Stockfish Development Versions are build automatically if there are changes on the master branch in the git repository (https://github.com/official-stockfish/Stockfish). Use it at your own risk.
They are compiled with gcc 5.4.0/mingw 5.3.1 on Ubuntu 16.04.
2.Jan.2016: BMI build now slightly faster due to profiling.
Development builds for macOS are available at stockfishchess.de
Read the FAQ or leave a comment.

goto Gull
!! latest version !!


Windows x64 for Haswell CPUs
Windows x64 for modern computers
Windows x64
Windows 32
Linux x64 for Haswell CPUs
Linux x64 for modern computers
Linux x64
Author: Stéphane Nicolet
Date: Tue Jan 17 14:50:03 2017 +0100
Timestamp: 1484661003

Update some comments (#973)

Use somewhat more precise comments in a couple of places.

No functional change.

Windows x64 for Haswell CPUs
Windows x64 for modern computers
Windows x64
Windows 32
Linux x64 for Haswell CPUs
Linux x64 for modern computers
Linux x64
Author: Stefano Cardanobile
Date: Fri Jan 13 09:17:48 2017 +0100
Timestamp: 1484295468

Introduce lazy evaluation

After we have taken into account all cheap evaluation
terms, we check whether the score exceeds a given threshold.
If this is the case, we return a scaled down evaluation.

STC:
LLR: 3.35 (-2.94,2.94) [0.00,5.00]
Total: 12575 W: 2316 L: 2122 D: 8137

LTC:
LLR: 2.95 (-2.94,2.94) [0.00,5.00]
Total: 67480 W: 9016 L: 8677 D: 49787

Current version is the one rewritten by ceebo
further edited by me.

Bench: 5367704
Windows x64 for Haswell CPUs
Windows x64 for modern computers
Windows x64
Windows 32
Linux x64 for Haswell CPUs
Linux x64 for modern computers
Linux x64
Author: loco-loco
Date: Thu Jan 12 08:46:46 2017 +0100
Timestamp: 1484207206

Removing CM parameter from the Stats struct. (#970)

After the history simplifications, we are only using Value Stats for CounterMoveHistory table. Therefore the parameter CM is not necessary.

No functional change.
Windows x64 for Haswell CPUs
Windows x64 for modern computers
Windows x64
Windows 32
Linux x64 for Haswell CPUs
Linux x64 for modern computers
Linux x64
Author: Stéphane Nicolet
Date: Wed Jan 11 18:11:17 2017 +0100
Timestamp: 1484154677

Check for overflow in Score * int multiplication (#969)

Add asserts to check for overflow in Score * int multiplication.

There is no overflow in current master, but it would be easy to
create one as the scale of the current eval does not leave many
spare bits. For instance, adding the following unused variables
in master at the end of evaluate() (line 882 of evaluate.cpp)
overflows:

Score s1 = score * 4; // no overflow
Score s2 = score * 5; // overflow

Assertion failed: (eg_value(result) == (i * eg_value(s))),
function operator*, file ./types.h, line 336.

Same md5 checksum as current master for non debug compiles.

No functional change.
Windows x64 for Haswell CPUs
Windows x64 for modern computers
Windows x64
Windows 32
Linux x64 for Haswell CPUs
Linux x64 for modern computers
Linux x64
Author: Rocky640
Date: Wed Jan 11 08:56:38 2017 +0100
Timestamp: 1484121398

StormDanger CleanUp (#964)

Order the enum and the array the same way they appear around line 250.
Makes it much easier to follow.

Add comments in the array definition and critical rows.
Use same terminology as elsewhere in pawns.cpp

No functional change.
Windows x64 for Haswell CPUs
Windows x64 for modern computers
Windows x64
Windows 32
Linux x64 for Haswell CPUs
Linux x64 for modern computers
Linux x64
Author: Joost VandeVondele
Date: Wed Jan 11 08:46:29 2017 +0100
Timestamp: 1484120789

Adjust copyright headers to 2017 (#965)

No functional change.
Windows x64 for Haswell CPUs
Windows x64 for modern computers
Windows x64
Windows 32
Linux x64 for Haswell CPUs
Linux x64 for modern computers
Linux x64
Author: Marco Costalba
Date: Wed Jan 11 08:44:06 2017 +0100
Timestamp: 1484120646

Travis: fix bench fetch in case of PR (#968)

When Travis tests a PR, a commit merge is created
but master branch is not updated, although HEAD is.

No functional change.
Windows x64 for Haswell CPUs
Windows x64 for modern computers
Windows x64
Windows 32
Linux x64 for Haswell CPUs
Linux x64 for modern computers
Linux x64
Author: lucasart
Date: Tue Jan 10 08:47:56 2017 +0100
Timestamp: 1484034476

Rename FromTo -> History (#963)

Previously, we had duplicated History:

- one with (piece,to) called History
- one with (from,to) called FromTo

Now that we have only one, rename it to History, which is the generally accepted
name in the chess programming litterature for this technique.

Also correct some comments that had not been updated since the introduction of CMH.

No functional change.
Windows x64 for Haswell CPUs
Windows x64 for modern computers
Windows x64
Windows 32
Linux x64 for Haswell CPUs
Linux x64 for modern computers
Linux x64
Author: lucasart
Date: Mon Jan 9 15:50:12 2017 +0100
Timestamp: 1483973412

Remove HistoryStats

STC:
LLR: 3.44 (-2.94,2.94) [-3.00,1.00]
Total: 120831 W: 21572 L: 21594 D: 77665

LTC:
LLR: 2.96 (-2.94,2.94) [-3.00,1.00]
Total: 26565 W: 3519 L: 3406 D: 19640

bench 5920493
Windows x64 for Haswell CPUs
Windows x64 for modern computers
Windows x64
Windows 32
Linux x64 for Haswell CPUs
Linux x64 for modern computers
Linux x64
Author: Marco Costalba
Date: Mon Jan 9 15:37:09 2017 +0100
Timestamp: 1483972629

Fix previous patch for OS X (#961)

Use posix version of sed that is available on all
platforms.

No functional change.
Windows x64 for Haswell CPUs
Windows x64 for modern computers
Windows x64
Windows 32
Linux x64 for Haswell CPUs
Linux x64 for modern computers
Linux x64
Author: Marco Costalba
Date: Mon Jan 9 12:51:28 2017 +0100
Timestamp: 1483962688

Trevis CI: use commit bench number as a reference

No functional change.
Windows x64 for Haswell CPUs
Windows x64 for modern computers
Windows x64
Windows 32
Linux x64 for Haswell CPUs
Linux x64 for modern computers
Linux x64
Author: Joost VandeVondele
Date: Mon Jan 9 10:30:57 2017 +0100
Timestamp: 1483954257

New shell scripts for testing, used for travis CI (#957)

Perform more complex verification and validation.

- signature.sh : extract and optionally compare Bench/Signature/Node count.
- perft.sh : verify perft counts for a number of positions.
- instrumented.sh : run a few commands or uci sequences through valgrind/sanitizer instrumented binaries.
- reprosearch.sh : verify reproducibility of search.

These script can be used from directly from the command line in the src directory.

Update travis script to use these shell scripts.

No functional change.
Windows x64 for Haswell CPUs
Windows x64 for modern computers
Windows x64
Windows 32
Linux x64 for Haswell CPUs
Linux x64 for modern computers
Linux x64
Author: ElbertoOne
Date: Mon Jan 9 10:26:50 2017 +0100
Timestamp: 1483954010

Simplified select best thread (#958)

Only select best thread if score is better and depth equal or larger.

STC (7 threads): http://tests.stockfishchess.org/tests/view/586a4d090ebc5903140c64b2
LLR: 2.95 (-2.94,2.94) [-3.00,1.00]
Total: 120297 W: 18652 L: 18682 D: 82963

LTC (7 threads): http://tests.stockfishchess.org/tests/view/586e31b30ebc5903140c663d
LLR: 2.96 (-2.94,2.94) [-3.00,1.00]
Total: 57187 W: 7035 L: 6959 D: 43193

bench: 4940355
Windows x64 for Haswell CPUs
Windows x64 for modern computers
Windows x64
Windows 32
Linux x64 for Haswell CPUs
Linux x64 for modern computers
Linux x64
Author: Marco Costalba
Date: Sat Jan 7 10:14:39 2017 +0100
Timestamp: 1483780479

Drop Stats c'tors

Now taht we correctly value-initialize Thread objects,
we don't need c'tors anymore because tables will be
zero-initialized by the compier when Thread object
is instanced.

Verified that we have no errors with Valgrind.

No functional change.
Windows x64 for Haswell CPUs
Windows x64 for modern computers
Windows x64
Windows 32
Linux x64 for Haswell CPUs
Linux x64 for modern computers
Linux x64
Author: Joost VandeVondele
Date: Sat Jan 7 10:02:43 2017 +0100
Timestamp: 1483779763

Correctly zero-initialize MainThread

It can be used uninitialized in time management.
Fixes all valgrind errors on './stockfish go wtime 8000 btime 8000 winc 500 binc 500'

This is one (of the many) quirks of C++. There is a subtle difference between:

new Foo
new Foo()

The first statement calls the default constructor (that in case of a POD leaves data members
uninitialized), the second one performs a value-initialization (that in case of POD is
equivalent to a zero-initialization)

See:
http://stackoverflow.com/questions/620137/do-the-parentheses-after-the-type-name-make-a-difference-with-new
http://stackoverflow.com/questions/5116541/difference-between-creating-object-with-or-without

No functional change.
Windows x64 for Haswell CPUs
Windows x64 for modern computers
Windows x64
Windows 32
Linux x64 for Haswell CPUs
Linux x64 for modern computers
Linux x64
Author: Joost VandeVondele
Date: Fri Jan 6 10:43:18 2017 +0100
Timestamp: 1483695798

Zero init fromToStats in constructor. (#953)

Extend commit fe99de to fromToStats, which fixes the last valgrind errors on
a simple 'go depth 12' at startup.

No functional change.
Windows x64 for Haswell CPUs
Windows x64 for modern computers
Windows x64
Windows 32
Linux x64 for Haswell CPUs
Linux x64 for modern computers
Linux x64
Author: pb00068
Date: Thu Jan 5 09:00:41 2017 +0100
Timestamp: 1483603241

Rejoin lines that belong to HalfDensity map (#952)

No functional change.
Windows x64 for Haswell CPUs
Windows x64 for modern computers
Windows x64
Windows 32
Linux x64 for Haswell CPUs
Linux x64 for modern computers
Linux x64
Author: Marco Costalba
Date: Thu Jan 5 08:50:17 2017 +0100
Timestamp: 1483602617

Correct zero-init of Thread data members

If not explicitly initialized in a class constructor,
then all data members are default-initialized when
the corresponing struct/class is instanced.

For array and built-in types (int, char, etc..)
default-initialization is a no-op and we need to
explicitly zero them.

No functional change.
Windows x64 for Haswell CPUs
Windows x64 for modern computers
Windows x64
Windows 32
Linux x64 for Haswell CPUs
Linux x64 for modern computers
Linux x64
Author: Joost VandeVondele
Date: Mon Jan 2 09:36:43 2017 +0100
Timestamp: 1483346203

Use consistent variable names for counterMoveStats (#949)

Unify naming in movepick and search, by adopting the latter convention (cmh,fmh,fmh2).

No functional change.
Windows x64 for Haswell CPUs
Windows x64 for modern computers
Windows x64
Windows 32
Linux x64 for Haswell CPUs
Linux x64 for modern computers
Linux x64
Author: Stéphane Nicolet
Date: Mon Jan 2 09:33:40 2017 +0100
Timestamp: 1483346020

Simplify unstoppable again (#950)

Assign a small bonus for our passed pawns when the opponent has no
pieces left.

STC:
LLR: 2.95 (-2.94,2.94) [-3.00,1.00]
Total: 51124 W: 9036 L: 8966 D: 33122

LTC:
LLR: 2.95 (-2.94,2.94) [-3.00,1.00]
Total: 114560 W: 14604 L: 14604 D: 85352

Bench: 4940355
Windows x64 for Haswell CPUs
Windows x64 for modern computers
Windows x64
Windows 32
Linux x64 for Haswell CPUs
Linux x64 for modern computers
Linux x64
Author: Jonathan Calovski
Date: Sun Jan 1 11:28:50 2017 +0100
Timestamp: 1483266530

Tweak best thread selection logic

STC 7 threads:
LLR: 2.96 (-2.94,2.94) [0.00,5.00]
Total: 26881 W: 4161 L: 3941 D: 18779
http://tests.stockfishchess.org/tests/view/58667a830ebc5903140c632f

LTC 7 threads:
LLR: 2.95 (-2.94,2.94) [0.00,5.00]
Total: 22988 W: 2767 L: 2583 D: 17638
http://tests.stockfishchess.org/tests/view/586722690ebc5903140c636d

bench: 5468995
Windows x64 for Haswell CPUs
Windows x64 for modern computers
Windows x64
Windows 32
Linux x64 for Haswell CPUs
Linux x64 for modern computers
Linux x64
Author: lucasart
Date: Sun Jan 1 11:11:52 2017 +0100
Timestamp: 1483265512

WDL: rename WDLCursedLoss into WDLBlessedLoss

Tested using syzygy bench method:

- 2016 random positions ranging between 3 and 10 pieces
- each searched using bench at depth=10

Same node count (and no speed regression).

No functional change.
Windows x64 for Haswell CPUs
Windows x64 for modern computers
Windows x64
Windows 32
Linux x64 for Haswell CPUs
Linux x64 for modern computers
Linux x64
Author: Aram Tumanian
Date: Sun Jan 1 11:04:24 2017 +0100
Timestamp: 1483265064

Don't clear EasyMove in search()

EasyMove is cleared after every iteration of the
search if the 3rd move in the PV of the main thread
changes from the previous iteration. Therefore,
clearing EasyMove during a search iteration may be
excessive. The tests show that this is indeed unnecessary.
In the new version the EasyMove variable is used only in
the Thread::search function.

STC
LLR: 2.95 (-2.94,2.94) [-3.00,1.00]
Total: 47719 W: 8438 L: 8362 D: 30919

LTC
LLR: 2.95 (-2.94,2.94) [-3.00,1.00]
Total: 122841 W: 15448 L: 15457 D: 91936

bench: 5468995
Windows x64 for Haswell CPUs
Windows x64 for modern computers
Windows x64
Windows 32
Linux x64 for Haswell CPUs
Linux x64 for modern computers
Linux x64
Author: Sergei Antonov
Date: Sun Jan 1 10:56:46 2017 +0100
Timestamp: 1483264606

Threefold repetition detection

Implement a threefold repetition detection. Below are the examples of
problems fixed by this change.

Loosing move in a drawn position.
position fen 8/k7/3p4/p2P1p2/P2P1P2/8/8/K7 w - - 0 1 moves a1a2 a7a8 a2a1
The old code suggested a loosing move "bestmove a8a7", the new code suggests "bestmove a8b7" leading to a draw.

Incorrect evaluation (happened in a real game in TCEC Season 9).
position fen 4rbkr/1q3pp1/b3pn2/7p/1pN5/1P1BBP1P/P1R2QP1/3R2K1 w - - 5 31 moves e3d4 h8h6 d4e3
The old code evaluated it as "cp 0", the new code evaluation is around "cp -50" which is adequate.

Brings 0.5-1 ELO gain. Passes [-3.00,1.00].

STC: http://tests.stockfishchess.org/tests/view/584ece040ebc5903140c5aea
LLR: 2.96 (-2.94,2.94) [-3.00,1.00]
Total: 47744 W: 8537 L: 8461 D: 30746

LTC: http://tests.stockfishchess.org/tests/view/584f134d0ebc5903140c5b37
LLR: 2.96 (-2.94,2.94) [-3.00,1.00]
Total: 36775 W: 4739 L: 4639 D: 27397

Patch has been rewritten into current form for simplification and
logic slightly changed so that return a draw score if the position
repeats once earlier but after or at the root, or repeats twice
strictly before the root. In its original form, repetition at root
was not returned as an immediate draw.

After retestimng testing both version with SPRT[-3, 1], both passed
succesfully, but this version was chosen becuase more natural. There is
an argument about MultiPV in which an extended draw at root may be sensible.
See discussion here:

https://github.com/official-stockfish/Stockfish/pull/925

For documentation, current version passed both at STC and LTC:

STC
LLR: 2.96 (-2.94,2.94) [-3.00,1.00]
Total: 51562 W: 9314 L: 9245 D: 33003

LTC
LLR: 2.96 (-2.94,2.94) [-3.00,1.00]
Total: 115663 W: 14904 L: 14906 D: 85853

bench: 5468995
Windows x64 for Haswell CPUs
Windows x64 for modern computers
Windows x64
Windows 32
Linux x64 for Haswell CPUs
Linux x64 for modern computers
Linux x64
Author: Alain SAVARD
Date: Sat Dec 31 14:15:57 2016 +0100
Timestamp: 1483190157

Small eval cleanup and renaming

Non-functional changes

a) splitting the threat array to avoid using an enum
b) reorder the scores according to functions where they are used.
c) declarations in evaluate_pieces after the const(s) like elsewhere
d) more compact definitions of KingFlank,
now that we need it also for the PanwLessFlank penalty.
e) reuse CenterFiles in evaluate_space
f) move one line inside next popcount

No functional change.

next page >