Marco Costalba [Sat, 2 May 2009 12:39:23 +0000 (14:39 +0200)]
Setup to use Callgrind profiler
Disabled by default.
No functional change.
Signed-off-by: Marco Costalba <mcostalba@gmail.com>
Marco Costalba [Sat, 2 May 2009 12:38:26 +0000 (14:38 +0200)]
Restore development versioning and LSN filtering
Signed-off-by: Marco Costalba <mcostalba@gmail.com>
Marco Costalba [Sun, 3 May 2009 06:49:34 +0000 (07:49 +0100)]
Stockfish 1.3.1
Mainteinance version to fix broken Glaurung 2
book reading.
Signed-off-by: Marco Costalba <mcostalba@gmail.com>
Marco Costalba [Sun, 3 May 2009 06:28:37 +0000 (07:28 +0100)]
Revert Glaurung 1 book compatibility patch
It breaks also Glaurung 2 book parsing.
We really need to work on book.cpp, but for now just
leave compatibility just for Glaurung 2 books.
Signed-off-by: Marco Costalba <mcostalba@gmail.com>
Marco Costalba [Sat, 2 May 2009 09:09:53 +0000 (10:09 +0100)]
Stockfish 1.3
Signed-off-by: Marco Costalba <mcostalba@gmail.com>
Marco Costalba [Sat, 2 May 2009 10:14:45 +0000 (11:14 +0100)]
Fix a very old UCI option parsing bug
We currently fail on an option with a sapece in the name,
as example
setoption name Clear Hash
returns error message "Option Clear not found". This
patch fixes this off-by-one type bug.
Thanks to Joona for spotting this.
Signed-off-by: Marco Costalba <mcostalba@gmail.com>
Marco Costalba [Fri, 1 May 2009 09:27:03 +0000 (11:27 +0200)]
Nicely simplify MovePicker::pick_move_from_list
It is a positive fall back from previous patch.
No functional change.
Signed-off-by: Marco Costalba <mcostalba@gmail.com>
Marco Costalba [Fri, 1 May 2009 08:17:34 +0000 (10:17 +0200)]
Sort moves just after scoring
Instead of a delayed selection sort so that the highest
score move is picked up from the list when needed, sort all
the moves up front just after score them.
Selection sort is O(n*n) while std::sort is O(n*log n), it
is true that delayed selection allows us to just pick the move
until a cut off occurs or up to a given limit (12), but with
an average of 30 non capture-moves delayed pick become slower
just after 5-6 moves and we now pick up to 12.
Profiling seem to prove this idea and movepick.cpp is now 10%
faster.
Also tests seem to confirm this:
After 700 games at 1+0: Mod vs Orig +178 -160 =362 +9 ELO
Signed-off-by: Marco Costalba <mcostalba@gmail.com>
Marco Costalba [Thu, 30 Apr 2009 14:16:35 +0000 (15:16 +0100)]
Update polyglot.ini
Upadte to new parameters and parameters values.
Signed-off-by: Marco Costalba <mcostalba@gmail.com>
Marco Costalba [Thu, 30 Apr 2009 07:18:20 +0000 (09:18 +0200)]
Stockfish 1.3 rc1
Signed-off-by: Marco Costalba <mcostalba@gmail.com>
Marco Costalba [Thu, 30 Apr 2009 06:55:38 +0000 (08:55 +0200)]
Do not razor after a null move
We don't want to return unproven null move fails high, so
that if a position is so good that null move fails high we
want to check this with real do_move() / undo_move() test,
not just razoring the position because, from the opponent
point of view, is very bad.
These are tests results at 1+0
Mod vs Orig +252 -264 =483 49.40%
Mod vs Toga II 1.4.1SE +365 -325 =309 52.00%
So it seems a very slightly regression regarding orig version (but
withing error bar) and a nice increase against Toga that is what we
are interested most. Orig version scores 49.75% against Toga, so
we welcome this change ;-)
Signed-off-by: Marco Costalba <mcostalba@gmail.com>
Marco Costalba [Wed, 29 Apr 2009 13:53:37 +0000 (15:53 +0200)]
Fix assignment of pv[0] when creating root move list
It is bogusly assigned from moves[i].move instead of mlist[i].move
or equivalently to moves[count].move that it seem more clear to me.
Bug is hidden while all the moves are included, in this default case
moves[i].move and mlist[i].move are the same variable.
Also a bit of cleanup while there.
Signed-off-by: Marco Costalba <mcostalba@gmail.com>
Marco Costalba [Tue, 28 Apr 2009 06:51:11 +0000 (08:51 +0200)]
Merge Joona's razoring tweaks
After proof testing on 3 different engines these
are the results:
Stockfish - Toga II 1.4.1SE +130 -132 =132 49.75%
Stockfish - Deep Sieng 3.0 +145 -110 =150 54.45%
Stockfish - HIARCS 12 MP +94 -149 =150 43.00%
So it seems no regressions occurs, although also no
improvment. But anyhow this patch increases Stockfish
strenght against itself, so merge it.
Note that this patch not only adds back razoring at depth
one, but also increases razor depth limit from 3 to 4
because hard coded depth 4 limit is no more overwritten
by UCI parameter that otherwise defaults to 3.
Marco Costalba [Tue, 28 Apr 2009 06:47:26 +0000 (08:47 +0200)]
Hardcode depth limit for selective search
Because futility margins array has a fixed size we cannot
arbitrarly choose or change the SelectiveDepth parameter,
otherwise we have a crash for values bigger then array size.
On the other hand tweaking of this parameter requires some
modification to the hardcoded margins, so makes sense to hard
code also this very bounded one.
Who wants to experiment is of course free to change the sources.
No functional change.
Signed-off-by: Marco Costalba <mcostalba@gmail.com>
Marco Costalba [Mon, 27 Apr 2009 17:00:45 +0000 (18:00 +0100)]
Fix a warning under MSVC
Somehow silly warning C4800:
'int' :forcing value to bool 'true' or 'false'(performance warning)
Signed-off-by: Marco Costalba <mcostalba@gmail.com>
Marco Costalba [Mon, 27 Apr 2009 15:39:59 +0000 (17:39 +0200)]
Micro optimize Position::move_is_check()
More then optimization it is worth a bit of better
code self documenting.
No functional change.
Signed-off-by: Marco Costalba <mcostalba@gmail.com>
Marco Costalba [Mon, 27 Apr 2009 14:35:31 +0000 (16:35 +0200)]
Inline Position::move_is_capture()
This is a very hot path function, profiling on Intel compiler
shows that inlining cuts in half the overhead.
No functional change.
Signed-off-by: Marco Costalba <mcostalba@gmail.com>
Marco Costalba [Mon, 27 Apr 2009 12:48:14 +0000 (13:48 +0100)]
Retire timeoday.cpp
Move the only function gettimeofday in misc.cpp
where is used.
This avoids polluting the global namespace.
No functional change.
Signed-off-by: Marco Costalba <mcostalba@gmail.com>
Marco Costalba [Mon, 27 Apr 2009 12:38:47 +0000 (13:38 +0100)]
Small cleanup in misc.cpp
No functional change.
Signed-off-by: Marco Costalba <mcostalba@gmail.com>
Marco Costalba [Mon, 27 Apr 2009 12:21:49 +0000 (13:21 +0100)]
Code style cleanup in transposition table code
Assorted fixes but no functional change.
Signed-off-by: Marco Costalba <mcostalba@gmail.com>
Marco Costalba [Mon, 27 Apr 2009 09:29:33 +0000 (10:29 +0100)]
Fix a bogus assert in tt.cpp
Max hash size is 4096 MB, not 1024 MB, see the corresponding
"Hash" UCI parameter in ucioption.cpp
No functional change.
Signed-off-by: Marco Costalba <mcostalba@gmail.com>
Marco Costalba [Mon, 27 Apr 2009 09:12:34 +0000 (10:12 +0100)]
Convert piece.cpp to C++
No functional change.
Signed-off-by: Marco Costalba <mcostalba@gmail.com>
Marco Costalba [Mon, 27 Apr 2009 08:45:15 +0000 (09:45 +0100)]
Remove an useless comparison in futility pruning
Currently futility is allowed when depth < SelectiveDepth
and SelectiveDepth is 7*OnePly, so the comprison is
always true.
Patch could introduce a functional change only if
we choose to increase SelectiveDepth.
Currently no functional change.
Signed-off-by: Marco Costalba <mcostalba@gmail.com>
Marco Costalba [Sun, 19 Apr 2009 16:23:30 +0000 (17:23 +0100)]
Small code tidy up and test results
When testing at 1'+0" time control results are still
reasonably good. We have made two sessions on two
different PC.
After 840 games Mod - Orig: +221 -194 =425 +10 ELO (two CPU)
After 935 games Mod - Orig: +246 -222 =467 +9 ELO (single CPU)
So it seems that with fast CPU and/or longer time controls
benefits of the patch are a bit reduced. This could be due
to the fact that only 3% of nodes are pruned by razoring at
depth one and these nodes are very swallow ones, mostly get
pruned anyway with only a slightly additional cost, even
without performing any do_move() call.
Another reason is that sometime (0,3%% of cases) a possible
good move is missed typically in positions when moving side
gives check, as example in the following one
3r2k1/pbpp1nbp/1p6/3P3q/6RP/1P4P1/P4Pb1/3Q2K1 w - -
The winning move Rxg7+ is missed.
Bottom line is that patch seems good for blitz times, perhaps
also for longer times. We should test against a third engine
(Toga ?) to have a final answer regarding this new setup.
Signed-off-by: Marco Costalba <mcostalba@gmail.com>
Joona Kiiski [Thu, 9 Apr 2009 04:29:13 +0000 (07:29 +0300)]
Razor again at depth one
Some time ago it was found by Marco Costalba that it's better
to disable razoring at depth one, because given the very low
evaluation of the node, futility pruning would already do
the job at very low cost and avoiding missing important moves.
Now enable razoring there again, but only when our quickly evaluated
material advantage is more than a rook. The idea is to try razoring
only when it's extremely likely that it will succeed.
Extreme lightning speed test show promising result:
Orig - Mod: +1285 =1495 -1348
This needs to be tested with longer time controls though.
Signed-off-by: Marco Costalba <mcostalba@gmail.com>
Joona Kiiski [Wed, 8 Apr 2009 14:35:36 +0000 (17:35 +0300)]
Make futility and razor margins more tunable
Restructure RazorMargins and FutilityMargins arrays so that their
values can be more easily tuned.
Add RazorApprMargins array which replaces razorAtDepthOne concept,
because setting RazorApprMargin very high value at ply one is
same as not razoring there at all.
Comment out setting razoring and futility margins through uci to
avoid errors while tuning.
Signed-off-by: Marco Costalba <mcostalba@gmail.com>
Marco Costalba [Sat, 25 Apr 2009 14:26:06 +0000 (15:26 +0100)]
Do not hardcode default values of UCI variables
If a variable will be populated reading an UCI option
then do not hard code its default values.
This avoids misleadings when reading the sources.
No functional change.
Signed-off-by: Marco Costalba <mcostalba@gmail.com>
Marco Costalba [Mon, 20 Apr 2009 07:10:27 +0000 (09:10 +0200)]
Greatly speedup has_mate_threat()
Instead of loop across all legal moves to find a mate
loop across possible check moves only.
This reduces more then 10 times the number of moves to
check for a possible mate.
Also rename generate_checks() in generate_non_capture_checks()
so to better clarify what the function does.
No functional change.
Signed-off-by: Marco Costalba <mcostalba@gmail.com>
Marco Costalba [Sun, 19 Apr 2009 19:09:53 +0000 (21:09 +0200)]
Micro optimize generate_piece_checks() take 2
Add some missing bits of this patch.
Signed-off-by: Marco Costalba <mcostalba@gmail.com>
Joona Kiiski [Sun, 19 Apr 2009 16:46:07 +0000 (19:46 +0300)]
Add missing header file to make stockfish compile with latest Inter C++ Compiler under Linux (memcpy needs cstring)
Correct some references glaurung -> stockfish in Makefile
Signed-off-by: Marco Costalba <mcostalba@gmail.com>
Marco Costalba [Sun, 19 Apr 2009 15:44:12 +0000 (16:44 +0100)]
Simplify Position::is_mate()
Should be a bit faster too.
No functional change.
Signed-off-by: Marco Costalba <mcostalba@gmail.com>
Marco Costalba [Sat, 18 Apr 2009 15:21:44 +0000 (16:21 +0100)]
Micro optimize generate_piece_checks()
Avoid calculating piece attacks if there aren't
available check sqaures for the given piece.
About 15% of cases. Not a biggie but still something
especially in the middle game where king is well covered
inside his castle.
No functional change.
Signed-off-by: Marco Costalba <mcostalba@gmail.com>
Marco Costalba [Sat, 18 Apr 2009 13:19:19 +0000 (14:19 +0100)]
MovePicker: retire per square MVV/LVA ordering
Is not used anyway and in case we need it again we
can resurrect from git archives.
Signed-off-by: Marco Costalba <mcostalba@gmail.com>
Marco Costalba [Sat, 18 Apr 2009 13:03:33 +0000 (14:03 +0100)]
In qsearch store the cut move in TT
And upon reentering the same position try it as first.
Normally qsearch moves order is already very good, first move
is the cut off in almost 90% of cases. With this patch, we get
a cut off on TT move of 98%.
Another good side effect is that we don't generate captures
and/or checks when we already have a TT move.
Unfortunatly we found a TT move only in 1% of cases. So real
impact of this patch is relatively low.
Signed-off-by: Marco Costalba <mcostalba@gmail.com>
Marco Costalba [Thu, 16 Apr 2009 06:35:26 +0000 (08:35 +0200)]
Merge Joona's new aspiration window search
It seems very positive.
After 999 games at 1'+0" result is: +249 -216 =534 +11 ELO
And after another 456 games we still have: +122 -113 =221
Marco Costalba [Tue, 14 Apr 2009 10:36:34 +0000 (12:36 +0200)]
Introduce move_pawns() helper in movegen.cpp
This let us to have more readable code keeping the
same speed.
No functional change.
Signed-off-by: Marco Costalba <mcostalba@gmail.com>
Marco Costalba [Thu, 16 Apr 2009 08:02:01 +0000 (10:02 +0200)]
Remove failHigh/Low bits from IterationInfoType
We don't use that info anyway.
Also document a little more new aspiration window code.
Signed-off-by: Marco Costalba <mcostalba@gmail.com>
Joona Kiiski [Wed, 15 Apr 2009 15:14:06 +0000 (18:14 +0300)]
Restore calling insert_pv after search is aborted + small clean-up
Restore old behaviour that after search is aborted we call insert_pv,
before breaking out from the iterative deepening loop.
Remove one useless FIXME and document other better.
Signed-off-by: Marco Costalba <mcostalba@gmail.com>
Marco Costalba [Wed, 15 Apr 2009 10:44:55 +0000 (12:44 +0200)]
Little code style tweaks
Let the code be more conformant to current style.
No functional change.
Signed-off-by: Marco Costalba <mcostalba@gmail.com>
Marco Costalba [Wed, 15 Apr 2009 08:20:35 +0000 (10:20 +0200)]
Revert previous patch as per Joona request
Joona says patch gives bad results after testing,
so revert for now.
Signed-off-by: Marco Costalba <mcostalba@gmail.com>
Joona Kiiski [Sun, 12 Apr 2009 21:27:07 +0000 (00:27 +0300)]
Improve handling of fail-highs in assumed PV
Check all fail highs in assumed PV with greater care (fruit/Toga already does this).
Add a flag when aspiration search fails high at ply 1 to prevent search to
be terminated prematurely.
Signed-off-by: Marco Costalba <mcostalba@gmail.com>
Joona Kiiski [Sun, 12 Apr 2009 20:30:40 +0000 (23:30 +0300)]
Revert "Implement a fallback system when aspiration search fails low and we are out of time."
This reverts commit
55dd98f7c717b94a659931cd20e088334b1cf7a6.
Revert fallback-system for root_search
Signed-off-by: Marco Costalba <mcostalba@gmail.com>
Joona Kiiski [Sun, 12 Apr 2009 20:25:05 +0000 (23:25 +0300)]
Implement a fallback system when aspiration search fails low and we are out of time.
However also this patch is immediately reverted. For three reasons:
1) the case it affects is very rare (and then we are likely to lose anyway),
so we can well live without this.
2) Because the case is so rare it's hard to test this change properly.
3) To perform fallback search, we must reset so many global variables that this
patch is very likely both buggy and extremely bad style.
Consider including this again if we clean-up global variables one day...
Signed-off-by: Marco Costalba <mcostalba@gmail.com>
Joona Kiiski [Sun, 12 Apr 2009 16:49:09 +0000 (19:49 +0300)]
Revert "Implement bestValue in root_search."
This reverts commit
9a39f93f35254787b7b57980019dde276a89c48c.
Revert bestValue in root_search
Signed-off-by: Marco Costalba <mcostalba@gmail.com>
Joona Kiiski [Sun, 12 Apr 2009 16:42:47 +0000 (19:42 +0300)]
Implement bestValue in root_search.
However just after finished writing this patch I realized that this
is not the way to go. So this will be immediately reverted.
(Just save this here in git in case I change my mind later :) )
Signed-off-by: Marco Costalba <mcostalba@gmail.com>
Joona Kiiski [Sat, 11 Apr 2009 13:46:35 +0000 (16:46 +0300)]
Dynamic aspiration search without research.
Implement system where aspiration search window is calculated using
values from previous iterations.
And then some crazy experimental stuff: If search fails low at the root,
don't widen window, but continue and hope we will find a better move
with given window. If search fails high at the root, cut immediately,
add some more time and start new iteration.
Note: this patch is not complete implementation, but a first test
for this idea. There are many FIXMEs left around. Most importantly
how to deal with the situation when we don't have any move!
Signed-off-by: Marco Costalba <mcostalba@gmail.com>
Marco Costalba [Sun, 12 Apr 2009 15:45:33 +0000 (16:45 +0100)]
Re-fix square.h warning to avoid a compile error under MSVC
This fix adds ugliness to an already ugly previous fix...hopefully
it is the last one :-(
Signed-off-by: Marco Costalba <mcostalba@gmail.com>
Marco Costalba [Sun, 12 Apr 2009 16:05:09 +0000 (18:05 +0200)]
Fix a gcc warning due to order of initialization in Option
Move idx declaration before minValue and maxValue and silence
this last warning.
Signed-off-by: Marco Costalba <mcostalba@gmail.com>
Marco Costalba [Sun, 12 Apr 2009 16:01:06 +0000 (18:01 +0200)]
Fix a compile error with Intel icc
To make std::sort() work operator<() should be
declared const.
Signed-off-by: Marco Costalba <mcostalba@gmail.com>
Marco Costalba [Sun, 12 Apr 2009 15:58:30 +0000 (17:58 +0200)]
Fix a warning under Intel compiler in square.h
We need to cast to char the whole expression...
Signed-off-by: Marco Costalba <mcostalba@gmail.com>
Marco Costalba [Sun, 12 Apr 2009 00:09:03 +0000 (01:09 +0100)]
Fix a very nasty conversion bug in Option c'tor
Sometimes C++ can be really bad!
In this case an hard coded c string selects Option c'tor
with int argument instead of the std::string one becuase
it is considered a better matching by the compiler.
Fix the bug changing the argument type from std::string to
const char* so to be a better match then the int one.
Signed-off-by: Marco Costalba <mcostalba@gmail.com>
Marco Costalba [Sat, 11 Apr 2009 12:04:33 +0000 (14:04 +0200)]
Store UCI options of type CHECK according to C++ convention
Store boolean values as "1" and "0" instead of "true" and "false"
and convert back to UCI protocol convention when needed.
This is simpler then the other way around.
Signed-off-by: Marco Costalba <mcostalba@gmail.com>
Marco Costalba [Sat, 11 Apr 2009 11:28:51 +0000 (13:28 +0200)]
Restore original UCI option printing order
For each option store its index so to be printed
according to insertion order.
Signed-off-by: Marco Costalba <mcostalba@gmail.com>
Marco Costalba [Sat, 11 Apr 2009 10:46:47 +0000 (12:46 +0200)]
Use a map instead of a vector to store UCI options
Apart from the teoretical speed increase, the main reason
of this patch is a good amount of code cleanup.
Note that now UCI options are printed in alphabetical
order and not in insertion order as before. Next patch
will take care of restoring old behaviour.
Signed-off-by: Marco Costalba <mcostalba@gmail.com>
Marco Costalba [Tue, 7 Apr 2009 07:54:32 +0000 (09:54 +0200)]
Remove unused currentMoveCaptureValue from search stack
No functional change.
Signed-off-by: Marco Costalba <mcostalba@gmail.com>
Marco Costalba [Fri, 10 Apr 2009 08:49:29 +0000 (09:49 +0100)]
Less aggressive null move dynamic reduction
In null move search do not jump directly in
qsearch() from depth(4*OnePly), but only
from depth(3*OnePly).
After 999 games at 1+0: +248 -224 =527 +8ELO
Signed-off-by: Marco Costalba <mcostalba@gmail.com>
Marco Costalba [Fri, 3 Apr 2009 19:34:22 +0000 (20:34 +0100)]
Revert dynamic LMR
It doesn't seem to work against Toga. After more then 400 games
we are at -13 ELO while, without it we are at + 5 ELO after 1000
games.
So revert for now to keep code simple.
Signed-off-by: Marco Costalba <mcostalba@gmail.com>
Marco Costalba [Mon, 30 Mar 2009 10:12:52 +0000 (12:12 +0200)]
Fix compile in the debug mode
Due to previous patches we end up with a compile
error in debug mode.
No functional change.
Signed-off-by: Marco Costalba <mcostalba@gmail.com>
Marco Costalba [Mon, 30 Mar 2009 10:07:45 +0000 (12:07 +0200)]
Silence idiotic warning on two's complement of an unsigned
MSVC gives:
warning C4146: unary minus operator applied to unsigned type,
result still unsigned
When finds -b where b is an unsigned integer. So rewrite the two's
complement in a way to avoid the warning. Theoretically the new
version is slower, but in practice changes nothing.
Signed-off-by: Marco Costalba <mcostalba@gmail.com>
Marco Costalba [Mon, 30 Mar 2009 09:55:13 +0000 (11:55 +0200)]
Silence a bunch of warnings under MSVC /W4
Still some remain, but are really the silly ones.
No functional change.
Signed-off-by: Marco Costalba <mcostalba@gmail.com>
Marco Costalba [Tue, 31 Mar 2009 12:57:44 +0000 (14:57 +0200)]
Revert setting a flag when TT value equals static evaluation
Strangely enough it seems that optimization doesn't work.
After 760 games at 1+0: +155 -184 =421 -13 ELO
Probably the overhead, although small, for setting the flag
is not compensated by the saved evaluation call.
This could be due to the fact that after a TT value is stored,
if and when we hit the position again the stored TT value is
actually used as a cut-off so that we don't need to go on
with another search and evaluation is avoided in any case.
No functional change.
Signed-off-by: Marco Costalba <mcostalba@gmail.com>
Marco Costalba [Mon, 30 Mar 2009 08:09:27 +0000 (09:09 +0100)]
An VALUE_TYPE_EVAL score cannot overwrite an entry
If we want to store a value of type VALUE_TYPE_EVAL for
a given position and we found an already exsisting entry
for the same position then we skip.
We don't want to overwrite a more valuable score with a
lesser one. Note that also in case the exsisting entry is
of VALUE_TYPE_EVAL type the overwrite is unuseful because
we would store the same score again.
Signed-off-by: Marco Costalba <mcostalba@gmail.com>
Marco Costalba [Mon, 30 Mar 2009 07:54:09 +0000 (08:54 +0100)]
Remember when TT value equals static evaluation value
When the stored TT value equals the static value set a
proper flag so to not call evaluation() if we hit the
same position again but use the stored TT value instead.
This is another trick to avoid calling costly evaluation()
in qsearch.
No functional change.
Signed-off-by: Marco Costalba <mcostalba@gmail.com>
Marco Costalba [Mon, 30 Mar 2009 07:29:02 +0000 (08:29 +0100)]
Document TTEntry and move layouts
No functional change.
Signed-off-by: Marco Costalba <mcostalba@gmail.com>
Marco Costalba [Sun, 29 Mar 2009 16:22:10 +0000 (17:22 +0100)]
A move needs 17 bits not 19
Fix a bug in the way a move is stored and read in a TT entry.
We use a mask of 19 bits insteaad of 17 so that the last
two bits in the TT entry end up to be random data.
This bug will bite us when we will use these two until now
unused bits.
Signed-off-by: Marco Costalba <mcostalba@gmail.com>
Marco Costalba [Fri, 27 Mar 2009 18:18:21 +0000 (19:18 +0100)]
Remove some obsolete code in movepick.cpp
This fixes some warning under Intel compiler.
No functional change.
Signed-off-by: Marco Costalba <mcostalba@gmail.com>
Marco Costalba [Fri, 27 Mar 2009 14:30:45 +0000 (15:30 +0100)]
Revert storing of TT when returning from "stand pat"
After testing it seems patch is bad:
After 999 games 1+0: +242 -271 =486 -10 ELO
So restore saving of TT at the end but using new Joona
idea of storing as VALUE_TYPE_UPPER/VALUE_TYPE_LOWER instead
of VALUE_TYPE_EXACT.
Some optimization is still possible but better test new ideas
one by one.
Signed-off-by: Marco Costalba <mcostalba@gmail.com>
Marco Costalba [Tue, 24 Mar 2009 14:48:14 +0000 (15:48 +0100)]
Cache evaluation score in qsearch
Instead of just drop evaluation score after stand pat
logic save it in TT so to be reused if the same position
occurs again.
Note that we NEVER use the cached value apart to avoid an
evaluation call, in particulary we never return to caller
after a succesful tt hit.
To accomodate this a new value type VALUE_TYPE_EVAL has been
introduced so that ok_to_use_TT() always returns false.
With this patch we cut about 15% of total evaluation calls.
Signed-off-by: Marco Costalba <mcostalba@gmail.com>
Marco Costalba [Tue, 24 Mar 2009 13:34:46 +0000 (14:34 +0100)]
Let to toggle dynamic LMR
It is now disabled by default due to bad results
against a pool of engines...more testing is needed tough.
Signed-off-by: Marco Costalba <mcostalba@gmail.com>
Marco Costalba [Tue, 24 Mar 2009 10:26:43 +0000 (11:26 +0100)]
In qsearch update TT only if returning from stand pat
This is the only "correct" exact value we can store.
Otherwise there could be spurious failed high/low nodes.
Signed-off-by: Marco Costalba <mcostalba@gmail.com>
Marco Costalba [Tue, 24 Mar 2009 09:50:13 +0000 (10:50 +0100)]
When asked for position key print it as an hex value
Merged from Glaurung current development snapshot.
Signed-off-by: Marco Costalba <mcostalba@gmail.com>
Marco Costalba [Tue, 24 Mar 2009 09:49:02 +0000 (10:49 +0100)]
Wait at least until iteration 3 before to stop the search
It was 2 before.
Merged from Glaurung current development snapshot.
Signed-off-by: Marco Costalba <mcostalba@gmail.com>
Marco Costalba [Tue, 24 Mar 2009 09:40:22 +0000 (10:40 +0100)]
Let zobrist keys to be compatible with Glaurung 1
Some changes to the zobrist keys, to make them identical
to those used by Glaurung 1.
The only purpose is to make it possible for both programs
to use the same opening book.
Merged from Glaurung current development snapshot.
Signed-off-by: Marco Costalba <mcostalba@gmail.com>
Marco Costalba [Tue, 24 Mar 2009 09:30:27 +0000 (10:30 +0100)]
Fixed a sliding attack bitboard bug in 32-bit mode
This is what prevented USE_32BIT_ATTACKS from working
on some architectures (like PowerPC).
Merged from Glaurung current development snapshot.
Signed-off-by: Marco Costalba <mcostalba@gmail.com>
Marco Costalba [Mon, 23 Mar 2009 14:30:20 +0000 (15:30 +0100)]
Fix a bug in insert_pv() where minimum depth is zero
We implicitly considered the minimum depth stored in TT
to be Depth(0), but because we store values in TT also in
qsearch() where depth is < 0, we need to use a negative
number as minimum depth.
Bug spotted by Joona Kiiski.
Signed-off-by: Marco Costalba <mcostalba@gmail.com>
Marco Costalba [Mon, 23 Mar 2009 11:02:15 +0000 (12:02 +0100)]
Revert odd depths razoring
I have just made a new rule that no modification
that increases pruning is allowed if after 1000 games
ELO is not increased by at least 10 point (was +5 in this case)
Yes, I like this kind of nosense rules :-)
Marco Costalba [Mon, 23 Mar 2009 10:58:28 +0000 (11:58 +0100)]
More aggressive dynamic LMR
Previous setup didn't change anything
After 996 games 1+0: +267 -261 =468 +2 ELO
Now with this new setup we have
After 999 games 1+0: +277 -245 =477 +11 ELO
Seems reasonable...
Signed-off-by: Marco Costalba <mcostalba@gmail.com>
Marco Costalba [Fri, 20 Mar 2009 11:33:30 +0000 (12:33 +0100)]
LMR dynamic reduction
Reduce of two plies near the leafs and when we still
have enough depth to go so to limit horizon effects.
Signed-off-by: Marco Costalba <mcostalba@gmail.com>
Marco Costalba [Sun, 22 Mar 2009 13:49:18 +0000 (14:49 +0100)]
Always print a best move when requested
Little fix merged from iPhone Glaurung.
Signed-off-by: Marco Costalba <mcostalba@gmail.com>
Marco Costalba [Sun, 22 Mar 2009 13:32:00 +0000 (14:32 +0100)]
Better castle move detector in move_to_san()
Merged from iPhone Glaurung.
Signed-off-by: Marco Costalba <mcostalba@gmail.com>
Marco Costalba [Sun, 22 Mar 2009 13:18:25 +0000 (14:18 +0100)]
Fix a smal bug in Position::from_fen
We could fail to parse an en-passant position
in same cases.
Merged from iPhone Glaurung.
Signed-off-by: Marco Costalba <mcostalba@gmail.com>
Marco Costalba [Sun, 22 Mar 2009 12:55:08 +0000 (13:55 +0100)]
Merge new pawn storm evaluation
More accuracy in pawn storm evaluation
directly from iPhone Glaurung.
Signed-off-by: Marco Costalba <mcostalba@gmail.com>
Marco Costalba [Sun, 22 Mar 2009 12:11:24 +0000 (13:11 +0100)]
Fix a small bug in king safety
Merged from iPhone Glaurung.
Signed-off-by: Marco Costalba <mcostalba@gmail.com>
Marco Costalba [Sun, 22 Mar 2009 12:06:29 +0000 (13:06 +0100)]
Merge KBPP vs KB endgame from iPhone Glaurung
Signed-off-by: Marco Costalba <mcostalba@gmail.com>
Marco Costalba [Sat, 21 Mar 2009 13:51:31 +0000 (14:51 +0100)]
Fix: In qsearch do not use TT value when in a PV node
We already do like this in search_pv(), so extend
also in qsearch().
Bug spotted by Joona Kiiski.
Signed-off-by: Marco Costalba <mcostalba@gmail.com>
Marco Costalba [Wed, 18 Mar 2009 14:36:03 +0000 (15:36 +0100)]
Big headers cleanup
No functional change.
Signed-off-by: Marco Costalba <mcostalba@gmail.com>
Marco Costalba [Mon, 16 Mar 2009 12:59:41 +0000 (13:59 +0100)]
Safe guard some wild and ugly casts
These casts are needed but plain ugly, at least be
sure they don't hide any subtle conversion bug.
Signed-off-by: Marco Costalba <mcostalba@gmail.com>
Marco Costalba [Mon, 16 Mar 2009 07:02:33 +0000 (08:02 +0100)]
Partially revert pawns storm bug fix
Try to save space and use the minimum size
possible.
In particular restore int16_t for values and int8_t
for halfOpenFiles.
Use int16_t for storm values insted of int and also
instead of original buggy and too small int8_t.
Signed-off-by: Marco Costalba <mcostalba@gmail.com>
Marco Costalba [Sun, 15 Mar 2009 17:07:28 +0000 (18:07 +0100)]
Silence a good bunch of Intel warnings
Note that some pawns and material info has been switched
to int from int8_t.
This is a waste of space but it is not clear if we have a
faster or slower code (or nothing changed), some test should be
needed.
Few warnings still are alive.
Signed-off-by: Marco Costalba <mcostalba@gmail.com>
Marco Costalba [Sun, 15 Mar 2009 17:06:40 +0000 (18:06 +0100)]
Fix an overflow bug in pawns stormValue
These fields are defined as int8_t but values bigger
then 127 are stored there so that we silently overflow.
Fix bringing up all the fields to a sane int type. This
will increase memory usage, but apart from being safe, it is
not clear if code is slower or faster. Test is needed.
Signed-off-by: Marco Costalba <mcostalba@gmail.com>
Marco Costalba [Sun, 15 Mar 2009 16:02:09 +0000 (17:02 +0100)]
Fix a silly warning on Intel compiler
No functional change.
Signed-off-by: Marco Costalba <mcostalba@gmail.com>
Marco Costalba [Sun, 15 Mar 2009 15:44:12 +0000 (16:44 +0100)]
Revert NULL move beta corrections
After testing result is bad -25 ELO
Marco Costalba [Fri, 13 Mar 2009 14:39:22 +0000 (15:39 +0100)]
Retire Null Driven IID
It does not seem to clearly improve things and
in any case is disabled by default, so retire for now.
Signed-off-by: Marco Costalba <mcostalba@gmail.com>
Marco Costalba [Fri, 13 Mar 2009 14:26:02 +0000 (15:26 +0100)]
Merge Joona Kiiski NULL search beta correction
Prune more moves after a null search because of
a lower beta limit then in main search.
In test positions reduces the searched nodes of 30% !!!!
Signed-off-by: Marco Costalba <mcostalba@gmail.com>
Marco Costalba [Fri, 13 Mar 2009 10:20:24 +0000 (11:20 +0100)]
Merge Joona Kiiski evaluation tweaks
Merge tewaks to many evaluation parameters
by Joona Kiiski.
After test they seem good!
Signed-off-by: Marco Costalba <mcostalba@gmail.com>
Marco Costalba [Sat, 7 Mar 2009 19:38:05 +0000 (20:38 +0100)]
Micro optimize move_is_check()
No functional change.
Signed-off-by: Marco Costalba <mcostalba@gmail.com>
Marco Costalba [Sat, 7 Mar 2009 19:08:45 +0000 (20:08 +0100)]
Micro optimize pl_move_is_legal()
No functional change.
Signed-off-by: Marco Costalba <mcostalba@gmail.com>
Marco Costalba [Fri, 6 Mar 2009 19:39:43 +0000 (20:39 +0100)]
Micro optimize previous patch
Also remove some Intel warnings, not all :-(
No functional change.
Signed-off-by: Marco Costalba <mcostalba@gmail.com>
Marco Costalba [Fri, 6 Mar 2009 15:29:46 +0000 (16:29 +0100)]
Introduce evaluate_pieces<>() to remove redundancy
No functional change.
Signed-off-by: Marco Costalba <mcostalba@gmail.com>
Marco Costalba [Wed, 4 Mar 2009 22:11:23 +0000 (23:11 +0100)]
Fix compile error with inlines under gcc and Intel
It seems that these compilers do not like inline functions
that call a template when template definition is not in scope.
So move functions from header to in *.cpp file
Signed-off-by: Marco Costalba <mcostalba@gmail.com>