13. f3 a4 14. Qc2 Nxd2 15. Bxd2
dxc4 16. Qxc4 b5 17. Qc3 Qb6 18. Rac1 e5 19. Nd3 exd4 20. Qxc6 Qxc6 21. Rxc6
Bd7 22. Rcc1 Be6 23. Bb4 Rae8 24. Bxe7 Rxe7 25. Nb4 Bc4 26. Bf1 Rd7 27. Rd2 Re8
28. Rcd1 Rc7 29. Ra1 Rd8 30. Rc1 Rdd7 31. Rcd1 Re7 32. Ra1 Nd5 33. Nc2 Ne3
34. Nxd4 Bxa2 35. Nxb5 Rc5 36. Nd4 Bf7 37. Kf2 g6 38. Rd3 Nxf1 39. Kxf1 Rc4
40. b3 axb3 41. Nxb3 Kf8 42. Rd8+ Re8 43. Rxe8+ Bxe8 44. Kf2 Ke7 45. Ra7+ Bd7
46. Ke1 Rc3 47. Rb7 Rc2 48. Kd1 Rc4 49. Kd2 Kd6 50. Kd3 Rc7 51. Rxc7 Kxc7
52. Kd4 Kd6 53. Nc5 Bb5 54. e4 Be2 55. e5+ Ke7 56. Ke3 Bd1 57. Kf4 h6 58. h3
Kf7 59. h4 Bc2 60. Na6 Ba4 61. Nc7 Bd7 62. h5 g5+ 63. Ke3 Ba4 64. f4 Bd1
65. fxg5 hxg5 66. h6 Kg6 67. e6 f4+ 68. gxf4 gxf4+ 69. Kxf4 Bh5 70. Ke5 Kh7
71. Kf6 Kxh6 72. Na6 Bg4 73. e7 Bh5 74. Nc7 Bg6 75. Nd5 Be8 76. Ne3 Kh7 77. Nc4
Kh6 78. Nd2 Kh5 79. Nf3 Kg4 80. Nd4 Bh5 81. Ne6 Be8 82. Nc5 Kf3 83. Kf5 Ke3
84. Ke5 Ke2 85. Kf4 Kd2 86. Ne4+ Kd3 87. Ke5 Ke3 88. Nf6 Bf7 89. Nd5+ Kf3
90. Kf5 Ke2 91. Ke4 Be8 92. Nc3+ Kd2 93. Kd4 Kc2 94. Nd5 Kd1 95. Nf6 Bf7
96. Ne4 Be8 97. Ke3 Kc2 98. Nd6 Bd7 99. Kd4 Kd1 100. Kd3 Ba4 101. Nc4 Bb5
102. Kc3 Be8 103. Nb2+ Ke1 104. Kd3 Kf2 105. Nd1+ 1/2-1/2

No functional change (just because bench does not change)
2013-03-18 20:40:06 +01:00
Marco Costalba 0586b51f9c Further increase SEE prune depth
After 16000 games at 60+0.05
ELO: 2.89 +-5.4 (95%) LOS: 96.5%
Total: 16000 W: 2775 L: 2642 D: 10583

bench: 5442365
2013-03-16 11:20:03 +01:00
Marco Costalba d2eeef89aa Revert "Check for easy move just once"
This reverts commit a24da071f0

Seems a regression when tested against 2.3.1

With this patch, have after 20000 games at 60+0.05, we have

ELO: 13.42 +-4.8 (95%) LOS: 100.0%
Total: 20000 W: 3746 L: 2974 D: 13280

Instead with the patch reverted:

ELO: 16.62 +-4.8 (95%) LOS: 100.0%
Total: 20000 W: 3816 L: 2860 D: 13324

Although we are within error bounds here we take the conservative
approach of not introducing changes that are not proved stronger
It doesn't mean that the change shall be weaker, simply that we
don't want to take any risk.

No functional change.
2013-03-16 11:12:35 +01:00
RyanTaker 70d20326b0 Improved Readability of Material
This is a non-functional change that simply changes the look
of the code to help clarity.

No functional change.
2013-03-15 09:14:00 +01:00
jundery ccf4ec6768 Do more work between prefetch and querying transposition table
More time to load the cache line before access

No functional change.
2013-03-12 19:58:32 +01:00
Gary Linscott a24da071f0 Check for easy move just once
Here the rational seems to be that if after one try easy
move detection fails then the easy move is not so easy :-)

After 15563 games at 60+0.05
ELO: 3.04 +-5.5 (95%) LOS: 97.0%
Total: 15563 W: 2664 L: 2528 D: 10371

No functional change.
2013-03-11 22:23:19 +01:00
Gary Linscott 3698d9aa55 Be more aggressive on trying to finish iterations
Increase MaxRatio to use more time when in trouble.

After 16000 games at 60+0.05
ELO: 4.89 +-5.4 (95%) LOS: 99.9%
Total: 16000 W: 2700 L: 2475 D: 10825

No functional change.
2013-03-11 19:07:55 +01:00
Marco Costalba 10429dd616 Increase see prune depth
This seems good at short TC controls.

After 10000 games at 20+0.05
ELO: 9.56 +-6.8 (95%) LOS: 100.0%
Total: 10000 W: 1949 L: 1674 D: 6377

Testing at long TC and regression testing is still
ongoing. So this is a bit speculative commit and
could be reverted in the future.

Also re-testing at long TC the SEE pruning in PV nodes
seems less effective (perhaps even a regression, but
still ongoing) so disabled for now.

bench: 4968764
2013-03-04 09:41:56 +01:00
Marco Costalba db322e6a63 Revert "Store moves sent with "position" UCI command"
This reverts commit 0d68b523a3.

After easy move semplification this machinery is not
needed anymore (because of we don't need to know if a
root move is a recapture)

No functional change.
2013-03-04 09:29:46 +01:00
Marco Costalba 45dba12c5b Simplify "easy move" detection
Detect a move as easy only if it is the only one ;-)
or if is much better than remaining ones after we
have spent 20% of search time.

Tests are ongoing, but it seems this semplification
stands. Anyhow it is experimental for now and could
be reverted/improved with further work Gary is
testing right now.

No functional change.
2013-03-04 09:27:00 +01:00
Marco Costalba ccad601389 Avoid locking/unlocking in a tight loop
After previous patch if split point master is
waiting for job and "Use Sleeping Threads" is
false (our condition for official releases) then
it will lock/unlock splitPoint mutex in a super
tight loop badly affecting performance.

Rewrite the code to lock only when we are about
to finish. Note that race condition on slavesMask
is anyhow fixed.

No functional change.
2013-03-04 09:07:48 +01:00
jundery d165d5af91 Fix race condition where idle_loop() gets called from Split()
SplitPoint member slavesMask wasn't read under lock

No functional change.
2013-03-04 08:52:24 +01:00
jhellis3 3ce43c20de Stop search if only 1 legal move
There is no point searching a move that is forced.
It wastes time while allowing computer opponents to
fill hash with 100% accuracy.

[edit: Condition moved together with "easy move" ones]

Bench identical: 4922272
2013-03-04 08:30:55 +01:00
Marco Costalba def50020ad Fix easy re-capture case
We detect an easy move as a recapture with an
high margin on the second best move.

Unfortunatly the recapture detection is broken
becuase we identify as a recapture any move that
follows an opponent's previous capture !

This patch fix the logic to correctly detect a
real re-capture.

No functional change.
2013-03-02 13:20:40 +01:00
Marco Costalba 0d68b523a3 Store moves sent with "position" UCI command
Store all the game moves until current position.

This will be used by next patch.

No functional change.
2013-03-02 13:08:50 +01:00
Marco Costalba 0e1ad3ad33 Rename sp to splitPoint
Still keep 'sp' name when used as local
variable with limited scope.

From Jundery.

No functional change.
2013-03-01 09:44:19 +01:00
jundery 0fc9a01933 Remove strange use of the ternary operator
Note that we read shared data without lock
protection, so code is theoretically prone to
torn reads. But, first splitPoint pointer
never changes, and alpha is of integer type so
it is read in a single DWORD access.

No functional change.
2013-03-01 08:05:47 +01:00
jundery 68d1bebd8e Split() clean up locking
Only unlock and relock when idle_loop() is actually called

No functional change
2013-03-01 07:57:18 +01:00
Marco Costalba 57b6df4874 Merge Lucas's "SEE pruning at PV nodes"
bench: 4922272
2013-02-27 08:14:00 +01:00
Marco Costalba bc38efd128 Remove pruning condition on alpha
Further simplifying on Lucas's idea, seems reliable
in tests:

ELO: 2.15 +-7 (95%) LOS: 84.9%
Total: 9999 W: 1831 L: 1769 D: 6399
2013-02-27 08:07:26 +01:00
Lucas Braesch 335b57b5ef Prune negative SEE moves also in PV nodes
This patch is actually the sum of two contributions that
have been tested independently:

1) Pruning of negative SEE moves in PV

After 10000 games at 20+0.05
ELO: 5.18 +-7 (95%) LOS: 99.2%
Total: 10000 W: 1952 L: 1803 D: 6245

2) Remove of bestValue > VALUE_MATED_IN_MAX_PLY condition

After 23000 games at 20+0.05
ELO: 1.63 +-4 (95%) LOS: 88.1%
Total: 23000 W: 4232 L: 4124 D: 14644

The whole patch as been re-tested at long TC with positive results:

After 10000 games at 60+0.05
ELO: 4.31 +-7 (95%) LOS: 98.3%
Total: 10000 W: 1765 L: 1641 D: 6594
2013-02-27 08:04:32 +01:00
Marco Costalba dbd28bc7f8 Avoid a tricky line in shelter_storm()
kf = (kf == FILE_A) ? kf++ : ....

is tricky becuase kf is updated twice and it happens
to do the right thing just by accident.

Rewrite in a better way.

Spotted by pdimov

No functional change.
2013-02-23 19:27:32 +01:00
Marco Costalba ccf21f5595 Convert Readme to markdown format
Looks better on GitHub, that supports this format.

No functional change.
2013-02-23 17:05:28 +01:00
Marco Costalba 1d1fcf80a0 Use DD-MM-YY as date format
In engine name and version number.

No functional change.
2013-02-21 07:19:48 +01:00
Marco Costalba 373503f4a9 Statically link std libraries under mingw
Allows for easier redistribution.

No functional change.
2013-02-21 05:51:04 +01:00
Marco Costalba 9a1d5f0f1d Merge Gary's bishop_pin patch
Give a bonus if a bishop can pin a piece or can
give a discovered check through an x-ray attack.

Seems good after 24000 games at 15"+0.05 (single thread):

ELO: 12.30 +- 99%: 5.79 95%: 4.40 LOS: 100.00%
Total: 24000 W: 4931 L: 4082 D: 14987

bench: 4917064
2013-02-20 12:39:09 +01:00
jundery d5e49a3ad4 Print leading zeroes in hash keys
And convert to uppercase. Reset the stream to dec too.

[Edit: Also fixed the hash key in Position::pretty()]
2013-02-19 20:06:01 +01:00
Gary Linscott ea4e22be1d Merge branch 'master' into bishop_pin_clop 2013-02-19 10:31:52 -05:00
Gary Linscott dc8c6ea2d1 Bring back original bonus 2013-02-19 10:31:50 -05:00
Marco Costalba c5ec94d0f1 Update copyright year
No functional change.
2013-02-19 07:54:14 +01:00
Gary Linscott d570260a28 Back to CLOP average values 2013-02-16 22:36:58 -05:00
Marco Costalba 76caef8ba1 Account for gamePly after each move
Rename startPosPly to gamePly and increment/decrement
the variable after each do/undo move. This adds a little
overhead in do_move() but we will need to have the
game ply during the search for the next patches now
under test.

Currently we don't increment gamePly in do_null_move()
becuase it is not needed at the moment. Could change
in the future.

As a nice side effect we can now remove an hack in
startpos_ply_counter().

No functional change.
2013-02-16 12:44:17 +01:00
Marco Costalba 5a156df719 Merge Gary's king safety tweak
Still well within error bars, so probably not a big improvement,
but may be worthwhile. I will let the test keep running. The idea
for the tweak came from the TCEC game against Houdini. Stockfish saw
it as a draw, well past when it should have seen problems. King safety
was off, since it was QN and pawns only, but in fact the king was
quite vulnerable.

After 8000 games at 60/1 (Gary's test)

ELO: -0.43 +- 99%: 10.02 95%: 7.62
Wins: 1235 Losses: 1245 Draws: 5520 Total: 8000

PGN of game against houdini. Moves 55-59 it was seeing a draw, and
Houdini was seeing a good sized advantage for black. With the change,
Stockfish now recognizes that moving the king there is a bad idea.

[Event "nTCEC - Stage 1 - Season 1"]
[Site "http://www.tcec-chess.net"]
[Date "2013.02.07"]
[Round "4.2"]
[White "Stockfish 2.31"]
[Black "Houdini 3.0"]
[Result "0-1"]
[Variant "normal"]

1. Nf3 Nf6 2. c4 e6 3. Nc3 Bb4 4. Qc2 O-O 5. a3 Bxc3 6. Qxc3 b6 7. b4 a5 8. Bb2
axb4 9. axb4 Rxa1+ 10. Bxa1 Na6 11. e3 Qe7 12. b5 Nc5 13. Qc2 Bb7 14. Be2 d6
15. O-O Ra8 16. Bb2 h6 17. Ra1 Rxa1+ 18. Bxa1 e5 19. Qa2 Be4 20. Ne1 Bg6
21. Bb2 Kh7 22. f3 Nfd7 23. Bc3 h5 24. Qa1 h4 25. Kf1 Qf6 26. Qa7 Qd8 27. Qa2
e4 28. Qa1 h3 29. g3 exf3 30. Nxf3 Bf5 31. Bd4 g6 32. Kf2 Kg8 33. Qa3 Bg4
34. d3 Qc8 35. Bxc5 bxc5 36. Ke1 Qb7 37. e4 Bxf3 38. Bxf3 Ne5 39. Be2 Qc8
40. Qa6 Qd8 41. Qa5 Kh7 42. Kd1 Kg7 43. Qa7 Kg8 44. Qa5 Kh7 45. Qa7 Kg7 46. Qa5
Qb8 47. Kd2 Kh7 48. Kc2 Kg8 49. Qa6 Qd8 50. Qa5 Qf6 51. Qe1 Kg7 52. Qf1 Qe6
53. Qe1 Qd7 54. Qc1 Qe8 55. Kc3 Qa8 56. Kb3 Kh7 57. Qa3 Qd8 58. Qc1 c6 59. Qc3
Qb6 60. Ka4 Qb7 61. Qd2 Nd7 62. Qc3 Qa7+ 63. Kb3 Kg8 64. Bg4 cxb5 65. cxb5 Nb6
66. Bxh3 Qa4+ 67. Kb2 Qd1 68. Ka3 Qe2 69. Kb3 Qh5 70. Bg2 Qxh2 71. Qf6 Qxg3
72. Bf1 Qe3 73. Kc2 Na4 74. b6 Nxb6 75. Qd8+ Kg7 76. Qxb6 Qf2+ 77. Kc3 Qxf1
78. Qxd6 Qf6+ 79. Qxf6+ Kxf6 80. Kc4 g5 81. Kxc5 Ke5 82. d4+ Kxe4 83. d5 g4
84. d6 g3 85. d7 g2 86. d8=Q g1=Q+ 87. Kb5 Qb1+ 88. Ka4 Qa2+ 89. Kb4 f5
90. Qe8+ Kf4 91. Qg6 Qd5 92. Qg1 Qe4+ 93. Ka5 Qe2 94. Qg8 Kf3 95. Qd5+ Qe4
96. Qd1+ Kg2 97. Qd2+ Kf1 98. Qh2 f4 99. Qh3+ Ke2 100. Qg4+ Kd2 101. Qg5 Kc2
102. Qh4 0-1

bench: 5518286
2013-02-15 16:25:33 +01:00
Marco Costalba e0dfb0bc34 Further speed up bitbase generation
Another trick, along the same lines of previous
patch. This time we first check positions with
white side to move that, becuase we start with
pawn on rank 7, are easily classified as wins,
then black ones.

Number of cycles reduced to 15 !

Becuase now it is faster we can remove a lot of
code to detect theoretical draws. We will calculate
them anyhow, although a bit slower, but the speed
up trick more than compensates it.

Verified that generated bitbases match original ones.

No functional change.
2013-02-15 11:58:33 +01:00
Gary Linscott 3922ac2fe4 Add new clop tuned value 2013-02-14 20:29:24 -05:00
Gary Linscott 66a1a77487 Merge branch 'master' into bishop_pin_clop 2013-02-13 21:41:15 -05:00
Gary Linscott cd0ecace19 Revert "Use CLOP mean value instead of max"
This reverts commit d0c2faa5fd.
2013-02-13 21:40:38 -05:00
Marco Costalba 10d29add18 Speedup KPK bitbase of 25%
Change the way the index is coded so that
now looping from 0 to IndexMax generates
the pawns from RANK_7 down to RANK2.

Becuase positions with pawns at RANK_7
are easily classified as wins/draws, this
small trick allows to reduce the number
of needed iterations from 30 down to 26!

No functional change.

Signed-off-by: Marco Costalba <mcostalba@gmail.com>
2013-02-13 20:13:56 +01:00
Marco Costalba f2950ae206 Simplify bitbase.cpp
Use a std::vector to store positions and
rearrange KPKPosition.

No functional change.

Signed-off-by: Marco Costalba <mcostalba@gmail.com>
2013-02-13 20:13:44 +01:00
Gary Linscott d0c2faa5fd Use CLOP mean value instead of max 2013-02-13 12:21:16 -05:00
Gary Linscott 671f24ff35 Merge branch 'simplify_eval' into bishop_pin_clop 2013-02-13 12:19:54 -05:00
Gary Linscott f32f899467 CLOP tuned 2013-02-13 00:12:02 -05:00
Gary Linscott 17b71fe51d Add clop parameters 2013-02-12 00:10:21 -05:00
Gary Linscott 30c2e3828a Merge branch 'master' into simplify_eval 2013-02-11 10:26:25 -05:00
Gary Linscott a7cdf7299a Bishop pins only 2013-02-11 10:26:18 -05:00
Marco Costalba 733d0099b2 Rename and de-templetize sort()
Rename to insertion_sort so to avoid confusion
with std::sort, also move it to movepicker.cpp
and use the bit slower std::stable_sort in
search.cpp where it is used in not performance
critical paths.

No functional change.
2013-02-11 00:09:21 +01:00
Marco Costalba 0be7b8c542 Further simplify first_entry()
We can encode the ClusterSize directly in the
hashMask, this allows to skip the left shift.

There is no real change, but bench number is now
different because instead of using the lowest order
bits of the key to index the start of the cluster,
now we don't use the last two lsb bits that are
always set to zero (cluster size is 4). So for
instance, if 10 bits are used to index the cluster,
instead of bits [9..0] now we use bits [11..2].
This changes the positions that end up in the same
cluster affecting TT hits and so bench is different.

Also some renaming while there.

bench: 5383795
2013-02-09 16:37:20 +01:00
Marco Costalba c698362680 Microptimize first_entry() for 32bits
Do a 32bit bitwise 'and' instead of a 64bit
subtract and bitwise 'and'.

This is possible because even in the biggest
hash table case (8GB) the number of entries
is 2^29 so storable in an unsigned int.

No functional change.
2013-02-09 10:55:38 +01:00
Marco Costalba fe3352665b Retire TTCluster and simplify TT
Also some renaming while there.

No functional change.
2013-02-09 10:55:20 +01:00
Marco Costalba 3f44be9baa Simplify move_to_san()
Nicely simplify disambiguation code.

No functional change.
2013-02-09 07:43:53 +01:00
Gary Linscott 0c06b4509d Slight tweak to king safety. Bench: 5534531 2013-02-08 08:53:13 -05:00
Marco Costalba e5bc79fb9c Retire slavesPositions
Save the current active position in each Thread
instead of keeping a centralized array in struct
SplitPoint.

This allow to skip a memset() call at each split.

No functional change.
2013-02-08 11:45:33 +01:00
Marco Costalba 880726c13a Add const qualifer to go()
Obsolete renmant of when position was directly
passed to the search instead of being copied
for the main thread as is now.

From Jundery.

No functional change.
2013-02-08 10:07:09 +01:00
Marco Costalba 50a7200b18 Workaround value-initialization in MSVC
The syntax splitPoints() should force the compiler to
value-initialize the array and because there is no
user defined c'tor it falls back on zero-initialization.

Unfortunatly this is broken in MSVC compilers, because
value initialization for non-POD types is not supported,
so left splitPoints un-initialized and add in split()
initialization of slavesPositions, that is the only
member not already set at split time.

This fixes an assert under MSVC when running with
more than one thread.

Spotted and reported by Jundery.

No functional change.
2013-02-08 09:20:40 +01:00
Gary Linscott c67fb8ef04 Add KBPKP endgame
It is a draw if pawns are on G or B files, weaker pawn is
on rank 7 and bishop can't attack the pawn.

No functional change (because it is very rare and does not appear in bench)
2013-02-07 06:51:55 +01:00
Marco Costalba 14c2c1395b Change slave_available() API
To return a pointer to the available
thread instead of a bool. This allows
to simplify the core loop in split().

No functional change.

Signed-off-by: Marco Costalba <mcostalba@gmail.com>
2013-02-06 20:48:26 +01:00
jundery 88c3670edf Rename posKey stored in the transposition table
[Edit: Slightly extended by me]

No functional change.
2013-02-06 08:03:37 +01:00
jundery f69c185e02 Add const qualifer to check_is_dangerous
No functional change.
2013-02-06 07:49:40 +01:00
Marco Costalba bf706c4a4f Slightly change split() API
This function "returns" two values: bestValue and bestMove

Instead of returning one and passing as pointer the other
be consistent and pass as pointers both.

No functional change.
2013-02-05 06:35:38 +01:00
Marco Costalba 1a414cd9cb Derive ThreadPool from std::vector
Prefer sub-classing to composition in this case.

No functional change.
2013-02-04 22:59:20 +01:00
Marco Costalba 91427c8242 Move split() under Thread
Previous renaming patch suggested this reformat:
when a better naming leads to a better code!

No functional change.
2013-02-04 22:17:04 +01:00
Marco Costalba b8c5ea869c Some renaming in split()
Naming suggested by jundery.

No functional change.
2013-02-04 22:00:41 +01:00
Marco Costalba 6560e4cc5b Be clear about not LMR the ttMove
Currently a ttMove is reduced with ss->reduction = DEPTH_ZERO,
so it is actually not reduced (as it should be), but the
trick works just becuase it happens that ttMove is the first
to be tried and

reduction(depth, 1)

Always returns zero. So explicitly forbid reduction of ttMove
in the LMR condition. This is much clear and self-documented.

No functional change.
2013-02-03 11:14:21 +01:00
Marco Costalba 08d615cc95 Templetize score_xxx() functions
So to be style-wise aligned with the corresponding
generate() functions.

No functional change.
2013-02-03 10:11:12 +01:00
Marco Costalba 1c4e6d7ea2 Rename prevents_move() to refutes()
Better! From DiscoCheck.

No functional change.
2013-02-03 09:23:04 +01:00
Marco Costalba 5f58db8c99 Correctly score enpassant captures
Surprisingly this rare case was not considered
when scoring a capture.

Also take in account that in the promotion case
we gain a new piece (typically a queen) but we
lose the promoting pawn.

These small issues were present since Glaurung times!

Found while browsing DiscoCheck sources

bench: 5400063
2013-02-03 09:08:32 +01:00
Marco Costalba ddbe6082c4 Unify History and Gains under a single Stats class
Handling of History and Gains is almost the same, with
the exception of the update logic, so unify both
classes under a single Stats struct.

No functional change.
2013-02-02 17:45:09 +01:00
Marco Costalba 53051eefc7 Retire history.h
And move the contents to movepick.cpp, where they are
mostly used.

Idea from DiscoCheck.

No functional change (bench 5379503)
2013-02-02 16:54:35 +01:00
Gary Linscott a72710c660 Simplify eval take 2. Bench 5097444 2013-02-02 10:19:59 -05:00
Marco Costalba 9f94d22801 Restore "fail-low of reduced" and close regression
This reverts "Threat Extensions" and is the last of
this revert series.

In single-thread tests we should now be on par with 2.3.1
2013-02-02 07:16:33 +01:00
Marco Costalba 0901e12102 Revert "Simplify Evaluation"
This reverts commit 496c7497cb
2013-02-02 06:44:04 +01:00
Marco Costalba 58c9fbacc7 Revert "Extend full 3 fold detection to PvNodes" 2013-02-02 06:41:05 +01:00
Marco Costalba 483c98a69e Rewrite do_castle_move()
And handle the castle directly in do/undo_move().
This allow to greatly simplify the code.

Here the beast is the nasty Chess960 that is
really tricky to get it right because could be
that 'from' and 'to' squares are the same or
that king's 'to' square is rook's 'from' square.

Anyhow should work: verified on all Chess960
starting positions.

No functional and no speed change also in Chess960.
2013-01-28 13:40:47 +01:00
Marco Costalba 2218a5836a Rewrite do_null_move()
Use a more traditional approach, along the same lines
of do_move().

It is true that we copy more in do_null_move(), but we
save the work in undo_null_move(). Speed test shows the
new code to be even a bit faster.

No functional change.
2013-01-27 12:15:02 +01:00
Marco Costalba 76a0d3c05a Get rid of some locals in do_castle_move()
Rewrite the logic to get rid of kBefore and rBefore.

No functional change.
2013-01-27 11:03:55 +01:00
Marco Costalba 52cab06fff Don't prefetch if not needed
Prefetch access to hash tables only in case we
have changed pawn or material hash keys.

No functional change.
2013-01-27 10:19:11 +01:00
Marco Costalba bd87ab9ff5 Retire generate_king_moves()
We have only one call place so inline its content.
BTW, function is already declared as FORCE_INLINE.

Also some small refactoring while there.

No functional change.
2013-01-26 22:43:58 +01:00
Gary Linscott 57797822f8 Bring back just bishop pins 2013-01-26 15:35:00 -05:00
Marco Costalba 7062db7cb2 Clarify slavesMask usage
When a thread is allocated a bit is set in slavesMask.
This bit corresponds to the thread's index field that,
because it happens to be the position in the threads
array, eventually it is equal to the loop index 'i'.

But instead of relying on this 'coincidence', explicitly
use the 'idx' field so to clarify slavesMask usage.

Backported from c++11 branch.

No functional change.
2013-01-26 14:38:51 +01:00
Marco Costalba 166cc0292c Revert "Further push singular extension"
This reverts commit 4c91dbc28e

Seems a regression on extended test by both Gary and me.
2013-01-26 10:20:46 +01:00
Marco Costalba 496c7497cb Merge branch 'simplify_eval' of https://github.com/glinscott/Stockfish
Test results are looking good after 12500 games.

ELO: 6.55 +- 99%: 8.02 95%: 6.09
LOS: 99.99%
Wins: 1968 Losses: 1732 Draws: 8813

Also, here are the noise.py results, which seem to have stabilized:
Games: 12526 , result: [1969, 1734, 8823]
Estimated ELO: 6.94963842777
Noise as function of number of games:
['81.89', '565.26', '110.87', '104.39', '38.22', '49.98', '18.56', '16.76',
'11.02', '8.90', '17.36', '9.84', '10.81', '5.13', '6.22', '3.32', '5.83',
'7.21', '15.27', '1.63', '4.04', '9.51', '0.54', '0.75', '1.06', '2.93',
'4.59', '6.85', '13.62', '9.87', '14.74', '20.46', '22.18', '24.33', '31.02',
'34.99', '35.22', '33.22', '32.46', '37.02', '29.10', '36.34', '42.11', '39.33',
'26.16', '28.25', '35.42', '31.04', '29.26', '23.91', '22.52', '23.49', '20.00',
'24.39', '17.22', '16.50', '10.69', '9.15', '9.57', '4.77', '6.67', '3.87', '2.57',
'2.84', '2.60', '3.32', '2.08', '2.93', '4.47', '4.41', '4.83', '4.86', '6.40',
'5.98', '6.10', '6.83', '5.83', '6.22', '5.71', '8.52', '9.25', '5.98', '7.52',
'7.76', '8.76', '8.55', '8.64', '7.19', '5.83', '4.59', '4.77', '4.26', '4.98',
'5.29', '5.41', '4.92', '5.59']

bench: 5229106
2013-01-26 10:18:36 +01:00
Gary Linscott e83b9075ff Simplify evaluation 2013-01-24 08:54:13 -05:00
Marco Costalba 6950d07bf4 Small reformat of split()
No functional chhange.

Signed-off-by: Marco Costalba <mcostalba@gmail.com>
2013-01-21 23:31:33 +01:00
Marco Costalba 054d117d25 Fix an idiotic icc warning
Intel Compiler has 'invented' this pearl:

warning #1476: field uses tail padding of a base class

Just becuase we have subclassed MainThread and added
the field 'bool thinking'.

Pure nosense. Silence the warning.

No functional change.
2013-01-20 17:36:24 +01:00
Marco Costalba 62b32a4737 Futher renaming in thread.cpp
No functional change.
2013-01-20 17:35:55 +01:00
Marco Costalba 588670e8d2 Big renaming in thread stuff
No functional change.

Signed-off-by: Marco Costalba <mcostalba@gmail.com>
2013-01-16 20:00:05 +01:00
Marco Costalba c465f4c4df Fix race while exiting
Fix again TimerThread::idle_loop() to prevent a
theoretical race with 'exit' flag in ~Thread().

Indeed in Thread d'tor we raise 'exit' and then
call notify() that is lock protected, so we
have to check again for 'exit' before going to
sleep in idle_loop().

Also same change in Thread::idle_loop() where we
now check for 'exit' before to go to sleep.

No functional change.

Signed-off-by: Marco Costalba <mcostalba@gmail.com>
2013-01-16 19:58:55 +01:00
Lucas Braesch 8737b26a23 Remove Threat Extension
Great code simplification: - instead do not futility
prune threat refutations. allows_move() is therefore removed.

4000 games at 50,000 nodes/move:
1085-989-1926 [51.2%] LOS=98.3%

4000 games in 10"+0.1"
756-751-2493 [50.1%] LOS=55.1%

EDIT: I have retested the patch of Lucas in a slightly different form
(without pruning in PvNode) and test mre or less confirms that
60 lines of code are totally unuseful:

After 6195 games at 15"+0.05"
1333 - 1325 - 3537 ELO 0

bench 5140990
2013-01-15 17:03:56 +01:00
Marco Costalba 78a9531773 Fix a bug in timer loop
Silly logic bug introduced in dda7de17e7

Timer thread, when msec = 0, instead of going
to sleep, calls check_time() in an endless loop.

Spotted and reported by snino64 due to abnormally
high CPU usage.

No functional change.
2013-01-14 19:32:30 +01:00
Marco Costalba d1143794a0 Polymorphic Thread hierarchy
Subclass MainThread and TimerThread and declare
idle_loop() virtual. This allow us to cleanly
remove a good bunch of hacks, relying on C++
polymorphism to do the job.

No functional change.
2013-01-14 02:01:37 +01:00
Marco Costalba e70eae2c91 Don't use do_sleep flag
Rename it is_finished and use it only in main
thread to signal search is finished. This allows
us to simplify the complex SMP logic.

Ultra tricky patch: deep test is required under
wide conditions like pondering on and option
"Use Sleeping Threads" set to false.

No functional change.
2013-01-14 00:02:32 +01:00
Marco Costalba 99ae47716a Re-add the hack
This reverts commit 869c924410

I misunderstood here. Actually it can happen that
thread is created but still not entered idle_loop
and at the same time start_searching() is called.

Becuase 'do_sleep' is set start_searching() will
set it to false and start the search, but when,
at last, the thread enters idle_loop(), resets
the flag and goes to sleep: not what we want.

Revert the hack waiting for a better solution
in the next patches.

No functional change.
2013-01-13 23:56:04 +01:00
Marco Costalba dda7de17e7 Retire set_timer()
Also assure in Thread::timer_loop() that when
timer interval is 0 (timer is disabled) we
never call check_time()

No functional change.
2013-01-13 18:24:43 +01:00
Marco Costalba 869c924410 Retire obsolete race hack
This hack was introduced in d282cf6964
to workaround a race with start_searching(),
but these days is no more needed.

No functional change.
2013-01-13 17:05:30 +01:00
Marco Costalba ea6c1f7a17 Retire Threads wake_up() and sleep()
These functions are used in just one place.
And generalize wait_for_stop()

No functional change.
2013-01-13 16:57:40 +01:00
Marco Costalba 81cd7d787e Rename wake_up() to notify_one()
To align to C++ std::thread conventions.

No functional change.
2013-01-13 16:43:26 +01:00
Marco Costalba a523cea772 Unify 'ponderhit' handling
Finally we can now merge the 'ponderhit' case with
'stop' and 'quit'.

The patches have been done step by step to help debugging
becuase this is really tricky code.

No functional change.
2013-01-13 14:39:41 +01:00
Marco Costalba a66a7c3870 Small change to "ponderhit" handling
Reset Limits.ponder only if search continue, but if
we are going to stop the search there is no need
(and is also confusing) to clear the 'ponder' flag.

This mimics the behaviour upon rceiving 'stop' when
pondering.

No functional change.
2013-01-13 14:32:30 +01:00
Marco Costalba 89a89eb605 Simplify and rename wait_for_stop_or_ponderhit()
Setting stopOnPonderhit is now done by the caller.

No functional change.
2013-01-13 14:15:19 +01:00
Marco Costalba 3b14b17664 Simplify a condition in search()
And rearrange best value update in case of SpNode.

No functional change.
2013-01-13 13:17:16 +01:00
Marco Costalba 3a836dab59 Clarify SAN disambiguation in case of a pinned piece
In SAN notation when two pieces of the same type can
move to a given destination square, a disambiguation
additional info (like starting file) shall be added
to the SAN move.

If one of the two pieces is pinned, the corresponding
move _could_ be illegal and in this case disambiguation
is not needed. But to be pinned alone it is not enough
to deduce that the move is illegal, for instance in this
position:

R3rk2/2r6/8/8/8/8/8/K7 b - - 0 1

The move Rc8 is ambiguous although the rook in e8 is pinned
and the correct SAN notation should be Rcc8.

No functional change.
2013-01-12 13:21:14 +01:00
Marco Costalba e1191b35e8 Async 'stop' command
Don't wait for the search to finish after a 'stop'
command, but keep processing the GUI input if any.

Also explicitly wake up the main thread (that could be
sleeping) after a 'stop' or 'quit' command and do not
rely on wait_for_search_finished() doing it for us.

This patch cleans up the code and functions's definitions,
but it is risky and needs a good test under different
conditions to be sure it does not introduces hungs up.

No functional change.
2013-01-12 12:06:55 +01:00
Marco Costalba edce2a8448 Revert so called "fromNull patch"
Revert patch c581b7ea36

Seems a regression after testing from Gary:
ELO: 7.24 +- 99%: 17.03 95%: 12.93
LOS: 97.86%
Wins: 439 Losses: 381 Draws: 1962

And mine:
After 5410 games at 15"+0.05
Wins: 936 Losses: 1141 Draws: 3333  ELO -13

Moreover we know that there is a regression in the range
of patches which include the fromNull patch.

Probably this is not the only regression since 2.3.1 and
perhaps the idea under fromNull is good, but at the moment,
while in deep regression hunting, better to be on the safe
side and revert it entirely.

My guess on why this is a regression is that using the
negated evaluation of previous ply in case of null search
fails to take in account the king safety asymmetry between
the two colors. This is of course just a guess.

bench 5503830
2013-01-06 23:06:20 +01:00
Marco Costalba 9b1cf3cf43 Have fun with union in book.cpp
Fancy way to use an union to map polyglot
zobrist keys in one go.

Also some renaming while there.

No functional change.
2013-01-06 12:06:19 +01:00
Marco Costalba bff65a211f Retire 'Cowardice' and 'Aggressiveness' UCI options
They are not self-describing and create a lot of user
requests about them.

Given that the values are already well tuned there
is no need to expose them as UCI options.

No functional change.
2013-01-04 17:11:24 +01:00
Marco Costalba 2d60995c00 Retire 'mate in x' hack
Sometimes is faster, but not always and on very long mates
produces strange scores probably due to truncation of PV
artifacts.

So simply perform normal search also in case of UCI 'mate x'
command, with the only difference that when a mate in x is
found search returns immediately.

No functional change.
2013-01-04 16:30:46 +01:00
Marco Costalba 0454bbc54f Don't exit if unable to find bench file
Now that we can call 'bench' command also from
interactive terminal it makes no more sense to
exit the application if the user types a wrong
file name.

No functional change.
2013-01-04 14:52:21 +01:00
Marco Costalba 900e2d4e1e Teach file_to_char() about upper/lower case
This allows to further simplify Position::fen()

No functional change.
2013-01-04 14:45:04 +01:00
Marco Costalba 9d1151575d Reformat FEN construction
Simplify and shrink code.

No functional change.
2013-01-04 13:38:14 +01:00
Marco Costalba 193741218c Remove some obsolete asserts on TT values
Now that insert_pv_in_tt() stores VALUE_NONE in
TT's position evaluation those 2 asserts are
obselete.

No functional change.
2012-12-31 17:20:39 +01:00
Marco Costalba 896420b166 Allow to pass a 'seed' to RKISS
This somewhat simplifies the code.

Suggested by Lucas Braesch.

No functional change.
2012-12-31 11:59:53 +01:00
Marco Costalba 009a0f88e0 Micro-optimization in evaluate_space()
Since &-ing with SpaceMask restricts the set to the home
half of the board, it is possible to use just one popcount
instead of 2 by shifting "safe" to the other half of the
board. This gives a small speedup especially on systems
where hardware popcount is not available.

Patch kindly sent by Richard Vida.

No functional change.
2012-12-31 11:34:18 +01:00
Marco Costalba e1d681458e Add 'mate' limit to 'bench' command
It is now possible to run SF on a 'mate in x' testsuite.

For instance in case of a file with fen strings of
positions with mate in 10 we can now 'bench' on it:

stockfish bench 128 1 10 mate_in_10.epd mate

No functional change.
2012-12-30 15:49:02 +01:00
Marco Costalba ce063f59cd Handle UCI command "mate in x moves"
Following a user request I added the handling of UCI:

go mate x

Currently we just return from a PV node if x moves have been
done. Probably not the best approach. I have looked at Fruit/Toga
sources and there is even simpler: engine falls back on a fixed
depth search.

No functional change.
2012-12-30 14:43:23 +01:00
Marco Costalba 3cf6471738 Revert evaluation cache
And return on using TT as backing store for position
evaluations.

Tests (even on single thread) show eval cache was a regression.
In multi thread result should be even worst because eval cache
is a per-thread struct, while TT is shared.

After 4957 games at 15"+0.05 (single thread)
eval cache vs master 969 - 1093 - 2895  -9 ELO

So previous reported result of +18 ELO was probably due to an
issue in the testing framework (a bug in cutechess-cli) that
has been fixed in the meanwhile.

bench: 5386711
2012-12-27 13:57:17 +01:00
Marco Costalba f78b68b7ff Add list of legal moves to Position::pretty()
Along the same lines of previous patch now we add
the list of the legal moves in the given position.

No functional change.
2012-12-27 11:34:48 +01:00
Marco Costalba e9ab7353de Add checkers info to Position::pretty()
In case current position is under check, list the
squares of the checker(s) pieces.

This should satisfy a specific user request.

No functional change.
2012-12-26 18:28:45 +01:00
Marco Costalba 9d1978e217 Remove two obsolete asserts in prevents_move
Now that this function is called also to calculate
move's extensions the asserts are no more valid.

No functional change.
2012-12-26 12:27:41 +01:00
Marco Costalba db097921bc Rename yields_to_threat and prevents_threat
Follow Lucas suggestions and better name these
two functions.

No functional change.
2012-12-26 12:21:59 +01:00
Marco Costalba 894c43a1d6 Introduce Null Threat extension
In case of null search at low depths returns a fail low
due to a threat then, rather than return beta-1 (to cause
a re-search at full depth in the parent node), we set a flag
threatExtension = true (false by default) that will cause
moves that prevent the threat to be extended of one ply
in the following search.

Idea and patch is by Lucas Braesch.

Lucas also did the tests:
1500 games in 5"+0.05":
SF_threatExtension vs SF_20121222: 366 - 331 - 803 [51.2%] LOS=90.8%

3000 games in 10"+0.1":
SF_threatExtension vs SF_20121222: 610 - 559 - 1831 [50.8%] LOS=93.2%

Tests confirmed by Gary after 10570 games,
ELO: 2.79 +- 99%: 8.72 95%: 6.63
LOS: 94.08%
Wins: 1523 Losses: 1438 Draws: 7607

And finally by me at 15"+0.05, single thread, 3824 games
threatExtension vs master 768 - 692 - 2364  +7 ELO

bench 4918443
2012-12-25 19:17:27 +01:00
Marco Costalba b5b799b5ab Fix a couple of extra spaces
No functional change.
2012-12-25 18:48:41 +01:00
Marco Costalba e82382703c Retire Position::in_check()
It is redundant with Position::checkers()

No functional change.
2012-12-25 17:59:35 +01:00
Marco Costalba 3b49aeb4f2 Retire Position::move_is_legal()
Use the new contains() method of struct MoveList

No functional change.
2012-12-25 11:51:08 +01:00
Marco Costalba 423c6d8a8a Small tweak in is_pseudo_legal()
This is difficult code becuase a bug here could lead
to very subtle crashes in case of SMP games where we
have TT move corruption due to concurrent access.

Anyhow I have fully verified te code throwing at it
random moves. It shoudl work.

No functional change.
2012-12-25 11:31:32 +01:00
Marco Costalba 158014b39d Introduce namespace Pawns
And retire old struct PawnTable along the same lines
of previous patch.

No functional change.
2012-12-22 11:38:36 +01:00
Marco Costalba 231f62baf7 Introduce namespace Material
And retire old struct MaterialTable simplifying the code.

No functional change.
2012-12-16 12:58:39 +01:00
Marco Costalba 52bbf372bb Don't need to check for bestValue < beta to split
With rearrangement of fail high code this condition
is no more necessary.

Found by Jörg Oster.

No fuctional change.
2012-12-15 13:11:10 +01:00
Marco Costalba 3ddf91d9d1 Remove an extra semicolon
No functional change.
2012-12-15 11:20:04 +01:00
Marco Costalba a2f46446cf Revert store of distinct upper and lower bounds
Test by Joona prooves the new feature don't value 70 added lines of code.

Grand totals after 10040 games (crashes: 0) for tt_both

master_9edc7 - 6a93488_6a934: 1756 - 1688 - 6596 ELO +2 (+- 2.7)

Confirmed by test of Gary:

After 8680 games:
ELO: 0.80 +- 99%: 9.62 95%: 7.31
LOS: 65.38%
Wins: 1288 Losses: 1268 Draws: 6130

Thanks a lot to both for testing it !!!

bench 5149248
2012-12-15 11:18:52 +01:00
Marco Costalba 9edc7d6958 Merge branch 'eval_cache'
Unusually good result. Defenitly needs further verifications.

After 2160 games at 15"+0.05
Mod vs Orig 486 - 367 - 1307 ELO +19

bench: 6261882
2012-12-10 09:26:02 +01:00
Marco Costalba 55db871472 Fix comparison with alpha, not beta
This silly bug seems the reason of the unsual bench
value.

bench: 6261882
2012-12-09 14:19:21 +01:00
Marco Costalba da98a45bcb Ensure valueLower <= valueUpper
In case a TTEntry stores both an upper and a lower bound
ensure that upper bound is not smaller than lower bound.

bench 1813815
2012-12-09 14:14:44 +01:00
Marco Costalba feeafb0a50 Store distinct upper and lower bound scores
This is more complex than what I'd like but I
was unable to split in small chunks.

Here we add 2 slots to TTEntry (valueUpper and depthUpper)
so that sizeof(TTEntry) returns to the original 16 bytes
and we can pack exactly 4 entries in a 64 bytes cache line.

Now we save an upper bound score alongside a lower (exact)
score. The idea is to increase TT cut-offs rates becuase
there is now an higher probability for a node to use TT info.

This patch is highly experimental and probably needs further
steps as is hinted by an unrealistic bench number:

bench: 2022385
2012-12-09 13:15:50 +01:00
Marco Costalba 22c557ca7c Micro-optimize color_of()
In almost all cases we already know in advance that
color_of() argument is different from NO_PIECE.

So avoid the check for NO_PIECE in color_of() and
test at caller site in the very few places where
this case could occur.

As a nice side effect, this patch fixes a (bogus)
warning under some versions of gcc.

No functional change.

Signed-off-by: Marco Costalba <mcostalba@gmail.com>
2012-12-08 12:32:48 +01:00
thaspel 6a93488291 Update Readme.txt now that we support 64 threads
No functional change.

Signed-off-by: Marco Costalba <mcostalba@gmail.com>
2012-12-04 19:57:16 +01:00
Marco Costalba 6fa83f5188 Merge branch 'eval_cache'
Use an eval cache instead of TT to store node
position evaluations.

It is already an improvment and, because it frees
two TT entry slots, paves the way to extend TT to
store both upper and lower bounds.

After 4855 games, single thread, 15"+0.05
Mod vs Orig 1165 -920 - 2770 ELO +18

bench: 5149248
2012-12-04 08:05:15 +01:00
Marco Costalba ce248e7920 Increase MAX_THREADS to 64
And document why this is an hard limit. It
seems for some (lucky) people 32 threads
are not enough.

No functional change.

Signed-off-by: Marco Costalba <mcostalba@gmail.com>
2012-12-04 07:57:15 +01:00
Marco Costalba 23bdd06442 Reintroduce eval optimizaion from null search
Now that conversion to eval cache is finished
we can reintroduce this optimization.

bench: 5149248
2012-12-01 18:11:01 +01:00
Marco Costalba 98cd8239cc Don't save eval score in TT
This patch completes the removal of eval info
in TT table.

No functional change.
2012-12-01 15:19:50 +01:00
Marco Costalba 2a585b63b8 Don't use TT just to save a node evaluation
In search(), after we evalute the position, in case there
isn't any TT entry we create one with just the evaluation
score.

This patches removes that code. The reason becuase the patch
deserves a single commit it is becuase introduces a (very small)
functional change due to the fact that the total number of
TT stores is less now and this slightly alters the TT hits
of our benchmark.

bench: 4983262
2012-12-01 15:13:08 +01:00
Marco Costalba a5ea3a202e Don't read eval from TT anymore
Rely fully on eval cache. Note that we still save eval
info to TT, this is not needed at this moment and will be
removed in future patches. We keep it so to have a "non
functional change" patch.

No functional change.
2012-12-01 15:01:00 +01:00
Marco Costalba 69de670353 Retire eval optimization from null search
Simplify things preparing for further changes.

bench: 4983282
2012-12-01 14:48:13 +01:00
Marco Costalba 4e5d834e8e Add eval cache infrastructure
With this patch series we want to introduce a per-thread
evaluation cache to store node evaluation and do not
rely anymore on the TT table for this.

This patch just introduces the infrastructure.

No functional change.
2012-12-01 14:01:20 +01:00
Marco Costalba 4502917665 Don't double check for move legality
In case of a RootNode or a SpNode move has
been already checked for legality so we can
skip a redundant check.

Spotted by Frank Genot.

No functional change.
2012-11-27 20:05:03 +01:00
Marco Costalba 5af8179647 Update bestValue when futility pruning
In qsearch we should update the bestValue as we do
in case of futilityValue < beta, also when pruning
moves with non-positive see.

Spotted by Lucas Braesch

Bench: 5695710
2012-11-26 16:13:36 +01:00
Marco Costalba 55df3fa2d7 Avoid spamming the GUI in multipv search
Send the PV lines to GUI only once at the end of the
PV search loop or just in case of long searches.

We need to sync also sending of "currmove" info to
avoid sending info on current move without first
informing the GUI on the PV line we are searching on.

No functional change.
2012-11-18 11:46:37 +01:00
Marco Costalba 8367cf15da Triviality in move_gives_check()
It seems even a bit faster, although handling of the special
cases is not the hot path.

No functional change.
2012-11-17 13:08:04 +01:00
Marco Costalba 942989939a Better document fail-high condition
At this point we have already verified (value > alpha)
and this implies, in case of a non-PV node, where search
window size is zero, that value >= beta.

This is not so self-evident, so document the code with
an assert condition.

No functional change.
2012-11-17 12:50:10 +01:00
Marco Costalba 239d7b3fd1 Don't access std::cout from Position::print()
Let the caller to decide where to redirect (cout or cerr) the
ASCII representation of the position. Rename the function to
reflect this.

Renamed also from_fen() and to_fen() to set() and fen() respectively.

No functional change.
2012-11-11 13:30:20 +01:00
Marco Costalba 116234d6c6 Restore old BOUND_EXACT logic in qsearch
In case a PvNode node has a static evaluation above alpha but
no available moves we want to flag the node as BOUND_EXACT,
not as BOUND_UPPER as is currently.

The behaviour was recently introduced with patch d471c49700
of 3/10/2012

Spotted by Hongzhi Cheng.

bench: 5558464
2012-11-11 11:52:11 +01:00
Marco Costalba c45a4e0b48 Revert "Don't prune discovered checks"
Both Lucas re-test and Jean-Francois confirrm it
is a regression.

Here Jean-Francois's results after 3600 games :
Score of 96d3b1c92b vs 3b87314: 690 - 729 - 2181 [0.495] 3600
ELO: -3.86 +- 99%: 14.94 95%: 11.35
LOS: 15.03%
Wins: 690 Losses: 729 Draws: 2181 Total: 3600

Bench: 5404066
2012-11-11 11:20:39 +01:00
Marco Costalba 96d3b1c92b Don't prune discovered checks
Don't prune and eventually extend check moves of type
DISCO_CHECK (pun intended, Lucas will understand :-) ).

Patch from Lucas Braesch that has also tested it:

Result: 879-661-2137, score=52.96%, LOS=100.00% (time 10"+0.1")

I have started a verification test right now.

bench: 6004966
2012-11-07 18:29:56 +01:00
Marco Costalba 3b87314331 Previous patch test results
From Jean-Francois's

Final result after 5000 games :
Score of c581b7e vs a878312: 1163 - 970 - 2867 [0.519] 5000
ELO: 13.35 +- 99%: 12.71 95%: 9.65
LOS: 100.00%
Wins: 1163 Losses: 970 Draws: 2867 Total: 5000

From me

After 3266 games at 20"+0,05
Score of c581b7e vs a878312: 612 - 607 - 2047

So no regression at longer TC and perhaps a little gain at
fast TC.
2012-11-07 18:07:03 +01:00
Marco Costalba c581b7ea36 Another attempt at evaluation shortcut
In this case we try a rather drastic
approach: we simply don't futility prune
in qsearch when arriving from a null move.

So we save evaluating and also save to mess
with eval margins at all because margin is used
only in futility.

Also accuracy should not be affected, actually it
improves because we don't prune anything anymore.

bench: 5404066
2012-11-05 16:12:28 +01:00
Marco Costalba a5b1f4774f Temporary revert previous patch
Performs well at very short TC of 40/4+0.05 (courtesy of Jean-Francois):
Wins: 2503 Losses: 2146 Draws: 5581 Total: 10230 +12 ELO

But is poor at longer TC of 20"+0.05
Wins: 321 Losses: 373 Draws: 1141 Total: 1808 -10 ELO

The patch was clearly a tradoff between speed and accuracy and
the most interesting part of it are test results that can be
commented as follows:

- A short TC is very sensible to any speed increase
- A longer TC is more sensible to accuracy and less to speed

So a patch that does not change speed is suitable to be tested at
short TC, while a speed/accuracy compromise patch is IMO better to
be tested at longer TC to verify loss of accuracy can be tolerated.

In this case the revert is only temporary. We will come back again
once we will be able to preserve the evaluation margin.

bench: 5809010
2012-11-05 07:41:28 +01:00
Marco Costalba 37e9802411 Skip evaluate() call after a null move
Reuse the evaluation of the parent with inverted sign and
set margin to zero (this is an hack!).

This is done only in qsearch where almost 15% of calls are
from a null move. In normal search the number of nodes where

(ss-1)->currentMove == MOVE_NULL

is almost zero and so there is no need of using this trick.

The big advantage of this patch is a speed-up due to skipped
evaluate() calls, that are very costly.

Functionality is of course affected and we will need to proper
test it later. For now we just register a 3-4% speed up.

Suggested by Hongzhi Cheng.

bench: 5051328
2012-11-04 10:48:14 +01:00
Marco Costalba a87831230d Correctly fix "break from split" patch
In case split cut-offs we return with still
some moves to go but we really want to break
out from the loop now.

No functional change.
2012-11-03 17:09:18 +01:00
Marco Costalba dd5b3086f5 Relax constrain in prevents_threat()
When testing if a move blocks the threat path there is no
reason to require the threat to be a slider. Indeed threat
can be a double pawn push like in this example:

r1bq1rk1/ppp1np1p/4n1p1/3p4/3P2Q1/2P1B3/PPBN2PP/R4RK1 w - - 0 16

Where white's move Rf6 blocks the threat f5.

As a nice side effect we can retire the now useless helper
piece_is_slider().

This patch kicks in only very rare cases, indeed the bench is
still the same!

bench: 5809010
2012-11-03 15:57:20 +01:00
Marco Costalba 47f988f05f Sync connected_threat() and yields_to_threat()
Rename stuff so to sync as much as posisble the
two related functions.

No functional change.
2012-11-03 15:36:12 +01:00
Marco Costalba 07989712af Don't 'break' upon returning from split()
There is no guarantee that split() consumes all the node's
moves. Indeed split() can return without performing any job
for instance because MAX_SPLITPOINTS_PER_THREAD is reached
or becuase no available threads are found (this latter case
is much more common).

So search must continue in those cases and we cannot force
exiting from move's loop.

Bug introduced by 1ac417edb8 of 5/10/2012

Spotted by Frank Genot.

No functional change.
2012-11-03 14:54:38 +01:00
Marco Costalba d0d69a5358 Remove a redundant condition in connected_moves()
If a previous move attacks the king (with the piece
of the threat move removed) then must be a discovered
check, otherwise it means that first move gave check
and we were not able to do a null move.

Also renamed stuff to better document the function's
context.

No functional change.
2012-11-03 14:24:29 +01:00
Marco Costalba 972de506a0 Relax constrain in connected_moves()
When testing if a piece is moving through the squares
vacated by a previous move there is no reason to require
the piece to be a slider, indeed we can have a double
pawn push like in this example:

r1q2rk1/2p1bppp/2Pp4/pN5b/Q1P1p3/4B2P/PP1R1PP1/1K5R w - - 3 18

Where black's move f5 is connected to previous move Be7 that
frees the path.

Or we can have a castle move:

r1bqkb1r/pppp1ppp/2n1pn2/1B6/4P3/2N2N2/PPPP1PPP/R1BQK2R b KQkq - 5 1

Where a previous move Bb5 allows the white to castle king side.

This time patch is mine ;-)

new bench: 5809010
2012-11-03 13:34:04 +01:00
Marco Costalba ed1574e46c Reformat connected_moves()
Prepare code for the next patch that
will affect functionality.

No functional change.
2012-11-03 13:27:44 +01:00
Marco Costalba 52f55179a8 Fix an off-by-one bug in multi pv print
We send to GUI multi-pv info after each cycle,
not just once at the end of the PV loop. This is
because at high depths a single root search can
be very slow and we want to update the gui as
soon as we have a new PV score.

Idea is good but implementation is broken because
sort() takes as arguments a pointer to the first
element and one past the last element.

So fix the bug and rename sort arguments to better
reflect their meaning.

Another hit by Hongzhi Cheng.  Impressive!

No functional change.
2012-11-03 00:36:40 +01:00
Marco Costalba bbdf9e4737 Fix a condition in connected_moves()
When checking if the moving piece p1 in a previous
move m1 defends the destination square of a move m2
we have to use the occupancy with the from square of
m2 removed so to take in account the case in which
f2 will block an x-ray attack from p1.

For instance in this position:
r2k3r/p1pp1pb1/qn3np1/1N2P3/1p3P2/2B5/PPP3QP/R3K2R b KQ - 1 9

The move eXf6 is connected to the previous move Bc3 that
defends the destination square f6.

With this patch we have about 10% more moves detected as
'connected'. Anyhow the absolute number is very low, about
4000 more moves out of 6M nodes searched.

Another issue spotted by Hongzhi "Hawk Eye" Cheng ;-)

new bench: 5757373
2012-11-02 17:18:38 +01:00
Marco Costalba 94ecdef8ac Micro-optimize pop_lsb() for 64bit case
On Intel, perhaps due to 'lea' instruction this way of
zeroing the lsb of *b seems faster than a shift+negate.

On perft (where any speed difference is magnified) I
got a 6% speed up on my Intel i5 64bit.

Suggested by Hongzhi Cheng.

No functional change.
2012-11-02 12:11:49 +01:00
Marco Costalba e3b0327812 Fix a warning under MSVC
Compiler complies that 'cnt' is initialized but
unused (in !CheckThreeFold case). Moving the
definition of 'cnt'out of the loop  seems to do
the trick.

No functional change.
2012-11-02 11:43:23 +01:00
Marco Costalba c039103b31 Pass InCheck as template parameter of qsearch()
Instead of use a variable so to resolve many conditions
already at compile time. In quiesce is also where we
have most of the InCheck nodes and is one of the most
performance critical code paths.

Speed up of 1.5% with Clang and 1% with gcc

Suggested by Hongzhi Cheng.

No functional change.
2012-11-01 18:45:38 +01:00
Marco Costalba fe1cbe2638 Use correct occupancy in connected_threat()
When checking if a move defends the threatened piece
we correctly remove from the occupancy bitboard the
moved piece. This patch removes from the occupancy also
the threatening piece so to consider the cases of moves
that defend the threatened piece x-raying through the
threat move.

As example in this position:
r3k2r/p1ppqp2/Bn4p1/3p1n2/4P1N1/5Q1P/PPP2P1P/R3K2R w KQkq - 1 10

The threat black move is dxe4. With this patch we include
(and so don't prune) white's Bb7 that would be pruned otherwise.

The number of affected position is very low, around 1% of cases,
so we don't expect ELO changes, neverthless this is the logical
and natural thing to do.

Patch suggested by Hongzhicheng.

new bench: 5323798
2012-10-30 20:27:07 +01:00
Marco Costalba 4e31c39a64 Retire move_attacks_square()
There is only one call site. This patch is a
preparation for the next one that will affect
functionality.

No functional change.
2012-10-30 20:03:35 +01:00
Marco Costalba 13f90a30ef Get rid of ReducedStateInfo struct
ReducedStateInfo is a redundant struct that is also
prone to errors, indeed must be updated any time is
updated StateInfo. It is a trick to partial copy a
StateInfo object in do_move().

This patch takes advantage of builtin macro offsetof()
to directly calculate the number of quad words to copy.
Note that we still use memcpy to do the actual job of
copying the (48 bytes) of data.

Idea by Richard Vida.

No functional and no performance change.
2012-10-29 08:07:17 +01:00
Marco Costalba ad1941fd00 Creative formatting in uci.cpp
Have some fun breaking the indentation rules :-)

No functional change.
2012-10-28 11:28:14 +01:00
Gary Linscott cee6336515 Detect drawish KQKP endings
Based off of the rules from the wikipedia page,
here: http://en.wikipedia.org/wiki/Queen_versus_pawn_endgame.

bench does not change: 5312693 but patch is real, has been
tested on specific positions.
2012-10-28 10:17:57 +01:00
Marco Costalba b3d030294b Reformat check_is_dangerous()
And shuffle some code at search.cpp tail.

No functional change.
2012-10-27 15:07:20 +02:00
Marco Costalba cd80762c13 Use std::stack instead of fixed size array
Only in not performance critical code like pretty_pv(),
otherwise continue to use the good old C-style arrays
like in extract/insert PV where I have done some code
refactoring anyhow.

No functional change.
2012-10-27 13:31:23 +02:00
Marco Costalba 00a853950f Fix broken uci notation for promotions
Silly typo (introduced in e304db9d1e) completely
messed up move notation in case of promotions causing
"Illegal move" warning in cutechess-cli.

Reported by Jörg Oster.

No functional change.
2012-10-26 16:06:47 +02:00
Marco Costalba 4c7a71a44b Fix asserts due to TT access races
In multi-threads runs with debug on we experience some
asserts due to the fact that TT access is intrinsecally
racy and its contents cannot be always trusted so must
be validated before to be used and this is what the
patch does.

No functional case.
2012-10-26 12:41:12 +02:00
Marco Costalba c594b989c0 Extend full 3 fold detection to PvNodes
And restore old behaviour of not returning from a RootNode
without updating RootMoves[].

Also renamed is_draw() template parameters to reflect a
'positive' logic (Check instead of Skip) that is easier
to follow.

New bench: 5312693
2012-10-26 11:56:33 +02:00
Marco Costalba 71f37ac1aa Merge pull request #34 from jromang/repetition
Improve 3 fold repetition detection
2012-10-26 02:01:10 -07:00
Jean-Francois Romang 77c91ac1ba Full three fold repetition detection only at root node 2012-10-25 15:57:44 +02:00
Jean-Francois Romang 5436d98fc5 Enable true 3 fold detection in search 2012-10-25 06:54:05 +00:00
Jean-Francois Romang 0587c5b605 Allow full repetition detection
Based on sshivaji 6ee19aa5389ce60181907ba53bbb50642f2d5657 commit
2012-10-25 06:28:55 +00:00
Marco Costalba c1f4000426 Fix an assert when we stop the search
When signal 'stop' is raised we return bestValue
that could be still set at -VALUE_INFINITE and
this triggers an assert. Fix it by returning
a value we know for sure is not +-VALUE_INFINITE.

Reported by 平岡拓也 Hiraoka.

No functional change.
2012-10-25 00:07:59 +02:00
Marco Costalba 22715259a0 Rename RootPosition and shuffle think()
Just slightly code reshuffles, noting interesting here...

No functional change.
2012-10-24 15:01:39 +02:00
Marco Costalba 4eda653a56 Drop Chess960 and UCIMultiPV globals and rename MultiPV
No functional change.
2012-10-24 14:31:33 +02:00
Marco Costalba e8e5b9f537 Wrap in a class Skill Level code
Note that the actual pickup is done in the class
d'tor so to be sure it is always triggered, even
in case of a sudden exit due to a 'stop' signal.

No functional change.
2012-10-24 12:52:29 +02:00
Marco Costalba 9c2b3faec4 Shuffle aspiration window loop
No functional change.
2012-10-24 11:35:53 +02:00
Marco Costalba 70b1b79264 Retire refine_eval()
Inline its content and better comment what it does through
some renaming.

No functional change.
2012-10-22 10:03:53 +02:00
Marco Costalba 5fc8b27db9 Don't copy a full Position object in print()
Function move_to_san() requires the Position to be
passed by referenced because a do/undo move is done
inside the function to detect a possible mate and to
add to the san string the corresponding '#' suffix.

Instead of passing a copy of current position pass
directly the original position object after const
casting it. This has the advantage to avoid a costly
Position copy, on the down side a bench test could
report different searched nodes if print(move) is
used, due to the additionals do_move() calls.

No functional change.
2012-10-22 00:55:16 +02:00
Marco Costalba dbbbd3880c Don't need to init board[] with NO_PIECE
Now that NO_PIECE == 0 the common memset() will
do the work.

No functional change.
2012-10-22 00:38:12 +02:00
Marco Costalba e40b06a050 Change NO_PIECE value and shrink PieceValue[]
This requires changing color_of() definition.

No functional change.
2012-10-21 11:50:56 +02:00
Marco Costalba e304db9d1e Use self-describing constants instead of numbers
And remove now useless comments.

No functional change.
2012-10-21 11:16:21 +02:00
Marco Costalba 6b909b2343 Move RootColor from Eval to Search
No functional change.
2012-10-21 09:12:02 +02:00
Marco Costalba 55bd27b8f0 Contempt factor: use DrawValue also in case of stealmates
Spotted by Jörg Oster.

No functional change (when contempt factor is not used).
2012-10-20 11:02:37 +02:00
Marco Costalba 1018474853 Fix compatibility with old Windows 95 and 98
Report and patches by bnemias.

No functional change.

Signed-off-by: Marco Costalba <mcostalba@gmail.com>
2012-10-20 10:48:26 +02:00
Marco Costalba c11edd3b0b Merge pull request #33 from daylen/master
Further improve OS X compatibility

Change the minimum supported version from 10.6 to 10.0

No functional change.
2012-10-20 01:07:27 -07:00
Daylen Yang f0298862f5 Further improve compatibility when compiling on OS X
Change the minimum supported version from 10.6 to 10.0
2012-10-16 00:13:16 -07:00
Marco Costalba 739d23f2a3 Fix Makefile for PowerPC with prefetch enabled
Existing Makefile is buggy for PowerPC, it has no
SSE, yet it is given it if Prefetch is enabled,
because it isn't ARMv7.

Patch from Matthew Brades.

No functional change.
2012-10-15 01:13:41 +02:00
Marco Costalba 3aa2d6db18 Document why is safe ttValue == VALUE_NONE
We can have ttValue == VALUE_NONE when we use a TT
slot to just save a position static evaluation, but
in this case we also save DEPTH_NONE so to avoid
using the ttValue in search. This happens to work,
but due to a number of lucky and tricky cases that
we now documnet through a bunch of asserts and a
little change to value_from_tt() that has no real
effect but clarifing the code.

No functional change.
2012-10-14 12:47:16 +02:00
Marco Costalba 6a75291ab1 Set TT size to 32 MB during 'bench'
On some platforms 128 MB of RAM for TT is too much,
so run 'bench' with the default 32 MB size.

No functional change although of course now 'bench'
reports a different number: 5545018
2012-10-14 09:02:17 +02:00
Marco Costalba eb1a4f11fa Move all Contempt Factor code to search.cpp
Where it is used.

No functional change.
2012-10-13 14:49:01 +02:00
Marco Costalba d023be5a21 Retire BitCount8Bit[] table
Use popcount() instead in the only calling place.
It is used only at initialization so there is no
speed regression and anyhow even initialization
itself is not slowed down: magic bitboard setup
stays around 175 msec on my slow 32bit Core Duo.

No functional change.
2012-10-13 14:24:01 +02:00
Marco Costalba c2cd75843e Use new 64 bit De Bruijn BitScan
Allows to sync 32 and 64 bits implementations.

Idea by Kim Walisch, reported by Gerd Isenberg:
http://talkchess.com/forum/viewtopic.php?t=45554

No functional change.
2012-10-13 13:45:45 +02:00
Marco Costalba 189b6fc270 Retire can_return_tt() and rewirte TT-hit code
Simplify the code and doing this introduce a couple
of (very small) functional changes:

- Always compare to depth even in "mate value" condition
- TT cut-off in qsearch also in case of PvNode, as in search

Verified against regression with 2500 games at 30"+0.05
on 2 threads: 451 - 444 - 1602

Functional changed: new bench is 5544977
2012-10-13 11:45:14 +02:00
Marco Costalba 6c8663341e Scale contempt factor to zero at endgame
Contempt Factor is more effective at opening/middle game
and seems harmful at endgame phase. See:

http://chessprogramming.wikispaces.com/Contempt+Factor

http://web.archive.org/web/20070707023203/www.brucemo.com/compchess/programming/contempt.htm

Therefore we scale down the contempt factor while going
on with the game so to reach zero at endgame phase.

No functional change.
2012-10-13 10:40:38 +02:00
Marco Costalba aba152ea3a Fix a minor bug in search
As Joona says: "The problem is that when doing full
window search (-VALUE_INFINITE, VALUE_INFINITE), and
pruning all the moves will return fail low which is
mate score, because only clause touching alpha is
"mate distance pruning". So we are returning mate score
although we are just pruning all the moves. In reality
there probably is no mate in sight.

Bug spotted and fixed by Joona.
2012-10-11 21:12:43 +02:00
Jean-Francois Romang 7f9ebf8e86 ARM lsb/msb assembly
Implement lsb/msb using armv7 assembly instructions.
msb is the easiest one, using a gcc intrinsic that generates
code using the ARM's clz instruction. lsb is also using this
clz instruction, but with the help of ARM's 'rbit' (bit
reversing) instruction. This leads to a >2% speed gain.

I also renamed 'arm-32' to the more meaningfull 'armv7' in the Makefile

No functional change.
2012-10-11 21:01:52 +02:00
Jean-Francois Romang 4e7da9be3d Introduce arm-32 ARCH in Makefile
No functional change.
2012-10-11 07:34:48 +02:00
Marco Costalba 46c01b5083 Retire is_dangerous() and inline its content
No functional change.
2012-10-10 08:13:36 +02:00
Marco Costalba dae843d4d6 Rearrange prefetch code
No functional change.
2012-10-08 11:43:47 +02:00
Marco Costalba 78fe0cfb8d Merge pull request #29 from RyanTaker/patch-3
Add Contempt Factor to Polyglot.ini
2012-10-07 15:36:08 -07:00
RyanTaker d77d555c72 Add Contempt Factor in Polyglot
The contempt factor was previously not included in polyglot.ini
2012-10-07 09:49:55 -07:00
Marco Costalba bbaef048cd Sync qsearch with search
Port to qsearch() the same changes we recently
added to search().

Overall this search refactoring series shows
almost 2% speed up on gcc compile.

No functional change.
2012-10-07 13:15:41 +02:00
Daylen Yang 7e2d49368d Improve compatibility with older versions of Mac OS X
Use the -mmacosx-version-min flag to support older versions of Mac OS X
(like version 10.6 and 10.7) when compiled on a machine running version
10.8.
2012-10-06 17:56:12 -07:00
Marco Costalba 954fc950d9 Fix POPCNT support on mingw 64
When using asm 'popcnt' instruction the given
operand registers must be of the same type.

No functional change.
2012-10-06 13:01:44 +02:00
Marco Costalba d777c4d789 Fix mingw compile with ARCH=x86-64
When using the Makefile (as for the mingw case),
IS_64BIT and USE_BSFQ are already set with
ARCH=x86-64 and do not need to be redefined.

No functional change.

Signed-off-by: Marco Costalba <mcostalba@gmail.com>
2012-10-06 12:34:09 +02:00
Marco Costalba cedbd3332a Fix Contempt Factor implementation
First disable Contempt Factor during analysis, then
calculate the modified draw score from the point of
view of the player, so from the point of view of
RootPosition color.

Thanks to Ryan Taker for suggesting the fixes.

No functional change.
2012-10-06 10:12:34 +02:00
Marco Costalba ff9ca3e76e Fix fallouts from previous patch
These kind of arch specific code is really nasty
to make it right becuase you need to verify on
all the platforms.

Now should compile properly also on ARM

Reported by Jean-Francois.

No functional change.
2012-10-06 09:09:27 +02:00
Jean-Francois Romang b8948e84b8 Allow prefetching on non-x86 platforms with gcc
In particular on ARM processors. Original patch by
Jean-Francois, sligtly modified by me to preserve
the meaning of NO_PREFETCH flag.

Verified with gcc, clang and icc that prefetch instruction
is correctly created.

No functional change.
2012-10-06 00:43:16 +02:00
Marco Costalba 1ac417edb8 Retire futility_move_count()
And remove (bestValue < beta) condition from
moves loop.

No functional change.
2012-10-05 18:24:38 +02:00
Marco Costalba d471c49700 Rewrite search best value update
A simplification and also a small speed-up of
about 1% mainly due to reducing calls to
thisThread->cutoff_occurred().

Worst case split point recovering time after a
cut-off occurred is limited to 3 msec on my slow
PC, and usually is below 1 msec, so it seems safe
to remove the cutoff_occurred() check.

No functional change.
2012-10-05 13:53:05 +02:00
Marco Costalba c9f9262a49 Add experimental contempt factor
This is very crude and very basic: simply in case
of a draw for repetition or 50 moves rule return
a negative score instead of zero according to the
contempt factor (in centipawns). If contempt is
positive engine will try to avoid draws (to use
with weaker opponents), if negative engine will
try to draw. If zero (default) there are no changes.

No functional change.
2012-10-05 08:28:23 +02:00
Marco Costalba bd7a0d4ce4 Retire EasyMoveMargin
Use a value related to PawnValue instead.

This is a different patch from previous one because
could affect game play and skill levels, although
in a mostly unmeasurable way. Indeed thresold has
been raised so easy move is a bit harder to trigger
and skill level is a bit more prone to blunders.

No functional change.

Signed-off-by: Marco Costalba <mcostalba@gmail.com>
2012-10-03 18:59:24 +02:00
Marco Costalba 561eb34aea Don't hide thresolds values
Show the real value in the code, not hide it
behind a variable name, especially when there
is only once occurence.

No functional change.

Signed-off-by: Marco Costalba <mcostalba@gmail.com>
2012-10-03 18:56:36 +02:00
Marco Costalba 4c91dbc28e Further push singular extension
Extend for an extra half-ply in case the node is (probably)
going to fail high. In this case the added overhead is limited.

A novelity is the way this patch has been tested: Always in
self-play but with a much longer TC to allow the singular
extension to fully kick in and also (my impression) to have
less noisy results.

Ater 1015 games on my QUAD at 60"+0.05
Mod vs Orig 173 - 150 - 692 ELO +8
2012-10-02 06:39:27 +02:00
Marco Costalba 1b6b711c44 Further rearrange search()
No functional change.
2012-10-01 10:44:04 +02:00
Marco Costalba faaa1f1116 Don't push on the stack 200KB in one go
This could be a limit on some platforms (as it seems
to be in Native Client). Patch from a SF fork on github:

https://github.com/ccherng/Stockfish/commit/47374afd6fdfabd9de183a7a67d645daad45fb21

No functional change.
2012-09-30 11:35:14 +02:00
Marco Costalba ed0fb0b05f Add support for node limited search
Handle also the SMP case. This has been quite tricky, not
trivial to enforce the node limit in SMP case becuase
with "helpful master" concept we can have recursive split
points and we cannot lock them all at once so there is the
risk of counting the same nodes more than once.

Anyhow this patch should be race free and counted nodes are
correct.

No functional change.
2012-09-30 10:19:22 +02:00
Marco Costalba e5463eb3ae Skip some useless initializations in search()
And rearrange a bit the initialization code. Still
some polishing to do, but it is a first step.

No functional change.
2012-09-29 23:12:39 +02:00
Marco Costalba d53c928261 Don't need to early check PV moves for legality
As long as isPvMove (renamed to pvMove) is set after
legality check, we can postpone legality even in PV case.

Patch aligns the PV case with the common non-pv one.

No functional change.
2012-09-29 18:05:49 +02:00
Marco Costalba bc8f5fe0bf Drop a magic in book.cpp
Mask out move's spacial flags without relying
on internal Move representation.

No functional change.
2012-09-22 11:19:10 +02:00
Marco Costalba 9204a60dbb Tweaks to bitcount functions
Seems even a bit faster now (almost 1% in 32bit case).

No functional change.
2012-09-22 10:59:33 +02:00
Marco Costalba 1cb2722c95 Restore development version 2012-09-22 00:37:18 +02:00
Marco Costalba 3caeabf73b Stockfish 2.3.1
Stockfish bench signature is: 5423738
2012-09-22 00:20:44 +02:00
Gary Linscott fdbe8006e0 Bonus for rook/queen attacking pawns on same rank
Patch and tuning by Gary Linscott from an idea of Ryan Taker.

Double tested by Gary:

Wins: 3390 Losses: 2972 Draws: 11323
LOS: 99.999992%
ELO: 8.213465 +- 99%: 6.746506 95%: 5.124415
Win%: 51.181792 +- 99%: 0.969791 95%: 0.736740

And by me:

After 5612 games 1255  1085  3271 +11 ELO
2012-09-21 23:25:13 +02:00
Marco Costalba 09acdac56b Fix compile on 64 bits
Reported by Quocvuong82.

No functional change.
2012-09-20 19:25:27 +02:00
Marco Costalba e4a0482e43 Simplify BSFTable initialization
No functional change.

Signed-off-by: Marco Costalba <mcostalba@gmail.com>
2012-09-19 20:47:53 +02:00
Marco Costalba 22a5f91aa7 Fix crash under Chess 960
We have a crash with this position:

rkqbnnbr/pppppppp/8/8/8/8/PPPPPPPP/RKQBNNBR w HAha -

What happens is that even if we are castling QUEEN_SIDE,
in this case we have kfrom (B8) < kto (C8) so the loop
that checks for attackers runs forever leading to a crash.

The fix is to check for (kto > kfrom) instead of
Side == KING_SIDE, but this is slower in the normal case of
ortodhox chess, so rewrite generate_castle() to handle the
chess960 case as a template parameter and allow the compiler
to optimize out the comparison in case of normal chess.

Reported by Ray Banks.
2012-09-16 14:14:55 +02:00
Marco Costalba 9ce7469846 Rename class Book to PolyglotBook
And move struct BookEntry out of the header where it is
not needed.

No functional change.
2012-09-16 10:32:57 +02:00
Marco Costalba 4b7dbb3922 Fix KpsK endgame
Broken by commit a44c5cf4f7 of 3 /12 / 2011 that
was labeled "No functional change" because our 'bench'
test didn't triggered that particular endgame. Indeed
we need to run a specific bench on a set of endgames
position when touching endgame.cpp because normal bench
does not cover endgames properly.

Found by MSVC 2012 code analyzer.
2012-09-16 08:57:26 +02:00
Marco Costalba e0035e9ca9 Restore development version
No functional change.
2012-09-15 11:02:08 +02:00
45 changed files with 3414 additions and 3840 deletions
+18 -29
View File
@@ -1,5 +1,4 @@
1. Introduction ### Overview
---------------
Stockfish is a free UCI chess engine derived from Glaurung 2.1. It is Stockfish is a free UCI chess engine derived from Glaurung 2.1. It is
not a complete chess program and requires some UCI-compatible GUI not a complete chess program and requires some UCI-compatible GUI
@@ -8,46 +7,37 @@ Partner or Fritz) in order to be used comfortably. Read the
documentation for your GUI of choice for information about how to use documentation for your GUI of choice for information about how to use
Stockfish with it. Stockfish with it.
This version of Stockfish supports up to 32 CPUs, but has not been This version of Stockfish supports up to 64 CPUs. The engine defaults
tested thoroughly with more than 4. The program tries to detect the to one search thread it is therefore recommended to inspect the value of
number of CPUs on your computer and sets the number of search threads the *Threads* UCI parameter, and to make sure it equals the number of CPU
accordingly, but please be aware that the detection is not always cores on your computer.
correct. It is therefore recommended to inspect the value of the
"Threads" UCI parameter, and to make sure it equals the number of CPU
cores on your computer. If you are using more than eight threads, it is
recommended to raise the value of the "Min Split Depth" UCI parameter to
7.
2. Files ### Files
--------
This distribution of Stockfish consists of the following files: This distribution of Stockfish consists of the following files:
* Readme.txt, the file you are currently reading. * Readme.md, the file you are currently reading.
* Copying.txt, a text file containing the GNU General Public License. * Copying.txt, a text file containing the GNU General Public License.
* src/, a subdirectory containing the full source code, including a * src, a subdirectory containing the full source code, including a Makefile
Makefile that can be used to compile Stockfish on Unix-like systems. that can be used to compile Stockfish on Unix-like systems. For further
For further information about how to compile Stockfish yourself read information about how to compile Stockfish yourself read section below.
section 4 below.
* polyglot.ini, for using Stockfish with Fabien Letouzey's PolyGlot * polyglot.ini, for using Stockfish with Fabien Letouzey's PolyGlot
adapter. adapter.
3. Opening books ### Opening books
----------------
This version of Stockfish has support for PolyGlot opening books. For This version of Stockfish has support for PolyGlot opening books. For
information about how to create such books, consult the PolyGlot information about how to create such books, consult the PolyGlot
documentation. The book file can be selected by setting the "Book File" documentation. The book file can be selected by setting the *Book File*
UCI parameter. UCI parameter.
4. Compiling it yourself ### Compiling it yourself
------------------------
On Unix-like systems, it should be possible to compile Stockfish On Unix-like systems, it should be possible to compile Stockfish
directly from the source code with the included Makefile. directly from the source code with the included Makefile.
@@ -55,17 +45,16 @@ directly from the source code with the included Makefile.
Stockfish has support for 32 or 64-bit CPUs, the hardware POPCNT Stockfish has support for 32 or 64-bit CPUs, the hardware POPCNT
instruction, big-endian machines such as Power PC, and other platforms. instruction, big-endian machines such as Power PC, and other platforms.
In general it is recommended to run 'make help' to see a list of make In general it is recommended to run `make help` to see a list of make
targets with corresponding descriptions. When not using Makefile to targets with corresponding descriptions. When not using Makefile to
compile (for instance with Microsoft MSVC) you need to manually compile (for instance with Microsoft MSVC) you need to manually
set/unset some switches in the compiler command line; see file "types.h" set/unset some switches in the compiler command line; see file *types.h*
for a quick reference. for a quick reference.
5. Terms of use ### Terms of use
---------------
Stockfish is free, and distributed under the GNU General Public License Stockfish is free, and distributed under the **GNU General Public License**
(GPL). Essentially, this means that you are free to do almost exactly (GPL). Essentially, this means that you are free to do almost exactly
what you want with the program, including distributing it among your what you want with the program, including distributing it among your
friends, making it available for download from your web site, selling friends, making it available for download from your web site, selling
@@ -78,4 +67,4 @@ to where the source code can be found. If you make any changes to the
source code, these changes must also be made available under the GPL. source code, these changes must also be made available under the GPL.
For full details, read the copy of the GPL found in the file named For full details, read the copy of the GPL found in the file named
Copying.txt. *Copying.txt*
+10 -6
View File
@@ -1,4 +1,3 @@
[PolyGlot] [PolyGlot]
EngineDir = . EngineDir = .
@@ -15,21 +14,25 @@ ResignScore = 600
[Engine] [Engine]
Use Search Log = false Write Debug Log = false
Write Search Log = false
Search Log Filename = SearchLog.txt Search Log Filename = SearchLog.txt
Book File = book.bin Book File = book.bin
Best Book Move = false Best Book Move = false
Mobility (Middle Game) = 100 Contempt Factor = 0
Mobility (Midgame) = 100
Mobility (Endgame) = 100 Mobility (Endgame) = 100
Passed Pawns (Middle Game) = 100 Pawn Structure (Midgame) = 100
Pawn Structure (Endgame) = 100
Passed Pawns (Midgame) = 100
Passed Pawns (Endgame) = 100 Passed Pawns (Endgame) = 100
Space = 100 Space = 100
Aggressiveness = 100 Aggressiveness = 100
Cowardice = 100 Cowardice = 100
Min Split Depth = 4 Min Split Depth = 0
Max Threads per Split Point = 5 Max Threads per Split Point = 5
Threads = 1 Threads = 1
Use Sleeping Threads = false Idle Threads Sleep = false
Hash = 128 Hash = 128
Ponder = true Ponder = true
OwnBook = false OwnBook = false
@@ -39,5 +42,6 @@ Emergency Move Horizon = 40
Emergency Base Time = 200 Emergency Base Time = 200
Emergency Move Time = 70 Emergency Move Time = 70
Minimum Thinking Time = 20 Minimum Thinking Time = 20
Slow Mover = 100
UCI_Chess960 = false UCI_Chess960 = false
UCI_AnalyseMode = false UCI_AnalyseMode = false
+71 -20
View File
@@ -1,6 +1,6 @@
# Stockfish, a UCI chess playing engine derived from Glaurung 2.1 # Stockfish, a UCI chess playing engine derived from Glaurung 2.1
# Copyright (C) 2004-2008 Tord Romstad (Glaurung author) # Copyright (C) 2004-2008 Tord Romstad (Glaurung author)
# Copyright (C) 2008-2012 Marco Costalba, Joona Kiiski, Tord Romstad # Copyright (C) 2008-2013 Marco Costalba, Joona Kiiski, Tord Romstad
# #
# Stockfish is free software: you can redistribute it and/or modify # Stockfish is free software: you can redistribute it and/or modify
# it under the terms of the GNU General Public License as published by # it under the terms of the GNU General Public License as published by
@@ -34,8 +34,9 @@ ifeq ($(UNAME),Haiku)
endif endif
BINDIR = $(PREFIX)/bin BINDIR = $(PREFIX)/bin
### Built-in benchmark for pgo-builds ### Built-in benchmark for pgo-builds and signature
PGOBENCH = ./$(EXE) bench 32 1 10 default depth PGOBENCH = ./$(EXE) bench 32 1 10 default depth
SIGNBENCH = ./$(EXE) bench
### Object files ### Object files
OBJS = benchmark.o bitbase.o bitboard.o book.o endgame.o evaluate.o main.o \ OBJS = benchmark.o bitbase.o bitboard.o book.o endgame.o evaluate.o main.o \
@@ -58,6 +59,7 @@ OBJS = benchmark.o bitbase.o bitboard.o book.o endgame.o evaluate.o main.o \
# bsfq = yes/no --- -DUSE_BSFQ --- Use bsfq x86_64 asm-instruction (only # bsfq = yes/no --- -DUSE_BSFQ --- Use bsfq x86_64 asm-instruction (only
# with GCC and ICC 64-bit) # with GCC and ICC 64-bit)
# popcnt = yes/no --- -DUSE_POPCNT --- Use popcnt x86_64 asm-instruction # popcnt = yes/no --- -DUSE_POPCNT --- Use popcnt x86_64 asm-instruction
# sse = yes/no --- -msse --- Use Intel Streaming SIMD Extensions
# #
# Note that Makefile is space sensitive, so when adding new architectures # Note that Makefile is space sensitive, so when adding new architectures
# or modifying existing flags, you have to make sure there are no extra spaces # or modifying existing flags, you have to make sure there are no extra spaces
@@ -77,6 +79,7 @@ ifeq ($(ARCH),general-64)
prefetch = no prefetch = no
bsfq = no bsfq = no
popcnt = no popcnt = no
sse = no
endif endif
ifeq ($(ARCH),general-32) ifeq ($(ARCH),general-32)
@@ -86,6 +89,7 @@ ifeq ($(ARCH),general-32)
prefetch = no prefetch = no
bsfq = no bsfq = no
popcnt = no popcnt = no
sse = no
endif endif
# x86-section # x86-section
@@ -96,6 +100,7 @@ ifeq ($(ARCH),x86-64)
prefetch = yes prefetch = yes
bsfq = yes bsfq = yes
popcnt = no popcnt = no
sse = yes
endif endif
ifeq ($(ARCH),x86-64-modern) ifeq ($(ARCH),x86-64-modern)
@@ -105,6 +110,7 @@ ifeq ($(ARCH),x86-64-modern)
prefetch = yes prefetch = yes
bsfq = yes bsfq = yes
popcnt = yes popcnt = yes
sse = yes
endif endif
ifeq ($(ARCH),x86-32) ifeq ($(ARCH),x86-32)
@@ -114,6 +120,7 @@ ifeq ($(ARCH),x86-32)
prefetch = yes prefetch = yes
bsfq = no bsfq = no
popcnt = no popcnt = no
sse = yes
endif endif
ifeq ($(ARCH),x86-32-old) ifeq ($(ARCH),x86-32-old)
@@ -123,6 +130,18 @@ ifeq ($(ARCH),x86-32-old)
prefetch = no prefetch = no
bsfq = no bsfq = no
popcnt = no popcnt = no
sse = no
endif
#arm section
ifeq ($(ARCH),armv7)
arch = armv7
os = any
bits = 32
prefetch = yes
bsfq = yes
popcnt = no
sse = no
endif endif
# osx-section # osx-section
@@ -133,6 +152,7 @@ ifeq ($(ARCH),osx-ppc-64)
prefetch = no prefetch = no
bsfq = no bsfq = no
popcnt = no popcnt = no
sse = no
endif endif
ifeq ($(ARCH),osx-ppc-32) ifeq ($(ARCH),osx-ppc-32)
@@ -142,6 +162,7 @@ ifeq ($(ARCH),osx-ppc-32)
prefetch = no prefetch = no
bsfq = no bsfq = no
popcnt = no popcnt = no
sse = no
endif endif
ifeq ($(ARCH),osx-x86-64) ifeq ($(ARCH),osx-x86-64)
@@ -151,6 +172,7 @@ ifeq ($(ARCH),osx-x86-64)
prefetch = yes prefetch = yes
bsfq = yes bsfq = yes
popcnt = no popcnt = no
sse = yes
endif endif
ifeq ($(ARCH),osx-x86-32) ifeq ($(ARCH),osx-x86-32)
@@ -160,6 +182,7 @@ ifeq ($(ARCH),osx-x86-32)
prefetch = yes prefetch = yes
bsfq = no bsfq = no
popcnt = no popcnt = no
sse = yes
endif endif
@@ -209,7 +232,7 @@ ifeq ($(COMP),clang)
endif endif
### 3.2 General compiler settings ### 3.2 General compiler settings
CXXFLAGS = -g -Wall -Wcast-qual -fno-exceptions -fno-rtti $(EXTRACXXFLAGS) CXXFLAGS = -Wall -Wcast-qual -fno-exceptions -fno-rtti $(EXTRACXXFLAGS)
ifeq ($(comp),gcc) ifeq ($(comp),gcc)
CXXFLAGS += -ansi -pedantic -Wno-long-long -Wextra -Wshadow CXXFLAGS += -ansi -pedantic -Wno-long-long -Wextra -Wshadow
@@ -220,7 +243,7 @@ ifeq ($(comp),mingw)
endif endif
ifeq ($(comp),icc) ifeq ($(comp),icc)
CXXFLAGS += -wd383,981,1418,1419,10187,10188,11505,11503 -Wcheck -Wabi -Wdeprecated -strict-ansi CXXFLAGS += -wd383,981,1418,1419,1476,10187,10188,11505,11503 -Wcheck -Wabi -Wdeprecated -strict-ansi
endif endif
ifeq ($(comp),clang) ifeq ($(comp),clang)
@@ -228,12 +251,16 @@ ifeq ($(comp),clang)
endif endif
ifeq ($(os),osx) ifeq ($(os),osx)
CXXFLAGS += -arch $(arch) CXXFLAGS += -arch $(arch) -mmacosx-version-min=10.0
endif endif
### 3.3 General linker settings ### 3.3 General linker settings
LDFLAGS = $(EXTRALDFLAGS) LDFLAGS = $(EXTRALDFLAGS)
ifeq ($(comp),mingw)
LDFLAGS += -static-libstdc++ -static-libgcc
endif
### On mingw use Windows threads, otherwise POSIX ### On mingw use Windows threads, otherwise POSIX
ifneq ($(comp),mingw) ifneq ($(comp),mingw)
# Haiku has pthreads in its libroot, so only link it in on other platforms # Haiku has pthreads in its libroot, so only link it in on other platforms
@@ -243,12 +270,14 @@ ifneq ($(comp),mingw)
endif endif
ifeq ($(os),osx) ifeq ($(os),osx)
LDFLAGS += -arch $(arch) LDFLAGS += -arch $(arch) -mmacosx-version-min=10.0
endif endif
### 3.4 Debugging ### 3.4 Debugging
ifeq ($(debug),no) ifeq ($(debug),no)
CXXFLAGS += -DNDEBUG CXXFLAGS += -DNDEBUG
else
CXXFLAGS += -g
endif endif
### 3.5 Optimization ### 3.5 Optimization
@@ -265,6 +294,10 @@ ifeq ($(optimize),yes)
CXXFLAGS += -mdynamic-no-pic CXXFLAGS += -mdynamic-no-pic
endif endif
endif endif
ifeq ($(arch),armv7)
CXXFLAGS += -fno-gcse
endif
endif endif
ifeq ($(comp),mingw) ifeq ($(comp),mingw)
@@ -301,8 +334,10 @@ endif
### 3.7 prefetch ### 3.7 prefetch
ifeq ($(prefetch),yes) ifeq ($(prefetch),yes)
ifeq ($(sse),yes)
CXXFLAGS += -msse CXXFLAGS += -msse
DEPENDFLAGS += -msse DEPENDFLAGS += -msse
endif
else else
CXXFLAGS += -DNO_PREFETCH CXXFLAGS += -DNO_PREFETCH
endif endif
@@ -322,6 +357,7 @@ endif
### needs access to the optimization flags. ### needs access to the optimization flags.
ifeq ($(comp),gcc) ifeq ($(comp),gcc)
ifeq ($(optimize),yes) ifeq ($(optimize),yes)
ifeq ($(debug),no)
GCC_MAJOR := `$(CXX) -dumpversion | cut -f1 -d.` GCC_MAJOR := `$(CXX) -dumpversion | cut -f1 -d.`
GCC_MINOR := `$(CXX) -dumpversion | cut -f2 -d.` GCC_MINOR := `$(CXX) -dumpversion | cut -f2 -d.`
ifeq (1,$(shell expr \( $(GCC_MAJOR) \> 4 \) \| \( $(GCC_MAJOR) \= 4 \& $(GCC_MINOR) \>= 5 \))) ifeq (1,$(shell expr \( $(GCC_MAJOR) \> 4 \) \| \( $(GCC_MAJOR) \= 4 \& $(GCC_MINOR) \>= 5 \)))
@@ -329,6 +365,7 @@ ifeq ($(comp),gcc)
LDFLAGS += $(CXXFLAGS) LDFLAGS += $(CXXFLAGS)
endif endif
endif endif
endif
endif endif
### ========================================================================== ### ==========================================================================
@@ -343,32 +380,34 @@ help:
@echo "" @echo ""
@echo "Supported targets:" @echo "Supported targets:"
@echo "" @echo ""
@echo "build > Build unoptimized version" @echo "build > Standard build"
@echo "profile-build > Build PGO-optimized version" @echo "signature-build > Standard build with embedded signature"
@echo "profile-build > PGO build"
@echo "signature-profile-build > PGO build with embedded signature"
@echo "strip > Strip executable" @echo "strip > Strip executable"
@echo "install > Install executable" @echo "install > Install executable"
@echo "clean > Clean up" @echo "clean > Clean up"
@echo "testrun > Make sample run"
@echo "" @echo ""
@echo "Supported archs:" @echo "Supported archs:"
@echo "" @echo ""
@echo "x86-64 > x86 64-bit" @echo "x86-64 > x86 64-bit"
@echo "x86-64-modern > x86 64-bit with runtime support for popcnt instruction" @echo "x86-64-modern > x86 64-bit with popcnt support"
@echo "x86-32 > x86 32-bit excluding old hardware without SSE-support" @echo "x86-32 > x86 32-bit with SSE support"
@echo "x86-32-old > x86 32-bit including also very old hardware" @echo "x86-32-old > x86 32-bit fall back for old hardware"
@echo "osx-ppc-64 > PPC-Mac OS X 64 bit" @echo "osx-ppc-64 > PPC-Mac OS X 64 bit"
@echo "osx-ppc-32 > PPC-Mac OS X 32 bit" @echo "osx-ppc-32 > PPC-Mac OS X 32 bit"
@echo "osx-x86-64 > x86-Mac OS X 64 bit" @echo "osx-x86-64 > x86-Mac OS X 64 bit"
@echo "osx-x86-32 > x86-Mac OS X 32 bit" @echo "osx-x86-32 > x86-Mac OS X 32 bit"
@echo "armv7 > ARMv7 32 bit"
@echo "general-64 > unspecified 64-bit" @echo "general-64 > unspecified 64-bit"
@echo "general-32 > unspecified 32-bit" @echo "general-32 > unspecified 32-bit"
@echo "" @echo ""
@echo "Supported comps:" @echo "Supported compilers:"
@echo "" @echo ""
@echo "gcc > Gnu compiler (default)" @echo "gcc > Gnu compiler (default)"
@echo "icc > Intel compiler"
@echo "mingw > Gnu compiler with MinGW under Windows" @echo "mingw > Gnu compiler with MinGW under Windows"
@echo "clang > LLVM Clang compiler" @echo "clang > LLVM Clang compiler"
@echo "icc > Intel compiler"
@echo "" @echo ""
@echo "Non-standard targets:" @echo "Non-standard targets:"
@echo "" @echo ""
@@ -376,10 +415,11 @@ help:
@echo "" @echo ""
@echo "Examples. If you don't know what to do, you likely want to run: " @echo "Examples. If you don't know what to do, you likely want to run: "
@echo "" @echo ""
@echo "make profile-build ARCH=x86-64 (This is for 64-bit systems)" @echo "make build ARCH=x86-64 (This is for 64-bit systems)"
@echo "make profile-build ARCH=x86-32 (This is for 32-bit systems)" @echo "make build ARCH=x86-32 (This is for 32-bit systems)"
@echo "" @echo ""
.PHONY: build profile-build embed-signature
build: build:
$(MAKE) ARCH=$(ARCH) COMP=$(COMP) config-sanity $(MAKE) ARCH=$(ARCH) COMP=$(COMP) config-sanity
$(MAKE) ARCH=$(ARCH) COMP=$(COMP) all $(MAKE) ARCH=$(ARCH) COMP=$(COMP) all
@@ -404,6 +444,18 @@ profile-build:
@echo "Step 4/4. Deleting profile data ..." @echo "Step 4/4. Deleting profile data ..."
$(MAKE) ARCH=$(ARCH) COMP=$(COMP) $(profile_clean) $(MAKE) ARCH=$(ARCH) COMP=$(COMP) $(profile_clean)
embed-signature:
@echo "Running benchmark for getting the signature ..."
@$(SIGNBENCH) 2>&1 | sed -n 's/Nodes searched : \(.*\)/\/string Version\/s\/"\\(.*\\)"\/"sig-\1"\//p' > sign.txt
@sed -f sign.txt misc.cpp > misc2.cpp
@mv misc2.cpp misc.cpp
@rm sign.txt
signature-build: build embed-signature
$(MAKE) ARCH=$(ARCH) COMP=$(COMP) all
signature-profile-build: build embed-signature profile-build
strip: strip:
strip $(EXE) strip $(EXE)
@@ -415,9 +467,6 @@ install:
clean: clean:
$(RM) $(EXE) $(EXE).exe *.o .depend *~ core bench.txt *.gcda $(RM) $(EXE) $(EXE).exe *.o .depend *~ core bench.txt *.gcda
testrun:
@$(PGOBENCH)
default: default:
help help
@@ -438,6 +487,7 @@ config-sanity:
@echo "prefetch: '$(prefetch)'" @echo "prefetch: '$(prefetch)'"
@echo "bsfq: '$(bsfq)'" @echo "bsfq: '$(bsfq)'"
@echo "popcnt: '$(popcnt)'" @echo "popcnt: '$(popcnt)'"
@echo "sse: '$(sse)'"
@echo "" @echo ""
@echo "Flags:" @echo "Flags:"
@echo "CXX: $(CXX)" @echo "CXX: $(CXX)"
@@ -449,12 +499,13 @@ config-sanity:
@test "$(debug)" = "yes" || test "$(debug)" = "no" @test "$(debug)" = "yes" || test "$(debug)" = "no"
@test "$(optimize)" = "yes" || test "$(optimize)" = "no" @test "$(optimize)" = "yes" || test "$(optimize)" = "no"
@test "$(arch)" = "any" || test "$(arch)" = "x86_64" || test "$(arch)" = "i386" || \ @test "$(arch)" = "any" || test "$(arch)" = "x86_64" || test "$(arch)" = "i386" || \
test "$(arch)" = "ppc64" || test "$(arch)" = "ppc" test "$(arch)" = "ppc64" || test "$(arch)" = "ppc" || test "$(arch)" = "armv7"
@test "$(os)" = "any" || test "$(os)" = "osx" @test "$(os)" = "any" || test "$(os)" = "osx"
@test "$(bits)" = "32" || test "$(bits)" = "64" @test "$(bits)" = "32" || test "$(bits)" = "64"
@test "$(prefetch)" = "yes" || test "$(prefetch)" = "no" @test "$(prefetch)" = "yes" || test "$(prefetch)" = "no"
@test "$(bsfq)" = "yes" || test "$(bsfq)" = "no" @test "$(bsfq)" = "yes" || test "$(bsfq)" = "no"
@test "$(popcnt)" = "yes" || test "$(popcnt)" = "no" @test "$(popcnt)" = "yes" || test "$(popcnt)" = "no"
@test "$(sse)" = "yes" || test "$(sse)" = "no"
@test "$(comp)" = "gcc" || test "$(comp)" = "icc" || test "$(comp)" = "mingw" || test "$(comp)" = "clang" @test "$(comp)" = "gcc" || test "$(comp)" = "icc" || test "$(comp)" = "mingw" || test "$(comp)" = "clang"
$(EXE): $(OBJS) $(EXE): $(OBJS)
+11 -8
View File
@@ -1,7 +1,7 @@
/* /*
Stockfish, a UCI chess playing engine derived from Glaurung 2.1 Stockfish, a UCI chess playing engine derived from Glaurung 2.1
Copyright (C) 2004-2008 Tord Romstad (Glaurung author) Copyright (C) 2004-2008 Tord Romstad (Glaurung author)
Copyright (C) 2008-2012 Marco Costalba, Joona Kiiski, Tord Romstad Copyright (C) 2008-2013 Marco Costalba, Joona Kiiski, Tord Romstad
Stockfish is free software: you can redistribute it and/or modify Stockfish is free software: you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by it under the terms of the GNU General Public License as published by
@@ -66,7 +66,7 @@ void benchmark(const Position& current, istream& is) {
vector<string> fens; vector<string> fens;
// Assign default values to missing arguments // Assign default values to missing arguments
string ttSize = (is >> token) ? token : "128"; string ttSize = (is >> token) ? token : "32";
string threads = (is >> token) ? token : "1"; string threads = (is >> token) ? token : "1";
string limit = (is >> token) ? token : "12"; string limit = (is >> token) ? token : "12";
string fenFile = (is >> token) ? token : "default"; string fenFile = (is >> token) ? token : "default";
@@ -82,6 +82,9 @@ void benchmark(const Position& current, istream& is) {
else if (limitType == "nodes") else if (limitType == "nodes")
limits.nodes = atoi(limit.c_str()); limits.nodes = atoi(limit.c_str());
else if (limitType == "mate")
limits.mate = atoi(limit.c_str());
else else
limits.depth = atoi(limit.c_str()); limits.depth = atoi(limit.c_str());
@@ -89,7 +92,7 @@ void benchmark(const Position& current, istream& is) {
fens.assign(Defaults, Defaults + 16); fens.assign(Defaults, Defaults + 16);
else if (fenFile == "current") else if (fenFile == "current")
fens.push_back(current.to_fen()); fens.push_back(current.fen());
else else
{ {
@@ -99,7 +102,7 @@ void benchmark(const Position& current, istream& is) {
if (!file.is_open()) if (!file.is_open())
{ {
cerr << "Unable to open file " << fenFile << endl; cerr << "Unable to open file " << fenFile << endl;
exit(EXIT_FAILURE); return;
} }
while (getline(file, fen)) while (getline(file, fen))
@@ -115,7 +118,7 @@ void benchmark(const Position& current, istream& is) {
for (size_t i = 0; i < fens.size(); i++) for (size_t i = 0; i < fens.size(); i++)
{ {
Position pos(fens[i], Options["UCI_Chess960"], Threads.main_thread()); Position pos(fens[i], Options["UCI_Chess960"], Threads.main());
cerr << "\nPosition: " << i + 1 << '/' << fens.size() << endl; cerr << "\nPosition: " << i + 1 << '/' << fens.size() << endl;
@@ -127,9 +130,9 @@ void benchmark(const Position& current, istream& is) {
} }
else else
{ {
Threads.start_searching(pos, limits, vector<Move>(), st); Threads.start_thinking(pos, limits, vector<Move>(), st);
Threads.wait_for_search_finished(); Threads.wait_for_think_finished();
nodes += Search::RootPosition.nodes_searched(); nodes += Search::RootPos.nodes_searched();
} }
} }
+86 -141
View File
@@ -1,7 +1,7 @@
/* /*
Stockfish, a UCI chess playing engine derived from Glaurung 2.1 Stockfish, a UCI chess playing engine derived from Glaurung 2.1
Copyright (C) 2004-2008 Tord Romstad (Glaurung author) Copyright (C) 2004-2008 Tord Romstad (Glaurung author)
Copyright (C) 2008-2012 Marco Costalba, Joona Kiiski, Tord Romstad Copyright (C) 2008-2013 Marco Costalba, Joona Kiiski, Tord Romstad
Stockfish is free software: you can redistribute it and/or modify Stockfish is free software: you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by it under the terms of the GNU General Public License as published by
@@ -18,12 +18,32 @@
*/ */
#include <cassert> #include <cassert>
#include <vector>
#include "bitboard.h" #include "bitboard.h"
#include "types.h" #include "types.h"
namespace { namespace {
// The possible pawns squares are 24, the first 4 files and ranks from 2 to 7
const unsigned IndexMax = 2*24*64*64; // stm * psq * wksq * bksq = 196608
// Each uint32_t stores results of 32 positions, one per bit
uint32_t KPKBitbase[IndexMax / 32];
// A KPK bitbase index is an integer in [0, IndexMax] range
//
// Information is mapped in a way that minimizes number of iterations:
//
// bit 0- 5: white king square (from SQ_A1 to SQ_H8)
// bit 6-11: black king square (from SQ_A1 to SQ_H8)
// bit 12: side to move (WHITE or BLACK)
// bit 13-14: white pawn file (from FILE_A to FILE_D)
// bit 15-17: white pawn RANK_7 - rank (from RANK_7 - RANK_7 to RANK_7 - RANK_2)
unsigned index(Color us, Square bksq, Square wksq, Square psq) {
return wksq + (bksq << 6) + (us << 12) + (file_of(psq) << 13) + ((RANK_7 - rank_of(psq)) << 15);
}
enum Result { enum Result {
INVALID = 0, INVALID = 0,
UNKNOWN = 1, UNKNOWN = 1,
@@ -35,196 +55,121 @@ namespace {
struct KPKPosition { struct KPKPosition {
Result classify_leaf(int idx); operator Result() const { return res; }
Result classify(int idx, Result db[]); Result classify_leaf(unsigned idx);
Result classify(const std::vector<KPKPosition>& db)
{ return us == WHITE ? classify<WHITE>(db) : classify<BLACK>(db); }
private: private:
template<Color Us> Result classify(const Result db[]) const; template<Color Us> Result classify(const std::vector<KPKPosition>& db);
template<Color Us> Bitboard k_attacks() const { Color us;
return Us == WHITE ? StepAttacksBB[W_KING][wksq] : StepAttacksBB[B_KING][bksq]; Square bksq, wksq, psq;
} Result res;
Bitboard p_attacks() const { return StepAttacksBB[W_PAWN][psq]; }
void decode_index(int idx);
Square wksq, bksq, psq;
Color stm;
}; };
// The possible pawns squares are 24, the first 4 files and ranks from 2 to 7 } // namespace
const int IndexMax = 2 * 24 * 64 * 64; // stm * wp_sq * wk_sq * bk_sq = 196608
// Each uint32_t stores results of 32 positions, one per bit
uint32_t KPKBitbase[IndexMax / 32];
int index(Square wksq, Square bksq, Square psq, Color stm);
}
uint32_t Bitbases::probe_kpk(Square wksq, Square wpsq, Square bksq, Color stm) { bool Bitbases::probe_kpk(Square wksq, Square wpsq, Square bksq, Color us) {
int idx = index(wksq, bksq, wpsq, stm); assert(file_of(wpsq) <= FILE_D);
return KPKBitbase[idx / 32] & (1 << (idx & 31));
unsigned idx = index(us, bksq, wksq, wpsq);
return KPKBitbase[idx / 32] & (1 << (idx & 0x1F));
} }
void Bitbases::init_kpk() { void Bitbases::init_kpk() {
Result db[IndexMax]; unsigned idx, repeat = 1;
KPKPosition pos; std::vector<KPKPosition> db(IndexMax);
int idx, bit, repeat = 1;
// Initialize table with known win / draw positions // Initialize db with known win / draw positions
for (idx = 0; idx < IndexMax; idx++) for (idx = 0; idx < IndexMax; idx++)
db[idx] = pos.classify_leaf(idx); db[idx].classify_leaf(idx);
// Iterate until all positions are classified (30 cycles needed) // Iterate through the positions until no more of the unknown positions can be
// changed to either wins or draws (15 cycles needed).
while (repeat) while (repeat)
for (repeat = idx = 0; idx < IndexMax; idx++) for (repeat = idx = 0; idx < IndexMax; idx++)
if (db[idx] == UNKNOWN && (db[idx] = pos.classify(idx, db)) != UNKNOWN) if (db[idx] == UNKNOWN && db[idx].classify(db) != UNKNOWN)
repeat = 1; repeat = 1;
// Map 32 position results into one KPKBitbase[] entry // Map 32 results into one KPKBitbase[] entry
for (idx = 0; idx < IndexMax / 32; idx++) for (idx = 0; idx < IndexMax; idx++)
for (bit = 0; bit < 32; bit++) if (db[idx] == WIN)
if (db[32 * idx + bit] == WIN) KPKBitbase[idx / 32] |= 1 << (idx & 0x1F);
KPKBitbase[idx] |= 1 << bit;
} }
namespace { namespace {
// A KPK bitbase index is an integer in [0, IndexMax] range Result KPKPosition::classify_leaf(unsigned idx) {
//
// Information is mapped in this way
//
// bit 0: side to move (WHITE or BLACK)
// bit 1- 6: black king square (from SQ_A1 to SQ_H8)
// bit 7-12: white king square (from SQ_A1 to SQ_H8)
// bit 13-14: white pawn file (from FILE_A to FILE_D)
// bit 15-17: white pawn rank - 1 (from RANK_2 - 1 to RANK_7 - 1)
int index(Square w, Square b, Square p, Color c) { wksq = Square((idx >> 0) & 0x3F);
bksq = Square((idx >> 6) & 0x3F);
assert(file_of(p) <= FILE_D); us = Color ((idx >> 12) & 0x01);
psq = File ((idx >> 13) & 0x03) | Rank(RANK_7 - (idx >> 15));
return c + (b << 1) + (w << 7) + (file_of(p) << 13) + ((rank_of(p) - 1) << 15);
}
void KPKPosition::decode_index(int idx) {
stm = Color(idx & 1);
bksq = Square((idx >> 1) & 63);
wksq = Square((idx >> 7) & 63);
psq = File((idx >> 13) & 3) | Rank((idx >> 15) + 1);
}
Result KPKPosition::classify_leaf(int idx) {
decode_index(idx);
// Check if two pieces are on the same square or if a king can be captured // Check if two pieces are on the same square or if a king can be captured
if ( wksq == psq || wksq == bksq || bksq == psq if ( wksq == psq || wksq == bksq || bksq == psq
|| (k_attacks<WHITE>() & bksq) || (StepAttacksBB[KING][wksq] & bksq)
|| (stm == WHITE && (p_attacks() & bksq))) || (us == WHITE && (StepAttacksBB[PAWN][psq] & bksq)))
return INVALID; return res = INVALID;
// The position is an immediate win if it is white to move and the white if (us == WHITE)
// pawn can be promoted without getting captured. {
// Immediate win if pawn can be promoted without getting captured
if ( rank_of(psq) == RANK_7 if ( rank_of(psq) == RANK_7
&& stm == WHITE
&& wksq != psq + DELTA_N && wksq != psq + DELTA_N
&& ( square_distance(bksq, psq + DELTA_N) > 1 && ( square_distance(bksq, psq + DELTA_N) > 1
||(k_attacks<WHITE>() & (psq + DELTA_N)))) ||(StepAttacksBB[KING][wksq] & (psq + DELTA_N))))
return WIN; return res = WIN;
}
// Immediate draw if is stalemate or king captures undefended pawn
else if ( !(StepAttacksBB[KING][bksq] & ~(StepAttacksBB[KING][wksq] | StepAttacksBB[PAWN][psq]))
|| (StepAttacksBB[KING][bksq] & psq & ~StepAttacksBB[KING][wksq]))
return res = DRAW;
// Check for known draw positions return res = UNKNOWN;
//
// Case 1: Stalemate
if ( stm == BLACK
&& !(k_attacks<BLACK>() & ~(k_attacks<WHITE>() | p_attacks())))
return DRAW;
// Case 2: King can capture undefended pawn
if ( stm == BLACK
&& (k_attacks<BLACK>() & psq & ~k_attacks<WHITE>()))
return DRAW;
// Case 3: Black king in front of white pawn
if ( bksq == psq + DELTA_N
&& rank_of(psq) < RANK_7)
return DRAW;
// Case 4: White king in front of pawn and black has opposition
if ( stm == WHITE
&& wksq == psq + DELTA_N
&& bksq == wksq + DELTA_N + DELTA_N
&& rank_of(psq) < RANK_5)
return DRAW;
// Case 5: Stalemate with rook pawn
if ( bksq == SQ_A8
&& file_of(psq) == FILE_A)
return DRAW;
// Case 6: White king trapped on the rook file
if ( file_of(wksq) == FILE_A
&& file_of(psq) == FILE_A
&& rank_of(wksq) > rank_of(psq)
&& bksq == wksq + 2)
return DRAW;
return UNKNOWN;
} }
template<Color Us> template<Color Us>
Result KPKPosition::classify(const Result db[]) const { Result KPKPosition::classify(const std::vector<KPKPosition>& db) {
// White to Move: If one move leads to a position classified as RESULT_WIN, // White to Move: If one move leads to a position classified as WIN, the result
// the result of the current position is RESULT_WIN. If all moves lead to // of the current position is WIN. If all moves lead to positions classified
// positions classified as RESULT_DRAW, the current position is classified // as DRAW, the current position is classified DRAW otherwise the current
// RESULT_DRAW otherwise the current position is classified as RESULT_UNKNOWN. // position is classified as UNKNOWN.
// //
// Black to Move: If one move leads to a position classified as RESULT_DRAW, // Black to Move: If one move leads to a position classified as DRAW, the result
// the result of the current position is RESULT_DRAW. If all moves lead to // of the current position is DRAW. If all moves lead to positions classified
// positions classified as RESULT_WIN, the position is classified RESULT_WIN. // as WIN, the position is classified WIN otherwise the current position is
// Otherwise, the current position is classified as RESULT_UNKNOWN. // classified UNKNOWN.
const Color Them = (Us == WHITE ? BLACK : WHITE);
Result r = INVALID; Result r = INVALID;
Bitboard b = k_attacks<Us>(); Bitboard b = StepAttacksBB[KING][Us == WHITE ? wksq : bksq];
while (b) while (b)
{ r |= Us == WHITE ? db[index(Them, bksq, pop_lsb(&b), psq)]
r |= Us == WHITE ? db[index(pop_lsb(&b), bksq, psq, BLACK)] : db[index(Them, pop_lsb(&b), wksq, psq)];
: db[index(wksq, pop_lsb(&b), psq, WHITE)];
if (Us == WHITE && (r & WIN))
return WIN;
if (Us == BLACK && (r & DRAW))
return DRAW;
}
if (Us == WHITE && rank_of(psq) < RANK_7) if (Us == WHITE && rank_of(psq) < RANK_7)
{ {
Square s = psq + DELTA_N; Square s = psq + DELTA_N;
r |= db[index(wksq, bksq, s, BLACK)]; // Single push r |= db[index(BLACK, bksq, wksq, s)]; // Single push
if (rank_of(s) == RANK_3 && s != wksq && s != bksq) if (rank_of(s) == RANK_3 && s != wksq && s != bksq)
r |= db[index(wksq, bksq, s + DELTA_N, BLACK)]; // Double push r |= db[index(BLACK, bksq, wksq, s + DELTA_N)]; // Double push
if (r & WIN)
return WIN;
} }
return r & UNKNOWN ? UNKNOWN : Us == WHITE ? DRAW : WIN; if (Us == WHITE)
return res = r & WIN ? WIN : r & UNKNOWN ? UNKNOWN : DRAW;
else
return res = r & DRAW ? DRAW : r & UNKNOWN ? UNKNOWN : WIN;
} }
Result KPKPosition::classify(int idx, Result db[]) { } // namespace
decode_index(idx);
return stm == WHITE ? classify<WHITE>(db) : classify<BLACK>(db);
}
}
+46 -70
View File
@@ -1,7 +1,7 @@
/* /*
Stockfish, a UCI chess playing engine derived from Glaurung 2.1 Stockfish, a UCI chess playing engine derived from Glaurung 2.1
Copyright (C) 2004-2008 Tord Romstad (Glaurung author) Copyright (C) 2004-2008 Tord Romstad (Glaurung author)
Copyright (C) 2008-2012 Marco Costalba, Joona Kiiski, Tord Romstad Copyright (C) 2008-2013 Marco Costalba, Joona Kiiski, Tord Romstad
Stockfish is free software: you can redistribute it and/or modify Stockfish is free software: you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by it under the terms of the GNU General Public License as published by
@@ -28,78 +28,70 @@
CACHE_LINE_ALIGNMENT CACHE_LINE_ALIGNMENT
Bitboard RMasks[64]; Bitboard RMasks[SQUARE_NB];
Bitboard RMagics[64]; Bitboard RMagics[SQUARE_NB];
Bitboard* RAttacks[64]; Bitboard* RAttacks[SQUARE_NB];
unsigned RShifts[64]; unsigned RShifts[SQUARE_NB];
Bitboard BMasks[64]; Bitboard BMasks[SQUARE_NB];
Bitboard BMagics[64]; Bitboard BMagics[SQUARE_NB];
Bitboard* BAttacks[64]; Bitboard* BAttacks[SQUARE_NB];
unsigned BShifts[64]; unsigned BShifts[SQUARE_NB];
Bitboard SquareBB[64]; Bitboard SquareBB[SQUARE_NB];
Bitboard FileBB[8]; Bitboard FileBB[FILE_NB];
Bitboard RankBB[8]; Bitboard RankBB[RANK_NB];
Bitboard AdjacentFilesBB[8]; Bitboard AdjacentFilesBB[FILE_NB];
Bitboard ThisAndAdjacentFilesBB[8]; Bitboard InFrontBB[COLOR_NB][RANK_NB];
Bitboard InFrontBB[2][8]; Bitboard StepAttacksBB[PIECE_NB][SQUARE_NB];
Bitboard StepAttacksBB[16][64]; Bitboard BetweenBB[SQUARE_NB][SQUARE_NB];
Bitboard BetweenBB[64][64]; Bitboard DistanceRingsBB[SQUARE_NB][8];
Bitboard DistanceRingsBB[64][8]; Bitboard ForwardBB[COLOR_NB][SQUARE_NB];
Bitboard ForwardBB[2][64]; Bitboard PassedPawnMask[COLOR_NB][SQUARE_NB];
Bitboard PassedPawnMask[2][64]; Bitboard PawnAttackSpan[COLOR_NB][SQUARE_NB];
Bitboard AttackSpanMask[2][64]; Bitboard PseudoAttacks[PIECE_TYPE_NB][SQUARE_NB];
Bitboard PseudoAttacks[6][64];
int SquareDistance[64][64]; int SquareDistance[SQUARE_NB][SQUARE_NB];
namespace { namespace {
// De Bruijn sequences. See chessprogramming.wikispaces.com/BitScan // De Bruijn sequences. See chessprogramming.wikispaces.com/BitScan
const uint64_t DeBruijn_64 = 0x218A392CD3D5DBFULL; const uint64_t DeBruijn_64 = 0x3F79D71B4CB0A89ULL;
const uint32_t DeBruijn_32 = 0x783A9B23; const uint32_t DeBruijn_32 = 0x783A9B23;
CACHE_LINE_ALIGNMENT CACHE_LINE_ALIGNMENT
int BSFTable[64];
int MS1BTable[256]; int MS1BTable[256];
Square BSFTable[SQUARE_NB];
Bitboard RTable[0x19000]; // Storage space for rook attacks Bitboard RTable[0x19000]; // Storage space for rook attacks
Bitboard BTable[0x1480]; // Storage space for bishop attacks Bitboard BTable[0x1480]; // Storage space for bishop attacks
uint8_t BitCount8Bit[256];
typedef unsigned (Fn)(Square, Bitboard); typedef unsigned (Fn)(Square, Bitboard);
void init_magics(Bitboard table[], Bitboard* attacks[], Bitboard magics[], void init_magics(Bitboard table[], Bitboard* attacks[], Bitboard magics[],
Bitboard masks[], unsigned shifts[], Square deltas[], Fn index); Bitboard masks[], unsigned shifts[], Square deltas[], Fn index);
FORCE_INLINE unsigned bsf_index(Bitboard b) {
// Matt Taylor's folding for 32 bit systems, extended to 64 bits by Kim Walisch
b ^= (b - 1);
return Is64Bit ? (b * DeBruijn_64) >> 58
: ((unsigned(b) ^ unsigned(b >> 32)) * DeBruijn_32) >> 26;
}
} }
/// lsb()/msb() finds the least/most significant bit in a nonzero bitboard. /// lsb()/msb() finds the least/most significant bit in a nonzero bitboard.
/// pop_lsb() finds and clears the least significant bit in a nonzero bitboard. /// pop_lsb() finds and clears the least significant bit in a nonzero bitboard.
#if !defined(USE_BSFQ) #ifndef USE_BSFQ
Square lsb(Bitboard b) { Square lsb(Bitboard b) { return BSFTable[bsf_index(b)]; }
if (Is64Bit)
return Square(BSFTable[((b & -b) * DeBruijn_64) >> 58]);
b ^= (b - 1);
uint32_t fold = unsigned(b) ^ unsigned(b >> 32);
return Square(BSFTable[(fold * DeBruijn_32) >> 26]);
}
Square pop_lsb(Bitboard* b) { Square pop_lsb(Bitboard* b) {
Bitboard bb = *b; Bitboard bb = *b;
*b = bb & (bb - 1); *b = bb & (bb - 1);
return BSFTable[bsf_index(bb)];
if (Is64Bit)
return Square(BSFTable[((bb & -bb) * DeBruijn_64) >> 58]);
bb ^= (bb - 1);
uint32_t fold = unsigned(bb) ^ unsigned(bb >> 32);
return Square(BSFTable[(fold * DeBruijn_32) >> 26]);
} }
Square msb(Bitboard b) { Square msb(Bitboard b) {
@@ -127,10 +119,10 @@ Square msb(Bitboard b) {
result += 8; result += 8;
} }
return Square(result + MS1BTable[b32]); return (Square)(result + MS1BTable[b32]);
} }
#endif // !defined(USE_BSFQ) #endif // ifndef USE_BSFQ
/// Bitboards::print() prints a bitboard in an easily readable format to the /// Bitboards::print() prints a bitboard in an easily readable format to the
@@ -162,8 +154,8 @@ void Bitboards::init() {
while (k < (2 << i)) while (k < (2 << i))
MS1BTable[k++] = i; MS1BTable[k++] = i;
for (Bitboard b = 0; b < 256; b++) for (int i = 0; i < 64; i++)
BitCount8Bit[b] = (uint8_t)popcount<Max15>(b); BSFTable[bsf_index(1ULL << i)] = Square(i);
for (Square s = SQ_A1; s <= SQ_H8; s++) for (Square s = SQ_A1; s <= SQ_H8; s++)
SquareBB[s] = 1ULL << s; SquareBB[s] = 1ULL << s;
@@ -178,10 +170,7 @@ void Bitboards::init() {
} }
for (File f = FILE_A; f <= FILE_H; f++) for (File f = FILE_A; f <= FILE_H; f++)
{
AdjacentFilesBB[f] = (f > FILE_A ? FileBB[f - 1] : 0) | (f < FILE_H ? FileBB[f + 1] : 0); AdjacentFilesBB[f] = (f > FILE_A ? FileBB[f - 1] : 0) | (f < FILE_H ? FileBB[f + 1] : 0);
ThisAndAdjacentFilesBB[f] = FileBB[f] | AdjacentFilesBB[f];
}
for (Rank r = RANK_1; r < RANK_8; r++) for (Rank r = RANK_1; r < RANK_8; r++)
InFrontBB[WHITE][r] = ~(InFrontBB[BLACK][r + 1] = InFrontBB[BLACK][r] | RankBB[r]); InFrontBB[WHITE][r] = ~(InFrontBB[BLACK][r + 1] = InFrontBB[BLACK][r] | RankBB[r]);
@@ -190,30 +179,17 @@ void Bitboards::init() {
for (Square s = SQ_A1; s <= SQ_H8; s++) for (Square s = SQ_A1; s <= SQ_H8; s++)
{ {
ForwardBB[c][s] = InFrontBB[c][rank_of(s)] & FileBB[file_of(s)]; ForwardBB[c][s] = InFrontBB[c][rank_of(s)] & FileBB[file_of(s)];
PassedPawnMask[c][s] = InFrontBB[c][rank_of(s)] & ThisAndAdjacentFilesBB[file_of(s)]; PawnAttackSpan[c][s] = InFrontBB[c][rank_of(s)] & AdjacentFilesBB[file_of(s)];
AttackSpanMask[c][s] = InFrontBB[c][rank_of(s)] & AdjacentFilesBB[file_of(s)]; PassedPawnMask[c][s] = ForwardBB[c][s] | PawnAttackSpan[c][s];
} }
for (Square s1 = SQ_A1; s1 <= SQ_H8; s1++) for (Square s1 = SQ_A1; s1 <= SQ_H8; s1++)
for (Square s2 = SQ_A1; s2 <= SQ_H8; s2++) for (Square s2 = SQ_A1; s2 <= SQ_H8; s2++)
SquareDistance[s1][s2] = std::max(file_distance(s1, s2), rank_distance(s1, s2));
for (Square s1 = SQ_A1; s1 <= SQ_H8; s1++)
for (int d = 1; d < 8; d++)
for (Square s2 = SQ_A1; s2 <= SQ_H8; s2++)
if (SquareDistance[s1][s2] == d)
DistanceRingsBB[s1][d - 1] |= s2;
for (int i = 0; i < 64; i++)
if (!Is64Bit) // Matt Taylor's folding trick for 32 bit systems
{ {
Bitboard b = 1ULL << i; SquareDistance[s1][s2] = std::max(file_distance(s1, s2), rank_distance(s1, s2));
b ^= b - 1; if (s1 != s2)
b ^= b >> 32; DistanceRingsBB[s1][SquareDistance[s1][s2] - 1] |= s2;
BSFTable[(uint32_t)(b * DeBruijn_32) >> 26] = i;
} }
else
BSFTable[((1ULL << i) * DeBruijn_64) >> 58] = i;
int steps[][9] = { {}, { 7, 9 }, { 17, 15, 10, 6, -6, -10, -15, -17 }, int steps[][9] = { {}, { 7, 9 }, { 17, 15, 10, 6, -6, -10, -15, -17 },
{}, {}, {}, { 9, 7, -7, -9, 8, 1, -1, -8 } }; {}, {}, {}, { 9, 7, -7, -9, 8, 1, -1, -8 } };
@@ -338,9 +314,9 @@ namespace {
// until we find the one that passes the verification test. // until we find the one that passes the verification test.
do { do {
do magics[s] = pick_random(rk, booster); do magics[s] = pick_random(rk, booster);
while (BitCount8Bit[(magics[s] * masks[s]) >> 56] < 6); while (popcount<Max15>((magics[s] * masks[s]) >> 56) < 6);
memset(attacks[s], 0, size * sizeof(Bitboard)); std::memset(attacks[s], 0, size * sizeof(Bitboard));
// A good magic must map every possible occupancy to an index that // A good magic must map every possible occupancy to an index that
// looks up the correct sliding attack in the attacks[s] database. // looks up the correct sliding attack in the attacks[s] database.
+111 -69
View File
@@ -1,7 +1,7 @@
/* /*
Stockfish, a UCI chess playing engine derived from Glaurung 2.1 Stockfish, a UCI chess playing engine derived from Glaurung 2.1
Copyright (C) 2004-2008 Tord Romstad (Glaurung author) Copyright (C) 2004-2008 Tord Romstad (Glaurung author)
Copyright (C) 2008-2012 Marco Costalba, Joona Kiiski, Tord Romstad Copyright (C) 2008-2013 Marco Costalba, Joona Kiiski, Tord Romstad
Stockfish is free software: you can redistribute it and/or modify Stockfish is free software: you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by it under the terms of the GNU General Public License as published by
@@ -18,7 +18,7 @@
along with this program. If not, see <http://www.gnu.org/licenses/>. along with this program. If not, see <http://www.gnu.org/licenses/>.
*/ */
#if !defined(BITBOARD_H_INCLUDED) #ifndef BITBOARD_H_INCLUDED
#define BITBOARD_H_INCLUDED #define BITBOARD_H_INCLUDED
#include "types.h" #include "types.h"
@@ -33,36 +33,56 @@ void print(Bitboard b);
namespace Bitbases { namespace Bitbases {
void init_kpk(); void init_kpk();
uint32_t probe_kpk(Square wksq, Square wpsq, Square bksq, Color stm); bool probe_kpk(Square wksq, Square wpsq, Square bksq, Color us);
} }
const Bitboard FileABB = 0x0101010101010101ULL;
const Bitboard FileBBB = FileABB << 1;
const Bitboard FileCBB = FileABB << 2;
const Bitboard FileDBB = FileABB << 3;
const Bitboard FileEBB = FileABB << 4;
const Bitboard FileFBB = FileABB << 5;
const Bitboard FileGBB = FileABB << 6;
const Bitboard FileHBB = FileABB << 7;
const Bitboard Rank1BB = 0xFF;
const Bitboard Rank2BB = Rank1BB << (8 * 1);
const Bitboard Rank3BB = Rank1BB << (8 * 2);
const Bitboard Rank4BB = Rank1BB << (8 * 3);
const Bitboard Rank5BB = Rank1BB << (8 * 4);
const Bitboard Rank6BB = Rank1BB << (8 * 5);
const Bitboard Rank7BB = Rank1BB << (8 * 6);
const Bitboard Rank8BB = Rank1BB << (8 * 7);
CACHE_LINE_ALIGNMENT CACHE_LINE_ALIGNMENT
extern Bitboard RMasks[64]; extern Bitboard RMasks[SQUARE_NB];
extern Bitboard RMagics[64]; extern Bitboard RMagics[SQUARE_NB];
extern Bitboard* RAttacks[64]; extern Bitboard* RAttacks[SQUARE_NB];
extern unsigned RShifts[64]; extern unsigned RShifts[SQUARE_NB];
extern Bitboard BMasks[64]; extern Bitboard BMasks[SQUARE_NB];
extern Bitboard BMagics[64]; extern Bitboard BMagics[SQUARE_NB];
extern Bitboard* BAttacks[64]; extern Bitboard* BAttacks[SQUARE_NB];
extern unsigned BShifts[64]; extern unsigned BShifts[SQUARE_NB];
extern Bitboard SquareBB[64]; extern Bitboard SquareBB[SQUARE_NB];
extern Bitboard FileBB[8]; extern Bitboard FileBB[FILE_NB];
extern Bitboard RankBB[8]; extern Bitboard RankBB[RANK_NB];
extern Bitboard AdjacentFilesBB[8]; extern Bitboard AdjacentFilesBB[FILE_NB];
extern Bitboard ThisAndAdjacentFilesBB[8]; extern Bitboard InFrontBB[COLOR_NB][RANK_NB];
extern Bitboard InFrontBB[2][8]; extern Bitboard StepAttacksBB[PIECE_NB][SQUARE_NB];
extern Bitboard StepAttacksBB[16][64]; extern Bitboard BetweenBB[SQUARE_NB][SQUARE_NB];
extern Bitboard BetweenBB[64][64]; extern Bitboard DistanceRingsBB[SQUARE_NB][8];
extern Bitboard DistanceRingsBB[64][8]; extern Bitboard ForwardBB[COLOR_NB][SQUARE_NB];
extern Bitboard ForwardBB[2][64]; extern Bitboard PassedPawnMask[COLOR_NB][SQUARE_NB];
extern Bitboard PassedPawnMask[2][64]; extern Bitboard PawnAttackSpan[COLOR_NB][SQUARE_NB];
extern Bitboard AttackSpanMask[2][64]; extern Bitboard PseudoAttacks[PIECE_TYPE_NB][SQUARE_NB];
extern Bitboard PseudoAttacks[6][64];
extern int SquareDistance[SQUARE_NB][SQUARE_NB];
const Bitboard DarkSquares = 0xAA55AA55AA55AA55ULL;
/// Overloads of bitwise operators between a Bitboard and a Square for testing /// Overloads of bitwise operators between a Bitboard and a Square for testing
/// whether a given bit is set in a bitboard, and for setting and clearing bits. /// whether a given bit is set in a bitboard, and for setting and clearing bits.
@@ -87,13 +107,34 @@ inline Bitboard operator^(Bitboard b, Square s) {
return b ^ SquareBB[s]; return b ^ SquareBB[s];
} }
/// more_than_one() returns true if in 'b' there is more than one bit set
inline bool more_than_one(Bitboard b) { inline bool more_than_one(Bitboard b) {
return b & (b - 1); return b & (b - 1);
} }
inline int square_distance(Square s1, Square s2) {
return SquareDistance[s1][s2];
}
inline int file_distance(Square s1, Square s2) {
return abs(file_of(s1) - file_of(s2));
}
inline int rank_distance(Square s1, Square s2) {
return abs(rank_of(s1) - rank_of(s2));
}
/// shift_bb() moves bitboard one step along direction Delta. Mainly for pawns.
template<Square Delta>
inline Bitboard shift_bb(Bitboard b) {
return Delta == DELTA_N ? b << 8 : Delta == DELTA_S ? b >> 8
: Delta == DELTA_NE ? (b & ~FileHBB) << 9 : Delta == DELTA_SE ? (b & ~FileHBB) >> 7
: Delta == DELTA_NW ? (b & ~FileABB) << 7 : Delta == DELTA_SW ? (b & ~FileABB) >> 9
: 0;
}
/// rank_bb() and file_bb() take a file or a square as input and return /// rank_bb() and file_bb() take a file or a square as input and return
/// a bitboard representing all squares on the given file or rank. /// a bitboard representing all squares on the given file or rank.
@@ -115,7 +156,7 @@ inline Bitboard file_bb(Square s) {
} }
/// adjacent_files_bb takes a file as input and returns a bitboard representing /// adjacent_files_bb() takes a file as input and returns a bitboard representing
/// all squares on the adjacent files. /// all squares on the adjacent files.
inline Bitboard adjacent_files_bb(File f) { inline Bitboard adjacent_files_bb(File f) {
@@ -123,30 +164,17 @@ inline Bitboard adjacent_files_bb(File f) {
} }
/// this_and_adjacent_files_bb takes a file as input and returns a bitboard /// in_front_bb() takes a color and a rank as input, and returns a bitboard
/// representing all squares on the given and adjacent files. /// representing all the squares on all ranks in front of the rank, from the
/// given color's point of view. For instance, in_front_bb(BLACK, RANK_3) will
inline Bitboard this_and_adjacent_files_bb(File f) { /// give all squares on ranks 1 and 2.
return ThisAndAdjacentFilesBB[f];
}
/// in_front_bb() takes a color and a rank or square as input, and returns a
/// bitboard representing all the squares on all ranks in front of the rank
/// (or square), from the given color's point of view. For instance,
/// in_front_bb(WHITE, RANK_5) will give all squares on ranks 6, 7 and 8, while
/// in_front_bb(BLACK, SQ_D3) will give all squares on ranks 1 and 2.
inline Bitboard in_front_bb(Color c, Rank r) { inline Bitboard in_front_bb(Color c, Rank r) {
return InFrontBB[c][r]; return InFrontBB[c][r];
} }
inline Bitboard in_front_bb(Color c, Square s) {
return InFrontBB[c][rank_of(s)];
}
/// between_bb() returns a bitboard representing all squares between two squares.
/// between_bb returns a bitboard representing all squares between two squares.
/// For instance, between_bb(SQ_C4, SQ_F7) returns a bitboard with the bits for /// For instance, between_bb(SQ_C4, SQ_F7) returns a bitboard with the bits for
/// square d5 and e6 set. If s1 and s2 are not on the same line, file or diagonal, /// square d5 and e6 set. If s1 and s2 are not on the same line, file or diagonal,
/// 0 is returned. /// 0 is returned.
@@ -156,7 +184,7 @@ inline Bitboard between_bb(Square s1, Square s2) {
} }
/// forward_bb takes a color and a square as input, and returns a bitboard /// forward_bb() takes a color and a square as input, and returns a bitboard
/// representing all squares along the line in front of the square, from the /// representing all squares along the line in front of the square, from the
/// point of view of the given color. Definition of the table is: /// point of view of the given color. Definition of the table is:
/// ForwardBB[c][s] = in_front_bb(c, s) & file_bb(s) /// ForwardBB[c][s] = in_front_bb(c, s) & file_bb(s)
@@ -166,27 +194,35 @@ inline Bitboard forward_bb(Color c, Square s) {
} }
/// passed_pawn_mask takes a color and a square as input, and returns a /// pawn_attack_span() takes a color and a square as input, and returns a bitboard
/// representing all squares that can be attacked by a pawn of the given color
/// when it moves along its file starting from the given square. Definition is:
/// PawnAttackSpan[c][s] = in_front_bb(c, s) & adjacent_files_bb(s);
inline Bitboard pawn_attack_span(Color c, Square s) {
return PawnAttackSpan[c][s];
}
/// passed_pawn_mask() takes a color and a square as input, and returns a
/// bitboard mask which can be used to test if a pawn of the given color on /// bitboard mask which can be used to test if a pawn of the given color on
/// the given square is a passed pawn. Definition of the table is: /// the given square is a passed pawn. Definition of the table is:
/// PassedPawnMask[c][s] = in_front_bb(c, s) & this_and_adjacent_files_bb(s) /// PassedPawnMask[c][s] = pawn_attack_span(c, s) | forward_bb(c, s)
inline Bitboard passed_pawn_mask(Color c, Square s) { inline Bitboard passed_pawn_mask(Color c, Square s) {
return PassedPawnMask[c][s]; return PassedPawnMask[c][s];
} }
/// attack_span_mask takes a color and a square as input, and returns a bitboard /// squares_of_color() returns a bitboard representing all squares with the same
/// representing all squares that can be attacked by a pawn of the given color /// color of the given square.
/// when it moves along its file starting from the given square. Definition is:
/// AttackSpanMask[c][s] = in_front_bb(c, s) & adjacent_files_bb(s);
inline Bitboard attack_span_mask(Color c, Square s) { inline Bitboard squares_of_color(Square s) {
return AttackSpanMask[c][s]; return DarkSquares & s ? DarkSquares : ~DarkSquares;
} }
/// squares_aligned returns true if the squares s1, s2 and s3 are aligned /// squares_aligned() returns true if the squares s1, s2 and s3 are aligned
/// either on a straight or on a diagonal line. /// either on a straight or on a diagonal line.
inline bool squares_aligned(Square s1, Square s2, Square s3) { inline bool squares_aligned(Square s1, Square s2, Square s3) {
@@ -195,15 +231,6 @@ inline bool squares_aligned(Square s1, Square s2, Square s3) {
} }
/// same_color_squares() returns a bitboard representing all squares with
/// the same color of the given square.
inline Bitboard same_color_squares(Square s) {
return Bitboard(0xAA55AA55AA55AA55ULL) & s ? 0xAA55AA55AA55AA55ULL
: ~0xAA55AA55AA55AA55ULL;
}
/// Functions for computing sliding attack bitboards. Function attacks_bb() takes /// Functions for computing sliding attack bitboards. Function attacks_bb() takes
/// a square and a bitboard of occupied squares as input, and returns a bitboard /// a square and a bitboard of occupied squares as input, and returns a bitboard
/// representing all squares attacked by Pt (bishop or rook) on the given square. /// representing all squares attacked by Pt (bishop or rook) on the given square.
@@ -231,7 +258,7 @@ inline Bitboard attacks_bb(Square s, Bitboard occ) {
/// lsb()/msb() finds the least/most significant bit in a nonzero bitboard. /// lsb()/msb() finds the least/most significant bit in a nonzero bitboard.
/// pop_lsb() finds and clears the least significant bit in a nonzero bitboard. /// pop_lsb() finds and clears the least significant bit in a nonzero bitboard.
#if defined(USE_BSFQ) #ifdef USE_BSFQ
# if defined(_MSC_VER) && !defined(__INTEL_COMPILER) # if defined(_MSC_VER) && !defined(__INTEL_COMPILER)
@@ -247,6 +274,21 @@ FORCE_INLINE Square msb(Bitboard b) {
return (Square) index; return (Square) index;
} }
# elif defined(__arm__)
FORCE_INLINE int lsb32(uint32_t v) {
__asm__("rbit %0, %1" : "=r"(v) : "r"(v));
return __builtin_clz(v);
}
FORCE_INLINE Square msb(Bitboard b) {
return (Square) (63 - __builtin_clzll(b));
}
FORCE_INLINE Square lsb(Bitboard b) {
return (Square) (uint32_t(b) ? lsb32(uint32_t(b)) : 32 + lsb32(uint32_t(b >> 32)));
}
# else # else
FORCE_INLINE Square lsb(Bitboard b) { // Assembly code by Heinz van Saanen FORCE_INLINE Square lsb(Bitboard b) { // Assembly code by Heinz van Saanen
@@ -265,11 +307,11 @@ FORCE_INLINE Square msb(Bitboard b) {
FORCE_INLINE Square pop_lsb(Bitboard* b) { FORCE_INLINE Square pop_lsb(Bitboard* b) {
const Square s = lsb(*b); const Square s = lsb(*b);
*b &= ~(1ULL << s); *b &= *b - 1;
return s; return s;
} }
#else // if !defined(USE_BSFQ) #else // if defined(USE_BSFQ)
extern Square msb(Bitboard b); extern Square msb(Bitboard b);
extern Square lsb(Bitboard b); extern Square lsb(Bitboard b);
@@ -277,4 +319,4 @@ extern Square pop_lsb(Bitboard* b);
#endif #endif
#endif // !defined(BITBOARD_H_INCLUDED) #endif // #ifndef BITBOARD_H_INCLUDED
+18 -25
View File
@@ -1,7 +1,7 @@
/* /*
Stockfish, a UCI chess playing engine derived from Glaurung 2.1 Stockfish, a UCI chess playing engine derived from Glaurung 2.1
Copyright (C) 2004-2008 Tord Romstad (Glaurung author) Copyright (C) 2004-2008 Tord Romstad (Glaurung author)
Copyright (C) 2008-2012 Marco Costalba, Joona Kiiski, Tord Romstad Copyright (C) 2008-2013 Marco Costalba, Joona Kiiski, Tord Romstad
Stockfish is free software: you can redistribute it and/or modify Stockfish is free software: you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by it under the terms of the GNU General Public License as published by
@@ -18,7 +18,7 @@
along with this program. If not, see <http://www.gnu.org/licenses/>. along with this program. If not, see <http://www.gnu.org/licenses/>.
*/ */
#if !defined(BITCOUNT_H_INCLUDED) #ifndef BITCOUNT_H_INCLUDED
#define BITCOUNT_H_INCLUDED #define BITCOUNT_H_INCLUDED
#include <cassert> #include <cassert>
@@ -33,8 +33,8 @@ enum BitCountType {
}; };
/// Determine at compile time the best popcount<> specialization according if /// Determine at compile time the best popcount<> specialization according if
/// platform is 32 or 64 bits, to the maximum number of nonzero bits to count or /// platform is 32 or 64 bits, to the maximum number of nonzero bits to count
/// use hardware popcnt instruction when available. /// and if hardware popcnt instruction is available.
const BitCountType Full = HasPopCnt ? CNT_HW_POPCNT : Is64Bit ? CNT_64 : CNT_32; const BitCountType Full = HasPopCnt ? CNT_HW_POPCNT : Is64Bit ? CNT_64 : CNT_32;
const BitCountType Max15 = HasPopCnt ? CNT_HW_POPCNT : Is64Bit ? CNT_64_MAX15 : CNT_32_MAX15; const BitCountType Max15 = HasPopCnt ? CNT_HW_POPCNT : Is64Bit ? CNT_64_MAX15 : CNT_32_MAX15;
@@ -44,19 +44,17 @@ template<BitCountType> inline int popcount(Bitboard);
template<> template<>
inline int popcount<CNT_64>(Bitboard b) { inline int popcount<CNT_64>(Bitboard b) {
b -= ((b>>1) & 0x5555555555555555ULL); b -= (b >> 1) & 0x5555555555555555ULL;
b = ((b>>2) & 0x3333333333333333ULL) + (b & 0x3333333333333333ULL); b = ((b >> 2) & 0x3333333333333333ULL) + (b & 0x3333333333333333ULL);
b = ((b>>4) + b) & 0x0F0F0F0F0F0F0F0FULL; b = ((b >> 4) + b) & 0x0F0F0F0F0F0F0F0FULL;
b *= 0x0101010101010101ULL; return (b * 0x0101010101010101ULL) >> 56;
return int(b >> 56);
} }
template<> template<>
inline int popcount<CNT_64_MAX15>(Bitboard b) { inline int popcount<CNT_64_MAX15>(Bitboard b) {
b -= (b>>1) & 0x5555555555555555ULL; b -= (b >> 1) & 0x5555555555555555ULL;
b = ((b>>2) & 0x3333333333333333ULL) + (b & 0x3333333333333333ULL); b = ((b >> 2) & 0x3333333333333333ULL) + (b & 0x3333333333333333ULL);
b *= 0x1111111111111111ULL; return (b * 0x1111111111111111ULL) >> 60;
return int(b >> 60);
} }
template<> template<>
@@ -66,10 +64,8 @@ inline int popcount<CNT_32>(Bitboard b) {
w -= (w >> 1) & 0x55555555; w -= (w >> 1) & 0x55555555;
v = ((v >> 2) & 0x33333333) + (v & 0x33333333); // 0-4 in 4 bits v = ((v >> 2) & 0x33333333) + (v & 0x33333333); // 0-4 in 4 bits
w = ((w >> 2) & 0x33333333) + (w & 0x33333333); w = ((w >> 2) & 0x33333333) + (w & 0x33333333);
v = ((v >> 4) + v) & 0x0F0F0F0F; // 0-8 in 8 bits v = ((v >> 4) + v + (w >> 4) + w) & 0x0F0F0F0F;
v += (((w >> 4) + w) & 0x0F0F0F0F); // 0-16 in 8 bits return (v * 0x01010101) >> 24;
v *= 0x01010101; // mul is fast on amd procs
return int(v >> 24);
} }
template<> template<>
@@ -79,15 +75,13 @@ inline int popcount<CNT_32_MAX15>(Bitboard b) {
w -= (w >> 1) & 0x55555555; w -= (w >> 1) & 0x55555555;
v = ((v >> 2) & 0x33333333) + (v & 0x33333333); // 0-4 in 4 bits v = ((v >> 2) & 0x33333333) + (v & 0x33333333); // 0-4 in 4 bits
w = ((w >> 2) & 0x33333333) + (w & 0x33333333); w = ((w >> 2) & 0x33333333) + (w & 0x33333333);
v += w; // 0-8 in 4 bits return ((v + w) * 0x11111111) >> 28;
v *= 0x11111111;
return int(v >> 28);
} }
template<> template<>
inline int popcount<CNT_HW_POPCNT>(Bitboard b) { inline int popcount<CNT_HW_POPCNT>(Bitboard b) {
#if !defined(USE_POPCNT) #ifndef USE_POPCNT
assert(false); assert(false);
return b != 0; // Avoid 'b not used' warning return b != 0; // Avoid 'b not used' warning
@@ -102,11 +96,10 @@ inline int popcount<CNT_HW_POPCNT>(Bitboard b) {
#else #else
unsigned long ret; __asm__("popcnt %1, %0" : "=r" (b) : "r" (b));
__asm__("popcnt %1, %0" : "=r" (ret) : "r" (b)); return b;
return ret;
#endif #endif
} }
#endif // !defined(BITCOUNT_H_INCLUDED) #endif // #ifndef BITCOUNT_H_INCLUDED
+56 -49
View File
@@ -1,7 +1,7 @@
/* /*
Stockfish, a UCI chess playing engine derived from Glaurung 2.1 Stockfish, a UCI chess playing engine derived from Glaurung 2.1
Copyright (C) 2004-2008 Tord Romstad (Glaurung author) Copyright (C) 2004-2008 Tord Romstad (Glaurung author)
Copyright (C) 2008-2012 Marco Costalba, Joona Kiiski, Tord Romstad Copyright (C) 2008-2013 Marco Costalba, Joona Kiiski, Tord Romstad
Stockfish is free software: you can redistribute it and/or modify Stockfish is free software: you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by it under the terms of the GNU General Public License as published by
@@ -35,8 +35,26 @@ using namespace std;
namespace { namespace {
// A Polyglot book is a series of "entries" of 16 bytes. All integers are
// stored in big-endian format, with highest byte first (regardless of size).
// The entries are ordered according to the key in ascending order.
struct Entry {
uint64_t key;
uint16_t move;
uint16_t count;
uint32_t learn;
};
// Random numbers from PolyGlot, used to compute book hash keys // Random numbers from PolyGlot, used to compute book hash keys
const Key PolyGlotRandoms[781] = { const union {
Key PolyGlotRandoms[781];
struct {
Key psq[12][64]; // [piece][square]
Key castle[4]; // [castle right]
Key enpassant[8]; // [file]
Key turn;
} Zobrist;
} PG = {{
0x9D39247E33776D41ULL, 0x2AF7398005AAA5C7ULL, 0x44DB015024623547ULL, 0x9D39247E33776D41ULL, 0x2AF7398005AAA5C7ULL, 0x44DB015024623547ULL,
0x9C15F73E62A76AE2ULL, 0x75834465489C0C89ULL, 0x3290AC3A203001BFULL, 0x9C15F73E62A76AE2ULL, 0x75834465489C0C89ULL, 0x3290AC3A203001BFULL,
0x0FBBAD1F61042279ULL, 0xE83A908FF2FB60CAULL, 0x0D7E765D58755C10ULL, 0x0FBBAD1F61042279ULL, 0xE83A908FF2FB60CAULL, 0x0D7E765D58755C10ULL,
@@ -298,60 +316,49 @@ namespace {
0x003A93D8B2806962ULL, 0x1C99DED33CB890A1ULL, 0xCF3145DE0ADD4289ULL, 0x003A93D8B2806962ULL, 0x1C99DED33CB890A1ULL, 0xCF3145DE0ADD4289ULL,
0xD0E4427A5514FB72ULL, 0x77C621CC9FB3A483ULL, 0x67A34DAC4356550BULL, 0xD0E4427A5514FB72ULL, 0x77C621CC9FB3A483ULL, 0x67A34DAC4356550BULL,
0xF8D626AAAF278509ULL 0xF8D626AAAF278509ULL
}; }};
// Offsets to the PolyGlotRandoms[] array of zobrist keys // polyglot_key() returns the PolyGlot hash key of the given position
const Key* ZobPiece = PolyGlotRandoms; Key polyglot_key(const Position& pos) {
const Key* ZobCastle = ZobPiece + 12 * 64; // Pieces * squares
const Key* ZobEnPassant = ZobCastle + 4; // Castle flags
const Key* ZobTurn = ZobEnPassant + 8; // Number of files
// book_key() returns the PolyGlot hash key of the given position Key key = 0;
uint64_t book_key(const Position& pos) {
uint64_t key = 0;
Bitboard b = pos.pieces(); Bitboard b = pos.pieces();
while (b) while (b)
{ {
// In PolyGlotRandoms[] pieces are stored in the following sequence:
// BP = 0, WP = 1, BN = 2, WN = 3, ... BK = 10, WK = 11
Square s = pop_lsb(&b); Square s = pop_lsb(&b);
Piece p = pos.piece_on(s); Piece p = pos.piece_on(s);
int pieceOfs = 2 * (type_of(p) - 1) + (color_of(p) == WHITE);
key ^= ZobPiece[64 * pieceOfs + s]; // PolyGlot pieces are: BP = 0, WP = 1, BN = 2, ... BK = 10, WK = 11
key ^= PG.Zobrist.psq[2 * (type_of(p) - 1) + (color_of(p) == WHITE)][s];
} }
b = pos.can_castle(ALL_CASTLES); b = pos.can_castle(ALL_CASTLES);
while (b) while (b)
key ^= ZobCastle[pop_lsb(&b)]; key ^= PG.Zobrist.castle[pop_lsb(&b)];
if (pos.ep_square() != SQ_NONE) if (pos.ep_square() != SQ_NONE)
key ^= ZobEnPassant[file_of(pos.ep_square())]; key ^= PG.Zobrist.enpassant[file_of(pos.ep_square())];
if (pos.side_to_move() == WHITE) if (pos.side_to_move() == WHITE)
key ^= ZobTurn[0]; key ^= PG.Zobrist.turn;
return key; return key;
} }
} // namespace } // namespace
Book::Book() { PolyglotBook::PolyglotBook() : rkiss(Time::now() % 10000) {}
for (int i = Time::now() % 10000; i > 0; i--) PolyglotBook::~PolyglotBook() { if (is_open()) close(); }
RKiss.rand<unsigned>(); // Make random number generation less deterministic
}
Book::~Book() { if (is_open()) close(); }
/// Book::operator>>() reads sizeof(T) chars from the file's binary byte stream /// operator>>() reads sizeof(T) chars from the file's binary byte stream and
/// and converts them in a number of type T. A Polyglot book stores numbers in /// converts them in a number of type T. A Polyglot book stores numbers in
/// big-endian format. /// big-endian format.
template<typename T> Book& Book::operator>>(T& n) { template<typename T> PolyglotBook& PolyglotBook::operator>>(T& n) {
n = 0; n = 0;
for (size_t i = 0; i < sizeof(T); i++) for (size_t i = 0; i < sizeof(T); i++)
@@ -360,15 +367,15 @@ template<typename T> Book& Book::operator>>(T& n) {
return *this; return *this;
} }
template<> Book& Book::operator>>(BookEntry& e) { template<> PolyglotBook& PolyglotBook::operator>>(Entry& e) {
return *this >> e.key >> e.move >> e.count >> e.learn; return *this >> e.key >> e.move >> e.count >> e.learn;
} }
/// Book::open() tries to open a book file with the given name after closing /// open() tries to open a book file with the given name after closing any
/// any exsisting one. /// exsisting one.
bool Book::open(const char* fName) { bool PolyglotBook::open(const char* fName) {
if (is_open()) // Cannot close an already closed file if (is_open()) // Cannot close an already closed file
close(); close();
@@ -381,22 +388,22 @@ bool Book::open(const char* fName) {
} }
/// Book::probe() tries to find a book move for the given position. If no move /// probe() tries to find a book move for the given position. If no move is
/// is found returns MOVE_NONE. If pickBest is true returns always the highest /// found returns MOVE_NONE. If pickBest is true returns always the highest
/// rated move, otherwise randomly chooses one, based on the move score. /// rated move, otherwise randomly chooses one, based on the move score.
Move Book::probe(const Position& pos, const string& fName, bool pickBest) { Move PolyglotBook::probe(const Position& pos, const string& fName, bool pickBest) {
if (fileName != fName && !open(fName.c_str())) if (fileName != fName && !open(fName.c_str()))
return MOVE_NONE; return MOVE_NONE;
BookEntry e; Entry e;
uint16_t best = 0; uint16_t best = 0;
unsigned sum = 0; unsigned sum = 0;
Move move = MOVE_NONE; Move move = MOVE_NONE;
uint64_t key = book_key(pos); Key key = polyglot_key(pos);
seekg(find_first(key) * sizeof(BookEntry), ios_base::beg); seekg(find_first(key) * sizeof(Entry), ios_base::beg);
while (*this >> e, e.key == key && good()) while (*this >> e, e.key == key && good())
{ {
@@ -406,7 +413,7 @@ Move Book::probe(const Position& pos, const string& fName, bool pickBest) {
// Choose book move according to its score. If a move has a very // Choose book move according to its score. If a move has a very
// high score it has higher probability to be choosen than a move // high score it has higher probability to be choosen than a move
// with lower score. Note that first entry is always chosen. // with lower score. Note that first entry is always chosen.
if ( (sum && RKiss.rand<unsigned>() % sum < e.count) if ( (sum && rkiss.rand<unsigned>() % sum < e.count)
|| (pickBest && e.count == best)) || (pickBest && e.count == best))
move = Move(e.move); move = Move(e.move);
} }
@@ -429,24 +436,24 @@ Move Book::probe(const Position& pos, const string& fName, bool pickBest) {
move = make<PROMOTION>(from_sq(move), to_sq(move), PieceType(pt + 1)); move = make<PROMOTION>(from_sq(move), to_sq(move), PieceType(pt + 1));
// Add 'special move' flags and verify it is legal // Add 'special move' flags and verify it is legal
for (MoveList<LEGAL> ml(pos); !ml.end(); ++ml) for (MoveList<LEGAL> it(pos); *it; ++it)
if (move == (ml.move() & 0x3FFF)) if (move == (*it ^ type_of(*it)))
return ml.move(); return *it;
return MOVE_NONE; return MOVE_NONE;
} }
/// Book::find_first() takes a book key as input, and does a binary search /// find_first() takes a book key as input, and does a binary search through
/// through the book file for the given key. Returns the index of the leftmost /// the book file for the given key. Returns the index of the leftmost book
/// book entry with the same key as the input. /// entry with the same key as the input.
size_t Book::find_first(uint64_t key) { size_t PolyglotBook::find_first(Key key) {
seekg(0, ios::end); // Move pointer to end, so tellg() gets file's size seekg(0, ios::end); // Move pointer to end, so tellg() gets file's size
size_t low = 0, mid, high = (size_t)tellg() / sizeof(BookEntry) - 1; size_t low = 0, mid, high = (size_t)tellg() / sizeof(Entry) - 1;
BookEntry e; Entry e;
assert(low <= high); assert(low <= high);
@@ -456,7 +463,7 @@ size_t Book::find_first(uint64_t key) {
assert(mid >= low && mid < high); assert(mid >= low && mid < high);
seekg(mid * sizeof(BookEntry), ios_base::beg); seekg(mid * sizeof(Entry), ios_base::beg);
*this >> e; *this >> e;
if (key <= e.key) if (key <= e.key)
+9 -21
View File
@@ -1,7 +1,7 @@
/* /*
Stockfish, a UCI chess playing engine derived from Glaurung 2.1 Stockfish, a UCI chess playing engine derived from Glaurung 2.1
Copyright (C) 2004-2008 Tord Romstad (Glaurung author) Copyright (C) 2004-2008 Tord Romstad (Glaurung author)
Copyright (C) 2008-2012 Marco Costalba, Joona Kiiski, Tord Romstad Copyright (C) 2008-2013 Marco Costalba, Joona Kiiski, Tord Romstad
Stockfish is free software: you can redistribute it and/or modify Stockfish is free software: you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by it under the terms of the GNU General Public License as published by
@@ -17,7 +17,7 @@
along with this program. If not, see <http://www.gnu.org/licenses/>. along with this program. If not, see <http://www.gnu.org/licenses/>.
*/ */
#if !defined(BOOK_H_INCLUDED) #ifndef BOOK_H_INCLUDED
#define BOOK_H_INCLUDED #define BOOK_H_INCLUDED
#include <fstream> #include <fstream>
@@ -26,32 +26,20 @@
#include "position.h" #include "position.h"
#include "rkiss.h" #include "rkiss.h"
class PolyglotBook : private std::ifstream {
/// A Polyglot book is a series of "entries" of 16 bytes. All integers are
/// stored highest byte first (regardless of size). The entries are ordered
/// according to key. Lowest key first.
struct BookEntry {
uint64_t key;
uint16_t move;
uint16_t count;
uint32_t learn;
};
class Book : private std::ifstream {
public: public:
Book(); PolyglotBook();
~Book(); ~PolyglotBook();
Move probe(const Position& pos, const std::string& fName, bool pickBest); Move probe(const Position& pos, const std::string& fName, bool pickBest);
private: private:
template<typename T> Book& operator>>(T& n); template<typename T> PolyglotBook& operator>>(T& n);
bool open(const char* fName); bool open(const char* fName);
size_t find_first(uint64_t key); size_t find_first(Key key);
RKISS RKiss; RKISS rkiss;
std::string fileName; std::string fileName;
}; };
#endif // !defined(BOOK_H_INCLUDED) #endif // #ifndef BOOK_H_INCLUDED
+187 -117
View File
@@ -1,7 +1,7 @@
/* /*
Stockfish, a UCI chess playing engine derived from Glaurung 2.1 Stockfish, a UCI chess playing engine derived from Glaurung 2.1
Copyright (C) 2004-2008 Tord Romstad (Glaurung author) Copyright (C) 2004-2008 Tord Romstad (Glaurung author)
Copyright (C) 2008-2012 Marco Costalba, Joona Kiiski, Tord Romstad Copyright (C) 2008-2013 Marco Costalba, Joona Kiiski, Tord Romstad
Stockfish is free software: you can redistribute it and/or modify Stockfish is free software: you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by it under the terms of the GNU General Public License as published by
@@ -31,7 +31,7 @@ namespace {
// Table used to drive the defending king towards the edge of the board // Table used to drive the defending king towards the edge of the board
// in KX vs K and KQ vs KR endgames. // in KX vs K and KQ vs KR endgames.
const int MateTable[64] = { const int MateTable[SQUARE_NB] = {
100, 90, 80, 70, 70, 80, 90, 100, 100, 90, 80, 70, 70, 80, 90, 100,
90, 70, 60, 50, 50, 60, 70, 90, 90, 70, 60, 50, 50, 60, 70, 90,
80, 60, 40, 30, 30, 40, 60, 80, 80, 60, 40, 30, 30, 40, 60, 80,
@@ -44,7 +44,7 @@ namespace {
// Table used to drive the defending king towards a corner square of the // Table used to drive the defending king towards a corner square of the
// right color in KBN vs K endgames. // right color in KBN vs K endgames.
const int KBNKMateTable[64] = { const int KBNKMateTable[SQUARE_NB] = {
200, 190, 180, 170, 160, 150, 140, 130, 200, 190, 180, 170, 160, 150, 140, 130,
190, 180, 170, 160, 150, 140, 130, 140, 190, 180, 170, 160, 150, 140, 130, 140,
180, 170, 155, 140, 140, 125, 140, 150, 180, 170, 155, 140, 140, 125, 140, 150,
@@ -89,16 +89,21 @@ namespace {
Endgames::Endgames() { Endgames::Endgames() {
add<KK>("KK");
add<KPK>("KPK"); add<KPK>("KPK");
add<KBK>("KBK");
add<KNK>("KNK");
add<KNNK>("KNNK"); add<KNNK>("KNNK");
add<KBNK>("KBNK"); add<KBNK>("KBNK");
add<KRKP>("KRKP"); add<KRKP>("KRKP");
add<KRKB>("KRKB"); add<KRKB>("KRKB");
add<KRKN>("KRKN"); add<KRKN>("KRKN");
add<KQKP>("KQKP");
add<KQKR>("KQKR"); add<KQKR>("KQKR");
add<KBBKN>("KBBKN"); add<KBBKN>("KBBKN");
add<KNPK>("KNPK"); add<KNPK>("KNPK");
add<KNPKB>("KNPKB");
add<KRPKR>("KRPKR"); add<KRPKR>("KRPKR");
add<KBPKB>("KBPKB"); add<KBPKB>("KBPKB");
add<KBPKN>("KBPKN"); add<KBPKN>("KBPKN");
@@ -128,28 +133,25 @@ template<>
Value Endgame<KXK>::operator()(const Position& pos) const { Value Endgame<KXK>::operator()(const Position& pos) const {
assert(pos.non_pawn_material(weakerSide) == VALUE_ZERO); assert(pos.non_pawn_material(weakerSide) == VALUE_ZERO);
assert(pos.piece_count(weakerSide, PAWN) == VALUE_ZERO); assert(!pos.count<PAWN>(weakerSide));
assert(!pos.checkers()); // Eval is never called when in check
// Stalemate detection with lone king // Stalemate detection with lone king
if ( pos.side_to_move() == weakerSide if (pos.side_to_move() == weakerSide && !MoveList<LEGAL>(pos).size())
&& !pos.in_check()
&& !MoveList<LEGAL>(pos).size()) {
return VALUE_DRAW; return VALUE_DRAW;
}
Square winnerKSq = pos.king_square(strongerSide); Square winnerKSq = pos.king_square(strongerSide);
Square loserKSq = pos.king_square(weakerSide); Square loserKSq = pos.king_square(weakerSide);
Value result = pos.non_pawn_material(strongerSide) Value result = pos.non_pawn_material(strongerSide)
+ pos.piece_count(strongerSide, PAWN) * PawnValueEg + pos.count<PAWN>(strongerSide) * PawnValueEg
+ MateTable[loserKSq] + MateTable[loserKSq]
+ DistanceBonus[square_distance(winnerKSq, loserKSq)]; + DistanceBonus[square_distance(winnerKSq, loserKSq)];
if ( pos.piece_count(strongerSide, QUEEN) if ( pos.count<QUEEN>(strongerSide)
|| pos.piece_count(strongerSide, ROOK) || pos.count<ROOK>(strongerSide)
|| pos.bishop_pair(strongerSide)) { || pos.bishop_pair(strongerSide))
result += VALUE_KNOWN_WIN; result += VALUE_KNOWN_WIN;
}
return strongerSide == pos.side_to_move() ? result : -result; return strongerSide == pos.side_to_move() ? result : -result;
} }
@@ -160,21 +162,21 @@ Value Endgame<KXK>::operator()(const Position& pos) const {
template<> template<>
Value Endgame<KBNK>::operator()(const Position& pos) const { Value Endgame<KBNK>::operator()(const Position& pos) const {
assert(pos.non_pawn_material(weakerSide) == VALUE_ZERO);
assert(pos.piece_count(weakerSide, PAWN) == VALUE_ZERO);
assert(pos.non_pawn_material(strongerSide) == KnightValueMg + BishopValueMg); assert(pos.non_pawn_material(strongerSide) == KnightValueMg + BishopValueMg);
assert(pos.piece_count(strongerSide, BISHOP) == 1); assert(pos.non_pawn_material(weakerSide) == VALUE_ZERO);
assert(pos.piece_count(strongerSide, KNIGHT) == 1); assert(pos.count<BISHOP>(strongerSide) == 1);
assert(pos.piece_count(strongerSide, PAWN) == 0); assert(pos.count<KNIGHT>(strongerSide) == 1);
assert(pos.count< PAWN>(strongerSide) == 0);
assert(pos.count< PAWN>(weakerSide ) == 0);
Square winnerKSq = pos.king_square(strongerSide); Square winnerKSq = pos.king_square(strongerSide);
Square loserKSq = pos.king_square(weakerSide); Square loserKSq = pos.king_square(weakerSide);
Square bishopSquare = pos.piece_list(strongerSide, BISHOP)[0]; Square bishopSq = pos.list<BISHOP>(strongerSide)[0];
// kbnk_mate_table() tries to drive toward corners A1 or H8, // kbnk_mate_table() tries to drive toward corners A1 or H8,
// if we have a bishop that cannot reach the above squares we // if we have a bishop that cannot reach the above squares we
// mirror the kings so to drive enemy toward corners A8 or H1. // mirror the kings so to drive enemy toward corners A8 or H1.
if (opposite_colors(bishopSquare, SQ_A1)) if (opposite_colors(bishopSq, SQ_A1))
{ {
winnerKSq = mirror(winnerKSq); winnerKSq = mirror(winnerKSq);
loserKSq = mirror(loserKSq); loserKSq = mirror(loserKSq);
@@ -194,25 +196,25 @@ Value Endgame<KPK>::operator()(const Position& pos) const {
assert(pos.non_pawn_material(strongerSide) == VALUE_ZERO); assert(pos.non_pawn_material(strongerSide) == VALUE_ZERO);
assert(pos.non_pawn_material(weakerSide) == VALUE_ZERO); assert(pos.non_pawn_material(weakerSide) == VALUE_ZERO);
assert(pos.piece_count(strongerSide, PAWN) == 1); assert(pos.count<PAWN>(strongerSide) == 1);
assert(pos.piece_count(weakerSide, PAWN) == 0); assert(pos.count<PAWN>(weakerSide ) == 0);
Square wksq, bksq, wpsq; Square wksq, bksq, wpsq;
Color stm; Color us;
if (strongerSide == WHITE) if (strongerSide == WHITE)
{ {
wksq = pos.king_square(WHITE); wksq = pos.king_square(WHITE);
bksq = pos.king_square(BLACK); bksq = pos.king_square(BLACK);
wpsq = pos.piece_list(WHITE, PAWN)[0]; wpsq = pos.list<PAWN>(WHITE)[0];
stm = pos.side_to_move(); us = pos.side_to_move();
} }
else else
{ {
wksq = ~pos.king_square(BLACK); wksq = ~pos.king_square(BLACK);
bksq = ~pos.king_square(WHITE); bksq = ~pos.king_square(WHITE);
wpsq = ~pos.piece_list(BLACK, PAWN)[0]; wpsq = ~pos.list<PAWN>(BLACK)[0];
stm = ~pos.side_to_move(); us = ~pos.side_to_move();
} }
if (file_of(wpsq) >= FILE_E) if (file_of(wpsq) >= FILE_E)
@@ -222,7 +224,7 @@ Value Endgame<KPK>::operator()(const Position& pos) const {
wpsq = mirror(wpsq); wpsq = mirror(wpsq);
} }
if (!Bitbases::probe_kpk(wksq, wpsq, bksq, stm)) if (!Bitbases::probe_kpk(wksq, wpsq, bksq, us))
return VALUE_DRAW; return VALUE_DRAW;
Value result = VALUE_KNOWN_WIN + PawnValueEg + Value(rank_of(wpsq)); Value result = VALUE_KNOWN_WIN + PawnValueEg + Value(rank_of(wpsq));
@@ -239,17 +241,17 @@ template<>
Value Endgame<KRKP>::operator()(const Position& pos) const { Value Endgame<KRKP>::operator()(const Position& pos) const {
assert(pos.non_pawn_material(strongerSide) == RookValueMg); assert(pos.non_pawn_material(strongerSide) == RookValueMg);
assert(pos.piece_count(strongerSide, PAWN) == 0);
assert(pos.non_pawn_material(weakerSide) == 0); assert(pos.non_pawn_material(weakerSide) == 0);
assert(pos.piece_count(weakerSide, PAWN) == 1); assert(pos.count<PAWN>(strongerSide) == 0);
assert(pos.count<PAWN>(weakerSide ) == 1);
Square wksq, wrsq, bksq, bpsq; Square wksq, wrsq, bksq, bpsq;
int tempo = (pos.side_to_move() == strongerSide); int tempo = (pos.side_to_move() == strongerSide);
wksq = pos.king_square(strongerSide); wksq = pos.king_square(strongerSide);
wrsq = pos.piece_list(strongerSide, ROOK)[0];
bksq = pos.king_square(weakerSide); bksq = pos.king_square(weakerSide);
bpsq = pos.piece_list(weakerSide, PAWN)[0]; wrsq = pos.list<ROOK>(strongerSide)[0];
bpsq = pos.list<PAWN>(weakerSide)[0];
if (strongerSide == BLACK) if (strongerSide == BLACK)
{ {
@@ -296,10 +298,10 @@ template<>
Value Endgame<KRKB>::operator()(const Position& pos) const { Value Endgame<KRKB>::operator()(const Position& pos) const {
assert(pos.non_pawn_material(strongerSide) == RookValueMg); assert(pos.non_pawn_material(strongerSide) == RookValueMg);
assert(pos.piece_count(strongerSide, PAWN) == 0); assert(pos.non_pawn_material(weakerSide ) == BishopValueMg);
assert(pos.non_pawn_material(weakerSide) == BishopValueMg); assert(pos.count<BISHOP>(weakerSide ) == 1);
assert(pos.piece_count(weakerSide, PAWN) == 0); assert(pos.count< PAWN>(weakerSide ) == 0);
assert(pos.piece_count(weakerSide, BISHOP) == 1); assert(pos.count< PAWN>(strongerSide) == 0);
Value result = Value(MateTable[pos.king_square(weakerSide)]); Value result = Value(MateTable[pos.king_square(weakerSide)]);
return strongerSide == pos.side_to_move() ? result : -result; return strongerSide == pos.side_to_move() ? result : -result;
@@ -312,20 +314,51 @@ template<>
Value Endgame<KRKN>::operator()(const Position& pos) const { Value Endgame<KRKN>::operator()(const Position& pos) const {
assert(pos.non_pawn_material(strongerSide) == RookValueMg); assert(pos.non_pawn_material(strongerSide) == RookValueMg);
assert(pos.piece_count(strongerSide, PAWN) == 0); assert(pos.non_pawn_material(weakerSide ) == KnightValueMg);
assert(pos.non_pawn_material(weakerSide) == KnightValueMg); assert(pos.count<KNIGHT>(weakerSide ) == 1);
assert(pos.piece_count(weakerSide, PAWN) == 0); assert(pos.count< PAWN>(weakerSide ) == 0);
assert(pos.piece_count(weakerSide, KNIGHT) == 1); assert(pos.count< PAWN>(strongerSide) == 0);
const int penalty[8] = { 0, 10, 14, 20, 30, 42, 58, 80 }; const int penalty[8] = { 0, 10, 14, 20, 30, 42, 58, 80 };
Square bksq = pos.king_square(weakerSide); Square bksq = pos.king_square(weakerSide);
Square bnsq = pos.piece_list(weakerSide, KNIGHT)[0]; Square bnsq = pos.list<KNIGHT>(weakerSide)[0];
Value result = Value(MateTable[bksq] + penalty[square_distance(bksq, bnsq)]); Value result = Value(MateTable[bksq] + penalty[square_distance(bksq, bnsq)]);
return strongerSide == pos.side_to_move() ? result : -result; return strongerSide == pos.side_to_move() ? result : -result;
} }
/// KQ vs KP. In general, a win for the stronger side, however, there are a few
/// important exceptions. Pawn on 7th rank, A,C,F or H file, with king next can
/// be a draw, so we scale down to distance between kings only.
template<>
Value Endgame<KQKP>::operator()(const Position& pos) const {
assert(pos.non_pawn_material(strongerSide) == QueenValueMg);
assert(pos.non_pawn_material(weakerSide ) == VALUE_ZERO);
assert(pos.count<PAWN>(strongerSide) == 0);
assert(pos.count<PAWN>(weakerSide ) == 1);
Square winnerKSq = pos.king_square(strongerSide);
Square loserKSq = pos.king_square(weakerSide);
Square pawnSq = pos.list<PAWN>(weakerSide)[0];
Value result = QueenValueEg
- PawnValueEg
+ DistanceBonus[square_distance(winnerKSq, loserKSq)];
if ( square_distance(loserKSq, pawnSq) == 1
&& relative_rank(weakerSide, pawnSq) == RANK_7)
{
File f = file_of(pawnSq);
if (f == FILE_A || f == FILE_C || f == FILE_F || f == FILE_H)
result = Value(DistanceBonus[square_distance(winnerKSq, loserKSq)]);
}
return strongerSide == pos.side_to_move() ? result : -result;
}
/// KQ vs KR. This is almost identical to KX vs K: We give the attacking /// KQ vs KR. This is almost identical to KX vs K: We give the attacking
/// king a bonus for having the kings close together, and for forcing the /// king a bonus for having the kings close together, and for forcing the
/// defending king towards the edge. If we also take care to avoid null move /// defending king towards the edge. If we also take care to avoid null move
@@ -335,9 +368,9 @@ template<>
Value Endgame<KQKR>::operator()(const Position& pos) const { Value Endgame<KQKR>::operator()(const Position& pos) const {
assert(pos.non_pawn_material(strongerSide) == QueenValueMg); assert(pos.non_pawn_material(strongerSide) == QueenValueMg);
assert(pos.piece_count(strongerSide, PAWN) == 0); assert(pos.non_pawn_material(weakerSide ) == RookValueMg);
assert(pos.non_pawn_material(weakerSide) == RookValueMg); assert(pos.count<PAWN>(strongerSide) == 0);
assert(pos.piece_count(weakerSide, PAWN) == 0); assert(pos.count<PAWN>(weakerSide ) == 0);
Square winnerKSq = pos.king_square(strongerSide); Square winnerKSq = pos.king_square(strongerSide);
Square loserKSq = pos.king_square(weakerSide); Square loserKSq = pos.king_square(weakerSide);
@@ -353,16 +386,16 @@ Value Endgame<KQKR>::operator()(const Position& pos) const {
template<> template<>
Value Endgame<KBBKN>::operator()(const Position& pos) const { Value Endgame<KBBKN>::operator()(const Position& pos) const {
assert(pos.piece_count(strongerSide, BISHOP) == 2); assert(pos.non_pawn_material(strongerSide) == 2 * BishopValueMg);
assert(pos.non_pawn_material(strongerSide) == 2*BishopValueMg); assert(pos.non_pawn_material(weakerSide ) == KnightValueMg);
assert(pos.piece_count(weakerSide, KNIGHT) == 1); assert(pos.count<BISHOP>(strongerSide) == 2);
assert(pos.non_pawn_material(weakerSide) == KnightValueMg); assert(pos.count<KNIGHT>(weakerSide ) == 1);
assert(!pos.pieces(PAWN)); assert(!pos.pieces(PAWN));
Value result = BishopValueEg; Value result = BishopValueEg;
Square wksq = pos.king_square(strongerSide); Square wksq = pos.king_square(strongerSide);
Square bksq = pos.king_square(weakerSide); Square bksq = pos.king_square(weakerSide);
Square nsq = pos.piece_list(weakerSide, KNIGHT)[0]; Square nsq = pos.list<KNIGHT>(weakerSide)[0];
// Bonus for attacking king close to defending king // Bonus for attacking king close to defending king
result += Value(DistanceBonus[square_distance(wksq, bksq)]); result += Value(DistanceBonus[square_distance(wksq, bksq)]);
@@ -377,17 +410,13 @@ Value Endgame<KBBKN>::operator()(const Position& pos) const {
} }
/// K and two minors vs K and one or two minors or K and two knights against /// Some cases of trivial draws
/// king alone are always draw. template<> Value Endgame<KK>::operator()(const Position&) const { return VALUE_DRAW; }
template<> template<> Value Endgame<KBK>::operator()(const Position&) const { return VALUE_DRAW; }
Value Endgame<KmmKm>::operator()(const Position&) const { template<> Value Endgame<KNK>::operator()(const Position&) const { return VALUE_DRAW; }
return VALUE_DRAW; template<> Value Endgame<KNNK>::operator()(const Position&) const { return VALUE_DRAW; }
} template<> Value Endgame<KmmKm>::operator()(const Position&) const { return VALUE_DRAW; }
template<>
Value Endgame<KNNK>::operator()(const Position&) const {
return VALUE_DRAW;
}
/// K, bishop and one or more pawns vs K. It checks for draws with rook pawns and /// K, bishop and one or more pawns vs K. It checks for draws with rook pawns and
/// a bishop of the wrong color. If such a draw is detected, SCALE_FACTOR_DRAW /// a bishop of the wrong color. If such a draw is detected, SCALE_FACTOR_DRAW
@@ -397,20 +426,20 @@ template<>
ScaleFactor Endgame<KBPsK>::operator()(const Position& pos) const { ScaleFactor Endgame<KBPsK>::operator()(const Position& pos) const {
assert(pos.non_pawn_material(strongerSide) == BishopValueMg); assert(pos.non_pawn_material(strongerSide) == BishopValueMg);
assert(pos.piece_count(strongerSide, BISHOP) == 1); assert(pos.count<BISHOP>(strongerSide) == 1);
assert(pos.piece_count(strongerSide, PAWN) >= 1); assert(pos.count< PAWN>(strongerSide) >= 1);
// No assertions about the material of weakerSide, because we want draws to // No assertions about the material of weakerSide, because we want draws to
// be detected even when the weaker side has some pawns. // be detected even when the weaker side has some pawns.
Bitboard pawns = pos.pieces(strongerSide, PAWN); Bitboard pawns = pos.pieces(strongerSide, PAWN);
File pawnFile = file_of(pos.piece_list(strongerSide, PAWN)[0]); File pawnFile = file_of(pos.list<PAWN>(strongerSide)[0]);
// All pawns are on a single rook file ? // All pawns are on a single rook file ?
if ( (pawnFile == FILE_A || pawnFile == FILE_H) if ( (pawnFile == FILE_A || pawnFile == FILE_H)
&& !(pawns & ~file_bb(pawnFile))) && !(pawns & ~file_bb(pawnFile)))
{ {
Square bishopSq = pos.piece_list(strongerSide, BISHOP)[0]; Square bishopSq = pos.list<BISHOP>(strongerSide)[0];
Square queeningSq = relative_square(strongerSide, pawnFile | RANK_8); Square queeningSq = relative_square(strongerSide, pawnFile | RANK_8);
Square kingSq = pos.king_square(weakerSide); Square kingSq = pos.king_square(weakerSide);
@@ -439,6 +468,29 @@ ScaleFactor Endgame<KBPsK>::operator()(const Position& pos) const {
return SCALE_FACTOR_DRAW; return SCALE_FACTOR_DRAW;
} }
} }
// All pawns on same B or G file? Then potential draw
if ( (pawnFile == FILE_B || pawnFile == FILE_G)
&& !(pos.pieces(PAWN) & ~file_bb(pawnFile))
&& pos.non_pawn_material(weakerSide) == 0
&& pos.count<PAWN>(weakerSide) >= 1)
{
// Get weaker pawn closest to opponent's queening square
Bitboard wkPawns = pos.pieces(weakerSide, PAWN);
Square weakerPawnSq = strongerSide == WHITE ? msb(wkPawns) : lsb(wkPawns);
Square strongerKingSq = pos.king_square(strongerSide);
Square weakerKingSq = pos.king_square(weakerSide);
Square bishopSq = pos.list<BISHOP>(strongerSide)[0];
// Draw if weaker pawn is on rank 7, bishop can't attack the pawn, and
// weaker king can stop opposing opponent's king from penetrating.
if ( relative_rank(strongerSide, weakerPawnSq) == RANK_7
&& opposite_colors(bishopSq, weakerPawnSq)
&& square_distance(weakerPawnSq, weakerKingSq) <= square_distance(weakerPawnSq, strongerKingSq))
return SCALE_FACTOR_DRAW;
}
return SCALE_FACTOR_NONE; return SCALE_FACTOR_NONE;
} }
@@ -449,10 +501,10 @@ template<>
ScaleFactor Endgame<KQKRPs>::operator()(const Position& pos) const { ScaleFactor Endgame<KQKRPs>::operator()(const Position& pos) const {
assert(pos.non_pawn_material(strongerSide) == QueenValueMg); assert(pos.non_pawn_material(strongerSide) == QueenValueMg);
assert(pos.piece_count(strongerSide, QUEEN) == 1); assert(pos.count<QUEEN>(strongerSide) == 1);
assert(pos.piece_count(strongerSide, PAWN) == 0); assert(pos.count< PAWN>(strongerSide) == 0);
assert(pos.piece_count(weakerSide, ROOK) == 1); assert(pos.count< ROOK>(weakerSide ) == 1);
assert(pos.piece_count(weakerSide, PAWN) >= 1); assert(pos.count< PAWN>(weakerSide ) >= 1);
Square kingSq = pos.king_square(weakerSide); Square kingSq = pos.king_square(weakerSide);
if ( relative_rank(weakerSide, kingSq) <= RANK_2 if ( relative_rank(weakerSide, kingSq) <= RANK_2
@@ -461,7 +513,7 @@ ScaleFactor Endgame<KQKRPs>::operator()(const Position& pos) const {
&& (pos.pieces(weakerSide, PAWN) & rank_bb(relative_rank(weakerSide, RANK_2))) && (pos.pieces(weakerSide, PAWN) & rank_bb(relative_rank(weakerSide, RANK_2)))
&& (pos.attacks_from<KING>(kingSq) & pos.pieces(weakerSide, PAWN))) && (pos.attacks_from<KING>(kingSq) & pos.pieces(weakerSide, PAWN)))
{ {
Square rsq = pos.piece_list(weakerSide, ROOK)[0]; Square rsq = pos.list<ROOK>(weakerSide)[0];
if (pos.attacks_from<PAWN>(rsq, strongerSide) & pos.pieces(weakerSide, PAWN)) if (pos.attacks_from<PAWN>(rsq, strongerSide) & pos.pieces(weakerSide, PAWN))
return SCALE_FACTOR_DRAW; return SCALE_FACTOR_DRAW;
} }
@@ -479,15 +531,15 @@ template<>
ScaleFactor Endgame<KRPKR>::operator()(const Position& pos) const { ScaleFactor Endgame<KRPKR>::operator()(const Position& pos) const {
assert(pos.non_pawn_material(strongerSide) == RookValueMg); assert(pos.non_pawn_material(strongerSide) == RookValueMg);
assert(pos.piece_count(strongerSide, PAWN) == 1);
assert(pos.non_pawn_material(weakerSide) == RookValueMg); assert(pos.non_pawn_material(weakerSide) == RookValueMg);
assert(pos.piece_count(weakerSide, PAWN) == 0); assert(pos.count<PAWN>(strongerSide) == 1);
assert(pos.count<PAWN>(weakerSide ) == 0);
Square wksq = pos.king_square(strongerSide); Square wksq = pos.king_square(strongerSide);
Square wrsq = pos.piece_list(strongerSide, ROOK)[0];
Square wpsq = pos.piece_list(strongerSide, PAWN)[0];
Square bksq = pos.king_square(weakerSide); Square bksq = pos.king_square(weakerSide);
Square brsq = pos.piece_list(weakerSide, ROOK)[0]; Square wrsq = pos.list<ROOK>(strongerSide)[0];
Square wpsq = pos.list<PAWN>(strongerSide)[0];
Square brsq = pos.list<ROOK>(weakerSide)[0];
// Orient the board in such a way that the stronger side is white, and the // Orient the board in such a way that the stronger side is white, and the
// pawn is on the left half of the board. // pawn is on the left half of the board.
@@ -597,12 +649,12 @@ template<>
ScaleFactor Endgame<KRPPKRP>::operator()(const Position& pos) const { ScaleFactor Endgame<KRPPKRP>::operator()(const Position& pos) const {
assert(pos.non_pawn_material(strongerSide) == RookValueMg); assert(pos.non_pawn_material(strongerSide) == RookValueMg);
assert(pos.piece_count(strongerSide, PAWN) == 2);
assert(pos.non_pawn_material(weakerSide) == RookValueMg); assert(pos.non_pawn_material(weakerSide) == RookValueMg);
assert(pos.piece_count(weakerSide, PAWN) == 1); assert(pos.count<PAWN>(strongerSide) == 2);
assert(pos.count<PAWN>(weakerSide ) == 1);
Square wpsq1 = pos.piece_list(strongerSide, PAWN)[0]; Square wpsq1 = pos.list<PAWN>(strongerSide)[0];
Square wpsq2 = pos.piece_list(strongerSide, PAWN)[1]; Square wpsq2 = pos.list<PAWN>(strongerSide)[1];
Square bksq = pos.king_square(weakerSide); Square bksq = pos.king_square(weakerSide);
// Does the stronger side have a passed pawn? // Does the stronger side have a passed pawn?
@@ -635,9 +687,9 @@ template<>
ScaleFactor Endgame<KPsK>::operator()(const Position& pos) const { ScaleFactor Endgame<KPsK>::operator()(const Position& pos) const {
assert(pos.non_pawn_material(strongerSide) == VALUE_ZERO); assert(pos.non_pawn_material(strongerSide) == VALUE_ZERO);
assert(pos.piece_count(strongerSide, PAWN) >= 2);
assert(pos.non_pawn_material(weakerSide) == VALUE_ZERO); assert(pos.non_pawn_material(weakerSide) == VALUE_ZERO);
assert(pos.piece_count(weakerSide, PAWN) == 0); assert(pos.count<PAWN>(strongerSide) >= 2);
assert(pos.count<PAWN>(weakerSide ) == 0);
Square ksq = pos.king_square(weakerSide); Square ksq = pos.king_square(weakerSide);
Bitboard pawns = pos.pieces(strongerSide, PAWN); Bitboard pawns = pos.pieces(strongerSide, PAWN);
@@ -648,7 +700,7 @@ ScaleFactor Endgame<KPsK>::operator()(const Position& pos) const {
// Does the defending king block the pawns? // Does the defending king block the pawns?
if ( square_distance(ksq, relative_square(strongerSide, SQ_A8)) <= 1 if ( square_distance(ksq, relative_square(strongerSide, SQ_A8)) <= 1
|| ( file_of(ksq) == FILE_A || ( file_of(ksq) == FILE_A
&& !in_front_bb(strongerSide, ksq) & pawns)) && !(in_front_bb(strongerSide, rank_of(ksq)) & pawns)))
return SCALE_FACTOR_DRAW; return SCALE_FACTOR_DRAW;
} }
// Are all pawns on the 'h' file? // Are all pawns on the 'h' file?
@@ -657,7 +709,7 @@ ScaleFactor Endgame<KPsK>::operator()(const Position& pos) const {
// Does the defending king block the pawns? // Does the defending king block the pawns?
if ( square_distance(ksq, relative_square(strongerSide, SQ_H8)) <= 1 if ( square_distance(ksq, relative_square(strongerSide, SQ_H8)) <= 1
|| ( file_of(ksq) == FILE_H || ( file_of(ksq) == FILE_H
&& !in_front_bb(strongerSide, ksq) & pawns)) && !(in_front_bb(strongerSide, rank_of(ksq)) & pawns)))
return SCALE_FACTOR_DRAW; return SCALE_FACTOR_DRAW;
} }
return SCALE_FACTOR_NONE; return SCALE_FACTOR_NONE;
@@ -672,15 +724,15 @@ template<>
ScaleFactor Endgame<KBPKB>::operator()(const Position& pos) const { ScaleFactor Endgame<KBPKB>::operator()(const Position& pos) const {
assert(pos.non_pawn_material(strongerSide) == BishopValueMg); assert(pos.non_pawn_material(strongerSide) == BishopValueMg);
assert(pos.piece_count(strongerSide, BISHOP) == 1); assert(pos.non_pawn_material(weakerSide ) == BishopValueMg);
assert(pos.piece_count(strongerSide, PAWN) == 1); assert(pos.count<BISHOP>(strongerSide) == 1);
assert(pos.non_pawn_material(weakerSide) == BishopValueMg); assert(pos.count<BISHOP>(weakerSide ) == 1);
assert(pos.piece_count(weakerSide, BISHOP) == 1); assert(pos.count< PAWN>(strongerSide) == 1);
assert(pos.piece_count(weakerSide, PAWN) == 0); assert(pos.count< PAWN>(weakerSide ) == 0);
Square pawnSq = pos.piece_list(strongerSide, PAWN)[0]; Square pawnSq = pos.list<PAWN>(strongerSide)[0];
Square strongerBishopSq = pos.piece_list(strongerSide, BISHOP)[0]; Square strongerBishopSq = pos.list<BISHOP>(strongerSide)[0];
Square weakerBishopSq = pos.piece_list(weakerSide, BISHOP)[0]; Square weakerBishopSq = pos.list<BISHOP>(weakerSide)[0];
Square weakerKingSq = pos.king_square(weakerSide); Square weakerKingSq = pos.king_square(weakerSide);
// Case 1: Defending king blocks the pawn, and cannot be driven away // Case 1: Defending king blocks the pawn, and cannot be driven away
@@ -727,21 +779,21 @@ template<>
ScaleFactor Endgame<KBPPKB>::operator()(const Position& pos) const { ScaleFactor Endgame<KBPPKB>::operator()(const Position& pos) const {
assert(pos.non_pawn_material(strongerSide) == BishopValueMg); assert(pos.non_pawn_material(strongerSide) == BishopValueMg);
assert(pos.piece_count(strongerSide, BISHOP) == 1); assert(pos.non_pawn_material(weakerSide ) == BishopValueMg);
assert(pos.piece_count(strongerSide, PAWN) == 2); assert(pos.count<BISHOP>(strongerSide) == 1);
assert(pos.non_pawn_material(weakerSide) == BishopValueMg); assert(pos.count<BISHOP>(weakerSide ) == 1);
assert(pos.piece_count(weakerSide, BISHOP) == 1); assert(pos.count< PAWN>(strongerSide) == 2);
assert(pos.piece_count(weakerSide, PAWN) == 0); assert(pos.count< PAWN>(weakerSide ) == 0);
Square wbsq = pos.piece_list(strongerSide, BISHOP)[0]; Square wbsq = pos.list<BISHOP>(strongerSide)[0];
Square bbsq = pos.piece_list(weakerSide, BISHOP)[0]; Square bbsq = pos.list<BISHOP>(weakerSide)[0];
if (!opposite_colors(wbsq, bbsq)) if (!opposite_colors(wbsq, bbsq))
return SCALE_FACTOR_NONE; return SCALE_FACTOR_NONE;
Square ksq = pos.king_square(weakerSide); Square ksq = pos.king_square(weakerSide);
Square psq1 = pos.piece_list(strongerSide, PAWN)[0]; Square psq1 = pos.list<PAWN>(strongerSide)[0];
Square psq2 = pos.piece_list(strongerSide, PAWN)[1]; Square psq2 = pos.list<PAWN>(strongerSide)[1];
Rank r1 = rank_of(psq1); Rank r1 = rank_of(psq1);
Rank r2 = rank_of(psq2); Rank r2 = rank_of(psq2);
Square blockSq1, blockSq2; Square blockSq1, blockSq2;
@@ -802,14 +854,14 @@ template<>
ScaleFactor Endgame<KBPKN>::operator()(const Position& pos) const { ScaleFactor Endgame<KBPKN>::operator()(const Position& pos) const {
assert(pos.non_pawn_material(strongerSide) == BishopValueMg); assert(pos.non_pawn_material(strongerSide) == BishopValueMg);
assert(pos.piece_count(strongerSide, BISHOP) == 1); assert(pos.non_pawn_material(weakerSide ) == KnightValueMg);
assert(pos.piece_count(strongerSide, PAWN) == 1); assert(pos.count<BISHOP>(strongerSide) == 1);
assert(pos.non_pawn_material(weakerSide) == KnightValueMg); assert(pos.count<KNIGHT>(weakerSide ) == 1);
assert(pos.piece_count(weakerSide, KNIGHT) == 1); assert(pos.count< PAWN>(strongerSide) == 1);
assert(pos.piece_count(weakerSide, PAWN) == 0); assert(pos.count< PAWN>(weakerSide ) == 0);
Square pawnSq = pos.piece_list(strongerSide, PAWN)[0]; Square pawnSq = pos.list<PAWN>(strongerSide)[0];
Square strongerBishopSq = pos.piece_list(strongerSide, BISHOP)[0]; Square strongerBishopSq = pos.list<BISHOP>(strongerSide)[0];
Square weakerKingSq = pos.king_square(weakerSide); Square weakerKingSq = pos.king_square(weakerSide);
if ( file_of(weakerKingSq) == file_of(pawnSq) if ( file_of(weakerKingSq) == file_of(pawnSq)
@@ -829,12 +881,12 @@ template<>
ScaleFactor Endgame<KNPK>::operator()(const Position& pos) const { ScaleFactor Endgame<KNPK>::operator()(const Position& pos) const {
assert(pos.non_pawn_material(strongerSide) == KnightValueMg); assert(pos.non_pawn_material(strongerSide) == KnightValueMg);
assert(pos.piece_count(strongerSide, KNIGHT) == 1); assert(pos.non_pawn_material(weakerSide ) == VALUE_ZERO);
assert(pos.piece_count(strongerSide, PAWN) == 1); assert(pos.count<KNIGHT>(strongerSide) == 1);
assert(pos.non_pawn_material(weakerSide) == VALUE_ZERO); assert(pos.count< PAWN>(strongerSide) == 1);
assert(pos.piece_count(weakerSide, PAWN) == 0); assert(pos.count< PAWN>(weakerSide ) == 0);
Square pawnSq = pos.piece_list(strongerSide, PAWN)[0]; Square pawnSq = pos.list<PAWN>(strongerSide)[0];
Square weakerKingSq = pos.king_square(weakerSide); Square weakerKingSq = pos.king_square(weakerSide);
if ( pawnSq == relative_square(strongerSide, SQ_A7) if ( pawnSq == relative_square(strongerSide, SQ_A7)
@@ -849,6 +901,24 @@ ScaleFactor Endgame<KNPK>::operator()(const Position& pos) const {
} }
/// K, knight and a pawn vs K and bishop. If knight can block bishop from taking
/// pawn, it's a win. Otherwise, drawn.
template<>
ScaleFactor Endgame<KNPKB>::operator()(const Position& pos) const {
Square pawnSq = pos.list<PAWN>(strongerSide)[0];
Square bishopSq = pos.list<BISHOP>(weakerSide)[0];
Square weakerKingSq = pos.king_square(weakerSide);
// King needs to get close to promoting pawn to prevent knight from blocking.
// Rules for this are very tricky, so just approximate.
if (forward_bb(strongerSide, pawnSq) & pos.attacks_from<BISHOP>(bishopSq))
return ScaleFactor(square_distance(weakerKingSq, pawnSq));
return SCALE_FACTOR_NONE;
}
/// K and a pawn vs K and a pawn. This is done by removing the weakest side's /// K and a pawn vs K and a pawn. This is done by removing the weakest side's
/// pawn and probing the KP vs K bitbase: If the weakest side has a draw without /// pawn and probing the KP vs K bitbase: If the weakest side has a draw without
/// the pawn, she probably has at least a draw with the pawn as well. The exception /// the pawn, she probably has at least a draw with the pawn as well. The exception
@@ -858,21 +928,21 @@ template<>
ScaleFactor Endgame<KPKP>::operator()(const Position& pos) const { ScaleFactor Endgame<KPKP>::operator()(const Position& pos) const {
assert(pos.non_pawn_material(strongerSide) == VALUE_ZERO); assert(pos.non_pawn_material(strongerSide) == VALUE_ZERO);
assert(pos.non_pawn_material(weakerSide) == VALUE_ZERO); assert(pos.non_pawn_material(weakerSide ) == VALUE_ZERO);
assert(pos.piece_count(WHITE, PAWN) == 1); assert(pos.count<PAWN>(WHITE) == 1);
assert(pos.piece_count(BLACK, PAWN) == 1); assert(pos.count<PAWN>(BLACK) == 1);
Square wksq = pos.king_square(strongerSide); Square wksq = pos.king_square(strongerSide);
Square bksq = pos.king_square(weakerSide); Square bksq = pos.king_square(weakerSide);
Square wpsq = pos.piece_list(strongerSide, PAWN)[0]; Square wpsq = pos.list<PAWN>(strongerSide)[0];
Color stm = pos.side_to_move(); Color us = pos.side_to_move();
if (strongerSide == BLACK) if (strongerSide == BLACK)
{ {
wksq = ~wksq; wksq = ~wksq;
bksq = ~bksq; bksq = ~bksq;
wpsq = ~wpsq; wpsq = ~wpsq;
stm = ~stm; us = ~us;
} }
if (file_of(wpsq) >= FILE_E) if (file_of(wpsq) >= FILE_E)
@@ -890,5 +960,5 @@ ScaleFactor Endgame<KPKP>::operator()(const Position& pos) const {
// Probe the KPK bitbase with the weakest side's pawn removed. If it's a draw, // Probe the KPK bitbase with the weakest side's pawn removed. If it's a draw,
// it's probably at least a draw even with the pawn. // it's probably at least a draw even with the pawn.
return Bitbases::probe_kpk(wksq, wpsq, bksq, stm) ? SCALE_FACTOR_NONE : SCALE_FACTOR_DRAW; return Bitbases::probe_kpk(wksq, wpsq, bksq, us) ? SCALE_FACTOR_NONE : SCALE_FACTOR_DRAW;
} }
+9 -4
View File
@@ -1,7 +1,7 @@
/* /*
Stockfish, a UCI chess playing engine derived from Glaurung 2.1 Stockfish, a UCI chess playing engine derived from Glaurung 2.1
Copyright (C) 2004-2008 Tord Romstad (Glaurung author) Copyright (C) 2004-2008 Tord Romstad (Glaurung author)
Copyright (C) 2008-2012 Marco Costalba, Joona Kiiski, Tord Romstad Copyright (C) 2008-2013 Marco Costalba, Joona Kiiski, Tord Romstad
Stockfish is free software: you can redistribute it and/or modify Stockfish is free software: you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by it under the terms of the GNU General Public License as published by
@@ -17,7 +17,7 @@
along with this program. If not, see <http://www.gnu.org/licenses/>. along with this program. If not, see <http://www.gnu.org/licenses/>.
*/ */
#if !defined(ENDGAME_H_INCLUDED) #ifndef ENDGAME_H_INCLUDED
#define ENDGAME_H_INCLUDED #define ENDGAME_H_INCLUDED
#include <map> #include <map>
@@ -33,15 +33,19 @@ enum EndgameType {
// Evaluation functions // Evaluation functions
KK, // K vs K
KBK, // KB vs K
KNK, // KN vs K
KNNK, // KNN vs K
KXK, // Generic "mate lone king" eval KXK, // Generic "mate lone king" eval
KBNK, // KBN vs K KBNK, // KBN vs K
KPK, // KP vs K KPK, // KP vs K
KRKP, // KR vs KP KRKP, // KR vs KP
KRKB, // KR vs KB KRKB, // KR vs KB
KRKN, // KR vs KN KRKN, // KR vs KN
KQKP, // KQ vs KP
KQKR, // KQ vs KR KQKR, // KQ vs KR
KBBKN, // KBB vs KN KBBKN, // KBB vs KN
KNNK, // KNN vs K
KmmKm, // K and two minors vs K and one or two minors KmmKm, // K and two minors vs K and one or two minors
@@ -57,6 +61,7 @@ enum EndgameType {
KBPPKB, // KBPP vs KB KBPPKB, // KBPP vs KB
KBPKN, // KBP vs KN KBPKN, // KBP vs KN
KNPK, // KNP vs K KNPK, // KNP vs K
KNPKB, // KNP vs KB
KPKP // KP vs KP KPKP // KP vs KP
}; };
@@ -117,4 +122,4 @@ public:
{ return eg = map(eg).count(key) ? map(eg)[key] : NULL; } { return eg = map(eg).count(key) ? map(eg)[key] : NULL; }
}; };
#endif // !defined(ENDGAME_H_INCLUDED) #endif // #ifndef ENDGAME_H_INCLUDED
+315 -339
View File
File diff suppressed because it is too large Load Diff
+3 -5
View File
@@ -1,7 +1,7 @@
/* /*
Stockfish, a UCI chess playing engine derived from Glaurung 2.1 Stockfish, a UCI chess playing engine derived from Glaurung 2.1
Copyright (C) 2004-2008 Tord Romstad (Glaurung author) Copyright (C) 2004-2008 Tord Romstad (Glaurung author)
Copyright (C) 2008-2012 Marco Costalba, Joona Kiiski, Tord Romstad Copyright (C) 2008-2013 Marco Costalba, Joona Kiiski, Tord Romstad
Stockfish is free software: you can redistribute it and/or modify Stockfish is free software: you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by it under the terms of the GNU General Public License as published by
@@ -17,7 +17,7 @@
along with this program. If not, see <http://www.gnu.org/licenses/>. along with this program. If not, see <http://www.gnu.org/licenses/>.
*/ */
#if !defined(EVALUATE_H_INCLUDED) #ifndef EVALUATE_H_INCLUDED
#define EVALUATE_H_INCLUDED #define EVALUATE_H_INCLUDED
#include "types.h" #include "types.h"
@@ -26,12 +26,10 @@ class Position;
namespace Eval { namespace Eval {
extern Color RootColor;
extern void init(); extern void init();
extern Value evaluate(const Position& pos, Value& margin); extern Value evaluate(const Position& pos, Value& margin);
extern std::string trace(const Position& pos); extern std::string trace(const Position& pos);
} }
#endif // !defined(EVALUATE_H_INCLUDED) #endif // #ifndef EVALUATE_H_INCLUDED
-72
View File
@@ -1,72 +0,0 @@
/*
Stockfish, a UCI chess playing engine derived from Glaurung 2.1
Copyright (C) 2004-2008 Tord Romstad (Glaurung author)
Copyright (C) 2008-2012 Marco Costalba, Joona Kiiski, Tord Romstad
Stockfish is free software: you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by
the Free Software Foundation, either version 3 of the License, or
(at your option) any later version.
Stockfish is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
GNU General Public License for more details.
You should have received a copy of the GNU General Public License
along with this program. If not, see <http://www.gnu.org/licenses/>.
*/
#if !defined(HISTORY_H_INCLUDED)
#define HISTORY_H_INCLUDED
#include <algorithm>
#include <cstring>
#include "types.h"
/// The History class stores statistics about how often different moves
/// have been successful or unsuccessful during the current search. These
/// statistics are used for reduction and move ordering decisions. History
/// entries are stored according only to moving piece and destination square,
/// in particular two moves with different origin but same destination and
/// same piece will be considered identical.
class History {
public:
void clear();
Value value(Piece p, Square to) const;
void add(Piece p, Square to, Value bonus);
Value gain(Piece p, Square to) const;
void update_gain(Piece p, Square to, Value g);
static const Value MaxValue = Value(2000);
private:
Value history[16][64]; // [piece][to_square]
Value maxGains[16][64]; // [piece][to_square]
};
inline void History::clear() {
memset(history, 0, 16 * 64 * sizeof(Value));
memset(maxGains, 0, 16 * 64 * sizeof(Value));
}
inline Value History::value(Piece p, Square to) const {
return history[p][to];
}
inline void History::add(Piece p, Square to, Value bonus) {
if (abs(history[p][to] + bonus) < MaxValue) history[p][to] += bonus;
}
inline Value History::gain(Piece p, Square to) const {
return maxGains[p][to];
}
inline void History::update_gain(Piece p, Square to, Value g) {
maxGains[p][to] = std::max(g, maxGains[p][to] - 1);
}
#endif // !defined(HISTORY_H_INCLUDED)
+2 -2
View File
@@ -1,7 +1,7 @@
/* /*
Stockfish, a UCI chess playing engine derived from Glaurung 2.1 Stockfish, a UCI chess playing engine derived from Glaurung 2.1
Copyright (C) 2004-2008 Tord Romstad (Glaurung author) Copyright (C) 2004-2008 Tord Romstad (Glaurung author)
Copyright (C) 2008-2012 Marco Costalba, Joona Kiiski, Tord Romstad Copyright (C) 2008-2013 Marco Costalba, Joona Kiiski, Tord Romstad
Stockfish is free software: you can redistribute it and/or modify Stockfish is free software: you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by it under the terms of the GNU General Public License as published by
@@ -34,7 +34,7 @@ int main(int argc, char* argv[]) {
UCI::init(Options); UCI::init(Options);
Bitboards::init(); Bitboards::init();
Zobrist::init(); Position::init();
Bitbases::init_kpk(); Bitbases::init_kpk();
Search::init(); Search::init();
Eval::init(); Eval::init();
+96 -80
View File
@@ -1,7 +1,7 @@
/* /*
Stockfish, a UCI chess playing engine derived from Glaurung 2.1 Stockfish, a UCI chess playing engine derived from Glaurung 2.1
Copyright (C) 2004-2008 Tord Romstad (Glaurung author) Copyright (C) 2004-2008 Tord Romstad (Glaurung author)
Copyright (C) 2008-2012 Marco Costalba, Joona Kiiski, Tord Romstad Copyright (C) 2008-2013 Marco Costalba, Joona Kiiski, Tord Romstad
Stockfish is free software: you can redistribute it and/or modify Stockfish is free software: you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by it under the terms of the GNU General Public License as published by
@@ -17,7 +17,7 @@
along with this program. If not, see <http://www.gnu.org/licenses/>. along with this program. If not, see <http://www.gnu.org/licenses/>.
*/ */
#include <algorithm> #include <algorithm> // For std::min
#include <cassert> #include <cassert>
#include <cstring> #include <cstring>
@@ -35,18 +35,32 @@ namespace {
const int NoPawnsSF[4] = { 6, 12, 32 }; const int NoPawnsSF[4] = { 6, 12, 32 };
// Polynomial material balance parameters // Polynomial material balance parameters
const Value RedundantQueenPenalty = Value(320); const Value RedundantQueen = Value(320);
const Value RedundantRookPenalty = Value(554); const Value RedundantRook = Value(554);
// pair pawn knight bishop rook queen
const int LinearCoefficients[6] = { 1617, -162, -1172, -190, 105, 26 }; const int LinearCoefficients[6] = { 1617, -162, -1172, -190, 105, 26 };
const int QuadraticCoefficientsSameColor[][8] = { const int QuadraticCoefficientsSameColor[][PIECE_TYPE_NB] = {
{ 7, 7, 7, 7, 7, 7 }, { 39, 2, 7, 7, 7, 7 }, { 35, 271, -4, 7, 7, 7 }, // pair pawn knight bishop rook queen
{ 7, 25, 4, 7, 7, 7 }, { -27, -2, 46, 100, 56, 7 }, { 58, 29, 83, 148, -3, -25 } }; { 7 }, // Bishop pair
{ 39, 2 }, // Pawn
{ 35, 271, -4 }, // Knight
{ 7, 105, 4, 7 }, // Bishop
{ -27, -2, 46, 100, 56 }, // Rook
{ 58, 29, 83, 148, -3, -25 } // Queen
};
const int QuadraticCoefficientsOppositeColor[][8] = { const int QuadraticCoefficientsOppositeColor[][PIECE_TYPE_NB] = {
{ 41, 41, 41, 41, 41, 41 }, { 37, 41, 41, 41, 41, 41 }, { 10, 62, 41, 41, 41, 41 }, // THEIR PIECES
{ 57, 64, 39, 41, 41, 41 }, { 50, 40, 23, -22, 41, 41 }, { 106, 101, 3, 151, 171, 41 } }; // pair pawn knight bishop rook queen
{ 41 }, // Bishop pair
{ 37, 41 }, // Pawn
{ 10, 62, 41 }, // Knight OUR PIECES
{ 57, 64, 39, 41 }, // Bishop
{ 50, 40, 23, -22, 41 }, // Rook
{ 106, 101, 3, 151, 171, 41 } // Queen
};
// Endgame evaluation and scaling functions accessed direcly and not through // Endgame evaluation and scaling functions accessed direcly and not through
// the function maps because correspond to more then one material hash key. // the function maps because correspond to more then one material hash key.
@@ -61,38 +75,74 @@ namespace {
// Helper templates used to detect a given material distribution // Helper templates used to detect a given material distribution
template<Color Us> bool is_KXK(const Position& pos) { template<Color Us> bool is_KXK(const Position& pos) {
const Color Them = (Us == WHITE ? BLACK : WHITE); const Color Them = (Us == WHITE ? BLACK : WHITE);
return pos.non_pawn_material(Them) == VALUE_ZERO return !pos.count<PAWN>(Them)
&& pos.piece_count(Them, PAWN) == 0 && pos.non_pawn_material(Them) == VALUE_ZERO
&& pos.non_pawn_material(Us) >= RookValueMg; && pos.non_pawn_material(Us) >= RookValueMg;
} }
template<Color Us> bool is_KBPsKs(const Position& pos) { template<Color Us> bool is_KBPsKs(const Position& pos) {
return pos.non_pawn_material(Us) == BishopValueMg return pos.non_pawn_material(Us) == BishopValueMg
&& pos.piece_count(Us, BISHOP) == 1 && pos.count<BISHOP>(Us) == 1
&& pos.piece_count(Us, PAWN) >= 1; && pos.count<PAWN >(Us) >= 1;
} }
template<Color Us> bool is_KQKRPs(const Position& pos) { template<Color Us> bool is_KQKRPs(const Position& pos) {
const Color Them = (Us == WHITE ? BLACK : WHITE); const Color Them = (Us == WHITE ? BLACK : WHITE);
return pos.piece_count(Us, PAWN) == 0 return !pos.count<PAWN>(Us)
&& pos.non_pawn_material(Us) == QueenValueMg && pos.non_pawn_material(Us) == QueenValueMg
&& pos.piece_count(Us, QUEEN) == 1 && pos.count<QUEEN>(Us) == 1
&& pos.piece_count(Them, ROOK) == 1 && pos.count<ROOK>(Them) == 1
&& pos.piece_count(Them, PAWN) >= 1; && pos.count<PAWN>(Them) >= 1;
}
/// imbalance() calculates imbalance comparing piece count of each
/// piece type for both colors.
template<Color Us>
int imbalance(const int pieceCount[][PIECE_TYPE_NB]) {
const Color Them = (Us == WHITE ? BLACK : WHITE);
int pt1, pt2, pc, v;
int value = 0;
// Redundancy of major pieces, formula based on Kaufman's paper
// "The Evaluation of Material Imbalances in Chess"
if (pieceCount[Us][ROOK] > 0)
value -= RedundantRook * (pieceCount[Us][ROOK] - 1)
+ RedundantQueen * pieceCount[Us][QUEEN];
// Second-degree polynomial material imbalance by Tord Romstad
for (pt1 = NO_PIECE_TYPE; pt1 <= QUEEN; pt1++)
{
pc = pieceCount[Us][pt1];
if (!pc)
continue;
v = LinearCoefficients[pt1];
for (pt2 = NO_PIECE_TYPE; pt2 <= pt1; pt2++)
v += QuadraticCoefficientsSameColor[pt1][pt2] * pieceCount[Us][pt2]
+ QuadraticCoefficientsOppositeColor[pt1][pt2] * pieceCount[Them][pt2];
value += pc * v;
}
return value;
} }
} // namespace } // namespace
namespace Material {
/// MaterialTable::probe() takes a position object as input, looks up a MaterialEntry /// Material::probe() takes a position object as input, looks up a MaterialEntry
/// object, and returns a pointer to it. If the material configuration is not /// object, and returns a pointer to it. If the material configuration is not
/// already present in the table, it is computed and stored there, so we don't /// already present in the table, it is computed and stored there, so we don't
/// have to recompute everything when the same material configuration occurs again. /// have to recompute everything when the same material configuration occurs again.
MaterialEntry* MaterialTable::probe(const Position& pos) { Entry* probe(const Position& pos, Table& entries, Endgames& endgames) {
Key key = pos.material_key(); Key key = pos.material_key();
MaterialEntry* e = entries[key]; Entry* e = entries[key];
// If e->key matches the position's material hash key, it means that we // If e->key matches the position's material hash key, it means that we
// have analysed this material configuration before, and we can simply // have analysed this material configuration before, and we can simply
@@ -100,10 +150,10 @@ MaterialEntry* MaterialTable::probe(const Position& pos) {
if (e->key == key) if (e->key == key)
return e; return e;
memset(e, 0, sizeof(MaterialEntry)); std::memset(e, 0, sizeof(Entry));
e->key = key; e->key = key;
e->factor[WHITE] = e->factor[BLACK] = (uint8_t)SCALE_FACTOR_NORMAL; e->factor[WHITE] = e->factor[BLACK] = (uint8_t)SCALE_FACTOR_NORMAL;
e->gamePhase = MaterialTable::game_phase(pos); e->gamePhase = game_phase(pos);
// Let's look if we have a specialized evaluation function for this // Let's look if we have a specialized evaluation function for this
// particular material configuration. First we look for a fixed // particular material configuration. First we look for a fixed
@@ -130,8 +180,8 @@ MaterialEntry* MaterialTable::probe(const Position& pos) {
assert((pos.pieces(WHITE, KNIGHT) | pos.pieces(WHITE, BISHOP))); assert((pos.pieces(WHITE, KNIGHT) | pos.pieces(WHITE, BISHOP)));
assert((pos.pieces(BLACK, KNIGHT) | pos.pieces(BLACK, BISHOP))); assert((pos.pieces(BLACK, KNIGHT) | pos.pieces(BLACK, BISHOP)));
if ( pos.piece_count(WHITE, BISHOP) + pos.piece_count(WHITE, KNIGHT) <= 2 if ( pos.count<BISHOP>(WHITE) + pos.count<KNIGHT>(WHITE) <= 2
&& pos.piece_count(BLACK, BISHOP) + pos.piece_count(BLACK, KNIGHT) <= 2) && pos.count<BISHOP>(BLACK) + pos.count<KNIGHT>(BLACK) <= 2)
{ {
e->evaluationFunction = &EvaluateKmmKm[pos.side_to_move()]; e->evaluationFunction = &EvaluateKmmKm[pos.side_to_move()];
return e; return e;
@@ -171,17 +221,17 @@ MaterialEntry* MaterialTable::probe(const Position& pos) {
if (npm_w + npm_b == VALUE_ZERO) if (npm_w + npm_b == VALUE_ZERO)
{ {
if (pos.piece_count(BLACK, PAWN) == 0) if (!pos.count<PAWN>(BLACK))
{ {
assert(pos.piece_count(WHITE, PAWN) >= 2); assert(pos.count<PAWN>(WHITE) >= 2);
e->scalingFunction[WHITE] = &ScaleKPsK[WHITE]; e->scalingFunction[WHITE] = &ScaleKPsK[WHITE];
} }
else if (pos.piece_count(WHITE, PAWN) == 0) else if (!pos.count<PAWN>(WHITE))
{ {
assert(pos.piece_count(BLACK, PAWN) >= 2); assert(pos.count<PAWN>(BLACK) >= 2);
e->scalingFunction[BLACK] = &ScaleKPsK[BLACK]; e->scalingFunction[BLACK] = &ScaleKPsK[BLACK];
} }
else if (pos.piece_count(WHITE, PAWN) == 1 && pos.piece_count(BLACK, PAWN) == 1) else if (pos.count<PAWN>(WHITE) == 1 && pos.count<PAWN>(BLACK) == 1)
{ {
// This is a special case because we set scaling functions // This is a special case because we set scaling functions
// for both colors instead of only one. // for both colors instead of only one.
@@ -191,82 +241,46 @@ MaterialEntry* MaterialTable::probe(const Position& pos) {
} }
// No pawns makes it difficult to win, even with a material advantage // No pawns makes it difficult to win, even with a material advantage
if (pos.piece_count(WHITE, PAWN) == 0 && npm_w - npm_b <= BishopValueMg) if (!pos.count<PAWN>(WHITE) && npm_w - npm_b <= BishopValueMg)
{ {
e->factor[WHITE] = (uint8_t) e->factor[WHITE] = (uint8_t)
(npm_w == npm_b || npm_w < RookValueMg ? 0 : NoPawnsSF[std::min(pos.piece_count(WHITE, BISHOP), 2)]); (npm_w == npm_b || npm_w < RookValueMg ? 0 : NoPawnsSF[std::min(pos.count<BISHOP>(WHITE), 2)]);
} }
if (pos.piece_count(BLACK, PAWN) == 0 && npm_b - npm_w <= BishopValueMg) if (!pos.count<PAWN>(BLACK) && npm_b - npm_w <= BishopValueMg)
{ {
e->factor[BLACK] = (uint8_t) e->factor[BLACK] = (uint8_t)
(npm_w == npm_b || npm_b < RookValueMg ? 0 : NoPawnsSF[std::min(pos.piece_count(BLACK, BISHOP), 2)]); (npm_w == npm_b || npm_b < RookValueMg ? 0 : NoPawnsSF[std::min(pos.count<BISHOP>(BLACK), 2)]);
} }
// Compute the space weight // Compute the space weight
if (npm_w + npm_b >= 2 * QueenValueMg + 4 * RookValueMg + 2 * KnightValueMg) if (npm_w + npm_b >= 2 * QueenValueMg + 4 * RookValueMg + 2 * KnightValueMg)
{ {
int minorPieceCount = pos.piece_count(WHITE, KNIGHT) + pos.piece_count(WHITE, BISHOP) int minorPieceCount = pos.count<KNIGHT>(WHITE) + pos.count<BISHOP>(WHITE)
+ pos.piece_count(BLACK, KNIGHT) + pos.piece_count(BLACK, BISHOP); + pos.count<KNIGHT>(BLACK) + pos.count<BISHOP>(BLACK);
e->spaceWeight = minorPieceCount * minorPieceCount; e->spaceWeight = make_score(minorPieceCount * minorPieceCount, 0);
} }
// Evaluate the material imbalance. We use PIECE_TYPE_NONE as a place holder // Evaluate the material imbalance. We use PIECE_TYPE_NONE as a place holder
// for the bishop pair "extended piece", this allow us to be more flexible // for the bishop pair "extended piece", this allow us to be more flexible
// in defining bishop pair bonuses. // in defining bishop pair bonuses.
const int pieceCount[2][8] = { const int pieceCount[COLOR_NB][PIECE_TYPE_NB] = {
{ pos.piece_count(WHITE, BISHOP) > 1, pos.piece_count(WHITE, PAWN), pos.piece_count(WHITE, KNIGHT), { pos.count<BISHOP>(WHITE) > 1, pos.count<PAWN>(WHITE), pos.count<KNIGHT>(WHITE),
pos.piece_count(WHITE, BISHOP) , pos.piece_count(WHITE, ROOK), pos.piece_count(WHITE, QUEEN) }, pos.count<BISHOP>(WHITE) , pos.count<ROOK>(WHITE), pos.count<QUEEN >(WHITE) },
{ pos.piece_count(BLACK, BISHOP) > 1, pos.piece_count(BLACK, PAWN), pos.piece_count(BLACK, KNIGHT), { pos.count<BISHOP>(BLACK) > 1, pos.count<PAWN>(BLACK), pos.count<KNIGHT>(BLACK),
pos.piece_count(BLACK, BISHOP) , pos.piece_count(BLACK, ROOK), pos.piece_count(BLACK, QUEEN) } }; pos.count<BISHOP>(BLACK) , pos.count<ROOK>(BLACK), pos.count<QUEEN >(BLACK) } };
e->value = (int16_t)((imbalance<WHITE>(pieceCount) - imbalance<BLACK>(pieceCount)) / 16); e->value = (int16_t)((imbalance<WHITE>(pieceCount) - imbalance<BLACK>(pieceCount)) / 16);
return e; return e;
} }
/// MaterialTable::imbalance() calculates imbalance comparing piece count of each /// Material::game_phase() calculates the phase given the current
/// piece type for both colors.
template<Color Us>
int MaterialTable::imbalance(const int pieceCount[][8]) {
const Color Them = (Us == WHITE ? BLACK : WHITE);
int pt1, pt2, pc, v;
int value = 0;
// Redundancy of major pieces, formula based on Kaufman's paper
// "The Evaluation of Material Imbalances in Chess"
if (pieceCount[Us][ROOK] > 0)
value -= RedundantRookPenalty * (pieceCount[Us][ROOK] - 1)
+ RedundantQueenPenalty * pieceCount[Us][QUEEN];
// Second-degree polynomial material imbalance by Tord Romstad
for (pt1 = NO_PIECE_TYPE; pt1 <= QUEEN; pt1++)
{
pc = pieceCount[Us][pt1];
if (!pc)
continue;
v = LinearCoefficients[pt1];
for (pt2 = NO_PIECE_TYPE; pt2 <= pt1; pt2++)
v += QuadraticCoefficientsSameColor[pt1][pt2] * pieceCount[Us][pt2]
+ QuadraticCoefficientsOppositeColor[pt1][pt2] * pieceCount[Them][pt2];
value += pc * v;
}
return value;
}
/// MaterialTable::game_phase() calculates the phase given the current
/// position. Because the phase is strictly a function of the material, it /// position. Because the phase is strictly a function of the material, it
/// is stored in MaterialEntry. /// is stored in MaterialEntry.
Phase MaterialTable::game_phase(const Position& pos) { Phase game_phase(const Position& pos) {
Value npm = pos.non_pawn_material(WHITE) + pos.non_pawn_material(BLACK); Value npm = pos.non_pawn_material(WHITE) + pos.non_pawn_material(BLACK);
@@ -274,3 +288,5 @@ Phase MaterialTable::game_phase(const Position& pos) {
: npm <= EndgameLimit ? PHASE_ENDGAME : npm <= EndgameLimit ? PHASE_ENDGAME
: Phase(((npm - EndgameLimit) * 128) / (MidgameLimit - EndgameLimit)); : Phase(((npm - EndgameLimit) * 128) / (MidgameLimit - EndgameLimit));
} }
} // namespace Material
+24 -67
View File
@@ -1,7 +1,7 @@
/* /*
Stockfish, a UCI chess playing engine derived from Glaurung 2.1 Stockfish, a UCI chess playing engine derived from Glaurung 2.1
Copyright (C) 2004-2008 Tord Romstad (Glaurung author) Copyright (C) 2004-2008 Tord Romstad (Glaurung author)
Copyright (C) 2008-2012 Marco Costalba, Joona Kiiski, Tord Romstad Copyright (C) 2008-2013 Marco Costalba, Joona Kiiski, Tord Romstad
Stockfish is free software: you can redistribute it and/or modify Stockfish is free software: you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by it under the terms of the GNU General Public License as published by
@@ -17,7 +17,7 @@
along with this program. If not, see <http://www.gnu.org/licenses/>. along with this program. If not, see <http://www.gnu.org/licenses/>.
*/ */
#if !defined(MATERIAL_H_INCLUDED) #ifndef MATERIAL_H_INCLUDED
#define MATERIAL_H_INCLUDED #define MATERIAL_H_INCLUDED
#include "endgame.h" #include "endgame.h"
@@ -25,96 +25,53 @@
#include "position.h" #include "position.h"
#include "types.h" #include "types.h"
const int MaterialTableSize = 8192; namespace Material {
/// Game phase /// Material::Entry contains various information about a material configuration.
enum Phase { /// It contains a material balance evaluation, a function pointer to a special
PHASE_ENDGAME = 0, /// endgame evaluation function (which in most cases is NULL, meaning that the
PHASE_MIDGAME = 128 /// standard evaluation function will be used), and "scale factors".
};
/// MaterialEntry is a class which contains various information about a
/// material configuration. It contains a material balance evaluation,
/// a function pointer to a special endgame evaluation function (which in
/// most cases is NULL, meaning that the standard evaluation function will
/// be used), and "scale factors" for black and white.
/// ///
/// The scale factors are used to scale the evaluation score up or down. /// The scale factors are used to scale the evaluation score up or down.
/// For instance, in KRB vs KR endgames, the score is scaled down by a factor /// For instance, in KRB vs KR endgames, the score is scaled down by a factor
/// of 4, which will result in scores of absolute value less than one pawn. /// of 4, which will result in scores of absolute value less than one pawn.
class MaterialEntry { struct Entry {
friend struct MaterialTable; Score material_value() const { return make_score(value, value); }
Score space_weight() const { return spaceWeight; }
public: Phase game_phase() const { return gamePhase; }
Score material_value() const; bool specialized_eval_exists() const { return evaluationFunction != NULL; }
Value evaluate(const Position& p) const { return (*evaluationFunction)(p); }
ScaleFactor scale_factor(const Position& pos, Color c) const; ScaleFactor scale_factor(const Position& pos, Color c) const;
int space_weight() const;
Phase game_phase() const;
bool specialized_eval_exists() const;
Value evaluate(const Position& pos) const;
private:
Key key; Key key;
int16_t value; int16_t value;
uint8_t factor[2]; uint8_t factor[COLOR_NB];
EndgameBase<Value>* evaluationFunction; EndgameBase<Value>* evaluationFunction;
EndgameBase<ScaleFactor>* scalingFunction[2]; EndgameBase<ScaleFactor>* scalingFunction[COLOR_NB];
int spaceWeight; Score spaceWeight;
Phase gamePhase; Phase gamePhase;
}; };
typedef HashTable<Entry, 8192> Table;
/// The MaterialTable class represents a material hash table. The most important Entry* probe(const Position& pos, Table& entries, Endgames& endgames);
/// method is probe(), which returns a pointer to a MaterialEntry object. Phase game_phase(const Position& pos);
struct MaterialTable { /// Material::scale_factor takes a position and a color as input, and
MaterialEntry* probe(const Position& pos);
static Phase game_phase(const Position& pos);
template<Color Us> static int imbalance(const int pieceCount[][8]);
HashTable<MaterialEntry, MaterialTableSize> entries;
Endgames endgames;
};
/// MaterialEntry::scale_factor takes a position and a color as input, and
/// returns a scale factor for the given color. We have to provide the /// returns a scale factor for the given color. We have to provide the
/// position in addition to the color, because the scale factor need not /// position in addition to the color, because the scale factor need not
/// to be a constant: It can also be a function which should be applied to /// to be a constant: It can also be a function which should be applied to
/// the position. For instance, in KBP vs K endgames, a scaling function /// the position. For instance, in KBP vs K endgames, a scaling function
/// which checks for draws with rook pawns and wrong-colored bishops. /// which checks for draws with rook pawns and wrong-colored bishops.
inline ScaleFactor MaterialEntry::scale_factor(const Position& pos, Color c) const { inline ScaleFactor Entry::scale_factor(const Position& pos, Color c) const {
if (!scalingFunction[c]) return !scalingFunction[c] || (*scalingFunction[c])(pos) == SCALE_FACTOR_NONE
return ScaleFactor(factor[c]); ? ScaleFactor(factor[c]) : (*scalingFunction[c])(pos);
ScaleFactor sf = (*scalingFunction[c])(pos);
return sf == SCALE_FACTOR_NONE ? ScaleFactor(factor[c]) : sf;
} }
inline Value MaterialEntry::evaluate(const Position& pos) const {
return (*evaluationFunction)(pos);
} }
inline Score MaterialEntry::material_value() const { #endif // #ifndef MATERIAL_H_INCLUDED
return make_score(value, value);
}
inline int MaterialEntry::space_weight() const {
return spaceWeight;
}
inline Phase MaterialEntry::game_phase() const {
return gamePhase;
}
inline bool MaterialEntry::specialized_eval_exists() const {
return evaluationFunction != NULL;
}
#endif // !defined(MATERIAL_H_INCLUDED)
+21 -61
View File
@@ -1,7 +1,7 @@
/* /*
Stockfish, a UCI chess playing engine derived from Glaurung 2.1 Stockfish, a UCI chess playing engine derived from Glaurung 2.1
Copyright (C) 2004-2008 Tord Romstad (Glaurung author) Copyright (C) 2004-2008 Tord Romstad (Glaurung author)
Copyright (C) 2008-2012 Marco Costalba, Joona Kiiski, Tord Romstad Copyright (C) 2008-2013 Marco Costalba, Joona Kiiski, Tord Romstad
Stockfish is free software: you can redistribute it and/or modify Stockfish is free software: you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by it under the terms of the GNU General Public License as published by
@@ -24,57 +24,41 @@
#include "misc.h" #include "misc.h"
#include "thread.h" #include "thread.h"
#if defined(__hpux)
# include <sys/pstat.h>
#endif
using namespace std; using namespace std;
/// Version number. If Version is left empty, then Tag plus current /// Version number. If Version is left empty, then compile date, in the
/// date (in the format YYMMDD) is used as a version number. /// format DD-MM-YY, is shown in engine_info.
static const string Version = "4";
static const string Version = "2.3";
static const string Tag = "";
/// engine_info() returns the full name of the current Stockfish version. /// engine_info() returns the full name of the current Stockfish version. This
/// This will be either "Stockfish YYMMDD" (where YYMMDD is the date when /// will be either "Stockfish <Tag> DD-MM-YY" (where DD-MM-YY is the date when
/// the program was compiled) or "Stockfish <version number>", depending /// the program was compiled) or "Stockfish <Version>", depending on whether
/// on whether Version is empty. /// Version is empty.
const string engine_info(bool to_uci) { const string engine_info(bool to_uci) {
const string months("Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec"); const string months("Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec");
const string cpu64(Is64Bit ? " 64bit" : "");
const string popcnt(HasPopCnt ? " SSE4.2" : "");
string month, day, year; string month, day, year;
stringstream s, date(__DATE__); // From compiler, format is "Sep 21 2008" stringstream s, date(__DATE__); // From compiler, format is "Sep 21 2008"
s << "Stockfish " << Version; s << "Stockfish " << Version << setfill('0');
if (Version.empty()) if (Version.empty())
{ {
date >> month >> day >> year; date >> month >> day >> year;
s << setw(2) << day << setw(2) << (1 + months.find(month) / 4) << year.substr(2);
s << Tag << setfill('0') << " " << year.substr(2)
<< setw(2) << (1 + months.find(month) / 4) << setw(2) << day;
} }
s << cpu64 << popcnt << (to_uci ? "\nid author ": " by ") s << (Is64Bit ? " 64" : "")
<< (HasPopCnt ? " SSE4.2" : "")
<< (to_uci ? "\nid author ": " by ")
<< "Tord Romstad, Marco Costalba and Joona Kiiski"; << "Tord Romstad, Marco Costalba and Joona Kiiski";
return s.str(); return s.str();
} }
/// Convert system time to milliseconds. That's all we need.
Time::point Time::now() {
sys_time_t t; system_time(&t); return time_to_msec(t);
}
/// Debug functions used mainly to collect run-time statistics /// Debug functions used mainly to collect run-time statistics
static uint64_t hits[2], means[2]; static uint64_t hits[2], means[2];
@@ -174,37 +158,12 @@ std::ostream& operator<<(std::ostream& os, SyncCout sc) {
void start_logger(bool b) { Logger::start(b); } void start_logger(bool b) { Logger::start(b); }
/// cpu_count() tries to detect the number of CPU cores
int cpu_count() {
#if defined(_WIN32) || defined(_WIN64)
SYSTEM_INFO s;
GetSystemInfo(&s);
return s.dwNumberOfProcessors;
#else
# if defined(_SC_NPROCESSORS_ONLN)
return sysconf(_SC_NPROCESSORS_ONLN);
# elif defined(__hpux)
struct pst_dynamic psd;
if (pstat_getdynamic(&psd, sizeof(psd), (size_t)1, 0) == -1)
return 1;
return psd.psd_proc_cnt;
# else
return 1;
# endif
#endif
}
/// timed_wait() waits for msec milliseconds. It is mainly an helper to wrap /// timed_wait() waits for msec milliseconds. It is mainly an helper to wrap
/// conversion from milliseconds to struct timespec, as used by pthreads. /// conversion from milliseconds to struct timespec, as used by pthreads.
void timed_wait(WaitCondition& sleepCond, Lock& sleepLock, int msec) { void timed_wait(WaitCondition& sleepCond, Lock& sleepLock, int msec) {
#if defined(_WIN32) || defined(_WIN64) #ifdef _WIN32
int tm = msec; int tm = msec;
#else #else
timespec ts, *tm = &ts; timespec ts, *tm = &ts;
@@ -221,24 +180,25 @@ void timed_wait(WaitCondition& sleepCond, Lock& sleepLock, int msec) {
/// prefetch() preloads the given address in L1/L2 cache. This is a non /// prefetch() preloads the given address in L1/L2 cache. This is a non
/// blocking function and do not stalls the CPU waiting for data to be /// blocking function and do not stalls the CPU waiting for data to be
/// loaded from memory, that can be quite slow. /// loaded from memory, that can be quite slow.
#if defined(NO_PREFETCH) #ifdef NO_PREFETCH
void prefetch(char*) {} void prefetch(char*) {}
#else #else
# include <xmmintrin.h>
void prefetch(char* addr) { void prefetch(char* addr) {
# if defined(__INTEL_COMPILER) || defined(__ICL) # if defined(__INTEL_COMPILER)
// This hack prevents prefetches to be optimized away by // This hack prevents prefetches to be optimized away by
// Intel compiler. Both MSVC and gcc seems not affected. // Intel compiler. Both MSVC and gcc seems not affected.
__asm__ (""); __asm__ ("");
# endif # endif
_mm_prefetch(addr, _MM_HINT_T2); # if defined(__INTEL_COMPILER) || defined(_MSC_VER)
_mm_prefetch(addr+64, _MM_HINT_T2); // 64 bytes ahead _mm_prefetch(addr, _MM_HINT_T0);
# else
__builtin_prefetch(addr);
# endif
} }
#endif #endif
+4 -5
View File
@@ -1,7 +1,7 @@
/* /*
Stockfish, a UCI chess playing engine derived from Glaurung 2.1 Stockfish, a UCI chess playing engine derived from Glaurung 2.1
Copyright (C) 2004-2008 Tord Romstad (Glaurung author) Copyright (C) 2004-2008 Tord Romstad (Glaurung author)
Copyright (C) 2008-2012 Marco Costalba, Joona Kiiski, Tord Romstad Copyright (C) 2008-2013 Marco Costalba, Joona Kiiski, Tord Romstad
Stockfish is free software: you can redistribute it and/or modify Stockfish is free software: you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by it under the terms of the GNU General Public License as published by
@@ -17,7 +17,7 @@
along with this program. If not, see <http://www.gnu.org/licenses/>. along with this program. If not, see <http://www.gnu.org/licenses/>.
*/ */
#if !defined(MISC_H_INCLUDED) #ifndef MISC_H_INCLUDED
#define MISC_H_INCLUDED #define MISC_H_INCLUDED
#include <fstream> #include <fstream>
@@ -27,7 +27,6 @@
#include "types.h" #include "types.h"
extern const std::string engine_info(bool to_uci = false); extern const std::string engine_info(bool to_uci = false);
extern int cpu_count();
extern void timed_wait(WaitCondition&, Lock&, int); extern void timed_wait(WaitCondition&, Lock&, int);
extern void prefetch(char* addr); extern void prefetch(char* addr);
extern void start_logger(bool b); extern void start_logger(bool b);
@@ -46,7 +45,7 @@ struct Log : public std::ofstream {
namespace Time { namespace Time {
typedef int64_t point; typedef int64_t point;
point now(); inline point now() { return system_time_to_msec(); }
} }
@@ -66,4 +65,4 @@ std::ostream& operator<<(std::ostream&, SyncCout);
#define sync_cout std::cout << io_lock #define sync_cout std::cout << io_lock
#define sync_endl std::endl << io_unlock #define sync_endl std::endl << io_unlock
#endif // !defined(MISC_H_INCLUDED) #endif // #ifndef MISC_H_INCLUDED
+90 -102
View File
@@ -1,7 +1,7 @@
/* /*
Stockfish, a UCI chess playing engine derived from Glaurung 2.1 Stockfish, a UCI chess playing engine derived from Glaurung 2.1
Copyright (C) 2004-2008 Tord Romstad (Glaurung author) Copyright (C) 2004-2008 Tord Romstad (Glaurung author)
Copyright (C) 2008-2012 Marco Costalba, Joona Kiiski, Tord Romstad Copyright (C) 2008-2013 Marco Costalba, Joona Kiiski, Tord Romstad
Stockfish is free software: you can redistribute it and/or modify Stockfish is free software: you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by it under the terms of the GNU General Public License as published by
@@ -24,15 +24,15 @@
/// Simple macro to wrap a very common while loop, no facny, no flexibility, /// Simple macro to wrap a very common while loop, no facny, no flexibility,
/// hardcoded names 'mlist' and 'from'. /// hardcoded names 'mlist' and 'from'.
#define SERIALIZE(b) while (b) (*mlist++).move = make_move(from, pop_lsb(&b)) #define SERIALIZE(b) while (b) (mlist++)->move = make_move(from, pop_lsb(&b))
/// Version used for pawns, where the 'from' square is given as a delta from the 'to' square /// Version used for pawns, where the 'from' square is given as a delta from the 'to' square
#define SERIALIZE_PAWNS(b, d) while (b) { Square to = pop_lsb(&b); \ #define SERIALIZE_PAWNS(b, d) while (b) { Square to = pop_lsb(&b); \
(*mlist++).move = make_move(to - (d), to); } (mlist++)->move = make_move(to - (d), to); }
namespace { namespace {
template<CastlingSide Side, bool Checks> template<CastlingSide Side, bool Checks, bool Chess960>
MoveStack* generate_castle(const Position& pos, MoveStack* mlist, Color us) { ExtMove* generate_castle(const Position& pos, ExtMove* mlist, Color us) {
if (pos.castle_impeded(us, Side) || !pos.can_castle(make_castle_right(us, Side))) if (pos.castle_impeded(us, Side) || !pos.can_castle(make_castle_right(us, Side)))
return mlist; return mlist;
@@ -44,20 +44,22 @@ namespace {
Square kto = relative_square(us, Side == KING_SIDE ? SQ_G1 : SQ_C1); Square kto = relative_square(us, Side == KING_SIDE ? SQ_G1 : SQ_C1);
Bitboard enemies = pos.pieces(~us); Bitboard enemies = pos.pieces(~us);
assert(!pos.in_check()); assert(!pos.checkers());
for (Square s = kto; s != kfrom; s += (Square)(Side == KING_SIDE ? -1 : 1)) const int K = Chess960 ? kto > kfrom ? -1 : 1
: Side == KING_SIDE ? -1 : 1;
for (Square s = kto; s != kfrom; s += (Square)K)
if (pos.attackers_to(s) & enemies) if (pos.attackers_to(s) & enemies)
return mlist; return mlist;
// Because we generate only legal castling moves we need to verify that // Because we generate only legal castling moves we need to verify that
// when moving the castling rook we do not discover some hidden checker. // when moving the castling rook we do not discover some hidden checker.
// For instance an enemy queen in SQ_A1 when castling rook is in SQ_B1. // For instance an enemy queen in SQ_A1 when castling rook is in SQ_B1.
if ( pos.is_chess960() if (Chess960 && (pos.attackers_to(kto, pos.pieces() ^ rfrom) & enemies))
&& (pos.attackers_to(kto, pos.pieces() ^ rfrom) & enemies))
return mlist; return mlist;
(*mlist++).move = make<CASTLE>(kfrom, rfrom); (mlist++)->move = make<CASTLE>(kfrom, rfrom);
if (Checks && !pos.move_gives_check((mlist - 1)->move, CheckInfo(pos))) if (Checks && !pos.move_gives_check((mlist - 1)->move, CheckInfo(pos)))
mlist--; mlist--;
@@ -66,42 +68,30 @@ namespace {
} }
template<Square Delta>
inline Bitboard move_pawns(Bitboard p) {
return Delta == DELTA_N ? p << 8
: Delta == DELTA_S ? p >> 8
: Delta == DELTA_NE ? (p & ~FileHBB) << 9
: Delta == DELTA_SE ? (p & ~FileHBB) >> 7
: Delta == DELTA_NW ? (p & ~FileABB) << 7
: Delta == DELTA_SW ? (p & ~FileABB) >> 9 : 0;
}
template<GenType Type, Square Delta> template<GenType Type, Square Delta>
inline MoveStack* generate_promotions(MoveStack* mlist, Bitboard pawnsOn7, inline ExtMove* generate_promotions(ExtMove* mlist, Bitboard pawnsOn7,
Bitboard target, const CheckInfo* ci) { Bitboard target, const CheckInfo* ci) {
Bitboard b = move_pawns<Delta>(pawnsOn7) & target; Bitboard b = shift_bb<Delta>(pawnsOn7) & target;
while (b) while (b)
{ {
Square to = pop_lsb(&b); Square to = pop_lsb(&b);
if (Type == CAPTURES || Type == EVASIONS || Type == NON_EVASIONS) if (Type == CAPTURES || Type == EVASIONS || Type == NON_EVASIONS)
(*mlist++).move = make<PROMOTION>(to - Delta, to, QUEEN); (mlist++)->move = make<PROMOTION>(to - Delta, to, QUEEN);
if (Type == QUIETS || Type == EVASIONS || Type == NON_EVASIONS) if (Type == QUIETS || Type == EVASIONS || Type == NON_EVASIONS)
{ {
(*mlist++).move = make<PROMOTION>(to - Delta, to, ROOK); (mlist++)->move = make<PROMOTION>(to - Delta, to, ROOK);
(*mlist++).move = make<PROMOTION>(to - Delta, to, BISHOP); (mlist++)->move = make<PROMOTION>(to - Delta, to, BISHOP);
(*mlist++).move = make<PROMOTION>(to - Delta, to, KNIGHT); (mlist++)->move = make<PROMOTION>(to - Delta, to, KNIGHT);
} }
// Knight-promotion is the only one that can give a direct check not // Knight-promotion is the only one that can give a direct check not
// already included in the queen-promotion. // already included in the queen-promotion.
if (Type == QUIET_CHECKS && (StepAttacksBB[W_KNIGHT][to] & ci->ksq)) if (Type == QUIET_CHECKS && (StepAttacksBB[W_KNIGHT][to] & ci->ksq))
(*mlist++).move = make<PROMOTION>(to - Delta, to, KNIGHT); (mlist++)->move = make<PROMOTION>(to - Delta, to, KNIGHT);
else else
(void)ci; // Silence a warning under MSVC (void)ci; // Silence a warning under MSVC
} }
@@ -111,7 +101,7 @@ namespace {
template<Color Us, GenType Type> template<Color Us, GenType Type>
MoveStack* generate_pawn_moves(const Position& pos, MoveStack* mlist, ExtMove* generate_pawn_moves(const Position& pos, ExtMove* mlist,
Bitboard target, const CheckInfo* ci) { Bitboard target, const CheckInfo* ci) {
// Compute our parametrized parameters at compile time, named according to // Compute our parametrized parameters at compile time, named according to
@@ -120,9 +110,9 @@ namespace {
const Bitboard TRank8BB = (Us == WHITE ? Rank8BB : Rank1BB); const Bitboard TRank8BB = (Us == WHITE ? Rank8BB : Rank1BB);
const Bitboard TRank7BB = (Us == WHITE ? Rank7BB : Rank2BB); const Bitboard TRank7BB = (Us == WHITE ? Rank7BB : Rank2BB);
const Bitboard TRank3BB = (Us == WHITE ? Rank3BB : Rank6BB); const Bitboard TRank3BB = (Us == WHITE ? Rank3BB : Rank6BB);
const Square UP = (Us == WHITE ? DELTA_N : DELTA_S); const Square Up = (Us == WHITE ? DELTA_N : DELTA_S);
const Square RIGHT = (Us == WHITE ? DELTA_NE : DELTA_SW); const Square Right = (Us == WHITE ? DELTA_NE : DELTA_SW);
const Square LEFT = (Us == WHITE ? DELTA_NW : DELTA_SE); const Square Left = (Us == WHITE ? DELTA_NW : DELTA_SE);
Bitboard b1, b2, dc1, dc2, emptySquares; Bitboard b1, b2, dc1, dc2, emptySquares;
@@ -137,8 +127,8 @@ namespace {
{ {
emptySquares = (Type == QUIETS || Type == QUIET_CHECKS ? target : ~pos.pieces()); emptySquares = (Type == QUIETS || Type == QUIET_CHECKS ? target : ~pos.pieces());
b1 = move_pawns<UP>(pawnsNotOn7) & emptySquares; b1 = shift_bb<Up>(pawnsNotOn7) & emptySquares;
b2 = move_pawns<UP>(b1 & TRank3BB) & emptySquares; b2 = shift_bb<Up>(b1 & TRank3BB) & emptySquares;
if (Type == EVASIONS) // Consider only blocking squares if (Type == EVASIONS) // Consider only blocking squares
{ {
@@ -157,16 +147,16 @@ namespace {
// promotion has been already generated among captures. // promotion has been already generated among captures.
if (pawnsNotOn7 & ci->dcCandidates) if (pawnsNotOn7 & ci->dcCandidates)
{ {
dc1 = move_pawns<UP>(pawnsNotOn7 & ci->dcCandidates) & emptySquares & ~file_bb(ci->ksq); dc1 = shift_bb<Up>(pawnsNotOn7 & ci->dcCandidates) & emptySquares & ~file_bb(ci->ksq);
dc2 = move_pawns<UP>(dc1 & TRank3BB) & emptySquares; dc2 = shift_bb<Up>(dc1 & TRank3BB) & emptySquares;
b1 |= dc1; b1 |= dc1;
b2 |= dc2; b2 |= dc2;
} }
} }
SERIALIZE_PAWNS(b1, UP); SERIALIZE_PAWNS(b1, Up);
SERIALIZE_PAWNS(b2, UP + UP); SERIALIZE_PAWNS(b2, Up + Up);
} }
// Promotions and underpromotions // Promotions and underpromotions
@@ -178,19 +168,19 @@ namespace {
if (Type == EVASIONS) if (Type == EVASIONS)
emptySquares &= target; emptySquares &= target;
mlist = generate_promotions<Type, RIGHT>(mlist, pawnsOn7, enemies, ci); mlist = generate_promotions<Type, Right>(mlist, pawnsOn7, enemies, ci);
mlist = generate_promotions<Type, LEFT>(mlist, pawnsOn7, enemies, ci); mlist = generate_promotions<Type, Left >(mlist, pawnsOn7, enemies, ci);
mlist = generate_promotions<Type, UP>(mlist, pawnsOn7, emptySquares, ci); mlist = generate_promotions<Type, Up>(mlist, pawnsOn7, emptySquares, ci);
} }
// Standard and en-passant captures // Standard and en-passant captures
if (Type == CAPTURES || Type == EVASIONS || Type == NON_EVASIONS) if (Type == CAPTURES || Type == EVASIONS || Type == NON_EVASIONS)
{ {
b1 = move_pawns<RIGHT>(pawnsNotOn7) & enemies; b1 = shift_bb<Right>(pawnsNotOn7) & enemies;
b2 = move_pawns<LEFT >(pawnsNotOn7) & enemies; b2 = shift_bb<Left >(pawnsNotOn7) & enemies;
SERIALIZE_PAWNS(b1, RIGHT); SERIALIZE_PAWNS(b1, Right);
SERIALIZE_PAWNS(b2, LEFT); SERIALIZE_PAWNS(b2, Left);
if (pos.ep_square() != SQ_NONE) if (pos.ep_square() != SQ_NONE)
{ {
@@ -199,7 +189,7 @@ namespace {
// An en passant capture can be an evasion only if the checking piece // An en passant capture can be an evasion only if the checking piece
// is the double pushed pawn and so is in the target. Otherwise this // is the double pushed pawn and so is in the target. Otherwise this
// is a discovery check and we are forced to do otherwise. // is a discovery check and we are forced to do otherwise.
if (Type == EVASIONS && !(target & (pos.ep_square() - UP))) if (Type == EVASIONS && !(target & (pos.ep_square() - Up)))
return mlist; return mlist;
b1 = pawnsNotOn7 & pos.attacks_from<PAWN>(pos.ep_square(), Them); b1 = pawnsNotOn7 & pos.attacks_from<PAWN>(pos.ep_square(), Them);
@@ -207,7 +197,7 @@ namespace {
assert(b1); assert(b1);
while (b1) while (b1)
(*mlist++).move = make<ENPASSANT>(pop_lsb(&b1), pos.ep_square()); (mlist++)->move = make<ENPASSANT>(pop_lsb(&b1), pos.ep_square());
} }
} }
@@ -216,12 +206,12 @@ namespace {
template<PieceType Pt, bool Checks> FORCE_INLINE template<PieceType Pt, bool Checks> FORCE_INLINE
MoveStack* generate_moves(const Position& pos, MoveStack* mlist, Color us, ExtMove* generate_moves(const Position& pos, ExtMove* mlist, Color us,
Bitboard target, const CheckInfo* ci) { Bitboard target, const CheckInfo* ci) {
assert(Pt != KING && Pt != PAWN); assert(Pt != KING && Pt != PAWN);
const Square* pl = pos.piece_list(us, Pt); const Square* pl = pos.list<Pt>(us);
for (Square from = *pl; from != SQ_NONE; from = *++pl) for (Square from = *pl; from != SQ_NONE; from = *++pl)
{ {
@@ -231,7 +221,7 @@ namespace {
&& !(PseudoAttacks[Pt][from] & target & ci->checkSq[Pt])) && !(PseudoAttacks[Pt][from] & target & ci->checkSq[Pt]))
continue; continue;
if (ci->dcCandidates && (ci->dcCandidates & from)) if (unlikely(ci->dcCandidates) && (ci->dcCandidates & from))
continue; continue;
} }
@@ -247,34 +237,37 @@ namespace {
} }
FORCE_INLINE MoveStack* generate_king_moves(const Position& pos, MoveStack* mlist, template<Color Us, GenType Type> FORCE_INLINE
Color us, Bitboard target) { ExtMove* generate_all(const Position& pos, ExtMove* mlist, Bitboard target,
Square from = pos.king_square(us); const CheckInfo* ci = NULL) {
Bitboard b = pos.attacks_from<KING>(from) & target;
SERIALIZE(b);
return mlist;
}
const bool Checks = Type == QUIET_CHECKS;
template<GenType Type> FORCE_INLINE mlist = generate_pawn_moves<Us, Type>(pos, mlist, target, ci);
MoveStack* generate_all_moves(const Position& pos, MoveStack* mlist, Color us, mlist = generate_moves<KNIGHT, Checks>(pos, mlist, Us, target, ci);
Bitboard target, const CheckInfo* ci = NULL) { mlist = generate_moves<BISHOP, Checks>(pos, mlist, Us, target, ci);
mlist = generate_moves< ROOK, Checks>(pos, mlist, Us, target, ci);
mlist = (us == WHITE ? generate_pawn_moves<WHITE, Type>(pos, mlist, target, ci) mlist = generate_moves< QUEEN, Checks>(pos, mlist, Us, target, ci);
: generate_pawn_moves<BLACK, Type>(pos, mlist, target, ci));
mlist = generate_moves<KNIGHT, Type == QUIET_CHECKS>(pos, mlist, us, target, ci);
mlist = generate_moves<BISHOP, Type == QUIET_CHECKS>(pos, mlist, us, target, ci);
mlist = generate_moves<ROOK, Type == QUIET_CHECKS>(pos, mlist, us, target, ci);
mlist = generate_moves<QUEEN, Type == QUIET_CHECKS>(pos, mlist, us, target, ci);
if (Type != QUIET_CHECKS && Type != EVASIONS) if (Type != QUIET_CHECKS && Type != EVASIONS)
mlist = generate_king_moves(pos, mlist, us, target);
if (Type != CAPTURES && Type != EVASIONS && pos.can_castle(us))
{ {
mlist = generate_castle<KING_SIDE, Type == QUIET_CHECKS>(pos, mlist, us); Square from = pos.king_square(Us);
mlist = generate_castle<QUEEN_SIDE, Type == QUIET_CHECKS>(pos, mlist, us); Bitboard b = pos.attacks_from<KING>(from) & target;
SERIALIZE(b);
}
if (Type != CAPTURES && Type != EVASIONS && pos.can_castle(Us))
{
if (pos.is_chess960())
{
mlist = generate_castle< KING_SIDE, Checks, true>(pos, mlist, Us);
mlist = generate_castle<QUEEN_SIDE, Checks, true>(pos, mlist, Us);
}
else
{
mlist = generate_castle< KING_SIDE, Checks, false>(pos, mlist, Us);
mlist = generate_castle<QUEEN_SIDE, Checks, false>(pos, mlist, Us);
}
} }
return mlist; return mlist;
@@ -294,38 +287,33 @@ namespace {
/// non-captures. Returns a pointer to the end of the move list. /// non-captures. Returns a pointer to the end of the move list.
template<GenType Type> template<GenType Type>
MoveStack* generate(const Position& pos, MoveStack* mlist) { ExtMove* generate(const Position& pos, ExtMove* mlist) {
assert(Type == CAPTURES || Type == QUIETS || Type == NON_EVASIONS); assert(Type == CAPTURES || Type == QUIETS || Type == NON_EVASIONS);
assert(!pos.in_check()); assert(!pos.checkers());
Color us = pos.side_to_move(); Color us = pos.side_to_move();
Bitboard target;
if (Type == CAPTURES) Bitboard target = Type == CAPTURES ? pos.pieces(~us)
target = pos.pieces(~us); : Type == QUIETS ? ~pos.pieces()
: Type == NON_EVASIONS ? ~pos.pieces(us) : 0;
else if (Type == QUIETS) return us == WHITE ? generate_all<WHITE, Type>(pos, mlist, target)
target = ~pos.pieces(); : generate_all<BLACK, Type>(pos, mlist, target);
else if (Type == NON_EVASIONS)
target = ~pos.pieces(us);
return generate_all_moves<Type>(pos, mlist, us, target);
} }
// Explicit template instantiations // Explicit template instantiations
template MoveStack* generate<CAPTURES>(const Position&, MoveStack*); template ExtMove* generate<CAPTURES>(const Position&, ExtMove*);
template MoveStack* generate<QUIETS>(const Position&, MoveStack*); template ExtMove* generate<QUIETS>(const Position&, ExtMove*);
template MoveStack* generate<NON_EVASIONS>(const Position&, MoveStack*); template ExtMove* generate<NON_EVASIONS>(const Position&, ExtMove*);
/// generate<QUIET_CHECKS> generates all pseudo-legal non-captures and knight /// generate<QUIET_CHECKS> generates all pseudo-legal non-captures and knight
/// underpromotions that give check. Returns a pointer to the end of the move list. /// underpromotions that give check. Returns a pointer to the end of the move list.
template<> template<>
MoveStack* generate<QUIET_CHECKS>(const Position& pos, MoveStack* mlist) { ExtMove* generate<QUIET_CHECKS>(const Position& pos, ExtMove* mlist) {
assert(!pos.in_check()); assert(!pos.checkers());
Color us = pos.side_to_move(); Color us = pos.side_to_move();
CheckInfo ci(pos); CheckInfo ci(pos);
@@ -347,21 +335,21 @@ MoveStack* generate<QUIET_CHECKS>(const Position& pos, MoveStack* mlist) {
SERIALIZE(b); SERIALIZE(b);
} }
return generate_all_moves<QUIET_CHECKS>(pos, mlist, us, ~pos.pieces(), &ci); return us == WHITE ? generate_all<WHITE, QUIET_CHECKS>(pos, mlist, ~pos.pieces(), &ci)
: generate_all<BLACK, QUIET_CHECKS>(pos, mlist, ~pos.pieces(), &ci);
} }
/// generate<EVASIONS> generates all pseudo-legal check evasions when the side /// generate<EVASIONS> generates all pseudo-legal check evasions when the side
/// to move is in check. Returns a pointer to the end of the move list. /// to move is in check. Returns a pointer to the end of the move list.
template<> template<>
MoveStack* generate<EVASIONS>(const Position& pos, MoveStack* mlist) { ExtMove* generate<EVASIONS>(const Position& pos, ExtMove* mlist) {
assert(pos.in_check()); assert(pos.checkers());
Square from, checksq;
int checkersCnt = 0; int checkersCnt = 0;
Color us = pos.side_to_move(); Color us = pos.side_to_move();
Square ksq = pos.king_square(us); Square ksq = pos.king_square(us), from = ksq /* For SERIALIZE */, checksq;
Bitboard sliderAttacks = 0; Bitboard sliderAttacks = 0;
Bitboard b = pos.checkers(); Bitboard b = pos.checkers();
@@ -400,29 +388,29 @@ MoveStack* generate<EVASIONS>(const Position& pos, MoveStack* mlist) {
// Generate evasions for king, capture and non capture moves // Generate evasions for king, capture and non capture moves
b = pos.attacks_from<KING>(ksq) & ~pos.pieces(us) & ~sliderAttacks; b = pos.attacks_from<KING>(ksq) & ~pos.pieces(us) & ~sliderAttacks;
from = ksq;
SERIALIZE(b); SERIALIZE(b);
if (checkersCnt > 1) if (checkersCnt > 1)
return mlist; // Double check, only a king move can save the day return mlist; // Double check, only a king move can save the day
// Generate blocking evasions or captures of the checking piece // Generate blocking evasions or captures of the checking piece
Bitboard target = between_bb(checksq, ksq) | pos.checkers(); Bitboard target = between_bb(checksq, ksq) | checksq;
return generate_all_moves<EVASIONS>(pos, mlist, us, target); return us == WHITE ? generate_all<WHITE, EVASIONS>(pos, mlist, target)
: generate_all<BLACK, EVASIONS>(pos, mlist, target);
} }
/// generate<LEGAL> generates all the legal moves in the given position /// generate<LEGAL> generates all the legal moves in the given position
template<> template<>
MoveStack* generate<LEGAL>(const Position& pos, MoveStack* mlist) { ExtMove* generate<LEGAL>(const Position& pos, ExtMove* mlist) {
MoveStack *end, *cur = mlist; ExtMove *end, *cur = mlist;
Bitboard pinned = pos.pinned_pieces(); Bitboard pinned = pos.pinned_pieces();
Square ksq = pos.king_square(pos.side_to_move()); Square ksq = pos.king_square(pos.side_to_move());
end = pos.in_check() ? generate<EVASIONS>(pos, mlist) end = pos.checkers() ? generate<EVASIONS>(pos, mlist)
: generate<NON_EVASIONS>(pos, mlist); : generate<NON_EVASIONS>(pos, mlist);
while (cur != end) while (cur != end)
if ( (pinned || from_sq(cur->move) == ksq || type_of(cur->move) == ENPASSANT) if ( (pinned || from_sq(cur->move) == ksq || type_of(cur->move) == ENPASSANT)
+12 -9
View File
@@ -1,7 +1,7 @@
/* /*
Stockfish, a UCI chess playing engine derived from Glaurung 2.1 Stockfish, a UCI chess playing engine derived from Glaurung 2.1
Copyright (C) 2004-2008 Tord Romstad (Glaurung author) Copyright (C) 2004-2008 Tord Romstad (Glaurung author)
Copyright (C) 2008-2012 Marco Costalba, Joona Kiiski, Tord Romstad Copyright (C) 2008-2013 Marco Costalba, Joona Kiiski, Tord Romstad
Stockfish is free software: you can redistribute it and/or modify Stockfish is free software: you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by it under the terms of the GNU General Public License as published by
@@ -17,7 +17,7 @@
along with this program. If not, see <http://www.gnu.org/licenses/>. along with this program. If not, see <http://www.gnu.org/licenses/>.
*/ */
#if !defined(MOVEGEN_H_INCLUDED) #ifndef MOVEGEN_H_INCLUDED
#define MOVEGEN_H_INCLUDED #define MOVEGEN_H_INCLUDED
#include "types.h" #include "types.h"
@@ -34,22 +34,25 @@ enum GenType {
class Position; class Position;
template<GenType> template<GenType>
MoveStack* generate(const Position& pos, MoveStack* mlist); ExtMove* generate(const Position& pos, ExtMove* mlist);
/// The MoveList struct is a simple wrapper around generate(), sometimes comes /// The MoveList struct is a simple wrapper around generate(), sometimes comes
/// handy to use this class instead of the low level generate() function. /// handy to use this class instead of the low level generate() function.
template<GenType T> template<GenType T>
struct MoveList { struct MoveList {
explicit MoveList(const Position& pos) : cur(mlist), last(generate<T>(pos, mlist)) {} explicit MoveList(const Position& pos) : cur(mlist), last(generate<T>(pos, mlist)) { last->move = MOVE_NONE; }
void operator++() { cur++; } void operator++() { cur++; }
bool end() const { return cur == last; } Move operator*() const { return cur->move; }
Move move() const { return cur->move; }
size_t size() const { return last - mlist; } size_t size() const { return last - mlist; }
bool contains(Move m) const {
for (const ExtMove* it(mlist); it != last; ++it) if (it->move == m) return true;
return false;
}
private: private:
MoveStack mlist[MAX_MOVES]; ExtMove mlist[MAX_MOVES];
MoveStack *cur, *last; ExtMove *cur, *last;
}; };
#endif // !defined(MOVEGEN_H_INCLUDED) #endif // #ifndef MOVEGEN_H_INCLUDED
+92 -79
View File
@@ -1,7 +1,7 @@
/* /*
Stockfish, a UCI chess playing engine derived from Glaurung 2.1 Stockfish, a UCI chess playing engine derived from Glaurung 2.1
Copyright (C) 2004-2008 Tord Romstad (Glaurung author) Copyright (C) 2004-2008 Tord Romstad (Glaurung author)
Copyright (C) 2008-2012 Marco Costalba, Joona Kiiski, Tord Romstad Copyright (C) 2008-2013 Marco Costalba, Joona Kiiski, Tord Romstad
Stockfish is free software: you can redistribute it and/or modify Stockfish is free software: you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by it under the terms of the GNU General Public License as published by
@@ -18,16 +18,14 @@
along with this program. If not, see <http://www.gnu.org/licenses/>. along with this program. If not, see <http://www.gnu.org/licenses/>.
*/ */
#include <algorithm>
#include <cassert> #include <cassert>
#include "movegen.h"
#include "movepick.h" #include "movepick.h"
#include "thread.h" #include "thread.h"
namespace { namespace {
enum Sequencer { enum Stages {
MAIN_SEARCH, CAPTURES_S1, KILLERS_S1, QUIETS_1_S1, QUIETS_2_S1, BAD_CAPTURES_S1, MAIN_SEARCH, CAPTURES_S1, KILLERS_S1, QUIETS_1_S1, QUIETS_2_S1, BAD_CAPTURES_S1,
EVASION, EVASIONS_S2, EVASION, EVASIONS_S2,
QSEARCH_0, CAPTURES_S3, QUIET_CHECKS_S3, QSEARCH_0, CAPTURES_S3, QUIET_CHECKS_S3,
@@ -37,14 +35,28 @@ namespace {
STOP STOP
}; };
// Our insertion sort, guaranteed to be stable, as is needed
void insertion_sort(ExtMove* begin, ExtMove* end)
{
ExtMove tmp, *p, *q;
for (p = begin + 1; p < end; ++p)
{
tmp = *p;
for (q = p; q != begin && *(q-1) < tmp; --q)
*q = *(q-1);
*q = tmp;
}
}
// Unary predicate used by std::partition to split positive scores from remaining // Unary predicate used by std::partition to split positive scores from remaining
// ones so to sort separately the two sets, and with the second sort delayed. // ones so to sort separately the two sets, and with the second sort delayed.
inline bool has_positive_score(const MoveStack& ms) { return ms.score > 0; } inline bool has_positive_score(const ExtMove& ms) { return ms.score > 0; }
// Picks and moves to the front the best move in the range [begin, end), // Picks and moves to the front the best move in the range [begin, end),
// it is faster than sorting all the moves in advance when moves are few, as // it is faster than sorting all the moves in advance when moves are few, as
// normally are the possible captures. // normally are the possible captures.
inline MoveStack* pick_best(MoveStack* begin, MoveStack* end) inline ExtMove* pick_best(ExtMove* begin, ExtMove* end)
{ {
std::swap(*begin, *std::max_element(begin, end)); std::swap(*begin, *std::max_element(begin, end));
return begin; return begin;
@@ -58,53 +70,40 @@ namespace {
/// search captures, promotions and some checks) and about how important good /// search captures, promotions and some checks) and about how important good
/// move ordering is at the current node. /// move ordering is at the current node.
MovePicker::MovePicker(const Position& p, Move ttm, Depth d, const History& h, MovePicker::MovePicker(const Position& p, Move ttm, Depth d, const HistoryStats& h,
Search::Stack* s, Value beta) : pos(p), H(h), depth(d) { Move* cm, Search::Stack* s) : pos(p), history(h), depth(d) {
assert(d > DEPTH_ZERO); assert(d > DEPTH_ZERO);
captureThreshold = 0;
cur = end = moves; cur = end = moves;
endBadCaptures = moves + MAX_MOVES - 1; endBadCaptures = moves + MAX_MOVES - 1;
countermoves = cm;
ss = s; ss = s;
if (p.in_check()) if (p.checkers())
phase = EVASION; stage = EVASION;
else else
{ stage = MAIN_SEARCH;
phase = MAIN_SEARCH;
killers[0].move = ss->killers[0];
killers[1].move = ss->killers[1];
// Consider sligtly negative captures as good if at low depth and far from beta
if (ss && ss->eval < beta - PawnValueMg && d < 3 * ONE_PLY)
captureThreshold = -PawnValueMg;
// Consider negative captures as good if still enough to reach beta
else if (ss && ss->eval > beta)
captureThreshold = beta - ss->eval;
}
ttMove = (ttm && pos.is_pseudo_legal(ttm) ? ttm : MOVE_NONE); ttMove = (ttm && pos.is_pseudo_legal(ttm) ? ttm : MOVE_NONE);
end += (ttMove != MOVE_NONE); end += (ttMove != MOVE_NONE);
} }
MovePicker::MovePicker(const Position& p, Move ttm, Depth d, const History& h, MovePicker::MovePicker(const Position& p, Move ttm, Depth d, const HistoryStats& h,
Square sq) : pos(p), H(h), cur(moves), end(moves) { Square sq) : pos(p), history(h), cur(moves), end(moves) {
assert(d <= DEPTH_ZERO); assert(d <= DEPTH_ZERO);
if (p.in_check()) if (p.checkers())
phase = EVASION; stage = EVASION;
else if (d > DEPTH_QS_NO_CHECKS) else if (d > DEPTH_QS_NO_CHECKS)
phase = QSEARCH_0; stage = QSEARCH_0;
else if (d > DEPTH_QS_RECAPTURES) else if (d > DEPTH_QS_RECAPTURES)
{ {
phase = QSEARCH_1; stage = QSEARCH_1;
// Skip TT move if is not a capture or a promotion, this avoids qsearch // Skip TT move if is not a capture or a promotion, this avoids qsearch
// tree explosion due to a possible perpetual check or similar rare cases // tree explosion due to a possible perpetual check or similar rare cases
@@ -114,7 +113,7 @@ MovePicker::MovePicker(const Position& p, Move ttm, Depth d, const History& h,
} }
else else
{ {
phase = RECAPTURE; stage = RECAPTURE;
recaptureSquare = sq; recaptureSquare = sq;
ttm = MOVE_NONE; ttm = MOVE_NONE;
} }
@@ -123,15 +122,15 @@ MovePicker::MovePicker(const Position& p, Move ttm, Depth d, const History& h,
end += (ttMove != MOVE_NONE); end += (ttMove != MOVE_NONE);
} }
MovePicker::MovePicker(const Position& p, Move ttm, const History& h, PieceType pt) MovePicker::MovePicker(const Position& p, Move ttm, const HistoryStats& h, PieceType pt)
: pos(p), H(h), cur(moves), end(moves) { : pos(p), history(h), cur(moves), end(moves) {
assert(!pos.in_check()); assert(!pos.checkers());
phase = PROBCUT; stage = PROBCUT;
// In ProbCut we generate only captures better than parent's captured piece // In ProbCut we generate only captures better than parent's captured piece
captureThreshold = PieceValue[Mg][pt]; captureThreshold = PieceValue[MG][pt];
ttMove = (ttm && pos.is_pseudo_legal(ttm) ? ttm : MOVE_NONE); ttMove = (ttm && pos.is_pseudo_legal(ttm) ? ttm : MOVE_NONE);
if (ttMove && (!pos.is_capture(ttMove) || pos.see(ttMove) <= captureThreshold)) if (ttMove && (!pos.is_capture(ttMove) || pos.see(ttMove) <= captureThreshold))
@@ -141,12 +140,10 @@ MovePicker::MovePicker(const Position& p, Move ttm, const History& h, PieceType
} }
/// MovePicker::score_captures(), MovePicker::score_noncaptures() and /// score() assign a numerical move ordering score to each move in a move list.
/// MovePicker::score_evasions() assign a numerical move ordering score /// The moves with highest scores will be picked first.
/// to each move in a move list. The moves with highest scores will be template<>
/// picked first by next_move(). void MovePicker::score<CAPTURES>() {
void MovePicker::score_captures() {
// Winning and equal captures in the main search are ordered by MVV/LVA. // Winning and equal captures in the main search are ordered by MVV/LVA.
// Suprisingly, this appears to perform slightly better than SEE based // Suprisingly, this appears to perform slightly better than SEE based
// move ordering. The reason is probably that in a position with a winning // move ordering. The reason is probably that in a position with a winning
@@ -162,83 +159,99 @@ void MovePicker::score_captures() {
// some SEE calls in case we get a cutoff (idea from Pablo Vazquez). // some SEE calls in case we get a cutoff (idea from Pablo Vazquez).
Move m; Move m;
for (MoveStack* it = moves; it != end; ++it) for (ExtMove* it = moves; it != end; ++it)
{ {
m = it->move; m = it->move;
it->score = PieceValue[Mg][pos.piece_on(to_sq(m))] it->score = PieceValue[MG][pos.piece_on(to_sq(m))]
- type_of(pos.piece_moved(m)); - type_of(pos.piece_moved(m));
if (type_of(m) == PROMOTION) if (type_of(m) == PROMOTION)
it->score += PieceValue[Mg][promotion_type(m)]; it->score += PieceValue[MG][promotion_type(m)] - PieceValue[MG][PAWN];
else if (type_of(m) == ENPASSANT)
it->score += PieceValue[MG][PAWN];
} }
} }
void MovePicker::score_noncaptures() { template<>
void MovePicker::score<QUIETS>() {
Move m; Move m;
for (MoveStack* it = moves; it != end; ++it) for (ExtMove* it = moves; it != end; ++it)
{ {
m = it->move; m = it->move;
it->score = H.value(pos.piece_moved(m), to_sq(m)); it->score = history[pos.piece_moved(m)][to_sq(m)];
} }
} }
void MovePicker::score_evasions() { template<>
void MovePicker::score<EVASIONS>() {
// Try good captures ordered by MVV/LVA, then non-captures if destination square // Try good captures ordered by MVV/LVA, then non-captures if destination square
// is not under attack, ordered by history value, then bad-captures and quiet // is not under attack, ordered by history value, then bad-captures and quiet
// moves with a negative SEE. This last group is ordered by the SEE score. // moves with a negative SEE. This last group is ordered by the SEE score.
Move m; Move m;
int seeScore; int seeScore;
if (end < moves + 2) for (ExtMove* it = moves; it != end; ++it)
return;
for (MoveStack* it = moves; it != end; ++it)
{ {
m = it->move; m = it->move;
if ((seeScore = pos.see_sign(m)) < 0) if ((seeScore = pos.see_sign(m)) < 0)
it->score = seeScore - History::MaxValue; // Be sure we are at the bottom it->score = seeScore - HistoryStats::Max; // At the bottom
else if (pos.is_capture(m)) else if (pos.is_capture(m))
it->score = PieceValue[Mg][pos.piece_on(to_sq(m))] it->score = PieceValue[MG][pos.piece_on(to_sq(m))]
- type_of(pos.piece_moved(m)) + History::MaxValue; - type_of(pos.piece_moved(m)) + HistoryStats::Max;
else else
it->score = H.value(pos.piece_moved(m), to_sq(m)); it->score = history[pos.piece_moved(m)][to_sq(m)];
} }
} }
/// MovePicker::generate_next() generates, scores and sorts the next bunch of moves, /// generate_next() generates, scores and sorts the next bunch of moves, when
/// when there are no more moves to try for the current phase. /// there are no more moves to try for the current phase.
void MovePicker::generate_next() { void MovePicker::generate_next() {
cur = moves; cur = moves;
switch (++phase) { switch (++stage) {
case CAPTURES_S1: case CAPTURES_S3: case CAPTURES_S4: case CAPTURES_S5: case CAPTURES_S6: case CAPTURES_S1: case CAPTURES_S3: case CAPTURES_S4: case CAPTURES_S5: case CAPTURES_S6:
end = generate<CAPTURES>(pos, moves); end = generate<CAPTURES>(pos, moves);
score_captures(); score<CAPTURES>();
return; return;
case KILLERS_S1: case KILLERS_S1:
cur = killers; cur = killers;
end = cur + 2; end = cur + 2;
killers[0].move = ss->killers[0];
killers[1].move = ss->killers[1];
killers[2].move = killers[3].move = MOVE_NONE;
// Be sure countermoves are different from killers
for (int i = 0; i < 2; i++)
if (countermoves[i] != cur->move && countermoves[i] != (cur+1)->move)
(end++)->move = countermoves[i];
if (countermoves[1] && countermoves[1] == countermoves[0]) // Due to SMP races
killers[3].move = MOVE_NONE;
return; return;
case QUIETS_1_S1: case QUIETS_1_S1:
endQuiets = end = generate<QUIETS>(pos, moves); endQuiets = end = generate<QUIETS>(pos, moves);
score_noncaptures(); score<QUIETS>();
end = std::partition(cur, end, has_positive_score); end = std::partition(cur, end, has_positive_score);
sort<MoveStack>(cur, end); insertion_sort(cur, end);
return; return;
case QUIETS_2_S1: case QUIETS_2_S1:
cur = end; cur = end;
end = endQuiets; end = endQuiets;
if (depth >= 3 * ONE_PLY) if (depth >= 3 * ONE_PLY)
sort<MoveStack>(cur, end); insertion_sort(cur, end);
return; return;
case BAD_CAPTURES_S1: case BAD_CAPTURES_S1:
@@ -249,7 +262,8 @@ void MovePicker::generate_next() {
case EVASIONS_S2: case EVASIONS_S2:
end = generate<EVASIONS>(pos, moves); end = generate<EVASIONS>(pos, moves);
score_evasions(); if (end > moves + 1)
score<EVASIONS>();
return; return;
case QUIET_CHECKS_S3: case QUIET_CHECKS_S3:
@@ -257,7 +271,7 @@ void MovePicker::generate_next() {
return; return;
case EVASION: case QSEARCH_0: case QSEARCH_1: case PROBCUT: case RECAPTURE: case EVASION: case QSEARCH_0: case QSEARCH_1: case PROBCUT: case RECAPTURE:
phase = STOP; stage = STOP;
case STOP: case STOP:
end = cur + 1; // Avoid another next_phase() call end = cur + 1; // Avoid another next_phase() call
return; return;
@@ -268,11 +282,10 @@ void MovePicker::generate_next() {
} }
/// MovePicker::next_move() is the most important method of the MovePicker class. /// next_move() is the most important method of the MovePicker class. It returns
/// It returns a new pseudo legal move every time it is called, until there /// a new pseudo legal move every time is called, until there are no more moves
/// are no more moves left. It picks the move with the biggest score from a list /// left. It picks the move with the biggest score from a list of generated moves
/// of generated moves taking care not to return the tt move if has already been /// taking care not returning the ttMove if has already been searched previously.
/// searched previously.
template<> template<>
Move MovePicker::next_move<false>() { Move MovePicker::next_move<false>() {
@@ -283,7 +296,7 @@ Move MovePicker::next_move<false>() {
while (cur == end) while (cur == end)
generate_next(); generate_next();
switch (phase) { switch (stage) {
case MAIN_SEARCH: case EVASION: case QSEARCH_0: case QSEARCH_1: case PROBCUT: case MAIN_SEARCH: case EVASION: case QSEARCH_0: case QSEARCH_1: case PROBCUT:
cur++; cur++;
@@ -293,9 +306,7 @@ Move MovePicker::next_move<false>() {
move = pick_best(cur++, end)->move; move = pick_best(cur++, end)->move;
if (move != ttMove) if (move != ttMove)
{ {
assert(captureThreshold <= 0); // Otherwise we cannot use see_sign() if (pos.see_sign(move) >= 0)
if (pos.see_sign(move) >= captureThreshold)
return move; return move;
// Losing capture, move it to the tail of the array // Losing capture, move it to the tail of the array
@@ -316,7 +327,9 @@ Move MovePicker::next_move<false>() {
move = (cur++)->move; move = (cur++)->move;
if ( move != ttMove if ( move != ttMove
&& move != killers[0].move && move != killers[0].move
&& move != killers[1].move) && move != killers[1].move
&& move != killers[2].move
&& move != killers[3].move)
return move; return move;
break; break;
@@ -359,6 +372,6 @@ Move MovePicker::next_move<false>() {
/// Version of next_move() to use at split point nodes where the move is grabbed /// Version of next_move() to use at split point nodes where the move is grabbed
/// from the split point's shared MovePicker object. This function is not thread /// from the split point's shared MovePicker object. This function is not thread
/// safe so should be lock protected by the caller. /// safe so must be lock protected by the caller.
template<> template<>
Move MovePicker::next_move<true>() { return ss->sp->mp->next_move<false>(); } Move MovePicker::next_move<true>() { return ss->splitPoint->movePicker->next_move<false>(); }
+61 -15
View File
@@ -1,7 +1,7 @@
/* /*
Stockfish, a UCI chess playing engine derived from Glaurung 2.1 Stockfish, a UCI chess playing engine derived from Glaurung 2.1
Copyright (C) 2004-2008 Tord Romstad (Glaurung author) Copyright (C) 2004-2008 Tord Romstad (Glaurung author)
Copyright (C) 2008-2012 Marco Costalba, Joona Kiiski, Tord Romstad Copyright (C) 2008-2013 Marco Costalba, Joona Kiiski, Tord Romstad
Stockfish is free software: you can redistribute it and/or modify Stockfish is free software: you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by it under the terms of the GNU General Public License as published by
@@ -17,15 +17,61 @@
along with this program. If not, see <http://www.gnu.org/licenses/>. along with this program. If not, see <http://www.gnu.org/licenses/>.
*/ */
#if !defined MOVEPICK_H_INCLUDED #ifndef MOVEPICK_H_INCLUDED
#define MOVEPICK_H_INCLUDED #define MOVEPICK_H_INCLUDED
#include "history.h" #include <algorithm> // For std::max
#include <cstring> // For std::memset
#include "movegen.h"
#include "position.h" #include "position.h"
#include "search.h" #include "search.h"
#include "types.h" #include "types.h"
/// The Stats struct stores moves statistics. According to the template parameter
/// the class can store History, Gains and Countermoves. History records how often
/// different moves have been successful or unsuccessful during the current search
/// and is used for reduction and move ordering decisions. Gains records the move's
/// best evaluation gain from one ply to the next and is used for pruning decisions.
/// Countermoves store the move that refute a previous one. Entries are stored
/// according only to moving piece and destination square, hence two moves with
/// different origin but same destination and piece will be considered identical.
template<bool Gain, typename T>
struct Stats {
static const Value Max = Value(2000);
const T* operator[](Piece p) const { return table[p]; }
void clear() { std::memset(table, 0, sizeof(table)); }
void update(Piece p, Square to, Move m) {
if (m == table[p][to].first)
return;
table[p][to].second = table[p][to].first;
table[p][to].first = m;
}
void update(Piece p, Square to, Value v) {
if (Gain)
table[p][to] = std::max(v, table[p][to] - 1);
else if (abs(table[p][to] + v) < Max)
table[p][to] += v;
}
private:
T table[PIECE_NB][SQUARE_NB];
};
typedef Stats< true, Value> GainsStats;
typedef Stats<false, Value> HistoryStats;
typedef Stats<false, std::pair<Move, Move> > CountermovesStats;
/// MovePicker class is used to pick one pseudo legal move at a time from the /// MovePicker class is used to pick one pseudo legal move at a time from the
/// current position. The most important method is next_move(), which returns a /// current position. The most important method is next_move(), which returns a
/// new pseudo legal move each time it is called, until there are no moves left, /// new pseudo legal move each time it is called, until there are no moves left,
@@ -38,27 +84,27 @@ class MovePicker {
MovePicker& operator=(const MovePicker&); // Silence a warning under MSVC MovePicker& operator=(const MovePicker&); // Silence a warning under MSVC
public: public:
MovePicker(const Position&, Move, Depth, const History&, Search::Stack*, Value); MovePicker(const Position&, Move, Depth, const HistoryStats&, Square);
MovePicker(const Position&, Move, Depth, const History&, Square); MovePicker(const Position&, Move, const HistoryStats&, PieceType);
MovePicker(const Position&, Move, const History&, PieceType); MovePicker(const Position&, Move, Depth, const HistoryStats&, Move*, Search::Stack*);
template<bool SpNode> Move next_move(); template<bool SpNode> Move next_move();
private: private:
void score_captures(); template<GenType> void score();
void score_noncaptures();
void score_evasions();
void generate_next(); void generate_next();
const Position& pos; const Position& pos;
const History& H; const HistoryStats& history;
Search::Stack* ss; Search::Stack* ss;
Move* countermoves;
Depth depth; Depth depth;
Move ttMove; Move ttMove;
MoveStack killers[2]; ExtMove killers[4];
Square recaptureSquare; Square recaptureSquare;
int captureThreshold, phase; int captureThreshold, stage;
MoveStack *cur, *end, *endQuiets, *endBadCaptures; ExtMove *cur, *end, *endQuiets, *endBadCaptures;
MoveStack moves[MAX_MOVES]; ExtMove moves[MAX_MOVES];
}; };
#endif // !defined(MOVEPICK_H_INCLUDED) #endif // #ifndef MOVEPICK_H_INCLUDED
+24 -32
View File
@@ -1,7 +1,7 @@
/* /*
Stockfish, a UCI chess playing engine derived from Glaurung 2.1 Stockfish, a UCI chess playing engine derived from Glaurung 2.1
Copyright (C) 2004-2008 Tord Romstad (Glaurung author) Copyright (C) 2004-2008 Tord Romstad (Glaurung author)
Copyright (C) 2008-2012 Marco Costalba, Joona Kiiski, Tord Romstad Copyright (C) 2008-2013 Marco Costalba, Joona Kiiski, Tord Romstad
Stockfish is free software: you can redistribute it and/or modify Stockfish is free software: you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by it under the terms of the GNU General Public License as published by
@@ -20,7 +20,7 @@
#include <cassert> #include <cassert>
#include <iomanip> #include <iomanip>
#include <sstream> #include <sstream>
#include <string> #include <stack>
#include "movegen.h" #include "movegen.h"
#include "notation.h" #include "notation.h"
@@ -28,7 +28,7 @@
using namespace std; using namespace std;
static const char* PieceToChar = " PNBRQK pnbrqk"; static const char* PieceToChar[COLOR_NB] = { " PNBRQK", " pnbrqk" };
/// score_to_uci() converts a value to a string suitable for use with the UCI /// score_to_uci() converts a value to a string suitable for use with the UCI
@@ -75,7 +75,7 @@ const string move_to_uci(Move m, bool chess960) {
string move = square_to_string(from) + square_to_string(to); string move = square_to_string(from) + square_to_string(to);
if (type_of(m) == PROMOTION) if (type_of(m) == PROMOTION)
move += PieceToChar[make_piece(BLACK, promotion_type(m))]; // Lower case move += PieceToChar[BLACK][promotion_type(m)]; // Lower case
return move; return move;
} }
@@ -89,9 +89,9 @@ Move move_from_uci(const Position& pos, string& str) {
if (str.length() == 5) // Junior could send promotion piece in uppercase if (str.length() == 5) // Junior could send promotion piece in uppercase
str[4] = char(tolower(str[4])); str[4] = char(tolower(str[4]));
for (MoveList<LEGAL> ml(pos); !ml.end(); ++ml) for (MoveList<LEGAL> it(pos); *it; ++it)
if (str == move_to_uci(ml.move(), pos.is_chess960())) if (str == move_to_uci(*it, pos.is_chess960()))
return ml.move(); return *it;
return MOVE_NONE; return MOVE_NONE;
} }
@@ -108,10 +108,9 @@ const string move_to_san(Position& pos, Move m) {
if (m == MOVE_NULL) if (m == MOVE_NULL)
return "(null)"; return "(null)";
assert(pos.move_is_legal(m)); assert(MoveList<LEGAL>(pos).contains(m));
Bitboard attackers; Bitboard others, b;
bool ambiguousMove, ambiguousFile, ambiguousRank;
string san; string san;
Color us = pos.side_to_move(); Color us = pos.side_to_move();
Square from = from_sq(m); Square from = from_sq(m);
@@ -125,33 +124,25 @@ const string move_to_san(Position& pos, Move m) {
{ {
if (pt != PAWN) if (pt != PAWN)
{ {
san = PieceToChar[pt]; // Upper case san = PieceToChar[WHITE][pt]; // Upper case
// Disambiguation if we have more then one piece with destination 'to' // Disambiguation if we have more then one piece of type 'pt' that can
// note that for pawns is not needed because starting file is explicit. // reach 'to' with a legal move.
ambiguousMove = ambiguousFile = ambiguousRank = false; others = b = (pos.attacks_from(pc, to) & pos.pieces(us, pt)) ^ from;
attackers = (pos.attacks_from(pc, to) & pos.pieces(us, pt)) ^ from; while (b)
while (attackers)
{ {
Square sq = pop_lsb(&attackers); Move move = make_move(pop_lsb(&b), to);
if (!pos.pl_move_is_legal(move, pos.pinned_pieces()))
// Pinned pieces are not included in the possible sub-set others ^= from_sq(move);
if (!pos.pl_move_is_legal(make_move(sq, to), pos.pinned_pieces()))
continue;
ambiguousFile |= file_of(sq) == file_of(from);
ambiguousRank |= rank_of(sq) == rank_of(from);
ambiguousMove = true;
} }
if (ambiguousMove) if (others)
{ {
if (!ambiguousFile) if (!(others & file_bb(from)))
san += file_to_char(file_of(from)); san += file_to_char(file_of(from));
else if (!ambiguousRank) else if (!(others & rank_bb(from)))
san += rank_to_char(rank_of(from)); san += rank_to_char(rank_of(from));
else else
@@ -167,7 +158,7 @@ const string move_to_san(Position& pos, Move m) {
san += square_to_string(to); san += square_to_string(to);
if (type_of(m) == PROMOTION) if (type_of(m) == PROMOTION)
san += string("=") + PieceToChar[promotion_type(m)]; san += string("=") + PieceToChar[WHITE][promotion_type(m)];
} }
if (pos.move_gives_check(m, CheckInfo(pos))) if (pos.move_gives_check(m, CheckInfo(pos)))
@@ -226,7 +217,7 @@ string pretty_pv(Position& pos, int depth, Value value, int64_t msecs, Move pv[]
const int64_t K = 1000; const int64_t K = 1000;
const int64_t M = 1000000; const int64_t M = 1000000;
StateInfo state[MAX_PLY_PLUS_2], *st = state; std::stack<StateInfo> st;
Move* m = pv; Move* m = pv;
string san, padding; string san, padding;
size_t length; size_t length;
@@ -261,7 +252,8 @@ string pretty_pv(Position& pos, int depth, Value value, int64_t msecs, Move pv[]
s << san << ' '; s << san << ' ';
length += san.length() + 1; length += san.length() + 1;
pos.do_move(*m++, *st++); st.push(StateInfo());
pos.do_move(*m++, st.top());
} }
while (m != pv) while (m != pv)
+3 -3
View File
@@ -1,7 +1,7 @@
/* /*
Stockfish, a UCI chess playing engine derived from Glaurung 2.1 Stockfish, a UCI chess playing engine derived from Glaurung 2.1
Copyright (C) 2004-2008 Tord Romstad (Glaurung author) Copyright (C) 2004-2008 Tord Romstad (Glaurung author)
Copyright (C) 2008-2012 Marco Costalba, Joona Kiiski, Tord Romstad Copyright (C) 2008-2013 Marco Costalba, Joona Kiiski, Tord Romstad
Stockfish is free software: you can redistribute it and/or modify Stockfish is free software: you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by it under the terms of the GNU General Public License as published by
@@ -17,7 +17,7 @@
along with this program. If not, see <http://www.gnu.org/licenses/>. along with this program. If not, see <http://www.gnu.org/licenses/>.
*/ */
#if !defined(NOTATION_H_INCLUDED) #ifndef NOTATION_H_INCLUDED
#define NOTATION_H_INCLUDED #define NOTATION_H_INCLUDED
#include <string> #include <string>
@@ -32,4 +32,4 @@ const std::string move_to_uci(Move m, bool chess960);
const std::string move_to_san(Position& pos, Move m); const std::string move_to_san(Position& pos, Move m);
std::string pretty_pv(Position& pos, int depth, Value score, int64_t msecs, Move pv[]); std::string pretty_pv(Position& pos, int depth, Value score, int64_t msecs, Move pv[]);
#endif // !defined(NOTATION_H_INCLUDED) #endif // #ifndef NOTATION_H_INCLUDED
+81 -87
View File
@@ -1,7 +1,7 @@
/* /*
Stockfish, a UCI chess playing engine derived from Glaurung 2.1 Stockfish, a UCI chess playing engine derived from Glaurung 2.1
Copyright (C) 2004-2008 Tord Romstad (Glaurung author) Copyright (C) 2004-2008 Tord Romstad (Glaurung author)
Copyright (C) 2008-2012 Marco Costalba, Joona Kiiski, Tord Romstad Copyright (C) 2008-2013 Marco Costalba, Joona Kiiski, Tord Romstad
Stockfish is free software: you can redistribute it and/or modify Stockfish is free software: you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by it under the terms of the GNU General Public License as published by
@@ -17,6 +17,7 @@
along with this program. If not, see <http://www.gnu.org/licenses/>. along with this program. If not, see <http://www.gnu.org/licenses/>.
*/ */
#include <algorithm>
#include <cassert> #include <cassert>
#include "bitboard.h" #include "bitboard.h"
@@ -30,49 +31,48 @@ namespace {
#define S(mg, eg) make_score(mg, eg) #define S(mg, eg) make_score(mg, eg)
// Doubled pawn penalty by opposed flag and file // Doubled pawn penalty by opposed flag and file
const Score DoubledPawnPenalty[2][8] = { const Score Doubled[2][FILE_NB] = {
{ S(13, 43), S(20, 48), S(23, 48), S(23, 48), { S(13, 43), S(20, 48), S(23, 48), S(23, 48),
S(23, 48), S(23, 48), S(20, 48), S(13, 43) }, S(23, 48), S(23, 48), S(20, 48), S(13, 43) },
{ S(13, 43), S(20, 48), S(23, 48), S(23, 48), { S(13, 43), S(20, 48), S(23, 48), S(23, 48),
S(23, 48), S(23, 48), S(20, 48), S(13, 43) }}; S(23, 48), S(23, 48), S(20, 48), S(13, 43) }};
// Isolated pawn penalty by opposed flag and file // Isolated pawn penalty by opposed flag and file
const Score IsolatedPawnPenalty[2][8] = { const Score Isolated[2][FILE_NB] = {
{ S(37, 45), S(54, 52), S(60, 52), S(60, 52), { S(37, 45), S(54, 52), S(60, 52), S(60, 52),
S(60, 52), S(60, 52), S(54, 52), S(37, 45) }, S(60, 52), S(60, 52), S(54, 52), S(37, 45) },
{ S(25, 30), S(36, 35), S(40, 35), S(40, 35), { S(25, 30), S(36, 35), S(40, 35), S(40, 35),
S(40, 35), S(40, 35), S(36, 35), S(25, 30) }}; S(40, 35), S(40, 35), S(36, 35), S(25, 30) }};
// Backward pawn penalty by opposed flag and file // Backward pawn penalty by opposed flag and file
const Score BackwardPawnPenalty[2][8] = { const Score Backward[2][FILE_NB] = {
{ S(30, 42), S(43, 46), S(49, 46), S(49, 46), { S(30, 42), S(43, 46), S(49, 46), S(49, 46),
S(49, 46), S(49, 46), S(43, 46), S(30, 42) }, S(49, 46), S(49, 46), S(43, 46), S(30, 42) },
{ S(20, 28), S(29, 31), S(33, 31), S(33, 31), { S(20, 28), S(29, 31), S(33, 31), S(33, 31),
S(33, 31), S(33, 31), S(29, 31), S(20, 28) }}; S(33, 31), S(33, 31), S(29, 31), S(20, 28) }};
// Pawn chain membership bonus by file // Pawn chain membership bonus by file
const Score ChainBonus[8] = { const Score ChainMember[FILE_NB] = {
S(11,-1), S(13,-1), S(13,-1), S(14,-1), S(11,-1), S(13,-1), S(13,-1), S(14,-1),
S(14,-1), S(13,-1), S(13,-1), S(11,-1) S(14,-1), S(13,-1), S(13,-1), S(11,-1)
}; };
// Candidate passed pawn bonus by rank // Candidate passed pawn bonus by rank
const Score CandidateBonus[8] = { const Score CandidatePassed[RANK_NB] = {
S( 0, 0), S( 6, 13), S(6,13), S(14,29), S( 0, 0), S( 6, 13), S(6,13), S(14,29),
S(34,68), S(83,166), S(0, 0), S( 0, 0) S(34,68), S(83,166), S(0, 0), S( 0, 0)
}; };
const Score PawnStructureWeight = S(233, 201); // Weakness of our pawn shelter in front of the king indexed by [rank]
const Value ShelterWeakness[RANK_NB] =
{ V(100), V(0), V(27), V(73), V(92), V(101), V(101) };
// Weakness of our pawn shelter in front of the king indexed by [king pawn][rank] // Danger of enemy pawns moving toward our king indexed by
const Value ShelterWeakness[2][8] = // [no friendly pawn | pawn unblocked | pawn blocked][rank of enemy pawn]
{ { V(141), V(0), V(38), V(102), V(128), V(141), V(141) }, const Value StormDanger[3][RANK_NB] = {
{ V( 61), V(0), V(16), V( 44), V( 56), V( 61), V( 61) } }; { V( 0), V(64), V(128), V(51), V(26) },
{ V(26), V(32), V( 96), V(38), V(20) },
// Danger of enemy pawns moving toward our king indexed by [pawn blocked][rank] { V( 0), V( 0), V( 64), V(25), V(13) }};
const Value StormDanger[2][8] =
{ { V(26), V(0), V(128), V(51), V(26) },
{ V(13), V(0), V( 64), V(25), V(13) } };
// Max bonus for king safety. Corresponds to start position with all the pawns // Max bonus for king safety. Corresponds to start position with all the pawns
// in front of the king and no enemy pawn on the horizont. // in front of the king and no enemy pawn on the horizont.
@@ -80,51 +80,14 @@ namespace {
#undef S #undef S
#undef V #undef V
}
template<Color Us>
/// PawnTable::probe() takes a position object as input, computes a PawnEntry Score evaluate(const Position& pos, Pawns::Entry* e) {
/// object, and returns a pointer to it. The result is also stored in a hash
/// table, so we don't have to recompute everything when the same pawn structure
/// occurs again.
PawnEntry* PawnTable::probe(const Position& pos) {
Key key = pos.pawn_key();
PawnEntry* e = entries[key];
// If e->key matches the position's pawn hash key, it means that we
// have analysed this pawn structure before, and we can simply return
// the information we found the last time instead of recomputing it.
if (e->key == key)
return e;
e->key = key;
e->passedPawns[WHITE] = e->passedPawns[BLACK] = 0;
e->kingSquares[WHITE] = e->kingSquares[BLACK] = SQ_NONE;
e->halfOpenFiles[WHITE] = e->halfOpenFiles[BLACK] = 0xFF;
Bitboard wPawns = pos.pieces(WHITE, PAWN);
Bitboard bPawns = pos.pieces(BLACK, PAWN);
e->pawnAttacks[WHITE] = ((wPawns & ~FileHBB) << 9) | ((wPawns & ~FileABB) << 7);
e->pawnAttacks[BLACK] = ((bPawns & ~FileHBB) >> 7) | ((bPawns & ~FileABB) >> 9);
e->value = evaluate_pawns<WHITE>(pos, wPawns, bPawns, e)
- evaluate_pawns<BLACK>(pos, bPawns, wPawns, e);
e->value = apply_weight(e->value, PawnStructureWeight);
return e;
}
/// PawnTable::evaluate_pawns() evaluates each pawn of the given color
template<Color Us>
Score PawnTable::evaluate_pawns(const Position& pos, Bitboard ourPawns,
Bitboard theirPawns, PawnEntry* e) {
const Color Them = (Us == WHITE ? BLACK : WHITE); const Color Them = (Us == WHITE ? BLACK : WHITE);
const Square Up = (Us == WHITE ? DELTA_N : DELTA_S);
const Square Right = (Us == WHITE ? DELTA_NE : DELTA_SW);
const Square Left = (Us == WHITE ? DELTA_NW : DELTA_SE);
Bitboard b; Bitboard b;
Square s; Square s;
@@ -132,7 +95,17 @@ Score PawnTable::evaluate_pawns(const Position& pos, Bitboard ourPawns,
Rank r; Rank r;
bool passed, isolated, doubled, opposed, chain, backward, candidate; bool passed, isolated, doubled, opposed, chain, backward, candidate;
Score value = SCORE_ZERO; Score value = SCORE_ZERO;
const Square* pl = pos.piece_list(Us, PAWN); const Square* pl = pos.list<PAWN>(Us);
Bitboard ourPawns = pos.pieces(Us, PAWN);
Bitboard theirPawns = pos.pieces(Them, PAWN);
e->passedPawns[Us] = 0;
e->kingSquares[Us] = SQ_NONE;
e->semiopenFiles[Us] = 0xFF;
e->pawnAttacks[Us] = shift_bb<Right>(ourPawns) | shift_bb<Left>(ourPawns);
e->pawnsOnSquares[Us][BLACK] = popcount<Max15>(ourPawns & DarkSquares);
e->pawnsOnSquares[Us][WHITE] = pos.count<PAWN>(Us) - e->pawnsOnSquares[Us][BLACK];
// Loop through all pawns of the current color and score each pawn // Loop through all pawns of the current color and score each pawn
while ((s = *pl++) != SQ_NONE) while ((s = *pl++) != SQ_NONE)
@@ -142,8 +115,8 @@ Score PawnTable::evaluate_pawns(const Position& pos, Bitboard ourPawns,
f = file_of(s); f = file_of(s);
r = rank_of(s); r = rank_of(s);
// This file cannot be half open // This file cannot be semi-open
e->halfOpenFiles[Us] &= ~(1 << f); e->semiopenFiles[Us] &= ~(1 << f);
// Our rank plus previous one. Used for chain detection // Our rank plus previous one. Used for chain detection
b = rank_bb(r) | rank_bb(Us == WHITE ? r - Rank(1) : r + Rank(1)); b = rank_bb(r) | rank_bb(Us == WHITE ? r - Rank(1) : r + Rank(1));
@@ -163,7 +136,7 @@ Score PawnTable::evaluate_pawns(const Position& pos, Bitboard ourPawns,
// be backward. If there are friendly pawns behind on adjacent files // be backward. If there are friendly pawns behind on adjacent files
// or if can capture an enemy pawn it cannot be backward either. // or if can capture an enemy pawn it cannot be backward either.
if ( !(passed | isolated | chain) if ( !(passed | isolated | chain)
&& !(ourPawns & attack_span_mask(Them, s)) && !(ourPawns & pawn_attack_span(Them, s))
&& !(pos.attacks_from<PAWN>(s, Us) & theirPawns)) && !(pos.attacks_from<PAWN>(s, Us) & theirPawns))
{ {
// We now know that there are no friendly pawns beside or behind this // We now know that there are no friendly pawns beside or behind this
@@ -175,22 +148,22 @@ Score PawnTable::evaluate_pawns(const Position& pos, Bitboard ourPawns,
// Note that we are sure to find something because pawn is not passed // Note that we are sure to find something because pawn is not passed
// nor isolated, so loop is potentially infinite, but it isn't. // nor isolated, so loop is potentially infinite, but it isn't.
while (!(b & (ourPawns | theirPawns))) while (!(b & (ourPawns | theirPawns)))
Us == WHITE ? b <<= 8 : b >>= 8; b = shift_bb<Up>(b);
// The friendly pawn needs to be at least two ranks closer than the // The friendly pawn needs to be at least two ranks closer than the
// enemy pawn in order to help the potentially backward pawn advance. // enemy pawn in order to help the potentially backward pawn advance.
backward = (b | (Us == WHITE ? b << 8 : b >> 8)) & theirPawns; backward = (b | shift_bb<Up>(b)) & theirPawns;
} }
assert(opposed | passed | (attack_span_mask(Us, s) & theirPawns)); assert(opposed | passed | (pawn_attack_span(Us, s) & theirPawns));
// A not passed pawn is a candidate to become passed if it is free to // A not passed pawn is a candidate to become passed if it is free to
// advance and if the number of friendly pawns beside or behind this // advance and if the number of friendly pawns beside or behind this
// pawn on adjacent files is higher or equal than the number of // pawn on adjacent files is higher or equal than the number of
// enemy pawns in the forward direction on the adjacent files. // enemy pawns in the forward direction on the adjacent files.
candidate = !(opposed | passed | backward | isolated) candidate = !(opposed | passed | backward | isolated)
&& (b = attack_span_mask(Them, s + pawn_push(Us)) & ourPawns) != 0 && (b = pawn_attack_span(Them, s + pawn_push(Us)) & ourPawns) != 0
&& popcount<Max15>(b) >= popcount<Max15>(attack_span_mask(Us, s) & theirPawns); && popcount<Max15>(b) >= popcount<Max15>(pawn_attack_span(Us, s) & theirPawns);
// Passed pawns will be properly scored in evaluation because we need // Passed pawns will be properly scored in evaluation because we need
// full attack info to evaluate passed pawns. Only the frontmost passed // full attack info to evaluate passed pawns. Only the frontmost passed
@@ -200,64 +173,83 @@ Score PawnTable::evaluate_pawns(const Position& pos, Bitboard ourPawns,
// Score this pawn // Score this pawn
if (isolated) if (isolated)
value -= IsolatedPawnPenalty[opposed][f]; value -= Isolated[opposed][f];
if (doubled) if (doubled)
value -= DoubledPawnPenalty[opposed][f]; value -= Doubled[opposed][f];
if (backward) if (backward)
value -= BackwardPawnPenalty[opposed][f]; value -= Backward[opposed][f];
if (chain) if (chain)
value += ChainBonus[f]; value += ChainMember[f];
if (candidate) if (candidate)
value += CandidateBonus[relative_rank(Us, s)]; value += CandidatePassed[relative_rank(Us, s)];
} }
return value; return value;
}
} // namespace
namespace Pawns {
/// probe() takes a position object as input, computes a Entry object, and returns
/// a pointer to it. The result is also stored in a hash table, so we don't have
/// to recompute everything when the same pawn structure occurs again.
Entry* probe(const Position& pos, Table& entries) {
Key key = pos.pawn_key();
Entry* e = entries[key];
if (e->key == key)
return e;
e->key = key;
e->value = evaluate<WHITE>(pos, e) - evaluate<BLACK>(pos, e);
return e;
} }
/// PawnEntry::shelter_storm() calculates shelter and storm penalties for the file /// Entry::shelter_storm() calculates shelter and storm penalties for the file
/// the king is on, as well as the two adjacent files. /// the king is on, as well as the two adjacent files.
template<Color Us> template<Color Us>
Value PawnEntry::shelter_storm(const Position& pos, Square ksq) { Value Entry::shelter_storm(const Position& pos, Square ksq) {
const Color Them = (Us == WHITE ? BLACK : WHITE); const Color Them = (Us == WHITE ? BLACK : WHITE);
Value safety = MaxSafetyBonus; Value safety = MaxSafetyBonus;
Bitboard b = pos.pieces(PAWN) & (in_front_bb(Us, ksq) | rank_bb(ksq)); Bitboard b = pos.pieces(PAWN) & (in_front_bb(Us, rank_of(ksq)) | rank_bb(ksq));
Bitboard ourPawns = b & pos.pieces(Us) & ~rank_bb(ksq); Bitboard ourPawns = b & pos.pieces(Us);
Bitboard theirPawns = b & pos.pieces(Them); Bitboard theirPawns = b & pos.pieces(Them);
Rank rkUs, rkThem; Rank rkUs, rkThem;
File kf = file_of(ksq); File kf = file_of(ksq);
kf = (kf == FILE_A) ? kf++ : (kf == FILE_H) ? kf-- : kf; kf = (kf == FILE_A) ? FILE_B : (kf == FILE_H) ? FILE_G : kf;
for (int f = kf - 1; f <= kf + 1; f++) for (int f = kf - 1; f <= kf + 1; f++)
{ {
// Shelter penalty is higher for the pawn in front of the king
b = ourPawns & FileBB[f]; b = ourPawns & FileBB[f];
rkUs = b ? rank_of(Us == WHITE ? lsb(b) : ~msb(b)) : RANK_1; rkUs = b ? relative_rank(Us, Us == WHITE ? lsb(b) : msb(b)) : RANK_1;
safety -= ShelterWeakness[f != kf][rkUs]; safety -= ShelterWeakness[rkUs];
// Storm danger is smaller if enemy pawn is blocked
b = theirPawns & FileBB[f]; b = theirPawns & FileBB[f];
rkThem = b ? rank_of(Us == WHITE ? lsb(b) : ~msb(b)) : RANK_1; rkThem = b ? relative_rank(Us, Us == WHITE ? lsb(b) : msb(b)) : RANK_1;
safety -= StormDanger[rkThem == rkUs + 1][rkThem]; safety -= StormDanger[rkUs == RANK_1 ? 0 : rkThem == rkUs + 1 ? 2 : 1][rkThem];
} }
return safety; return safety;
} }
/// PawnEntry::update_safety() calculates and caches a bonus for king safety. It is /// Entry::update_safety() calculates and caches a bonus for king safety. It is
/// called only when king square changes, about 20% of total king_safety() calls. /// called only when king square changes, about 20% of total king_safety() calls.
template<Color Us> template<Color Us>
Score PawnEntry::update_safety(const Position& pos, Square ksq) { Score Entry::update_safety(const Position& pos, Square ksq) {
kingSquares[Us] = ksq; kingSquares[Us] = ksq;
castleRights[Us] = pos.can_castle(Us); castleRights[Us] = pos.can_castle(Us);
@@ -283,5 +275,7 @@ Score PawnEntry::update_safety(const Position& pos, Square ksq) {
} }
// Explicit template instantiation // Explicit template instantiation
template Score PawnEntry::update_safety<WHITE>(const Position& pos, Square ksq); template Score Entry::update_safety<WHITE>(const Position& pos, Square ksq);
template Score PawnEntry::update_safety<BLACK>(const Position& pos, Square ksq); template Score Entry::update_safety<BLACK>(const Position& pos, Square ksq);
} // namespace Pawns
+33 -70
View File
@@ -1,7 +1,7 @@
/* /*
Stockfish, a UCI chess playing engine derived from Glaurung 2.1 Stockfish, a UCI chess playing engine derived from Glaurung 2.1
Copyright (C) 2004-2008 Tord Romstad (Glaurung author) Copyright (C) 2004-2008 Tord Romstad (Glaurung author)
Copyright (C) 2008-2012 Marco Costalba, Joona Kiiski, Tord Romstad Copyright (C) 2008-2013 Marco Costalba, Joona Kiiski, Tord Romstad
Stockfish is free software: you can redistribute it and/or modify Stockfish is free software: you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by it under the terms of the GNU General Public License as published by
@@ -17,38 +17,40 @@
along with this program. If not, see <http://www.gnu.org/licenses/>. along with this program. If not, see <http://www.gnu.org/licenses/>.
*/ */
#if !defined(PAWNS_H_INCLUDED) #ifndef PAWNS_H_INCLUDED
#define PAWNS_H_INCLUDED #define PAWNS_H_INCLUDED
#include "misc.h" #include "misc.h"
#include "position.h" #include "position.h"
#include "types.h" #include "types.h"
const int PawnTableSize = 16384; namespace Pawns {
/// PawnEntry is a class which contains various information about a pawn /// Pawns::Entry contains various information about a pawn structure. Currently,
/// structure. Currently, it only includes a middle game and an end game /// it only includes a middle game and end game pawn structure evaluation, and a
/// pawn structure evaluation, and a bitboard of passed pawns. We may want /// bitboard of passed pawns. We may want to add further information in the future.
/// to add further information in the future. A lookup to the pawn hash /// A lookup to the pawn hash table (performed by calling the probe function)
/// table (performed by calling the probe method in a PawnTable object) /// returns a pointer to an Entry object.
/// returns a pointer to a PawnEntry object.
class PawnEntry { struct Entry {
friend struct PawnTable; Score pawns_value() const { return value; }
Bitboard pawn_attacks(Color c) const { return pawnAttacks[c]; }
Bitboard passed_pawns(Color c) const { return passedPawns[c]; }
int pawns_on_same_color_squares(Color c, Square s) const { return pawnsOnSquares[c][!!(DarkSquares & s)]; }
int semiopen(Color c, File f) const { return semiopenFiles[c] & (1 << int(f)); }
int semiopen_on_side(Color c, File f, bool left) const {
public: return semiopenFiles[c] & (left ? ((1 << int(f)) - 1) : ~((1 << int(f+1)) - 1));
Score pawns_value() const; }
Bitboard pawn_attacks(Color c) const;
Bitboard passed_pawns(Color c) const;
int file_is_half_open(Color c, File f) const;
int has_open_file_to_left(Color c, File f) const;
int has_open_file_to_right(Color c, File f) const;
template<Color Us> template<Color Us>
Score king_safety(const Position& pos, Square ksq); Score king_safety(const Position& pos, Square ksq) {
return kingSquares[Us] == ksq && castleRights[Us] == pos.can_castle(Us)
? kingSafety[Us] : update_safety<Us>(pos, ksq);
}
private:
template<Color Us> template<Color Us>
Score update_safety(const Position& pos, Square ksq); Score update_safety(const Position& pos, Square ksq);
@@ -56,60 +58,21 @@ private:
Value shelter_storm(const Position& pos, Square ksq); Value shelter_storm(const Position& pos, Square ksq);
Key key; Key key;
Bitboard passedPawns[2]; Bitboard passedPawns[COLOR_NB];
Bitboard pawnAttacks[2]; Bitboard pawnAttacks[COLOR_NB];
Square kingSquares[2]; Square kingSquares[COLOR_NB];
int minKPdistance[2]; int minKPdistance[COLOR_NB];
int castleRights[2]; int castleRights[COLOR_NB];
Score value; Score value;
int halfOpenFiles[2]; int semiopenFiles[COLOR_NB];
Score kingSafety[2]; Score kingSafety[COLOR_NB];
int pawnsOnSquares[COLOR_NB][COLOR_NB];
}; };
typedef HashTable<Entry, 16384> Table;
/// The PawnTable class represents a pawn hash table. The most important Entry* probe(const Position& pos, Table& entries);
/// method is probe, which returns a pointer to a PawnEntry object.
struct PawnTable {
PawnEntry* probe(const Position& pos);
template<Color Us>
static Score evaluate_pawns(const Position& pos, Bitboard ourPawns,
Bitboard theirPawns, PawnEntry* e);
HashTable<PawnEntry, PawnTableSize> entries;
};
inline Score PawnEntry::pawns_value() const {
return value;
} }
inline Bitboard PawnEntry::pawn_attacks(Color c) const { #endif // #ifndef PAWNS_H_INCLUDED
return pawnAttacks[c];
}
inline Bitboard PawnEntry::passed_pawns(Color c) const {
return passedPawns[c];
}
inline int PawnEntry::file_is_half_open(Color c, File f) const {
return halfOpenFiles[c] & (1 << int(f));
}
inline int PawnEntry::has_open_file_to_left(Color c, File f) const {
return halfOpenFiles[c] & ((1 << int(f)) - 1);
}
inline int PawnEntry::has_open_file_to_right(Color c, File f) const {
return halfOpenFiles[c] & ~((1 << int(f+1)) - 1);
}
template<Color Us>
inline Score PawnEntry::king_safety(const Position& pos, Square ksq) {
return kingSquares[Us] == ksq && castleRights[Us] == pos.can_castle(Us)
? kingSafety[Us] : update_safety<Us>(pos, ksq);
}
#endif // !defined(PAWNS_H_INCLUDED)
+21 -14
View File
@@ -1,7 +1,7 @@
/* /*
Stockfish, a UCI chess playing engine derived from Glaurung 2.1 Stockfish, a UCI chess playing engine derived from Glaurung 2.1
Copyright (C) 2004-2008 Tord Romstad (Glaurung author) Copyright (C) 2004-2008 Tord Romstad (Glaurung author)
Copyright (C) 2008-2012 Marco Costalba, Joona Kiiski, Tord Romstad Copyright (C) 2008-2013 Marco Costalba, Joona Kiiski, Tord Romstad
Stockfish is free software: you can redistribute it and/or modify Stockfish is free software: you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by it under the terms of the GNU General Public License as published by
@@ -17,10 +17,10 @@
along with this program. If not, see <http://www.gnu.org/licenses/>. along with this program. If not, see <http://www.gnu.org/licenses/>.
*/ */
#if !defined(PLATFORM_H_INCLUDED) #ifndef PLATFORM_H_INCLUDED
#define PLATFORM_H_INCLUDED #define PLATFORM_H_INCLUDED
#if defined(_MSC_VER) #ifdef _MSC_VER
// Disable some silly and noisy warning from MSVC compiler // Disable some silly and noisy warning from MSVC compiler
#pragma warning(disable: 4127) // Conditional expression is constant #pragma warning(disable: 4127) // Conditional expression is constant
@@ -42,13 +42,15 @@ typedef unsigned __int64 uint64_t;
# include <inttypes.h> # include <inttypes.h>
#endif #endif
#if !defined(_WIN32) && !defined(_WIN64) // Linux - Unix #ifndef _WIN32 // Linux - Unix
# include <sys/time.h> # include <sys/time.h>
typedef timeval sys_time_t;
inline void system_time(sys_time_t* t) { gettimeofday(t, NULL); } inline int64_t system_time_to_msec() {
inline int64_t time_to_msec(const sys_time_t& t) { return t.tv_sec * 1000LL + t.tv_usec / 1000; } timeval t;
gettimeofday(&t, NULL);
return t.tv_sec * 1000LL + t.tv_usec / 1000;
}
# include <pthread.h> # include <pthread.h>
typedef pthread_mutex_t Lock; typedef pthread_mutex_t Lock;
@@ -65,18 +67,20 @@ typedef void*(*pt_start_fn)(void*);
# define cond_signal(x) pthread_cond_signal(&(x)) # define cond_signal(x) pthread_cond_signal(&(x))
# define cond_wait(x,y) pthread_cond_wait(&(x),&(y)) # define cond_wait(x,y) pthread_cond_wait(&(x),&(y))
# define cond_timedwait(x,y,z) pthread_cond_timedwait(&(x),&(y),z) # define cond_timedwait(x,y,z) pthread_cond_timedwait(&(x),&(y),z)
# define thread_create(x,f,t) !pthread_create(&(x),NULL,(pt_start_fn)f,t) # define thread_create(x,f,t) pthread_create(&(x),NULL,(pt_start_fn)f,t)
# define thread_join(x) pthread_join(x, NULL) # define thread_join(x) pthread_join(x, NULL)
#else // Windows and MinGW #else // Windows and MinGW
# include <sys/timeb.h> # include <sys/timeb.h>
typedef _timeb sys_time_t;
inline void system_time(sys_time_t* t) { _ftime(t); } inline int64_t system_time_to_msec() {
inline int64_t time_to_msec(const sys_time_t& t) { return t.time * 1000LL + t.millitm; } _timeb t;
_ftime(&t);
return t.time * 1000LL + t.millitm;
}
#if !defined(NOMINMAX) #ifndef NOMINMAX
# define NOMINMAX // disable macros min() and max() # define NOMINMAX // disable macros min() and max()
#endif #endif
@@ -92,6 +96,9 @@ typedef CRITICAL_SECTION Lock;
typedef HANDLE WaitCondition; typedef HANDLE WaitCondition;
typedef HANDLE NativeHandle; typedef HANDLE NativeHandle;
// On Windows 95 and 98 parameter lpThreadId my not be null
inline DWORD* dwWin9xKludge() { static DWORD dw; return &dw; }
# define lock_init(x) InitializeCriticalSection(&(x)) # define lock_init(x) InitializeCriticalSection(&(x))
# define lock_grab(x) EnterCriticalSection(&(x)) # define lock_grab(x) EnterCriticalSection(&(x))
# define lock_release(x) LeaveCriticalSection(&(x)) # define lock_release(x) LeaveCriticalSection(&(x))
@@ -101,9 +108,9 @@ typedef HANDLE NativeHandle;
# define cond_signal(x) SetEvent(x) # define cond_signal(x) SetEvent(x)
# define cond_wait(x,y) { lock_release(y); WaitForSingleObject(x, INFINITE); lock_grab(y); } # define cond_wait(x,y) { lock_release(y); WaitForSingleObject(x, INFINITE); lock_grab(y); }
# define cond_timedwait(x,y,z) { lock_release(y); WaitForSingleObject(x,z); lock_grab(y); } # define cond_timedwait(x,y,z) { lock_release(y); WaitForSingleObject(x,z); lock_grab(y); }
# define thread_create(x,f,t) (x = CreateThread(NULL,0,(LPTHREAD_START_ROUTINE)f,t,0,NULL), x != NULL) # define thread_create(x,f,t) (x = CreateThread(NULL,0,(LPTHREAD_START_ROUTINE)f,t,0,dwWin9xKludge()))
# define thread_join(x) { WaitForSingleObject(x, INFINITE); CloseHandle(x); } # define thread_join(x) { WaitForSingleObject(x, INFINITE); CloseHandle(x); }
#endif #endif
#endif // !defined(PLATFORM_H_INCLUDED) #endif // #ifndef PLATFORM_H_INCLUDED
+313 -550
View File
File diff suppressed because it is too large Load Diff
+90 -64
View File
@@ -1,7 +1,7 @@
/* /*
Stockfish, a UCI chess playing engine derived from Glaurung 2.1 Stockfish, a UCI chess playing engine derived from Glaurung 2.1
Copyright (C) 2004-2008 Tord Romstad (Glaurung author) Copyright (C) 2004-2008 Tord Romstad (Glaurung author)
Copyright (C) 2008-2012 Marco Costalba, Joona Kiiski, Tord Romstad Copyright (C) 2008-2013 Marco Costalba, Joona Kiiski, Tord Romstad
Stockfish is free software: you can redistribute it and/or modify Stockfish is free software: you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by it under the terms of the GNU General Public License as published by
@@ -17,10 +17,11 @@
along with this program. If not, see <http://www.gnu.org/licenses/>. along with this program. If not, see <http://www.gnu.org/licenses/>.
*/ */
#if !defined(POSITION_H_INCLUDED) #ifndef POSITION_H_INCLUDED
#define POSITION_H_INCLUDED #define POSITION_H_INCLUDED
#include <cassert> #include <cassert>
#include <cstddef>
#include "bitboard.h" #include "bitboard.h"
#include "types.h" #include "types.h"
@@ -29,7 +30,7 @@
/// The checkInfo struct is initialized at c'tor time and keeps info used /// The checkInfo struct is initialized at c'tor time and keeps info used
/// to detect if a move gives check. /// to detect if a move gives check.
class Position; class Position;
class Thread; struct Thread;
struct CheckInfo { struct CheckInfo {
@@ -37,21 +38,21 @@ struct CheckInfo {
Bitboard dcCandidates; Bitboard dcCandidates;
Bitboard pinned; Bitboard pinned;
Bitboard checkSq[8]; Bitboard checkSq[PIECE_TYPE_NB];
Square ksq; Square ksq;
}; };
/// The StateInfo struct stores information we need to restore a Position /// The StateInfo struct stores information we need to restore a Position
/// object to its previous state when we retract a move. Whenever a move /// object to its previous state when we retract a move. Whenever a move
/// is made on the board (by calling Position::do_move), an StateInfo object /// is made on the board (by calling Position::do_move), a StateInfo object
/// must be passed as a parameter. /// must be passed as a parameter.
struct StateInfo { struct StateInfo {
Key pawnKey, materialKey; Key pawnKey, materialKey;
Value npMaterial[2]; Value npMaterial[COLOR_NB];
int castleRights, rule50, pliesFromNull; int castleRights, rule50, pliesFromNull;
Score psqScore; Score psq;
Square epSquare; Square epSquare;
Key key; Key key;
@@ -60,13 +61,10 @@ struct StateInfo {
StateInfo* previous; StateInfo* previous;
}; };
struct ReducedStateInfo {
Key pawnKey, materialKey; /// When making a move the current StateInfo up to 'key' excluded is copied to
Value npMaterial[2]; /// the new one. Here we calculate the quad words (64bits) needed to be copied.
int castleRights, rule50, pliesFromNull; const size_t StateCopySize64 = offsetof(StateInfo, key) / sizeof(uint64_t) + 1;
Score psqScore;
Square epSquare;
};
/// The position data structure. A position consists of the following data: /// The position data structure. A position consists of the following data:
@@ -95,13 +93,14 @@ class Position {
public: public:
Position() {} Position() {}
Position(const Position& p, Thread* t) { *this = p; thisThread = t; } Position(const Position& p, Thread* t) { *this = p; thisThread = t; }
Position(const std::string& f, bool c960, Thread* t) { from_fen(f, c960, t); } Position(const std::string& f, bool c960, Thread* t) { set(f, c960, t); }
Position& operator=(const Position&); Position& operator=(const Position&);
static void init();
// Text input/output // Text input/output
void from_fen(const std::string& fen, bool isChess960, Thread* th); void set(const std::string& fen, bool isChess960, Thread* th);
const std::string to_fen() const; const std::string fen() const;
void print(Move m = MOVE_NONE) const; const std::string pretty(Move m = MOVE_NONE) const;
// Position representation // Position representation
Bitboard pieces() const; Bitboard pieces() const;
@@ -114,8 +113,8 @@ public:
Square king_square(Color c) const; Square king_square(Color c) const;
Square ep_square() const; Square ep_square() const;
bool is_empty(Square s) const; bool is_empty(Square s) const;
const Square* piece_list(Color c, PieceType pt) const; template<PieceType Pt> int count(Color c) const;
int piece_count(Color c, PieceType pt) const; template<PieceType Pt> const Square* list(Color c) const;
// Castling // Castling
int can_castle(CastleRight f) const; int can_castle(CastleRight f) const;
@@ -124,7 +123,6 @@ public:
Square castle_rook_square(Color c, CastlingSide s) const; Square castle_rook_square(Color c, CastlingSide s) const;
// Checking // Checking
bool in_check() const;
Bitboard checkers() const; Bitboard checkers() const;
Bitboard discovered_check_candidates() const; Bitboard discovered_check_candidates() const;
Bitboard pinned_pieces() const; Bitboard pinned_pieces() const;
@@ -139,8 +137,6 @@ public:
// Properties of moves // Properties of moves
bool move_gives_check(Move m, const CheckInfo& ci) const; bool move_gives_check(Move m, const CheckInfo& ci) const;
bool move_attacks_square(Move m, Square s) const;
bool move_is_legal(const Move m) const;
bool pl_move_is_legal(Move m, Bitboard pinned) const; bool pl_move_is_legal(Move m, Bitboard pinned) const;
bool is_pseudo_legal(const Move m) const; bool is_pseudo_legal(const Move m) const;
bool is_capture(Move m) const; bool is_capture(Move m) const;
@@ -159,10 +155,11 @@ public:
void do_move(Move m, StateInfo& st); void do_move(Move m, StateInfo& st);
void do_move(Move m, StateInfo& st, const CheckInfo& ci, bool moveIsCheck); void do_move(Move m, StateInfo& st, const CheckInfo& ci, bool moveIsCheck);
void undo_move(Move m); void undo_move(Move m);
template<bool Do> void do_null_move(StateInfo& st); void do_null_move(StateInfo& st);
void undo_null_move();
// Static exchange evaluation // Static exchange evaluation
int see(Move m) const; int see(Move m, int asymmThreshold = 0) const;
int see_sign(Move m) const; int see_sign(Move m) const;
// Accessing hash keys // Accessing hash keys
@@ -173,17 +170,16 @@ public:
// Incremental piece-square evaluation // Incremental piece-square evaluation
Score psq_score() const; Score psq_score() const;
Score psq_delta(Piece p, Square from, Square to) const;
Value non_pawn_material(Color c) const; Value non_pawn_material(Color c) const;
// Other properties of the position // Other properties of the position
Color side_to_move() const; Color side_to_move() const;
int startpos_ply_counter() const; int game_ply() const;
bool is_chess960() const; bool is_chess960() const;
Thread* this_thread() const; Thread* this_thread() const;
int64_t nodes_searched() const; int64_t nodes_searched() const;
void set_nodes_searched(int64_t n); void set_nodes_searched(int64_t n);
template<bool SkipRepetition> bool is_draw() const; bool is_draw() const;
// Position consistency check, for debugging // Position consistency check, for debugging
bool pos_is_ok(int* failedStep = NULL) const; bool pos_is_ok(int* failedStep = NULL) const;
@@ -192,12 +188,14 @@ public:
private: private:
// Initialization helpers (used while setting up a position) // Initialization helpers (used while setting up a position)
void clear(); void clear();
void put_piece(Piece p, Square s);
void set_castle_right(Color c, Square rfrom); void set_castle_right(Color c, Square rfrom);
// Helper template functions // Helper functions
template<bool Do> void do_castle_move(Move m); void do_castle(Square kfrom, Square kto, Square rfrom, Square rto);
template<bool FindPinned> Bitboard hidden_checkers() const; Bitboard hidden_checkers(Square ksq, Color c) const;
void put_piece(Square s, Color c, PieceType pt);
void remove_piece(Square s, Color c, PieceType pt);
void move_piece(Square from, Square to, Color c, PieceType pt);
// Computing hash keys from scratch (for initialization and debugging) // Computing hash keys from scratch (for initialization and debugging)
Key compute_key() const; Key compute_key() const;
@@ -209,20 +207,20 @@ private:
Value compute_non_pawn_material(Color c) const; Value compute_non_pawn_material(Color c) const;
// Board and pieces // Board and pieces
Piece board[64]; // [square] Piece board[SQUARE_NB];
Bitboard byTypeBB[8]; // [pieceType] Bitboard byTypeBB[PIECE_TYPE_NB];
Bitboard byColorBB[2]; // [color] Bitboard byColorBB[COLOR_NB];
int pieceCount[2][8]; // [color][pieceType] int pieceCount[COLOR_NB][PIECE_TYPE_NB];
Square pieceList[2][8][16]; // [color][pieceType][index] Square pieceList[COLOR_NB][PIECE_TYPE_NB][16];
int index[64]; // [square] int index[SQUARE_NB];
// Other info // Other info
int castleRightsMask[64]; // [square] int castleRightsMask[SQUARE_NB];
Square castleRookSquare[2][2]; // [color][side] Square castleRookSquare[COLOR_NB][CASTLING_SIDE_NB];
Bitboard castlePath[2][2]; // [color][side] Bitboard castlePath[COLOR_NB][CASTLING_SIDE_NB];
StateInfo startState; StateInfo startState;
int64_t nodes; int64_t nodes;
int startPosPly; int gamePly;
Color sideToMove; Color sideToMove;
Thread* thisThread; Thread* thisThread;
StateInfo* st; StateInfo* st;
@@ -277,12 +275,12 @@ inline Bitboard Position::pieces(Color c, PieceType pt1, PieceType pt2) const {
return byColorBB[c] & (byTypeBB[pt1] | byTypeBB[pt2]); return byColorBB[c] & (byTypeBB[pt1] | byTypeBB[pt2]);
} }
inline int Position::piece_count(Color c, PieceType pt) const { template<PieceType Pt> inline int Position::count(Color c) const {
return pieceCount[c][pt]; return pieceCount[c][Pt];
} }
inline const Square* Position::piece_list(Color c, PieceType pt) const { template<PieceType Pt> inline const Square* Position::list(Color c) const {
return pieceList[c][pt]; return pieceList[c][Pt];
} }
inline Square Position::ep_square() const { inline Square Position::ep_square() const {
@@ -334,16 +332,12 @@ inline Bitboard Position::checkers() const {
return st->checkersBB; return st->checkersBB;
} }
inline bool Position::in_check() const {
return st->checkersBB != 0;
}
inline Bitboard Position::discovered_check_candidates() const { inline Bitboard Position::discovered_check_candidates() const {
return hidden_checkers<false>(); return hidden_checkers(king_square(~sideToMove), sideToMove);
} }
inline Bitboard Position::pinned_pieces() const { inline Bitboard Position::pinned_pieces() const {
return hidden_checkers<true>(); return hidden_checkers(king_square(sideToMove), ~sideToMove);
} }
inline bool Position::pawn_is_passed(Color c, Square s) const { inline bool Position::pawn_is_passed(Color c, Square s) const {
@@ -354,10 +348,6 @@ inline Key Position::key() const {
return st->key; return st->key;
} }
inline Key Position::exclusion_key() const {
return st->key ^ Zobrist::exclusion;
}
inline Key Position::pawn_key() const { inline Key Position::pawn_key() const {
return st->pawnKey; return st->pawnKey;
} }
@@ -366,12 +356,8 @@ inline Key Position::material_key() const {
return st->materialKey; return st->materialKey;
} }
inline Score Position::psq_delta(Piece p, Square from, Square to) const {
return pieceSquareTable[p][to] - pieceSquareTable[p][from];
}
inline Score Position::psq_score() const { inline Score Position::psq_score() const {
return st->psqScore; return st->psq;
} }
inline Value Position::non_pawn_material(Color c) const { inline Value Position::non_pawn_material(Color c) const {
@@ -384,8 +370,8 @@ inline bool Position::is_passed_pawn_push(Move m) const {
&& pawn_is_passed(sideToMove, to_sq(m)); && pawn_is_passed(sideToMove, to_sq(m));
} }
inline int Position::startpos_ply_counter() const { inline int Position::game_ply() const {
return startPosPly + st->pliesFromNull; // HACK return gamePly;
} }
inline bool Position::opposite_bishops() const { inline bool Position::opposite_bishops() const {
@@ -430,4 +416,44 @@ inline Thread* Position::this_thread() const {
return thisThread; return thisThread;
} }
#endif // !defined(POSITION_H_INCLUDED) inline void Position::put_piece(Square s, Color c, PieceType pt) {
board[s] = make_piece(c, pt);
byTypeBB[ALL_PIECES] |= s;
byTypeBB[pt] |= s;
byColorBB[c] |= s;
index[s] = pieceCount[c][pt]++;
pieceList[c][pt][index[s]] = s;
}
inline void Position::move_piece(Square from, Square to, Color c, PieceType pt) {
// index[from] is not updated and becomes stale. This works as long
// as index[] is accessed just by known occupied squares.
Bitboard from_to_bb = SquareBB[from] ^ SquareBB[to];
byTypeBB[ALL_PIECES] ^= from_to_bb;
byTypeBB[pt] ^= from_to_bb;
byColorBB[c] ^= from_to_bb;
board[from] = NO_PIECE;
board[to] = make_piece(c, pt);
index[to] = index[from];
pieceList[c][pt][index[to]] = to;
}
inline void Position::remove_piece(Square s, Color c, PieceType pt) {
// WARNING: This is not a reversible operation. If we remove a piece in
// do_move() and then replace it in undo_move() we will put it at the end of
// the list and not in its original place, it means index[] and pieceList[]
// are not guaranteed to be invariant to a do_move() + undo_move() sequence.
byTypeBB[ALL_PIECES] ^= s;
byTypeBB[pt] ^= s;
byColorBB[c] ^= s;
/* board[s] = NO_PIECE; */ // Not needed, will be overwritten by capturing
Square lastSquare = pieceList[c][pt][--pieceCount[c][pt]];
index[lastSquare] = index[s];
pieceList[c][pt][index[lastSquare]] = lastSquare;
pieceList[c][pt][pieceCount[c][pt]] = SQ_NONE;
}
#endif // #ifndef POSITION_H_INCLUDED
+10 -10
View File
@@ -1,7 +1,7 @@
/* /*
Stockfish, a UCI chess playing engine derived from Glaurung 2.1 Stockfish, a UCI chess playing engine derived from Glaurung 2.1
Copyright (C) 2004-2008 Tord Romstad (Glaurung author) Copyright (C) 2004-2008 Tord Romstad (Glaurung author)
Copyright (C) 2008-2012 Marco Costalba, Joona Kiiski, Tord Romstad Copyright (C) 2008-2013 Marco Costalba, Joona Kiiski, Tord Romstad
Stockfish is free software: you can redistribute it and/or modify Stockfish is free software: you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by it under the terms of the GNU General Public License as published by
@@ -17,7 +17,7 @@
along with this program. If not, see <http://www.gnu.org/licenses/>. along with this program. If not, see <http://www.gnu.org/licenses/>.
*/ */
#if !defined(PSQTAB_H_INCLUDED) #ifndef PSQTAB_H_INCLUDED
#define PSQTAB_H_INCLUDED #define PSQTAB_H_INCLUDED
#include "types.h" #include "types.h"
@@ -29,16 +29,16 @@
/// a given square a (midgame, endgame) score pair is assigned. PSQT is defined /// a given square a (midgame, endgame) score pair is assigned. PSQT is defined
/// for white side, for black side the tables are symmetric. /// for white side, for black side the tables are symmetric.
static const Score PSQT[][64] = { static const Score PSQT[][SQUARE_NB] = {
{ }, { },
{ // Pawn { // Pawn
S( 0, 0), S( 0, 0), S( 0, 0), S( 0, 0), S(0, 0), S( 0, 0), S( 0, 0), S( 0, 0), S( 0, 0), S( 0, 0), S( 0, 0), S( 0, 0), S(0, 0), S( 0, 0), S( 0, 0), S( 0, 0),
S(-28,-8), S(-6,-8), S( 4,-8), S(14,-8), S(14,-8), S( 4,-8), S(-6,-8), S(-28,-8), S(-20,-8), S(-6,-8), S( 4,-8), S(14,-8), S(14,-8), S( 4,-8), S(-6,-8), S(-20,-8),
S(-28,-8), S(-6,-8), S( 9,-8), S(36,-8), S(36,-8), S( 9,-8), S(-6,-8), S(-28,-8), S(-20,-8), S(-6,-8), S( 9,-8), S(34,-8), S(34,-8), S( 9,-8), S(-6,-8), S(-20,-8),
S(-28,-8), S(-6,-8), S(17,-8), S(58,-8), S(58,-8), S(17,-8), S(-6,-8), S(-28,-8), S(-20,-8), S(-6,-8), S(17,-8), S(54,-8), S(54,-8), S(17,-8), S(-6,-8), S(-20,-8),
S(-28,-8), S(-6,-8), S(17,-8), S(36,-8), S(36,-8), S(17,-8), S(-6,-8), S(-28,-8), S(-20,-8), S(-6,-8), S(17,-8), S(34,-8), S(34,-8), S(17,-8), S(-6,-8), S(-20,-8),
S(-28,-8), S(-6,-8), S( 9,-8), S(14,-8), S(14,-8), S( 9,-8), S(-6,-8), S(-28,-8), S(-20,-8), S(-6,-8), S( 9,-8), S(14,-8), S(14,-8), S( 9,-8), S(-6,-8), S(-20,-8),
S(-28,-8), S(-6,-8), S( 4,-8), S(14,-8), S(14,-8), S( 4,-8), S(-6,-8), S(-28,-8), S(-20,-8), S(-6,-8), S( 4,-8), S(14,-8), S(14,-8), S( 4,-8), S(-6,-8), S(-20,-8),
S( 0, 0), S( 0, 0), S( 0, 0), S( 0, 0), S(0, 0), S( 0, 0), S( 0, 0), S( 0, 0) S( 0, 0), S( 0, 0), S( 0, 0), S( 0, 0), S(0, 0), S( 0, 0), S( 0, 0), S( 0, 0)
}, },
{ // Knight { // Knight
@@ -95,4 +95,4 @@ static const Score PSQT[][64] = {
#undef S #undef S
#endif // !defined(PSQTAB_H_INCLUDED) #endif // #ifndef PSQTAB_H_INCLUDED
+7 -11
View File
@@ -1,7 +1,7 @@
/* /*
Stockfish, a UCI chess playing engine derived from Glaurung 2.1 Stockfish, a UCI chess playing engine derived from Glaurung 2.1
Copyright (C) 2004-2008 Tord Romstad (Glaurung author) Copyright (C) 2004-2008 Tord Romstad (Glaurung author)
Copyright (C) 2008-2012 Marco Costalba, Joona Kiiski, Tord Romstad Copyright (C) 2008-2013 Marco Costalba, Joona Kiiski, Tord Romstad
Stockfish is free software: you can redistribute it and/or modify Stockfish is free software: you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by it under the terms of the GNU General Public License as published by
@@ -22,7 +22,7 @@
(at your option) any later version. (at your option) any later version.
*/ */
#if !defined(RKISS_H_INCLUDED) #ifndef RKISS_H_INCLUDED
#define RKISS_H_INCLUDED #define RKISS_H_INCLUDED
#include "types.h" #include "types.h"
@@ -43,14 +43,12 @@
class RKISS { class RKISS {
// Keep variables always together struct S { uint64_t a, b, c, d; } s; // Keep variables always together
struct S { uint64_t a, b, c, d; } s;
uint64_t rotate(uint64_t x, uint64_t k) const { uint64_t rotate(uint64_t x, uint64_t k) const {
return (x << k) | (x >> (64 - k)); return (x << k) | (x >> (64 - k));
} }
// Return 64 bit unsigned integer in between [0, 2^64 - 1]
uint64_t rand64() { uint64_t rand64() {
const uint64_t const uint64_t
@@ -61,18 +59,16 @@ class RKISS {
return s.d = e + s.a; return s.d = e + s.a;
} }
// Init seed and scramble a few rounds public:
void raninit() { RKISS(int seed = 73) {
s.a = 0xf1ea5eed; s.a = 0xf1ea5eed;
s.b = s.c = s.d = 0xd4e12c77; s.b = s.c = s.d = 0xd4e12c77;
for (int i = 0; i < 73; i++) for (int i = 0; i < seed; i++) // Scramble a few rounds
rand64(); rand64();
} }
public:
RKISS() { raninit(); }
template<typename T> T rand() { return T(rand64()); } template<typename T> T rand() { return T(rand64()); }
}; };
#endif // !defined(RKISS_H_INCLUDED) #endif // #ifndef RKISS_H_INCLUDED
+765 -736
View File
File diff suppressed because it is too large Load Diff
+12 -11
View File
@@ -1,7 +1,7 @@
/* /*
Stockfish, a UCI chess playing engine derived from Glaurung 2.1 Stockfish, a UCI chess playing engine derived from Glaurung 2.1
Copyright (C) 2004-2008 Tord Romstad (Glaurung author) Copyright (C) 2004-2008 Tord Romstad (Glaurung author)
Copyright (C) 2008-2012 Marco Costalba, Joona Kiiski, Tord Romstad Copyright (C) 2008-2013 Marco Costalba, Joona Kiiski, Tord Romstad
Stockfish is free software: you can redistribute it and/or modify Stockfish is free software: you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by it under the terms of the GNU General Public License as published by
@@ -17,7 +17,7 @@
along with this program. If not, see <http://www.gnu.org/licenses/>. along with this program. If not, see <http://www.gnu.org/licenses/>.
*/ */
#if !defined(SEARCH_H_INCLUDED) #ifndef SEARCH_H_INCLUDED
#define SEARCH_H_INCLUDED #define SEARCH_H_INCLUDED
#include <cstring> #include <cstring>
@@ -38,15 +38,16 @@ namespace Search {
/// has its own array of Stack objects, indexed by the current ply. /// has its own array of Stack objects, indexed by the current ply.
struct Stack { struct Stack {
SplitPoint* sp; SplitPoint* splitPoint;
int ply; int ply;
Move currentMove; Move currentMove;
Move excludedMove; Move excludedMove;
Move killers[2]; Move killers[2];
Depth reduction; Depth reduction;
Value eval; Value staticEval;
Value evalMargin; Value evalMargin;
int skipNullMove; int skipNullMove;
int futilityMoveCount;
}; };
@@ -56,12 +57,11 @@ struct Stack {
/// all non-pv moves. /// all non-pv moves.
struct RootMove { struct RootMove {
RootMove(){} // Needed by sort()
RootMove(Move m) : score(-VALUE_INFINITE), prevScore(-VALUE_INFINITE) { RootMove(Move m) : score(-VALUE_INFINITE), prevScore(-VALUE_INFINITE) {
pv.push_back(m); pv.push_back(MOVE_NONE); pv.push_back(m); pv.push_back(MOVE_NONE);
} }
bool operator<(const RootMove& m) const { return score < m.score; } bool operator<(const RootMove& m) const { return score > m.score; } // Ascending sort
bool operator==(const Move& m) const { return pv[0] == m; } bool operator==(const Move& m) const { return pv[0] == m; }
void extract_pv_from_tt(Position& pos); void extract_pv_from_tt(Position& pos);
@@ -79,10 +79,10 @@ struct RootMove {
struct LimitsType { struct LimitsType {
LimitsType() { memset(this, 0, sizeof(LimitsType)); } LimitsType() { std::memset(this, 0, sizeof(LimitsType)); }
bool use_time_management() const { return !(movetime | depth | nodes | infinite); } bool use_time_management() const { return !(mate | movetime | depth | nodes | infinite); }
int time[2], inc[2], movestogo, depth, nodes, movetime, infinite, ponder; int time[COLOR_NB], inc[COLOR_NB], movestogo, depth, nodes, movetime, mate, infinite, ponder;
}; };
@@ -98,7 +98,8 @@ typedef std::auto_ptr<std::stack<StateInfo> > StateStackPtr;
extern volatile SignalsType Signals; extern volatile SignalsType Signals;
extern LimitsType Limits; extern LimitsType Limits;
extern std::vector<RootMove> RootMoves; extern std::vector<RootMove> RootMoves;
extern Position RootPosition; extern Position RootPos;
extern Color RootColor;
extern Time::point SearchTime; extern Time::point SearchTime;
extern StateStackPtr SetupStates; extern StateStackPtr SetupStates;
@@ -108,4 +109,4 @@ extern void think();
} // namespace Search } // namespace Search
#endif // !defined(SEARCH_H_INCLUDED) #endif // #ifndef SEARCH_H_INCLUDED
+208 -250
View File
@@ -1,7 +1,7 @@
/* /*
Stockfish, a UCI chess playing engine derived from Glaurung 2.1 Stockfish, a UCI chess playing engine derived from Glaurung 2.1
Copyright (C) 2004-2008 Tord Romstad (Glaurung author) Copyright (C) 2004-2008 Tord Romstad (Glaurung author)
Copyright (C) 2008-2012 Marco Costalba, Joona Kiiski, Tord Romstad Copyright (C) 2008-2013 Marco Costalba, Joona Kiiski, Tord Romstad
Stockfish is free software: you can redistribute it and/or modify Stockfish is free software: you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by it under the terms of the GNU General Public License as published by
@@ -17,8 +17,8 @@
along with this program. If not, see <http://www.gnu.org/licenses/>. along with this program. If not, see <http://www.gnu.org/licenses/>.
*/ */
#include <algorithm> // For std::count
#include <cassert> #include <cassert>
#include <iostream>
#include "movegen.h" #include "movegen.h"
#include "search.h" #include "search.h"
@@ -29,101 +29,37 @@ using namespace Search;
ThreadPool Threads; // Global object ThreadPool Threads; // Global object
namespace { extern "C" { namespace {
// start_routine() is the C function which is called when a new thread // start_routine() is the C function which is called when a new thread
// is launched. It is a wrapper to member function pointed by start_fn. // is launched. It is a wrapper to the virtual function idle_loop().
long start_routine(Thread* th) { (th->*(th->start_fn))(); return 0; } extern "C" { long start_routine(ThreadBase* th) { th->idle_loop(); return 0; } }
} }
// Thread c'tor starts a newly-created thread of execution that will call // Helpers to launch a thread after creation and joining before delete. Must be
// the idle loop function pointed by start_fn going immediately to sleep. // outside Thread c'tor and d'tor because object shall be fully initialized
// when start_routine (and hence virtual idle_loop) is called and when joining.
Thread::Thread(Fn fn) { template<typename T> T* new_thread() {
T* th = new T();
is_searching = do_exit = false; thread_create(th->handle, start_routine, th); // Will go to sleep
maxPly = splitPointsCnt = 0; return th;
curSplitPoint = NULL;
start_fn = fn;
idx = Threads.size();
do_sleep = (fn != &Thread::main_loop); // Avoid a race with start_searching()
if (!thread_create(handle, start_routine, this))
{
std::cerr << "Failed to create thread number " << idx << std::endl;
::exit(EXIT_FAILURE);
} }
void delete_thread(ThreadBase* th) {
th->exit = true; // Search must be already finished
th->notify_one();
thread_join(th->handle); // Wait for thread termination
delete th;
}
} }
// Thread d'tor waits for thread termination before to return. // ThreadBase::notify_one() wakes up the thread when there is some work to do
Thread::~Thread() { void ThreadBase::notify_one() {
assert(do_sleep);
do_exit = true; // Search must be already finished
wake_up();
thread_join(handle); // Wait for thread termination
}
// Thread::timer_loop() is where the timer thread waits maxPly milliseconds and
// then calls check_time(). If maxPly is 0 thread sleeps until is woken up.
extern void check_time();
void Thread::timer_loop() {
while (!do_exit)
{
mutex.lock();
sleepCondition.wait_for(mutex, maxPly ? maxPly : INT_MAX);
mutex.unlock();
check_time();
}
}
// Thread::main_loop() is where the main thread is parked waiting to be started
// when there is a new search. Main thread will launch all the slave threads.
void Thread::main_loop() {
while (true)
{
mutex.lock();
do_sleep = true; // Always return to sleep after a search
is_searching = false;
while (do_sleep && !do_exit)
{
Threads.sleepCondition.notify_one(); // Wake up UI thread if needed
sleepCondition.wait(mutex);
}
mutex.unlock();
if (do_exit)
return;
is_searching = true;
Search::think();
assert(is_searching);
}
}
// Thread::wake_up() wakes up the thread, normally at the beginning of the search
// or, if "sleeping threads" is used at split time.
void Thread::wake_up() {
mutex.lock(); mutex.lock();
sleepCondition.notify_one(); sleepCondition.notify_one();
@@ -131,29 +67,89 @@ void Thread::wake_up() {
} }
// Thread::wait_for_stop_or_ponderhit() is called when the maximum depth is // ThreadBase::wait_for() set the thread to sleep until condition 'b' turns true
// reached while the program is pondering. The point is to work around a wrinkle
// in the UCI protocol: When pondering, the engine is not allowed to give a
// "bestmove" before the GUI sends it a "stop" or "ponderhit" command. We simply
// wait here until one of these commands (that raise StopRequest) is sent and
// then return, after which the bestmove and pondermove will be printed.
void Thread::wait_for_stop_or_ponderhit() { void ThreadBase::wait_for(volatile const bool& b) {
Signals.stopOnPonderhit = true;
mutex.lock(); mutex.lock();
while (!Signals.stop) sleepCondition.wait(mutex);; while (!b) sleepCondition.wait(mutex);
mutex.unlock(); mutex.unlock();
} }
// Thread c'tor just inits data but does not launch any thread of execution that
// instead will be started only upon c'tor returns.
Thread::Thread() /* : splitPoints() */ { // Value-initialization bug in MSVC
searching = false;
maxPly = splitPointsSize = 0;
activeSplitPoint = NULL;
activePosition = NULL;
idx = Threads.size();
}
// TimerThread::idle_loop() is where the timer thread waits msec milliseconds
// and then calls check_time(). If msec is 0 thread sleeps until is woken up.
extern void check_time();
void TimerThread::idle_loop() {
while (!exit)
{
mutex.lock();
if (!exit)
sleepCondition.wait_for(mutex, msec ? msec : INT_MAX);
mutex.unlock();
if (msec)
check_time();
}
}
// MainThread::idle_loop() is where the main thread is parked waiting to be started
// when there is a new search. Main thread will launch all the slave threads.
void MainThread::idle_loop() {
while (true)
{
mutex.lock();
thinking = false;
while (!thinking && !exit)
{
Threads.sleepCondition.notify_one(); // Wake up UI thread if needed
sleepCondition.wait(mutex);
}
mutex.unlock();
if (exit)
return;
searching = true;
Search::think();
assert(searching);
searching = false;
}
}
// Thread::cutoff_occurred() checks whether a beta cutoff has occurred in the // Thread::cutoff_occurred() checks whether a beta cutoff has occurred in the
// current active split point, or in some ancestor of the split point. // current active split point, or in some ancestor of the split point.
bool Thread::cutoff_occurred() const { bool Thread::cutoff_occurred() const {
for (SplitPoint* sp = curSplitPoint; sp; sp = sp->parent) for (SplitPoint* sp = activeSplitPoint; sp; sp = sp->parentSplitPoint)
if (sp->cutoff) if (sp->cutoff)
return true; return true;
@@ -164,46 +160,47 @@ bool Thread::cutoff_occurred() const {
// Thread::is_available_to() checks whether the thread is available to help the // Thread::is_available_to() checks whether the thread is available to help the
// thread 'master' at a split point. An obvious requirement is that thread must // thread 'master' at a split point. An obvious requirement is that thread must
// be idle. With more than two threads, this is not sufficient: If the thread is // be idle. With more than two threads, this is not sufficient: If the thread is
// the master of some active split point, it is only available as a slave to the // the master of some split point, it is only available as a slave to the slaves
// slaves which are busy searching the split point at the top of slaves split // which are busy searching the split point at the top of slaves split point
// point stack (the "helpful master concept" in YBWC terminology). // stack (the "helpful master concept" in YBWC terminology).
bool Thread::is_available_to(Thread* master) const { bool Thread::is_available_to(const Thread* master) const {
if (is_searching) if (searching)
return false; return false;
// Make a local copy to be sure doesn't become zero under our feet while // Make a local copy to be sure doesn't become zero under our feet while
// testing next condition and so leading to an out of bound access. // testing next condition and so leading to an out of bound access.
int spCnt = splitPointsCnt; int size = splitPointsSize;
// No active split points means that the thread is available as a slave for any // No split points means that the thread is available as a slave for any
// other thread otherwise apply the "helpful master" concept if possible. // other thread otherwise apply the "helpful master" concept if possible.
return !spCnt || (splitPoints[spCnt - 1].slavesMask & (1ULL << master->idx)); return !size || (splitPoints[size - 1].slavesMask & (1ULL << master->idx));
} }
// init() is called at startup. Initializes lock and condition variable and // init() is called at startup to create and launch requested threads, that will
// launches requested threads sending them immediately to sleep. We cannot use // go immediately to sleep due to 'sleepWhileIdle' set to true. We cannot use
// a c'tor becuase Threads is a static object and we need a fully initialized // a c'tor becuase Threads is a static object and we need a fully initialized
// engine at this point due to allocation of endgames in Thread c'tor. // engine at this point due to allocation of Endgames in Thread c'tor.
void ThreadPool::init() { void ThreadPool::init() {
timer = new Thread(&Thread::timer_loop); sleepWhileIdle = true;
threads.push_back(new Thread(&Thread::main_loop)); timer = new_thread<TimerThread>();
push_back(new_thread<MainThread>());
read_uci_options(); read_uci_options();
} }
// exit() cleanly terminates the threads before the program exits. // exit() cleanly terminates the threads before the program exits
void ThreadPool::exit() { void ThreadPool::exit() {
for (size_t i = 0; i < threads.size(); i++) delete_thread(timer); // As first because check_time() accesses threads data
delete threads[i];
delete timer; for (iterator it = begin(); it != end(); ++it)
delete_thread(*it);
} }
@@ -216,221 +213,182 @@ void ThreadPool::read_uci_options() {
maxThreadsPerSplitPoint = Options["Max Threads per Split Point"]; maxThreadsPerSplitPoint = Options["Max Threads per Split Point"];
minimumSplitDepth = Options["Min Split Depth"] * ONE_PLY; minimumSplitDepth = Options["Min Split Depth"] * ONE_PLY;
useSleepingThreads = Options["Use Sleeping Threads"];
size_t requested = Options["Threads"]; size_t requested = Options["Threads"];
assert(requested > 0); assert(requested > 0);
while (threads.size() < requested) // Value 0 has a special meaning: We determine the optimal minimum split depth
threads.push_back(new Thread(&Thread::idle_loop)); // automatically. Anyhow the minimumSplitDepth should never be under 4 plies.
if (!minimumSplitDepth)
minimumSplitDepth = (requested < 8 ? 4 : 7) * ONE_PLY;
else
minimumSplitDepth = std::max(4 * ONE_PLY, minimumSplitDepth);
while (threads.size() > requested) while (size() < requested)
push_back(new_thread<Thread>());
while (size() > requested)
{ {
delete threads.back(); delete_thread(back());
threads.pop_back(); pop_back();
} }
} }
// wake_up() is called before a new search to start the threads that are waiting // slave_available() tries to find an idle thread which is available as a slave
// on the sleep condition and to reset maxPly. When useSleepingThreads is set // for the thread 'master'.
// threads will be woken up at split time.
void ThreadPool::wake_up() const { Thread* ThreadPool::available_slave(const Thread* master) const {
for (size_t i = 0; i < threads.size(); i++) for (const_iterator it = begin(); it != end(); ++it)
{ if ((*it)->is_available_to(master))
threads[i]->maxPly = 0; return *it;
threads[i]->do_sleep = false;
if (!useSleepingThreads) return NULL;
threads[i]->wake_up();
}
}
// sleep() is called after the search finishes to ask all the threads but the
// main one to go waiting on a sleep condition.
void ThreadPool::sleep() const {
// Main thread will go to sleep by itself to avoid a race with start_searching()
for (size_t i = 1; i < threads.size(); i++)
threads[i]->do_sleep = true;
}
// available_slave_exists() tries to find an idle thread which is available as
// a slave for the thread 'master'.
bool ThreadPool::available_slave_exists(Thread* master) const {
for (size_t i = 0; i < threads.size(); i++)
if (threads[i]->is_available_to(master))
return true;
return false;
} }
// split() does the actual work of distributing the work at a node between // split() does the actual work of distributing the work at a node between
// several available threads. If it does not succeed in splitting the node // several available threads. If it does not succeed in splitting the node
// (because no idle threads are available, or because we have no unused split // (because no idle threads are available), the function immediately returns.
// point objects), the function immediately returns. If splitting is possible, a // If splitting is possible, a SplitPoint object is initialized with all the
// SplitPoint object is initialized with all the data that must be copied to the // data that must be copied to the helper threads and then helper threads are
// helper threads and then helper threads are told that they have been assigned // told that they have been assigned work. This will cause them to instantly
// work. This will cause them to instantly leave their idle loops and call // leave their idle loops and call search(). When all threads have returned from
// search(). When all threads have returned from search() then split() returns. // search() then split() returns.
template <bool Fake> template <bool Fake>
Value ThreadPool::split(Position& pos, Stack* ss, Value alpha, Value beta, void Thread::split(Position& pos, const Stack* ss, Value alpha, Value beta, Value* bestValue,
Value bestValue, Move* bestMove, Depth depth, Move* bestMove, Depth depth, Move threatMove, int moveCount,
Move threatMove, int moveCount, MovePicker* mp, int nodeType) { MovePicker* movePicker, int nodeType, bool cutNode) {
assert(pos.pos_is_ok()); assert(pos.pos_is_ok());
assert(bestValue > -VALUE_INFINITE); assert(*bestValue <= alpha && alpha < beta && beta <= VALUE_INFINITE);
assert(bestValue <= alpha); assert(*bestValue > -VALUE_INFINITE);
assert(alpha < beta); assert(depth >= Threads.minimumSplitDepth);
assert(beta <= VALUE_INFINITE); assert(searching);
assert(depth > DEPTH_ZERO); assert(splitPointsSize < MAX_SPLITPOINTS_PER_THREAD);
Thread* master = pos.this_thread();
if (master->splitPointsCnt >= MAX_SPLITPOINTS_PER_THREAD)
return bestValue;
// Pick the next available split point from the split point stack // Pick the next available split point from the split point stack
SplitPoint& sp = master->splitPoints[master->splitPointsCnt]; SplitPoint& sp = splitPoints[splitPointsSize];
sp.parent = master->curSplitPoint; sp.masterThread = this;
sp.master = master; sp.parentSplitPoint = activeSplitPoint;
sp.cutoff = false; sp.slavesMask = 1ULL << idx;
sp.slavesMask = 1ULL << master->idx;
sp.depth = depth; sp.depth = depth;
sp.bestValue = *bestValue;
sp.bestMove = *bestMove; sp.bestMove = *bestMove;
sp.threatMove = threatMove; sp.threatMove = threatMove;
sp.alpha = alpha; sp.alpha = alpha;
sp.beta = beta; sp.beta = beta;
sp.nodeType = nodeType; sp.nodeType = nodeType;
sp.bestValue = bestValue; sp.cutNode = cutNode;
sp.mp = mp; sp.movePicker = movePicker;
sp.moveCount = moveCount; sp.moveCount = moveCount;
sp.pos = &pos; sp.pos = &pos;
sp.nodes = 0; sp.nodes = 0;
sp.cutoff = false;
sp.ss = ss; sp.ss = ss;
assert(master->is_searching);
master->curSplitPoint = &sp;
int slavesCnt = 0;
// Try to allocate available threads and ask them to start searching setting // Try to allocate available threads and ask them to start searching setting
// is_searching flag. This must be done under lock protection to avoid concurrent // 'searching' flag. This must be done under lock protection to avoid concurrent
// allocation of the same slave by another master. // allocation of the same slave by another master.
Threads.mutex.lock();
sp.mutex.lock(); sp.mutex.lock();
mutex.lock();
for (size_t i = 0; i < threads.size() && !Fake; ++i) splitPointsSize++;
if (threads[i]->is_available_to(master)) activeSplitPoint = &sp;
activePosition = NULL;
size_t slavesCnt = 1; // This thread is always included
Thread* slave;
while ( (slave = Threads.available_slave(this)) != NULL
&& ++slavesCnt <= Threads.maxThreadsPerSplitPoint && !Fake)
{ {
sp.slavesMask |= 1ULL << i; sp.slavesMask |= 1ULL << slave->idx;
threads[i]->curSplitPoint = &sp; slave->activeSplitPoint = &sp;
threads[i]->is_searching = true; // Slave leaves idle_loop() slave->searching = true; // Slave leaves idle_loop()
slave->notify_one(); // Could be sleeping
if (useSleepingThreads)
threads[i]->wake_up();
if (++slavesCnt + 1 >= maxThreadsPerSplitPoint) // Master is always included
break;
} }
master->splitPointsCnt++;
mutex.unlock();
sp.mutex.unlock();
// Everything is set up. The master thread enters the idle loop, from which // Everything is set up. The master thread enters the idle loop, from which
// it will instantly launch a search, because its is_searching flag is set. // it will instantly launch a search, because its 'searching' flag is set.
// The thread will return from the idle loop when all slaves have finished // The thread will return from the idle loop when all slaves have finished
// their work at this split point. // their work at this split point.
if (slavesCnt || Fake) if (slavesCnt > 1 || Fake)
{ {
master->idle_loop(); sp.mutex.unlock();
Threads.mutex.unlock();
Thread::idle_loop(); // Force a call to base class idle_loop()
// In helpful master concept a master can help only a sub-tree of its split // In helpful master concept a master can help only a sub-tree of its split
// point, and because here is all finished is not possible master is booked. // point, and because here is all finished is not possible master is booked.
assert(!master->is_searching); assert(!searching);
} assert(!activePosition);
// We have returned from the idle loop, which means that all threads are // We have returned from the idle loop, which means that all threads are
// finished. Note that setting is_searching and decreasing splitPointsCnt is // finished. Note that setting 'searching' and decreasing splitPointsSize is
// done under lock protection to avoid a race with Thread::is_available_to(). // done under lock protection to avoid a race with Thread::is_available_to().
sp.mutex.lock(); // To protect sp.nodes Threads.mutex.lock();
mutex.lock(); sp.mutex.lock();
}
master->is_searching = true; searching = true;
master->splitPointsCnt--; splitPointsSize--;
master->curSplitPoint = sp.parent; activeSplitPoint = sp.parentSplitPoint;
activePosition = &pos;
pos.set_nodes_searched(pos.nodes_searched() + sp.nodes); pos.set_nodes_searched(pos.nodes_searched() + sp.nodes);
*bestMove = sp.bestMove; *bestMove = sp.bestMove;
*bestValue = sp.bestValue;
mutex.unlock();
sp.mutex.unlock(); sp.mutex.unlock();
Threads.mutex.unlock();
return sp.bestValue;
} }
// Explicit template instantiations // Explicit template instantiations
template Value ThreadPool::split<false>(Position&, Stack*, Value, Value, Value, Move*, Depth, Move, int, MovePicker*, int); template void Thread::split<false>(Position&, const Stack*, Value, Value, Value*, Move*, Depth, Move, int, MovePicker*, int, bool);
template Value ThreadPool::split<true>(Position&, Stack*, Value, Value, Value, Move*, Depth, Move, int, MovePicker*, int); template void Thread::split< true>(Position&, const Stack*, Value, Value, Value*, Move*, Depth, Move, int, MovePicker*, int, bool);
// set_timer() is used to set the timer to trigger after msec milliseconds. // wait_for_think_finished() waits for main thread to go to sleep then returns
// If msec is 0 then timer is stopped.
void ThreadPool::set_timer(int msec) { void ThreadPool::wait_for_think_finished() {
timer->mutex.lock(); MainThread* t = main();
timer->maxPly = msec;
timer->sleepCondition.notify_one(); // Wake up and restart the timer
timer->mutex.unlock();
}
// wait_for_search_finished() waits for main thread to go to sleep, this means
// search is finished. Then returns.
void ThreadPool::wait_for_search_finished() {
Thread* t = main_thread();
t->mutex.lock(); t->mutex.lock();
t->sleepCondition.notify_one(); // In case is waiting for stop or ponderhit while (t->thinking) sleepCondition.wait(t->mutex);
while (!t->do_sleep) sleepCondition.wait(t->mutex);
t->mutex.unlock(); t->mutex.unlock();
} }
// start_searching() wakes up the main thread sleeping in main_loop() so to start // start_thinking() wakes up the main thread sleeping in MainThread::idle_loop()
// a new search, then returns immediately. // so to start a new search, then returns immediately.
void ThreadPool::start_searching(const Position& pos, const LimitsType& limits, void ThreadPool::start_thinking(const Position& pos, const LimitsType& limits,
const std::vector<Move>& searchMoves, StateStackPtr& states) { const std::vector<Move>& searchMoves, StateStackPtr& states) {
wait_for_search_finished(); wait_for_think_finished();
SearchTime = Time::now(); // As early as possible SearchTime = Time::now(); // As early as possible
Signals.stopOnPonderhit = Signals.firstRootMove = false; Signals.stopOnPonderhit = Signals.firstRootMove = false;
Signals.stop = Signals.failedLowAtRoot = false; Signals.stop = Signals.failedLowAtRoot = false;
RootPosition = pos;
Limits = limits;
SetupStates = states; // Ownership transfer here
RootMoves.clear(); RootMoves.clear();
RootPos = pos;
Limits = limits;
if (states.get()) // If we don't set a new position, preserve current state
{
SetupStates = states; // Ownership transfer here
assert(!states.get());
}
for (MoveList<LEGAL> ml(pos); !ml.end(); ++ml) for (MoveList<LEGAL> it(pos); *it; ++it)
if (searchMoves.empty() || count(searchMoves.begin(), searchMoves.end(), ml.move())) if ( searchMoves.empty()
RootMoves.push_back(RootMove(ml.move())); || std::count(searchMoves.begin(), searchMoves.end(), *it))
RootMoves.push_back(RootMove(*it));
main_thread()->do_sleep = false; main()->thinking = true;
main_thread()->wake_up(); main()->notify_one(); // Starts main thread
} }
+68 -58
View File
@@ -1,7 +1,7 @@
/* /*
Stockfish, a UCI chess playing engine derived from Glaurung 2.1 Stockfish, a UCI chess playing engine derived from Glaurung 2.1
Copyright (C) 2004-2008 Tord Romstad (Glaurung author) Copyright (C) 2004-2008 Tord Romstad (Glaurung author)
Copyright (C) 2008-2012 Marco Costalba, Joona Kiiski, Tord Romstad Copyright (C) 2008-2013 Marco Costalba, Joona Kiiski, Tord Romstad
Stockfish is free software: you can redistribute it and/or modify Stockfish is free software: you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by it under the terms of the GNU General Public License as published by
@@ -17,7 +17,7 @@
along with this program. If not, see <http://www.gnu.org/licenses/>. along with this program. If not, see <http://www.gnu.org/licenses/>.
*/ */
#if !defined(THREAD_H_INCLUDED) #ifndef THREAD_H_INCLUDED
#define THREAD_H_INCLUDED #define THREAD_H_INCLUDED
#include <vector> #include <vector>
@@ -28,7 +28,7 @@
#include "position.h" #include "position.h"
#include "search.h" #include "search.h"
const int MAX_THREADS = 32; const int MAX_THREADS = 64; // Because SplitPoint::slavesMask is a uint64_t
const int MAX_SPLITPOINTS_PER_THREAD = 8; const int MAX_SPLITPOINTS_PER_THREAD = 8;
struct Mutex { struct Mutex {
@@ -56,22 +56,23 @@ private:
WaitCondition c; WaitCondition c;
}; };
class Thread; struct Thread;
struct SplitPoint { struct SplitPoint {
// Const data after split point has been setup // Const data after split point has been setup
const Position* pos; const Position* pos;
const Search::Stack* ss; const Search::Stack* ss;
Thread* masterThread;
Depth depth; Depth depth;
Value beta; Value beta;
int nodeType; int nodeType;
Thread* master;
Move threatMove; Move threatMove;
bool cutNode;
// Const pointers to shared data // Const pointers to shared data
MovePicker* mp; MovePicker* movePicker;
SplitPoint* parent; SplitPoint* parentSplitPoint;
// Shared data // Shared data
Mutex mutex; Mutex mutex;
@@ -85,84 +86,93 @@ struct SplitPoint {
}; };
/// ThreadBase struct is the base of the hierarchy from where we derive all the
/// specialized thread classes.
struct ThreadBase {
ThreadBase() : exit(false) {}
virtual ~ThreadBase() {}
virtual void idle_loop() = 0;
void notify_one();
void wait_for(volatile const bool& b);
Mutex mutex;
ConditionVariable sleepCondition;
NativeHandle handle;
volatile bool exit;
};
/// Thread struct keeps together all the thread related stuff like locks, state /// Thread struct keeps together all the thread related stuff like locks, state
/// and especially split points. We also use per-thread pawn and material hash /// and especially split points. We also use per-thread pawn and material hash
/// tables so that once we get a pointer to an entry its life time is unlimited /// tables so that once we get a pointer to an entry its life time is unlimited
/// and we don't have to care about someone changing the entry under our feet. /// and we don't have to care about someone changing the entry under our feet.
class Thread { struct Thread : public ThreadBase {
typedef void (Thread::* Fn) (); // Pointer to member function Thread();
virtual void idle_loop();
public:
Thread(Fn fn);
~Thread();
void wake_up();
bool cutoff_occurred() const; bool cutoff_occurred() const;
bool is_available_to(Thread* master) const; bool is_available_to(const Thread* master) const;
void idle_loop();
void main_loop(); template <bool Fake>
void timer_loop(); void split(Position& pos, const Search::Stack* ss, Value alpha, Value beta, Value* bestValue, Move* bestMove,
void wait_for_stop_or_ponderhit(); Depth depth, Move threatMove, int moveCount, MovePicker* movePicker, int nodeType, bool cutNode);
SplitPoint splitPoints[MAX_SPLITPOINTS_PER_THREAD]; SplitPoint splitPoints[MAX_SPLITPOINTS_PER_THREAD];
MaterialTable materialTable; Material::Table materialTable;
PawnTable pawnTable; Endgames endgames;
Pawns::Table pawnsTable;
Position* activePosition;
size_t idx; size_t idx;
int maxPly; int maxPly;
Mutex mutex; SplitPoint* volatile activeSplitPoint;
ConditionVariable sleepCondition; volatile int splitPointsSize;
NativeHandle handle; volatile bool searching;
Fn start_fn;
SplitPoint* volatile curSplitPoint;
volatile int splitPointsCnt;
volatile bool is_searching;
volatile bool do_sleep;
volatile bool do_exit;
}; };
/// ThreadPool class handles all the threads related stuff like init, starting, /// MainThread and TimerThread are derived classes used to characterize the two
/// special threads: the main one and the recurring timer.
struct MainThread : public Thread {
MainThread() : thinking(true) {} // Avoid a race with start_thinking()
virtual void idle_loop();
volatile bool thinking;
};
struct TimerThread : public ThreadBase {
TimerThread() : msec(0) {}
virtual void idle_loop();
int msec;
};
/// ThreadPool struct handles all the threads related stuff like init, starting,
/// parking and, the most important, launching a slave thread at a split point. /// parking and, the most important, launching a slave thread at a split point.
/// All the access to shared thread data is done through this class. /// All the access to shared thread data is done through this class.
class ThreadPool { struct ThreadPool : public std::vector<Thread*> {
public:
void init(); // No c'tor and d'tor, threads rely on globals that should void init(); // No c'tor and d'tor, threads rely on globals that should
void exit(); // be initialized and valid during the whole thread lifetime. void exit(); // be initialized and valid during the whole thread lifetime.
Thread& operator[](size_t id) { return *threads[id]; } MainThread* main() { return static_cast<MainThread*>((*this)[0]); }
bool use_sleeping_threads() const { return useSleepingThreads; }
int min_split_depth() const { return minimumSplitDepth; }
size_t size() const { return threads.size(); }
Thread* main_thread() { return threads[0]; }
void wake_up() const;
void sleep() const;
void read_uci_options(); void read_uci_options();
bool available_slave_exists(Thread* master) const; Thread* available_slave(const Thread* master) const;
void set_timer(int msec); void wait_for_think_finished();
void wait_for_search_finished(); void start_thinking(const Position&, const Search::LimitsType&,
void start_searching(const Position&, const Search::LimitsType&,
const std::vector<Move>&, Search::StateStackPtr&); const std::vector<Move>&, Search::StateStackPtr&);
template <bool Fake> bool sleepWhileIdle;
Value split(Position& pos, Search::Stack* ss, Value alpha, Value beta, Value bestValue, Move* bestMove, Depth minimumSplitDepth;
Depth depth, Move threatMove, int moveCount, MovePicker* mp, int nodeType); size_t maxThreadsPerSplitPoint;
private:
friend class Thread;
std::vector<Thread*> threads;
Thread* timer;
Mutex mutex; Mutex mutex;
ConditionVariable sleepCondition; ConditionVariable sleepCondition;
Depth minimumSplitDepth; TimerThread* timer;
int maxThreadsPerSplitPoint;
bool useSleepingThreads;
}; };
extern ThreadPool Threads; extern ThreadPool Threads;
#endif // !defined(THREAD_H_INCLUDED) #endif // #ifndef THREAD_H_INCLUDED
+2 -2
View File
@@ -1,7 +1,7 @@
/* /*
Stockfish, a UCI chess playing engine derived from Glaurung 2.1 Stockfish, a UCI chess playing engine derived from Glaurung 2.1
Copyright (C) 2004-2008 Tord Romstad (Glaurung author) Copyright (C) 2004-2008 Tord Romstad (Glaurung author)
Copyright (C) 2008-2012 Marco Costalba, Joona Kiiski, Tord Romstad Copyright (C) 2008-2013 Marco Costalba, Joona Kiiski, Tord Romstad
Stockfish is free software: you can redistribute it and/or modify Stockfish is free software: you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by it under the terms of the GNU General Public License as published by
@@ -29,7 +29,7 @@ namespace {
/// Constants /// Constants
const int MoveHorizon = 50; // Plan time management at most this many moves ahead const int MoveHorizon = 50; // Plan time management at most this many moves ahead
const float MaxRatio = 3.0f; // When in trouble, we can step over reserved time with this ratio const float MaxRatio = 7.0f; // When in trouble, we can step over reserved time with this ratio
const float StealRatio = 0.33f; // However we must not steal time from remaining moves over this ratio const float StealRatio = 0.33f; // However we must not steal time from remaining moves over this ratio
+3 -3
View File
@@ -1,7 +1,7 @@
/* /*
Stockfish, a UCI chess playing engine derived from Glaurung 2.1 Stockfish, a UCI chess playing engine derived from Glaurung 2.1
Copyright (C) 2004-2008 Tord Romstad (Glaurung author) Copyright (C) 2004-2008 Tord Romstad (Glaurung author)
Copyright (C) 2008-2012 Marco Costalba, Joona Kiiski, Tord Romstad Copyright (C) 2008-2013 Marco Costalba, Joona Kiiski, Tord Romstad
Stockfish is free software: you can redistribute it and/or modify Stockfish is free software: you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by it under the terms of the GNU General Public License as published by
@@ -17,7 +17,7 @@
along with this program. If not, see <http://www.gnu.org/licenses/>. along with this program. If not, see <http://www.gnu.org/licenses/>.
*/ */
#if !defined(TIMEMAN_H_INCLUDED) #ifndef TIMEMAN_H_INCLUDED
#define TIMEMAN_H_INCLUDED #define TIMEMAN_H_INCLUDED
/// The TimeManager class computes the optimal time to think depending on the /// The TimeManager class computes the optimal time to think depending on the
@@ -36,4 +36,4 @@ private:
int unstablePVExtraTime; int unstablePVExtraTime;
}; };
#endif // !defined(TIMEMAN_H_INCLUDED) #endif // #ifndef TIMEMAN_H_INCLUDED
+42 -62
View File
@@ -1,7 +1,7 @@
/* /*
Stockfish, a UCI chess playing engine derived from Glaurung 2.1 Stockfish, a UCI chess playing engine derived from Glaurung 2.1
Copyright (C) 2004-2008 Tord Romstad (Glaurung author) Copyright (C) 2004-2008 Tord Romstad (Glaurung author)
Copyright (C) 2008-2012 Marco Costalba, Joona Kiiski, Tord Romstad Copyright (C) 2008-2013 Marco Costalba, Joona Kiiski, Tord Romstad
Stockfish is free software: you can redistribute it and/or modify Stockfish is free software: you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by it under the terms of the GNU General Public License as published by
@@ -25,42 +25,32 @@
TranspositionTable TT; // Our global transposition table TranspositionTable TT; // Our global transposition table
TranspositionTable::TranspositionTable() {
size = generation = 0;
entries = NULL;
}
TranspositionTable::~TranspositionTable() {
delete [] entries;
}
/// TranspositionTable::set_size() sets the size of the transposition table, /// TranspositionTable::set_size() sets the size of the transposition table,
/// measured in megabytes. Transposition table consists of a power of 2 number of /// measured in megabytes. Transposition table consists of a power of 2 number
/// TTCluster and each cluster consists of ClusterSize number of TTEntries. Each /// of clusters and each cluster consists of ClusterSize number of TTEntry.
/// non-empty entry contains information of exactly one position.
void TranspositionTable::set_size(size_t mbSize) { void TranspositionTable::set_size(size_t mbSize) {
size_t newSize = 1ULL << msb((mbSize << 20) / sizeof(TTCluster)); assert(msb((mbSize << 20) / sizeof(TTEntry)) < 32);
if (newSize == size) uint32_t size = ClusterSize << msb((mbSize << 20) / sizeof(TTEntry[ClusterSize]));
if (hashMask == size - ClusterSize)
return; return;
size = newSize; hashMask = size - ClusterSize;
delete [] entries; free(mem);
entries = new (std::nothrow) TTCluster[size]; mem = calloc(size * sizeof(TTEntry) + CACHE_LINE_SIZE - 1, 1);
if (!entries) if (!mem)
{ {
std::cerr << "Failed to allocate " << mbSize std::cerr << "Failed to allocate " << mbSize
<< "MB for transposition table." << std::endl; << "MB for transposition table." << std::endl;
exit(EXIT_FAILURE); exit(EXIT_FAILURE);
} }
clear(); // Operator new is not guaranteed to initialize memory to zero table = (TTEntry*)((uintptr_t(mem) + CACHE_LINE_SIZE - 1) & ~(CACHE_LINE_SIZE - 1));
} }
@@ -70,7 +60,24 @@ void TranspositionTable::set_size(size_t mbSize) {
void TranspositionTable::clear() { void TranspositionTable::clear() {
memset(entries, 0, size * sizeof(TTCluster)); std::memset(table, 0, (hashMask + ClusterSize) * sizeof(TTEntry));
}
/// TranspositionTable::probe() looks up the current position in the
/// transposition table. Returns a pointer to the TTEntry or NULL if
/// position is not found.
const TTEntry* TranspositionTable::probe(const Key key) const {
const TTEntry* tte = first_entry(key);
uint32_t key32 = key >> 32;
for (unsigned i = 0; i < ClusterSize; i++, tte++)
if (tte->key() == key32)
return tte;
return NULL;
} }
@@ -82,60 +89,33 @@ void TranspositionTable::clear() {
/// more valuable than a TTEntry t2 if t1 is from the current search and t2 is from /// more valuable than a TTEntry t2 if t1 is from the current search and t2 is from
/// a previous search, or if the depth of t1 is bigger than the depth of t2. /// a previous search, or if the depth of t1 is bigger than the depth of t2.
void TranspositionTable::store(const Key posKey, Value v, Bound t, Depth d, Move m, Value statV, Value kingD) { void TranspositionTable::store(const Key key, Value v, Bound b, Depth d, Move m, Value statV, Value evalM) {
int c1, c2, c3; int c1, c2, c3;
TTEntry *tte, *replace; TTEntry *tte, *replace;
uint32_t posKey32 = posKey >> 32; // Use the high 32 bits as key inside the cluster uint32_t key32 = key >> 32; // Use the high 32 bits as key inside the cluster
tte = replace = first_entry(posKey); tte = replace = first_entry(key);
for (int i = 0; i < ClusterSize; i++, tte++) for (unsigned i = 0; i < ClusterSize; i++, tte++)
{ {
if (!tte->key() || tte->key() == posKey32) // Empty or overwrite old if (!tte->key() || tte->key() == key32) // Empty or overwrite old
{ {
// Preserve any existing ttMove if (!m)
if (m == MOVE_NONE) m = tte->move(); // Preserve any existing ttMove
m = tte->move();
tte->save(posKey32, v, t, d, m, generation, statV, kingD); replace = tte;
return; break;
} }
// Implement replace strategy // Implement replace strategy
c1 = (replace->generation() == generation ? 2 : 0); c1 = (replace->generation() == generation ? 2 : 0);
c2 = (tte->generation() == generation || tte->type() == BOUND_EXACT ? -2 : 0); c2 = (tte->generation() == generation || tte->bound() == BOUND_EXACT ? -2 : 0);
c3 = (tte->depth() < replace->depth() ? 1 : 0); c3 = (tte->depth() < replace->depth() ? 1 : 0);
if (c1 + c2 + c3 > 0) if (c1 + c2 + c3 > 0)
replace = tte; replace = tte;
} }
replace->save(posKey32, v, t, d, m, generation, statV, kingD);
} replace->save(key32, v, b, d, m, generation, statV, evalM);
/// TranspositionTable::probe() looks up the current position in the
/// transposition table. Returns a pointer to the TTEntry or NULL if
/// position is not found.
TTEntry* TranspositionTable::probe(const Key posKey) const {
uint32_t posKey32 = posKey >> 32;
TTEntry* tte = first_entry(posKey);
for (int i = 0; i < ClusterSize; i++, tte++)
if (tte->key() == posKey32)
return tte;
return NULL;
}
/// TranspositionTable::new_search() is called at the beginning of every new
/// search. It increments the "generation" variable, which is used to
/// distinguish transposition table entries from previous searches from
/// entries from the current search.
void TranspositionTable::new_search() {
generation++;
} }
+42 -60
View File
@@ -1,7 +1,7 @@
/* /*
Stockfish, a UCI chess playing engine derived from Glaurung 2.1 Stockfish, a UCI chess playing engine derived from Glaurung 2.1
Copyright (C) 2004-2008 Tord Romstad (Glaurung author) Copyright (C) 2004-2008 Tord Romstad (Glaurung author)
Copyright (C) 2008-2012 Marco Costalba, Joona Kiiski, Tord Romstad Copyright (C) 2008-2013 Marco Costalba, Joona Kiiski, Tord Romstad
Stockfish is free software: you can redistribute it and/or modify Stockfish is free software: you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by it under the terms of the GNU General Public License as published by
@@ -17,99 +17,81 @@
along with this program. If not, see <http://www.gnu.org/licenses/>. along with this program. If not, see <http://www.gnu.org/licenses/>.
*/ */
#if !defined(TT_H_INCLUDED) #ifndef TT_H_INCLUDED
#define TT_H_INCLUDED #define TT_H_INCLUDED
#include "misc.h" #include "misc.h"
#include "types.h" #include "types.h"
/// The TTEntry is the class of transposition table entries /// The TTEntry is the 128 bit transposition table entry, defined as below:
/// ///
/// A TTEntry needs 128 bits to be stored /// key: 32 bit
/// /// move: 16 bit
/// bit 0-31: key /// bound type: 8 bit
/// bit 32-63: data /// generation: 8 bit
/// bit 64-79: value /// value: 16 bit
/// bit 80-95: depth /// depth: 16 bit
/// bit 96-111: static value /// static value: 16 bit
/// bit 112-127: margin of static value /// static margin: 16 bit
///
/// the 32 bits of the data field are so defined
///
/// bit 0-15: move
/// bit 16-20: not used
/// bit 21-22: value type
/// bit 23-31: generation
class TTEntry { struct TTEntry {
public: void save(uint32_t k, Value v, Bound b, Depth d, Move m, int g, Value ev, Value em) {
void save(uint32_t k, Value v, Bound b, Depth d, Move m, int g, Value statV, Value statM) {
key32 = (uint32_t)k; key32 = (uint32_t)k;
move16 = (uint16_t)m; move16 = (uint16_t)m;
bound = (uint8_t)b; bound8 = (uint8_t)b;
generation8 = (uint8_t)g; generation8 = (uint8_t)g;
value16 = (int16_t)v; value16 = (int16_t)v;
depth16 = (int16_t)d; depth16 = (int16_t)d;
staticValue = (int16_t)statV; evalValue = (int16_t)ev;
staticMargin = (int16_t)statM; evalMargin = (int16_t)em;
} }
void set_generation(int g) { generation8 = (uint8_t)g; } void set_generation(uint8_t g) { generation8 = g; }
uint32_t key() const { return key32; } uint32_t key() const { return key32; }
Depth depth() const { return (Depth)depth16; } Depth depth() const { return (Depth)depth16; }
Move move() const { return (Move)move16; } Move move() const { return (Move)move16; }
Value value() const { return (Value)value16; } Value value() const { return (Value)value16; }
Bound type() const { return (Bound)bound; } Bound bound() const { return (Bound)bound8; }
int generation() const { return (int)generation8; } int generation() const { return (int)generation8; }
Value static_value() const { return (Value)staticValue; } Value eval_value() const { return (Value)evalValue; }
Value static_value_margin() const { return (Value)staticMargin; } Value eval_margin() const { return (Value)evalMargin; }
private: private:
uint32_t key32; uint32_t key32;
uint16_t move16; uint16_t move16;
uint8_t bound, generation8; uint8_t bound8, generation8;
int16_t value16, depth16, staticValue, staticMargin; int16_t value16, depth16, evalValue, evalMargin;
}; };
/// This is the number of TTEntry slots for each cluster /// A TranspositionTable consists of a power of 2 number of clusters and each
const int ClusterSize = 4; /// cluster consists of ClusterSize number of TTEntry. Each non-empty entry
/// contains information of exactly one position. Size of a cluster shall not be
/// bigger than a cache line size. In case it is less, it should be padded to
/// TTCluster consists of ClusterSize number of TTEntries. Size of TTCluster /// guarantee always aligned accesses.
/// must not be bigger than a cache line size. In case it is less, it should
/// be padded to guarantee always aligned accesses.
struct TTCluster {
TTEntry data[ClusterSize];
};
/// The transposition table class. This is basically just a huge array containing
/// TTCluster objects, and a few methods for writing and reading entries.
class TranspositionTable { class TranspositionTable {
TranspositionTable(const TranspositionTable&); static const unsigned ClusterSize = 4; // A cluster is 64 Bytes
TranspositionTable& operator=(const TranspositionTable&);
public: public:
TranspositionTable(); ~TranspositionTable() { free(mem); }
~TranspositionTable(); void new_search() { generation++; }
const TTEntry* probe(const Key key) const;
TTEntry* first_entry(const Key key) const;
void refresh(const TTEntry* tte) const;
void set_size(size_t mbSize); void set_size(size_t mbSize);
void clear(); void clear();
void store(const Key posKey, Value v, Bound type, Depth d, Move m, Value statV, Value kingD); void store(const Key key, Value v, Bound type, Depth d, Move m, Value statV, Value kingD);
TTEntry* probe(const Key posKey) const;
void new_search();
TTEntry* first_entry(const Key posKey) const;
void refresh(const TTEntry* tte) const;
private: private:
size_t size; uint32_t hashMask;
TTCluster* entries; TTEntry* table;
uint8_t generation; // Size must be not bigger then TTEntry::generation8 void* mem;
uint8_t generation; // Size must be not bigger than TTEntry::generation8
}; };
extern TranspositionTable TT; extern TranspositionTable TT;
@@ -119,9 +101,9 @@ extern TranspositionTable TT;
/// a cluster given a position. The lowest order bits of the key are used to /// a cluster given a position. The lowest order bits of the key are used to
/// get the index of the cluster. /// get the index of the cluster.
inline TTEntry* TranspositionTable::first_entry(const Key posKey) const { inline TTEntry* TranspositionTable::first_entry(const Key key) const {
return entries[((uint32_t)posKey) & (size - 1)].data; return table + ((uint32_t)key & hashMask);
} }
@@ -133,4 +115,4 @@ inline void TranspositionTable::refresh(const TTEntry* tte) const {
const_cast<TTEntry*>(tte)->set_generation(generation); const_cast<TTEntry*>(tte)->set_generation(generation);
} }
#endif // !defined(TT_H_INCLUDED) #endif // #ifndef TT_H_INCLUDED
+61 -111
View File
@@ -1,7 +1,7 @@
/* /*
Stockfish, a UCI chess playing engine derived from Glaurung 2.1 Stockfish, a UCI chess playing engine derived from Glaurung 2.1
Copyright (C) 2004-2008 Tord Romstad (Glaurung author) Copyright (C) 2004-2008 Tord Romstad (Glaurung author)
Copyright (C) 2008-2012 Marco Costalba, Joona Kiiski, Tord Romstad Copyright (C) 2008-2013 Marco Costalba, Joona Kiiski, Tord Romstad
Stockfish is free software: you can redistribute it and/or modify Stockfish is free software: you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by it under the terms of the GNU General Public License as published by
@@ -17,7 +17,7 @@
along with this program. If not, see <http://www.gnu.org/licenses/>. along with this program. If not, see <http://www.gnu.org/licenses/>.
*/ */
#if !defined(TYPES_H_INCLUDED) #ifndef TYPES_H_INCLUDED
#define TYPES_H_INCLUDED #define TYPES_H_INCLUDED
/// For Linux and OSX configuration is done automatically using Makefile. To get /// For Linux and OSX configuration is done automatically using Makefile. To get
@@ -35,13 +35,16 @@
/// | only in 64-bit mode. For compiling requires hardware with /// | only in 64-bit mode. For compiling requires hardware with
/// | popcnt support. /// | popcnt support.
#include <cassert>
#include <cctype> #include <cctype>
#include <climits> #include <climits>
#include <cstdlib> #include <cstdlib>
#include "platform.h" #include "platform.h"
#if defined(_WIN64) #define unlikely(x) (x) // For code annotation purposes
#if defined(_WIN64) && !defined(IS_64BIT)
# include <intrin.h> // MSVC popcnt and bsfq instrinsics # include <intrin.h> // MSVC popcnt and bsfq instrinsics
# define IS_64BIT # define IS_64BIT
# define USE_BSFQ # define USE_BSFQ
@@ -51,13 +54,18 @@
# include <nmmintrin.h> // Intel header for _mm_popcnt_u64() intrinsic # include <nmmintrin.h> // Intel header for _mm_popcnt_u64() intrinsic
#endif #endif
# if !defined(NO_PREFETCH) && (defined(__INTEL_COMPILER) || defined(_MSC_VER))
# include <xmmintrin.h> // Intel and Microsoft header for _mm_prefetch()
# endif
#define CACHE_LINE_SIZE 64
#if defined(_MSC_VER) || defined(__INTEL_COMPILER) #if defined(_MSC_VER) || defined(__INTEL_COMPILER)
# define CACHE_LINE_ALIGNMENT __declspec(align(64)) # define CACHE_LINE_ALIGNMENT __declspec(align(CACHE_LINE_SIZE))
#else #else
# define CACHE_LINE_ALIGNMENT __attribute__ ((aligned(64))) # define CACHE_LINE_ALIGNMENT __attribute__ ((aligned(CACHE_LINE_SIZE)))
#endif #endif
#if defined(_MSC_VER) #ifdef _MSC_VER
# define FORCE_INLINE __forceinline # define FORCE_INLINE __forceinline
#elif defined(__GNUC__) #elif defined(__GNUC__)
# define FORCE_INLINE inline __attribute__((always_inline)) # define FORCE_INLINE inline __attribute__((always_inline))
@@ -65,13 +73,13 @@
# define FORCE_INLINE inline # define FORCE_INLINE inline
#endif #endif
#if defined(USE_POPCNT) #ifdef USE_POPCNT
const bool HasPopCnt = true; const bool HasPopCnt = true;
#else #else
const bool HasPopCnt = false; const bool HasPopCnt = false;
#endif #endif
#if defined(IS_64BIT) #ifdef IS_64BIT
const bool Is64Bit = true; const bool Is64Bit = true;
#else #else
const bool Is64Bit = false; const bool Is64Bit = false;
@@ -82,26 +90,7 @@ typedef uint64_t Bitboard;
const int MAX_MOVES = 192; const int MAX_MOVES = 192;
const int MAX_PLY = 100; const int MAX_PLY = 100;
const int MAX_PLY_PLUS_2 = MAX_PLY + 2; const int MAX_PLY_PLUS_6 = MAX_PLY + 6;
const Bitboard FileABB = 0x0101010101010101ULL;
const Bitboard FileBBB = FileABB << 1;
const Bitboard FileCBB = FileABB << 2;
const Bitboard FileDBB = FileABB << 3;
const Bitboard FileEBB = FileABB << 4;
const Bitboard FileFBB = FileABB << 5;
const Bitboard FileGBB = FileABB << 6;
const Bitboard FileHBB = FileABB << 7;
const Bitboard Rank1BB = 0xFF;
const Bitboard Rank2BB = Rank1BB << (8 * 1);
const Bitboard Rank3BB = Rank1BB << (8 * 2);
const Bitboard Rank4BB = Rank1BB << (8 * 3);
const Bitboard Rank5BB = Rank1BB << (8 * 4);
const Bitboard Rank6BB = Rank1BB << (8 * 5);
const Bitboard Rank7BB = Rank1BB << (8 * 6);
const Bitboard Rank8BB = Rank1BB << (8 * 7);
/// A move needs 16 bits to be stored /// A move needs 16 bits to be stored
/// ///
@@ -115,29 +104,37 @@ const Bitboard Rank8BB = Rank1BB << (8 * 7);
/// while MOVE_NONE and MOVE_NULL have the same origin and destination square. /// while MOVE_NONE and MOVE_NULL have the same origin and destination square.
enum Move { enum Move {
MOVE_NONE = 0, MOVE_NONE,
MOVE_NULL = 65 MOVE_NULL = 65
}; };
enum MoveType { enum MoveType {
NORMAL = 0, NORMAL,
PROMOTION = 1 << 14, PROMOTION = 1 << 14,
ENPASSANT = 2 << 14, ENPASSANT = 2 << 14,
CASTLE = 3 << 14 CASTLE = 3 << 14
}; };
enum CastleRight { // Defined as in PolyGlot book hash key enum CastleRight { // Defined as in PolyGlot book hash key
CASTLES_NONE = 0, CASTLES_NONE,
WHITE_OO = 1, WHITE_OO,
WHITE_OOO = 2, WHITE_OOO = WHITE_OO << 1,
BLACK_OO = 4, BLACK_OO = WHITE_OO << 2,
BLACK_OOO = 8, BLACK_OOO = WHITE_OO << 3,
ALL_CASTLES = 15 ALL_CASTLES = WHITE_OO | WHITE_OOO | BLACK_OO | BLACK_OOO,
CASTLE_RIGHT_NB = 16
}; };
enum CastlingSide { enum CastlingSide {
KING_SIDE, KING_SIDE,
QUEEN_SIDE QUEEN_SIDE,
CASTLING_SIDE_NB = 2
};
enum Phase {
PHASE_ENDGAME,
PHASE_MIDGAME = 128,
MG = 0, EG = 1, PHASE_NB = 2
}; };
enum ScaleFactor { enum ScaleFactor {
@@ -148,9 +145,9 @@ enum ScaleFactor {
}; };
enum Bound { enum Bound {
BOUND_NONE = 0, BOUND_NONE,
BOUND_UPPER = 1, BOUND_UPPER,
BOUND_LOWER = 2, BOUND_LOWER,
BOUND_EXACT = BOUND_UPPER | BOUND_LOWER BOUND_EXACT = BOUND_UPPER | BOUND_LOWER
}; };
@@ -168,8 +165,6 @@ enum Value {
VALUE_ENSURE_INTEGER_SIZE_P = INT_MAX, VALUE_ENSURE_INTEGER_SIZE_P = INT_MAX,
VALUE_ENSURE_INTEGER_SIZE_N = INT_MIN, VALUE_ENSURE_INTEGER_SIZE_N = INT_MIN,
Mg = 0, Eg = 1,
PawnValueMg = 198, PawnValueEg = 258, PawnValueMg = 198, PawnValueEg = 258,
KnightValueMg = 817, KnightValueEg = 846, KnightValueMg = 817, KnightValueEg = 846,
BishopValueMg = 836, BishopValueEg = 857, BishopValueMg = 836, BishopValueEg = 857,
@@ -178,18 +173,20 @@ enum Value {
}; };
enum PieceType { enum PieceType {
NO_PIECE_TYPE = 0, ALL_PIECES = 0, NO_PIECE_TYPE, PAWN, KNIGHT, BISHOP, ROOK, QUEEN, KING,
PAWN = 1, KNIGHT = 2, BISHOP = 3, ROOK = 4, QUEEN = 5, KING = 6 ALL_PIECES = 0,
PIECE_TYPE_NB = 8
}; };
enum Piece { enum Piece {
NO_PIECE = 16, // color_of(NO_PIECE) == NO_COLOR NO_PIECE,
W_PAWN = 1, W_KNIGHT = 2, W_BISHOP = 3, W_ROOK = 4, W_QUEEN = 5, W_KING = 6, W_PAWN = 1, W_KNIGHT, W_BISHOP, W_ROOK, W_QUEEN, W_KING,
B_PAWN = 9, B_KNIGHT = 10, B_BISHOP = 11, B_ROOK = 12, B_QUEEN = 13, B_KING = 14 B_PAWN = 9, B_KNIGHT, B_BISHOP, B_ROOK, B_QUEEN, B_KING,
PIECE_NB = 16
}; };
enum Color { enum Color {
WHITE, BLACK, NO_COLOR WHITE, BLACK, NO_COLOR, COLOR_NB = 2
}; };
enum Depth { enum Depth {
@@ -199,7 +196,7 @@ enum Depth {
DEPTH_ZERO = 0 * ONE_PLY, DEPTH_ZERO = 0 * ONE_PLY,
DEPTH_QS_CHECKS = -1 * ONE_PLY, DEPTH_QS_CHECKS = -1 * ONE_PLY,
DEPTH_QS_NO_CHECKS = -2 * ONE_PLY, DEPTH_QS_NO_CHECKS = -2 * ONE_PLY,
DEPTH_QS_RECAPTURES = -5 * ONE_PLY, DEPTH_QS_RECAPTURES = -7 * ONE_PLY,
DEPTH_NONE = -127 * ONE_PLY DEPTH_NONE = -127 * ONE_PLY
}; };
@@ -215,6 +212,8 @@ enum Square {
SQ_A8, SQ_B8, SQ_C8, SQ_D8, SQ_E8, SQ_F8, SQ_G8, SQ_H8, SQ_A8, SQ_B8, SQ_C8, SQ_D8, SQ_E8, SQ_F8, SQ_G8, SQ_H8,
SQ_NONE, SQ_NONE,
SQUARE_NB = 64,
DELTA_N = 8, DELTA_N = 8,
DELTA_E = 1, DELTA_E = 1,
DELTA_S = -8, DELTA_S = -8,
@@ -229,11 +228,11 @@ enum Square {
}; };
enum File { enum File {
FILE_A, FILE_B, FILE_C, FILE_D, FILE_E, FILE_F, FILE_G, FILE_H FILE_A, FILE_B, FILE_C, FILE_D, FILE_E, FILE_F, FILE_G, FILE_H, FILE_NB
}; };
enum Rank { enum Rank {
RANK_1, RANK_2, RANK_3, RANK_4, RANK_5, RANK_6, RANK_7, RANK_8 RANK_1, RANK_2, RANK_3, RANK_4, RANK_5, RANK_6, RANK_7, RANK_8, RANK_NB
}; };
@@ -242,7 +241,7 @@ enum Rank {
/// for midgame value. Compiler is free to choose the enum type as long as can /// for midgame value. Compiler is free to choose the enum type as long as can
/// keep its data, so ensure Score to be an integer type. /// keep its data, so ensure Score to be an integer type.
enum Score { enum Score {
SCORE_ZERO = 0, SCORE_ZERO,
SCORE_ENSURE_INTEGER_SIZE_P = INT_MAX, SCORE_ENSURE_INTEGER_SIZE_P = INT_MAX,
SCORE_ENSURE_INTEGER_SIZE_N = INT_MIN SCORE_ENSURE_INTEGER_SIZE_N = INT_MIN
}; };
@@ -252,7 +251,7 @@ inline Score make_score(int mg, int eg) { return Score((mg << 16) + eg); }
/// Extracting the signed lower and upper 16 bits it not so trivial because /// Extracting the signed lower and upper 16 bits it not so trivial because
/// according to the standard a simple cast to short is implementation defined /// according to the standard a simple cast to short is implementation defined
/// and so is a right shift of a signed integer. /// and so is a right shift of a signed integer.
inline Value mg_value(Score s) { return Value(((s + 32768) & ~0xffff) / 0x10000); } inline Value mg_value(Score s) { return Value(((s + 0x8000) & ~0xffff) / 0x10000); }
/// On Intel 64 bit we have a small speed regression with the standard conforming /// On Intel 64 bit we have a small speed regression with the standard conforming
/// version, so use a faster code in this case that, although not 100% standard /// version, so use a faster code in this case that, although not 100% standard
@@ -309,38 +308,17 @@ inline Score operator/(Score s, int i) {
return make_score(mg_value(s) / i, eg_value(s) / i); return make_score(mg_value(s) / i, eg_value(s) / i);
} }
/// Weight score v by score w trying to prevent overflow
inline Score apply_weight(Score v, Score w) {
return make_score((int(mg_value(v)) * mg_value(w)) / 0x100,
(int(eg_value(v)) * eg_value(w)) / 0x100);
}
#undef ENABLE_OPERATORS_ON #undef ENABLE_OPERATORS_ON
#undef ENABLE_SAFE_OPERATORS_ON #undef ENABLE_SAFE_OPERATORS_ON
namespace Zobrist { extern Value PieceValue[PHASE_NB][PIECE_NB];
extern Key psq[2][8][64]; // [color][pieceType][square / piece count] struct ExtMove {
extern Key enpassant[8]; // [file]
extern Key castle[16]; // [castleRight]
extern Key side;
extern Key exclusion;
void init();
}
CACHE_LINE_ALIGNMENT
extern Score pieceSquareTable[16][64]; // [piece][square]
extern Value PieceValue[2][18]; // [Mg / Eg][piece / pieceType]
extern int SquareDistance[64][64]; // [square][square]
struct MoveStack {
Move move; Move move;
int score; int score;
}; };
inline bool operator<(const MoveStack& f, const MoveStack& s) { inline bool operator<(const ExtMove& f, const ExtMove& s) {
return f.score < s.score; return f.score < s.score;
} }
@@ -377,6 +355,7 @@ inline PieceType type_of(Piece p) {
} }
inline Color color_of(Piece p) { inline Color color_of(Piece p) {
assert(p != NO_PIECE);
return Color(p >> 3); return Color(p >> 3);
} }
@@ -413,24 +392,12 @@ inline bool opposite_colors(Square s1, Square s2) {
return ((s >> 3) ^ s) & 1; return ((s >> 3) ^ s) & 1;
} }
inline int file_distance(Square s1, Square s2) { inline char file_to_char(File f, bool tolower = true) {
return abs(file_of(s1) - file_of(s2)); return char(f - FILE_A + (tolower ? 'a' : 'A'));
}
inline int rank_distance(Square s1, Square s2) {
return abs(rank_of(s1) - rank_of(s2));
}
inline int square_distance(Square s1, Square s2) {
return SquareDistance[s1][s2];
}
inline char file_to_char(File f) {
return char(f - FILE_A + int('a'));
} }
inline char rank_to_char(Rank r) { inline char rank_to_char(Rank r) {
return char(r - RANK_1 + int('1')); return char(r - RANK_1 + '1');
} }
inline Square pawn_push(Color c) { inline Square pawn_push(Color c) {
@@ -459,7 +426,7 @@ inline Move make_move(Square from, Square to) {
template<MoveType T> template<MoveType T>
inline Move make(Square from, Square to, PieceType pt = KNIGHT) { inline Move make(Square from, Square to, PieceType pt = KNIGHT) {
return Move(to | (from << 6) | T | ((pt - KNIGHT) << 12)) ; return Move(to | (from << 6) | T | ((pt - KNIGHT) << 12));
} }
inline bool is_ok(Move m) { inline bool is_ok(Move m) {
@@ -473,21 +440,4 @@ inline const std::string square_to_string(Square s) {
return ch; return ch;
} }
/// Our insertion sort implementation, works with pointers and iterators and is #endif // #ifndef TYPES_H_INCLUDED
/// guaranteed to be stable, as is needed.
template<typename T, typename K>
void sort(K first, K last)
{
T tmp;
K p, q;
for (p = first + 1; p < last; p++)
{
tmp = *p;
for (q = p; q != first && *(q-1) < tmp; --q)
*q = *(q-1);
*q = tmp;
}
}
#endif // !defined(TYPES_H_INCLUDED)
+68 -100
View File
@@ -1,7 +1,7 @@
/* /*
Stockfish, a UCI chess playing engine derived from Glaurung 2.1 Stockfish, a UCI chess playing engine derived from Glaurung 2.1
Copyright (C) 2004-2008 Tord Romstad (Glaurung author) Copyright (C) 2004-2008 Tord Romstad (Glaurung author)
Copyright (C) 2008-2012 Marco Costalba, Joona Kiiski, Tord Romstad Copyright (C) 2008-2013 Marco Costalba, Joona Kiiski, Tord Romstad
Stockfish is free software: you can redistribute it and/or modify Stockfish is free software: you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by it under the terms of the GNU General Public License as published by
@@ -17,6 +17,7 @@
along with this program. If not, see <http://www.gnu.org/licenses/>. along with this program. If not, see <http://www.gnu.org/licenses/>.
*/ */
#include <iomanip>
#include <iostream> #include <iostream>
#include <sstream> #include <sstream>
#include <string> #include <string>
@@ -41,9 +42,9 @@ namespace {
// position just before to start searching). Needed by repetition draw detection. // position just before to start searching). Needed by repetition draw detection.
Search::StateStackPtr SetupStates; Search::StateStackPtr SetupStates;
void set_option(istringstream& up); void setoption(istringstream& up);
void set_position(Position& pos, istringstream& up); void position(Position& pos, istringstream& up);
void go(Position& pos, istringstream& up); void go(const Position& pos, istringstream& up);
} }
@@ -54,79 +55,33 @@ namespace {
void UCI::loop(const string& args) { void UCI::loop(const string& args) {
Position pos(StartFEN, false, Threads.main_thread()); // The root position Position pos(StartFEN, false, Threads.main()); // The root position
string cmd, token; string token, cmd = args;
while (token != "quit") do {
{ if (args.empty() && !getline(cin, cmd)) // Block here waiting for input
if (!args.empty())
cmd = args;
else if (!getline(cin, cmd)) // Block here waiting for input
cmd = "quit"; cmd = "quit";
istringstream is(cmd); istringstream is(cmd);
is >> skipws >> token; is >> skipws >> token;
if (token == "quit" || token == "stop") if (token == "quit" || token == "stop" || token == "ponderhit")
{
// GUI sends 'ponderhit' to tell us to ponder on the same move the
// opponent has played. In case Signals.stopOnPonderhit is set we are
// waiting for 'ponderhit' to stop the search (for instance because we
// already ran out of time), otherwise we should continue searching but
// switching from pondering to normal search.
if (token != "ponderhit" || Search::Signals.stopOnPonderhit)
{ {
Search::Signals.stop = true; Search::Signals.stop = true;
Threads.wait_for_search_finished(); // Cannot quit while threads are running Threads.main()->notify_one(); // Could be sleeping
} }
else
else if (token == "ponderhit")
{
// The opponent has played the expected move. GUI sends "ponderhit" if
// we were told to ponder on the same move the opponent has played. We
// should continue searching but switching from pondering to normal search.
Search::Limits.ponder = false; Search::Limits.ponder = false;
if (Search::Signals.stopOnPonderhit)
{
Search::Signals.stop = true;
Threads.main_thread()->wake_up(); // Could be sleeping
} }
} else if (token == "perft" && (is >> token)) // Read perft depth
else if (token == "go")
go(pos, is);
else if (token == "ucinewgame")
{ /* Avoid returning "Unknown command" */ }
else if (token == "isready")
sync_cout << "readyok" << sync_endl;
else if (token == "position")
set_position(pos, is);
else if (token == "setoption")
set_option(is);
else if (token == "d")
pos.print();
else if (token == "flip")
pos.flip();
else if (token == "eval")
sync_cout << Eval::trace(pos) << sync_endl;
else if (token == "bench")
benchmark(pos, is);
else if (token == "key")
sync_cout << "key: " << hex << pos.key()
<< "\nmaterial key: " << pos.material_key()
<< "\npawn key: " << pos.pawn_key() << sync_endl;
else if (token == "uci")
sync_cout << "id name " << engine_info(true)
<< "\n" << Options
<< "\nuciok" << sync_endl;
else if (token == "perft" && (is >> token)) // Read depth
{ {
stringstream ss; stringstream ss;
@@ -135,27 +90,48 @@ void UCI::loop(const string& args) {
benchmark(pos, ss); benchmark(pos, ss);
} }
else if (token == "key")
sync_cout << hex << uppercase << setfill('0')
<< "position key: " << setw(16) << pos.key()
<< "\nmaterial key: " << setw(16) << pos.material_key()
<< "\npawn key: " << setw(16) << pos.pawn_key()
<< dec << sync_endl;
else if (token == "uci")
sync_cout << "id name " << engine_info(true)
<< "\n" << Options
<< "\nuciok" << sync_endl;
else if (token == "eval")
{
Search::RootColor = pos.side_to_move(); // Ensure it is set
sync_cout << Eval::trace(pos) << sync_endl;
}
else if (token == "ucinewgame") { /* Avoid returning "Unknown command" */ }
else if (token == "go") go(pos, is);
else if (token == "position") position(pos, is);
else if (token == "setoption") setoption(is);
else if (token == "flip") pos.flip();
else if (token == "bench") benchmark(pos, is);
else if (token == "d") sync_cout << pos.pretty() << sync_endl;
else if (token == "isready") sync_cout << "readyok" << sync_endl;
else else
sync_cout << "Unknown command: " << cmd << sync_endl; sync_cout << "Unknown command: " << cmd << sync_endl;
if (!args.empty()) // Command line arguments have one-shot behaviour } while (token != "quit" && args.empty()); // Args have one-shot behaviour
{
Threads.wait_for_search_finished(); Threads.wait_for_think_finished(); // Cannot quit while search is running
break;
}
}
} }
namespace { namespace {
// set_position() is called when engine receives the "position" UCI command. // position() is called when engine receives the "position" UCI command.
// The function sets up the position described in the given fen string ("fen") // The function sets up the position described in the given fen string ("fen")
// or the starting position ("startpos") and then makes the moves given in the // or the starting position ("startpos") and then makes the moves given in the
// following move list ("moves"). // following move list ("moves").
void set_position(Position& pos, istringstream& is) { void position(Position& pos, istringstream& is) {
Move m; Move m;
string token, fen; string token, fen;
@@ -173,7 +149,7 @@ namespace {
else else
return; return;
pos.from_fen(fen, Options["UCI_Chess960"], Threads.main_thread()); pos.set(fen, Options["UCI_Chess960"], Threads.main());
SetupStates = Search::StateStackPtr(new std::stack<StateInfo>()); SetupStates = Search::StateStackPtr(new std::stack<StateInfo>());
// Parse move list (if any) // Parse move list (if any)
@@ -185,10 +161,10 @@ namespace {
} }
// set_option() is called when engine receives the "setoption" UCI command. The // setoption() is called when engine receives the "setoption" UCI command. The
// function updates the UCI option ("name") to the given value ("value"). // function updates the UCI option ("name") to the given value ("value").
void set_option(istringstream& is) { void setoption(istringstream& is) {
string token, name, value; string token, name, value;
@@ -210,10 +186,10 @@ namespace {
// go() is called when engine receives the "go" UCI command. The function sets // go() is called when engine receives the "go" UCI command. The function sets
// the thinking time and other parameters from the input string, and then starts // the thinking time and other parameters from the input string, and starts
// the search. // the search.
void go(Position& pos, istringstream& is) { void go(const Position& pos, istringstream& is) {
Search::LimitsType limits; Search::LimitsType limits;
vector<Move> searchMoves; vector<Move> searchMoves;
@@ -221,31 +197,23 @@ namespace {
while (is >> token) while (is >> token)
{ {
if (token == "wtime") if (token == "searchmoves")
is >> limits.time[WHITE];
else if (token == "btime")
is >> limits.time[BLACK];
else if (token == "winc")
is >> limits.inc[WHITE];
else if (token == "binc")
is >> limits.inc[BLACK];
else if (token == "movestogo")
is >> limits.movestogo;
else if (token == "depth")
is >> limits.depth;
else if (token == "nodes")
is >> limits.nodes;
else if (token == "movetime")
is >> limits.movetime;
else if (token == "infinite")
limits.infinite = true;
else if (token == "ponder")
limits.ponder = true;
else if (token == "searchmoves")
while (is >> token) while (is >> token)
searchMoves.push_back(move_from_uci(pos, token)); searchMoves.push_back(move_from_uci(pos, token));
else if (token == "wtime") is >> limits.time[WHITE];
else if (token == "btime") is >> limits.time[BLACK];
else if (token == "winc") is >> limits.inc[WHITE];
else if (token == "binc") is >> limits.inc[BLACK];
else if (token == "movestogo") is >> limits.movestogo;
else if (token == "depth") is >> limits.depth;
else if (token == "nodes") is >> limits.nodes;
else if (token == "movetime") is >> limits.movetime;
else if (token == "mate") is >> limits.mate;
else if (token == "infinite") limits.infinite = true;
else if (token == "ponder") limits.ponder = true;
} }
Threads.start_searching(pos, limits, searchMoves, SetupStates); Threads.start_thinking(pos, limits, searchMoves, SetupStates);
} }
} }
+12 -14
View File
@@ -1,7 +1,7 @@
/* /*
Stockfish, a UCI chess playing engine derived from Glaurung 2.1 Stockfish, a UCI chess playing engine derived from Glaurung 2.1
Copyright (C) 2004-2008 Tord Romstad (Glaurung author) Copyright (C) 2004-2008 Tord Romstad (Glaurung author)
Copyright (C) 2008-2012 Marco Costalba, Joona Kiiski, Tord Romstad Copyright (C) 2008-2013 Marco Costalba, Joona Kiiski, Tord Romstad
Stockfish is free software: you can redistribute it and/or modify Stockfish is free software: you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by it under the terms of the GNU General Public License as published by
@@ -51,31 +51,29 @@ bool CaseInsensitiveLess::operator() (const string& s1, const string& s2) const
/// init() initializes the UCI options to their hard coded default values /// init() initializes the UCI options to their hard coded default values
/// and initializes the default value of "Threads" and "Min Split Depth"
/// parameters according to the number of CPU cores detected.
void init(OptionsMap& o) { void init(OptionsMap& o) {
int cpus = std::min(cpu_count(), MAX_THREADS); o["Write Debug Log"] = Option(false, on_logger);
int msd = cpus < 8 ? 4 : 7; o["Write Search Log"] = Option(false);
o["Use Debug Log"] = Option(false, on_logger);
o["Use Search Log"] = Option(false);
o["Search Log Filename"] = Option("SearchLog.txt"); o["Search Log Filename"] = Option("SearchLog.txt");
o["Book File"] = Option("book.bin"); o["Book File"] = Option("book.bin");
o["Best Book Move"] = Option(false); o["Best Book Move"] = Option(false);
o["Mobility (Middle Game)"] = Option(100, 0, 200, on_eval); o["Contempt Factor"] = Option(0, -50, 50);
o["Mobility (Midgame)"] = Option(100, 0, 200, on_eval);
o["Mobility (Endgame)"] = Option(100, 0, 200, on_eval); o["Mobility (Endgame)"] = Option(100, 0, 200, on_eval);
o["Passed Pawns (Middle Game)"] = Option(100, 0, 200, on_eval); o["Pawn Structure (Midgame)"] = Option(100, 0, 200, on_eval);
o["Pawn Structure (Endgame)"] = Option(100, 0, 200, on_eval);
o["Passed Pawns (Midgame)"] = Option(100, 0, 200, on_eval);
o["Passed Pawns (Endgame)"] = Option(100, 0, 200, on_eval); o["Passed Pawns (Endgame)"] = Option(100, 0, 200, on_eval);
o["Space"] = Option(100, 0, 200, on_eval); o["Space"] = Option(100, 0, 200, on_eval);
o["Aggressiveness"] = Option(100, 0, 200, on_eval); o["Aggressiveness"] = Option(100, 0, 200, on_eval);
o["Cowardice"] = Option(100, 0, 200, on_eval); o["Cowardice"] = Option(100, 0, 200, on_eval);
o["Min Split Depth"] = Option(msd, 4, 7, on_threads); o["Min Split Depth"] = Option(0, 0, 12, on_threads);
o["Max Threads per Split Point"] = Option(5, 4, 8, on_threads); o["Max Threads per Split Point"] = Option(5, 4, 8, on_threads);
o["Threads"] = Option(cpus, 1, MAX_THREADS, on_threads); o["Threads"] = Option(1, 1, MAX_THREADS, on_threads);
o["Use Sleeping Threads"] = Option(false, on_threads); o["Idle Threads Sleep"] = Option(false);
o["Hash"] = Option(32, 4, 8192, on_hash_size); o["Hash"] = Option(32, 1, 8192, on_hash_size);
o["Clear Hash"] = Option(on_clear_hash); o["Clear Hash"] = Option(on_clear_hash);
o["Ponder"] = Option(true); o["Ponder"] = Option(true);
o["OwnBook"] = Option(false); o["OwnBook"] = Option(false);
+3 -3
View File
@@ -1,7 +1,7 @@
/* /*
Stockfish, a UCI chess playing engine derived from Glaurung 2.1 Stockfish, a UCI chess playing engine derived from Glaurung 2.1
Copyright (C) 2004-2008 Tord Romstad (Glaurung author) Copyright (C) 2004-2008 Tord Romstad (Glaurung author)
Copyright (C) 2008-2012 Marco Costalba, Joona Kiiski, Tord Romstad Copyright (C) 2008-2013 Marco Costalba, Joona Kiiski, Tord Romstad
Stockfish is free software: you can redistribute it and/or modify Stockfish is free software: you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by it under the terms of the GNU General Public License as published by
@@ -17,7 +17,7 @@
along with this program. If not, see <http://www.gnu.org/licenses/>. along with this program. If not, see <http://www.gnu.org/licenses/>.
*/ */
#if !defined(UCIOPTION_H_INCLUDED) #ifndef UCIOPTION_H_INCLUDED
#define UCIOPTION_H_INCLUDED #define UCIOPTION_H_INCLUDED
#include <map> #include <map>
@@ -66,4 +66,4 @@ void loop(const std::string&);
extern UCI::OptionsMap Options; extern UCI::OptionsMap Options;
#endif // !defined(UCIOPTION_H_INCLUDED) #endif // #ifndef UCIOPTION_H_INCLUDED