search

thomas-maeder / popeye

Popeye is a chess problem solving and testing software with strong support for fairy chess and heterodox genres. For more information cf. topic "Popeye (chess)" on http://en.wikipedia.org/
31 stars 14 forks source link

MagicSquares + RexInclusive #238

Closed adrianstori closed 7 months ago

adrianstori commented 5 years ago

Is there a way to make a king / royal piece also change colour when moving onto a magic square? I tried specifying combinations of "white magic Ka1" and "MagicSquares a2 type1 | type2" without success.

If currently not possible, adding such a feature will be welcome (any piece moving onto a magic square changes colour). I understand there may not be a large 'market' for this feature, but it would provide a more complete solution similar to other RexInclusive coverage instances.

(My problem in fact uses a magic board -- 64 magic squares --, where any piece landing -- moving or reborn -- onto a magic square changes colour: i.e., circe rebirths are also of the other colour -- but this can be accomplished with circe turncoats anyway.)

thomas-maeder commented 5 years ago

Yes, I think that there is a way.

There is the general assumption that the side to be checked has a king in stipulations that end in something related to check (including mate). But that assumption already isn't met in Royal Dynasty.

Could you provide an example problem, preferably simpler than something with 64 magic squares? :- )

adrianstori commented 5 years ago

Right now all I have is https://pdb.dieschwalbe.de/search.jsp?expression=PROBID%3D%27P1298971%27 (there is a more detailed solution in yacpdb http://www.yacpdb.org/#305789), which is definitely not an ideal testcase. I'll try to find time to concoct something simpler...

adrianstori commented 5 years ago

Perhaps this might do:

pieces black Kc1 stipulation hZa1 1 condition magicsquares rexinclusiv a1b1

1.Kc1-b1=w Kb1-a1=b z

thomas-maeder commented 5 years ago

The author's solution of P1298971 doesn't work.

  1. Kxa5=w[+wTa1=s]+ and 5. Kxb5=w[+wLf1=s]+ are not ??, but legal cross-checks.
thomas-maeder commented 5 years ago

Also, the rules are not clear: what happens if both sides have a king and the white king moves to a magic square?

In Royal Dynasty, no side would have a royal piece until one of the Black kings moves to a magic square. Is that the intention for Magic Squares Rex inclusive?

adrianstori commented 5 years ago

"The author's solution of P1298971 doesn't work. 5. Kxa5=w[+wTa1=s]+ and 5. Kxb5=w[+wLf1=s]+ are not ??, but legal cross-checks."

Cross-checks indeed (and beautiful ones too!), but this particular problem is Ohneschach, hence these moves here are illegal.

"Also, the rules are not clear: what happens if both sides have a king and the white king moves to a magic square?"

I don't claim authorship of the idea, just believe it's nice to complete/fill a visible 'hole' in the magic-squares scheme of things. I'd follow, in coding, the path of least resistance... :-)

thomas-maeder commented 4 years ago

Sorry for the wait ... Well, with Ohneschach, all bets are off. Ohneschach's definition is an inherently unsound, and meaningful results with Ohneschach (both on the board an the computer) are merely a result of luck.