summary |
shortlog | log |
commit |
commitdiff |
tree
first ⋅ prev ⋅ next
H.G. Muller [Fri, 18 Jan 2013 16:18:40 +0000]
Give reason when Lion captures are rejected as illegal
H.G. Muller [Fri, 18 Jan 2013 16:15:50 +0000]
Print static eval with thinking output
Also compute difEval from scratch, for error checking.
H.G. Muller [Fri, 18 Jan 2013 16:08:44 +0000]
Fix eval sign bug in (not-yet-used) null move
H.G. Muller [Fri, 18 Jan 2013 16:07:14 +0000]
Take somewhat more spacing for move sorting
The gaps in the move stack are now 50 in stead of 20.
H.G. Muller [Fri, 18 Jan 2013 16:05:30 +0000]
Fix bug in move sorting
Due to reverse comparison, oves were resorted on every iteration into the
static order, completely thwarting the purpose of (I)ID.
H.G. Muller [Fri, 18 Jan 2013 16:02:17 +0000]
Fix bug in PST initialization
The null table was not explicitly initialized, and was contaminated by
earlier initialization with another board size.
H.G. Muller [Fri, 18 Jan 2013 15:01:40 +0000]
Fix move bug for oce and for all
Also Falcons and Eagles captured like Knight if victim attacked by Lion.
H.G. Muller [Fri, 18 Jan 2013 14:59:18 +0000]
Claim win against illegal move if opponent is computer
Otherwise the game would hang in Two Machines mode, which is very undesirable
if the opponent is a connection adapter to a server.
H.G. Muller [Fri, 18 Jan 2013 14:55:41 +0000]
Give reason with Illegal move message in case of repeat
H.G. Muller [Fri, 18 Jan 2013 14:05:20 +0000]
Improve potential debug print
H.G. Muller [Fri, 18 Jan 2013 14:04:41 +0000]
Make sure King-capture move is printed
H.G. Muller [Fri, 18 Jan 2013 14:02:55 +0000]
Fix bug in Phoenix captures
Jumping pieces would capture as Knight when the victim was also attacked
b a Lion.
H.G. Muller [Fri, 18 Jan 2013 14:00:36 +0000]
Add Tenjiku and other variants
H.G. Muller [Thu, 12 Jul 2012 10:11:31 +0000]
Add Chess pieces
H.G. Muller [Thu, 12 Jul 2012 10:07:57 +0000]
Some work on incremental attack-map update
H.G. Muller [Thu, 12 Jul 2012 10:05:59 +0000]
Add mobility eval
Count moves during construction of atack map. The mobility weight is
based on the base value of the piece.
H.G. Muller [Thu, 12 Jul 2012 10:02:36 +0000]
Beter tuning of time control
Reduce target time, and keep a larger safety margin.
H.G. Muller [Thu, 12 Jul 2012 10:00:52 +0000]
Delete some disabled code
H.G. Muller [Mon, 9 Jul 2012 15:12:37 +0000]
Implement Dai Shogi
The board format was made variable, and the relevant parameters are set
during Init() according to the current variant (now passed as parameter).
LookUp() now searches the requested piece in multiple lists, depending
on the current variant. E.g. for Dai it searches daiPieces, followed by
chuPieces. The WB variant command is now interpreted, by comparing
the mentioned variat to the list of variants. Support for plain Knight
moves was added. The suppression of promotions after an initial deferral
was made subject to chuFlag.
H.G. Muller [Fri, 6 Jul 2012 20:17:01 +0000]
Version number 0.0
Add the myname feature to print name + this version.
H.G. Muller [Fri, 6 Jul 2012 20:15:08 +0000]
Some work on incremental update of attack map
H.G. Muller [Fri, 6 Jul 2012 07:45:41 +0000]
Recognize 'result' command in WB driver
Currently a no-op, but just to suppress Unknown Command errors.
H.G. Muller [Fri, 6 Jul 2012 07:43:15 +0000]
Implement printing of WB thinking output
H.G. Muller [Fri, 6 Jul 2012 07:38:26 +0000]
Add null move
The WB Alien null-move notation is recognized on input, and generated
on output. The internal representation is a move towards the same square
by a piece that has this capability. (This currently fails on a1, due
to 0 being used as representation for invalid moves!)
Only one null move is generated and added to the move list.
The null move is not searched yet.
H.G. Muller [Thu, 5 Jul 2012 21:45:15 +0000]
Implement repetition detection
A linear search through repStack, containing the hash keys, is used
to detect repetitions and declare them lost. Repetitions can already
start two ply ago, due to turn passing. A reversible move counter
is added for the purpose, and saved in the UndoInfo.
H.G. Muller [Thu, 5 Jul 2012 21:39:00 +0000]
Give pieces of same type same Zobrist key
Give different piece keys to white and black pieces of same type. Switch
to a system that assigns the key 'on demand', when the entry in the
list of piece descriptions is first used. Let Lookup return pointer
rather than number in list, so that future versions can search in multiple
lists (for other variants).
H.G. Muller [Thu, 5 Jul 2012 21:29:57 +0000]
Keep track of PV
Use the triangular-array method (Fairy-Max style) to keep track of the PV.
H.G. Muller [Thu, 5 Jul 2012 21:20:04 +0000]
Pass difEval on to root search
Upto now all searches started at 0.
H.G. Muller [Thu, 5 Jul 2012 21:13:26 +0000]
Add some debug prints
H.G. Muller [Thu, 5 Jul 2012 21:04:39 +0000]
Save hash key in UndoInfo
This to make it possible for the MakeMove and UnMake to update it
at game level during the ponder search.
H.G. Muller [Thu, 5 Jul 2012 20:52:48 +0000]
Fix alpha bug, promotion bug, highlighting and move printing bugs
Two crippling bugs fixed: alpha was not reset at the start of later
iterations, and promotions were added to the move-list of deeper nodes
due to spoiling of the global nonCapts.
Essential deferrals were not correctly flagged on input moves.
Printing move lists could miss some initial moves, wrecking highlighting.
H.G. Muller [Thu, 5 Jul 2012 20:36:49 +0000]
Add time control
A simple time control, finishing iterations, is added. Also a node counter.
H.G. Muller [Thu, 5 Jul 2012 19:53:37 +0000]
Fix signals for Linux, input-buffer overrun
H.G. Muller [Tue, 3 Jul 2012 13:00:23 +0000]
More developed version for WinBoard Alien Edition
This commit contains the code as it was further developed on Windows,
and tested under WB-Alien. Highlighting was added, and a 3-ply fixed-depth
search is invoked when it is the engines turn to move. Piece-square tables
have been added to draw pieces to the center, and the move generator respects
the special Lion-capture and promotion rules of Chu Shogi. Most of the time
things work, but there must be a bad bug, as sometimes it suddenly moves
with an opponent piece!
H.G. Muller [Tue, 3 Jul 2012 12:54:54 +0000]
Initial commit of Linux code
This embrionic code was developed on Linux. The move generator is
sort of working, but the multi-move feature is barely tested, because
the standard XBoard does not support that.