﻿﻿ MatPlus.Net

Website founded by
Milan Velimirović
in 2006

21:37 UTC
 ISC 2021

Remember me

 CHESS SOLVINGTournamentsRating lists1-Apr-2021

On behest (what ever that is) of Joost's problem in
the other thread (3 stipulations, 1 fairy piece type,
which doesn't even remotely seem to be a record :-),
my question: Which is the record problem in maximizing
stipulations?
Sorry, different fairy pieces do not count, it should
be fairy easy to make a meaningful problem with a
stud of m,n-leapers. Thus only stipulations count,
and the problem must be C+, preferrably in three
different programs.
It is explicitely allowed, though, to write a FORTRAN
program generating Popeye input for the Vielväterproblem,
combining random stipulations and hoping for a correct
problem popping up.

EDIT: Dmitri is of course correct, I mean side conditions,
not stipulations!

The word "stipulation" is conventionally reserved for the task the solver has to accomplish (i.e. checkmate black in two moves). The modifications to the rules of the game are usually called "fairy conditions".

The names of the modifications are absolutely arbitrary, you may call the combination of Atom + Mirror circe + Glasgow = Joost's chess, so the question doesn't make much sense.
And as far as modifications go, you may consider a crossword puzzle to be a chess problem with the rules of chess modified beyond recognition.

From a mathematical standpoint I admit you are right -
condition naming is arbitrary. From a purely practical
standpoint, though, one simply can count conditions,
especially when clicking them on in a testing program.

This is the current record holder in the yacpdb:
https://www.yacpdb.org/#330481

FYI: My composition isn't C+, since both WinChloe and Popeye don't know atom chess.

@Dmitri This got brought up during the Zoom meeting: the problem linked there doesn't have a solution; what's the one with the most conditions in YACPDB that has a solution provided?

I have added the author's solution to that problem, as printed in the Mat Plus 28:
http://www.matplus.net/milanvel/MP28.pdf

Other than this problem with 6 conditions in effect (7 total) there are several with 5 conditions by the same author.

Popeye seems to think that b is cooked:

1...Qc5*c3[+bRc5] + 2.Rc5*c3=Q[+wQg4][+wPc5] e4*f5=S =
1...Qc5*c3[+bRc5] + 2.Rc5*c3=Q[+wQg4][+wPc5] e4*f5=S[+bRh3] =
1...Qc5*c3[+bRc5] + 2.Rc5*c3=Q[+wQg5][+wPc5] e4*f5=S =
1...Qc5*c3[+bRc5] + 2.Rc5*c3=Q[+wQg7][+wPc5] e4*f5=S =

The word "stipulation" is still a bit ambiguous, with at least two more senses: I'd appreciate some help in nailing it down.

Sense 1: objective of the problem: e.g "hs#4"
Sense 2: (WinChloe/PDB) genre or a subdivision of genre: e.g. "hs", where (according to WinChloe) the *aim* is "#" and the "number of moves" is 4
Sense 3: the whole problem text: e.g. objective + fairy conditions + constraints (e.g. "mate with wPh2") + twin machinery

Has anyone got a coherent model for problem text structure please? It won't cover 100% of cases, but should get pretty close.

Supplementary question: do any of the databases count the number of fairy conditions that a problem has?

(10) Posted by Eric Huber [Sunday, Apr 25, 2021 14:20]

QUOTE
Has anyone got a coherent model for problem text structure please? It won't cover 100% of cases, but should get pretty close.

Details about "stipulation + aim + number of moves" for Winchloe, in French and English: http://christian.poisson.free.fr/problemesis/enonce.html

The recent "Fairy Chess Classification Project" on Julia's Fairies uses a slightly different terminology.
"A stipulation consists of two components, the Goal (the aim of a problem) and the Play (the way in which that aim is to be achieved)."
The definitions can be found at this link: https://juliasfairies.com/fairy-terms/fairy-classification-project/fairy-classification-text-layout/

QUOTE
Supplementary question: do any of the databases count the number of fairy conditions that a problem has?

None, but Peter Harris is the most likely record holder, with the problem quoted by Dmitri or with another "overloaded" problem.

QUOTE
Popeye seems to think that b is cooked:
1...Qc5*c3[+bRc5] + 2.Rc5*c3=Q[+wQg4][+wPc5] e4*f5=S =
1...Qc5*c3[+bRc5] + 2.Rc5*c3=Q[+wQg4][+wPc5] e4*f5=S[+bRh3] =
1...Qc5*c3[+bRc5] + 2.Rc5*c3=Q[+wQg5][+wPc5] e4*f5=S =
1...Qc5*c3[+bRc5] + 2.Rc5*c3=Q[+wQg7][+wPc5] e4*f5=S =

Older Popeye versions, until v4.61, didn't allow the move 1...Qc5*c3[+bRc5]
The issue is that the bR is reborn on c5, where the Sentinelles pawn should appear [+bPc5]