-
Something has changed, and now I get really weird 'solutions': 1.nPb7*c8=nQ [+nBe8] 2.nQc8*e8 [+nBh6] #
How would this be legal in CageCirce? As far as I can understand the condition, it means that a piece can only be reborn in a cage, i.e. have only capturing moves.
2009-07-22 16:24:29 UTC in Popeye (chess)
-
I was testing a trivial AUW scheme with CageCirce, and found the following:
begin
prot cage.out
pie whi ka1
bla kf8
neu pf7h7b7 bc8
stip ser-~3
cond cage
opt move
end
All move orders are found in 3 seconds.
begin
prot cage.out
pie whi ka1
bla kf8
neu pf7h7b7 bc8
stip ser-#3
cond cage
opt move
end
This hangs after the second move tested.
begin
prot cage.out
pie...
2009-07-21 13:45:35 UTC in Popeye (chess)
-
Currently there are a few non-named leapers implemented: 15, 16, 24, 25, 35, 36, 37. A generalised form of this would be useful, to avoid a whole list of these unnamed leapers. Suggested syntax: X,Y (comparable with X/Y for generalised hunters). The number-pieces could then even be removed altogether (although it's probably useful to keep them for backwards compatibility).
2009-07-13 19:33:46 UTC in Popeye (chess)
-
Now a=>b is implemented for fairies, it should be possible to check cavaliermajeur/leofamily proofgames, but the initial position should be changed for this to work. It's of course possible to use a=>b with the correct beginposition, but implementing the dia# stipulation for these conditions (are there any more where the initial game set is changed?) would be easier to use.
2009-06-28 18:41:24 UTC in Popeye (chess)
-
Lots of weird 'solutions' found with a fairy a=>b composition:
Rene J. Millour
Die Schwalbe Heft 203, october 2003
A: 8/3pP3/1N4pG/1pp2k1S/R5p1/7p/BP3p2/2KQS3 (10+8)
B: 8/3p4/1N4pG/1pp2k1S/R5p1/7p/BP6/2KQS3 (9+7)
a=>b3 6.1;1.1;1.1
Popeye Linux-2.6.29.4-167.fc11.i686.PAE-i686-32Bit v4.52 (1024 MB)
begin
fors 8/3pP3/1N4pG/1pp2k1S/R5p1/7p/BP3p2/2KQS3
stip a=>b3
fors...
2009-06-23 14:23:58 UTC in Popeye (chess)
-
>how would you indicate that any black castling is mutually exclusive with white king side castling?
Probably easiest by setting the mutual exclusiveness property to the king (comparable to nocastling):
option mutualexclusive h1e8
The name was just a suggestion, the exact name (and translations) are up to the developers of course.
2009-06-13 22:04:39 UTC in Popeye (chess)
-
Madrasi+Anticirce: If the capture is illegal because the rebirth square is occupied, a piece isn't paralysed. (e.g. a white knight on d3 isn't paralysed by a black knight on c1 if g8 is occupied).
Madrasi+MarsCirce: The paralysis starts from the rebirth square, not from the actual square (e.g. a black queen on c2 is paralysed by a white queen on a7).
There are probably other combinations...
2009-05-31 09:38:00 UTC in Popeye (chess)
-
Files py-engl.txt, py-deut.txt and py-fran.txt weren't fully updated either (header says 4.4, some new fairy pieces aren't in them, and perhaps more).
2009-05-31 09:34:08 UTC in Popeye (chess)
-
There seems to be something wrong with neutral bishops in Ghost chess:
Christian POISSON
Vlaicu CRISAN
Quartz 2001
White : Kd1
Black : Kb6 Ra5 Pb5a4
Neutral : Pc7d7
h‡2,5 (1+4+2) C+
2.1.1...
Fantômes
1…d8=Fn 2.Ta6 ç×d8=Cn(»Fnd8) 3.Ra5 Cnç6[Fnd8]‡
1…ç8=Dn 2.Ta7 d×ç8=Tn(»Dnç8) 3.Rb7 Tnb8[Dnç8]‡
The bishop solution isn't found by Popeye 4.51. When the white...
2009-05-27 08:18:30 UTC in Popeye (chess)
-
Odd combination of fairy rules results in illegal moves.
BeginProblem
Author Unknown
Origin Unknown
Protocol C:\apps\schaken\Fancy\py.txt
Pieces
White Ph7 Kd6 Qc1
Black Kd8 Qg8
Stipulation #1
Condition circedoubleag singlebox
Option Variation NoBoard
twin rem c1
EndProblem
In a), there should be no solution (hg8=X[+wQd1] is illegal under SB), but popeye finds 1.hg8=R[+wQd1]#...
2009-05-12 14:14:02 UTC in Popeye (chess)
