MatPlus.Net

 Website founded by
Milan Velimirović
in 2006

19:00 UTC
ISC 2024
 
  Forum*
 
 
 
 

Username:

Password:

Remember me

 
Forgot your
password?
Click here!
SIGN IN
to create your account if you don't already have one.
CHESS
SOLVING

Tournaments
Rating lists
1-Jan-2024

B P C F





 
 
MatPlus.Net Forum Fairies New stipulation: kiss in 2 moves
 
You can only view this page!
(1) Posted by Linden Lyons [Sunday, Mar 2, 2014 08:18]

New stipulation: kiss in 2 moves


I enjoyed reading about Vassily Ivanchuk's 2009 chess variant, "Kiss the Queen": http://en.chessbase.com/post/kiing-the-queen-ivanchuk-s-new-che-game/1

It isn't necessary to read the whole article, but I think that 'kissing' could be a new stipulation, e.g. 'white to play and kiss the bQ (or another black piece) in two moves' means that after two moves white must have at least one of its pieces adjacent to the bQ (or other black piece) which itself has no adjacent black pieces. I propose that the wK can't kiss by walking into check (unlike Ivanchuk's variant).

Here's an example to demonstrate some of the possibilities (not computer tested):

Linden Lyons
(example)
(= 4+8 )

White kisses bRc5 in 2 moves

My intended unpin key is 1 Qd2! (threat 2 Qxd5). Black can defend by having the rook run away: 1 ... Rc3 2 Qxc2; 1 ... Rb5 2 Qxa5; 1 ... Rc7 2 Kb8; 1 ... Rc8+ 2 Kb7. Black can also try to close off a white line: 1 ... Rd3 2 Bxd5; 1 ... Bd3 2 Rxd5; 1 ... d4 2 Qxd4+.

Note that 1 ... Rc3/Rc5 doesn't defeat the threat.

Incidentally, 1 Qd3? (threat 2 Qxd5) meets 1 ... Rb5!, 1 Qe4+? (threat 2 Qxd5) encounters 1 ... dxe4!, while 1 Qxf3? (threat 2 Qxd5) fails because 1 Qxf3? is mate. The game ends and white loses the opportunity to kiss the bRc5!

I suspect that problems of greater complexity are possible. I can also imagine 'self-kisses', 'reflex-kisses', and 'help-kisses'. Anyway, what do others think of this idea?
 
(Read Only)pid=11730
(2) Posted by Sarah Hornecker [Sunday, Mar 2, 2014 08:45]; edited by Sarah Hornecker [14-03-02]

EDIT: Nevermind, I missed the part where there can be no other adjacent piece. I can imagine the stipulation will allow it easily to show complex themes (AUW, etc.)
 
 
(Read Only)pid=11732
(3) Posted by Kevin Begley [Sunday, Mar 2, 2014 11:26]; edited by Kevin Begley [14-03-02]

@Linden,

Nice example problem, and an interesting idea...
This is perhaps better considered an aim (whereas a stipulation consists of many things: play [help-, direct-], aim [state- or move-], target [unit, square, etc], number of moves, etc)...

But, let's avoid the semantic issues (which consistently get us into trouble here), and instead focus specifically on your intended rules for your invention.

Here's my first question (and this is absolutely key, for new inventions): this idea seems to be defined based upon movement, and not based upon a state -- is that correct?

I wish every inventor would start with this important question... here's what it means:

Suppose I start from a position where two units are "kissing" -- can I stipulate a help-kiss in n moves (or do you consider the game over)?

If the game is over, you are defining a state -- such as checkmate, stalemate -- and this can be dangerous.
State-aims should generally only alter the form of check/checkmate, otherwise, the legality issues can lead to surprising inconsistencies (especially when you involve fairy elements).
The alternative is to define this as a "movement-aim" (such as capture, en passant, castling, etc), in which case the act of white making this move (in say "kiss-2"), satisfies the stipulation.

Note: the primary difference here concerns whether black can move his targeted unit into a kiss position.
For example, suppose we have a help-kiss-in 1 move... would you want to prevent the black unit from moving into a kiss position (should this be illegal)?
In movement-aims, black is not prohibited from moving into positions where white may exercise the movement.

If Yes, then I would encourage you make this a proper state-aim (read: redefine the nature of check -- as is done in the anti-Kings condition -- such that any kiss of a royal unit would constitute check).
Note, this has the added benefit of removing any targeting data, in your stipulation (the targets would be identified by the royal units).
However, note that this would completely erase any normal checks (so, it seems this is not your intent).

If No, then call it a movement-aim (which targets a specific unit).
However, then it would be legal for black to move into positions where a kiss occurs (it is not a state, therefore, legal for black to move into this).
Note: this does not alter the rules of check/checkmate, but may lead to displeasing situations where white moves to maintain a kiss, which black moved into (as in zugzwang problems).

There are hybrid options for defining such an aim, of course, but they tend to complicate programming efforts, and they also pose unexpectedly complex legality issues.
 
   
(Read Only)pid=11734
(4) Posted by Kevin Begley [Sunday, Mar 2, 2014 12:15]; edited by Kevin Begley [14-03-02]

Simple questions:
(= 1+1 )
help-kiss-1? (or is black stalemated?).

(= 1+1 )
help-kiss-1, 2 solutions? (or illegal position -- both players already lost?).

(= 1+1 )
help-kiss-1, 1 or 2 solutions? (suppose we add a white unit, and make this black's target?)
 
   
(Read Only)pid=11736
(5) Posted by Linden Lyons [Sunday, Mar 2, 2014 14:50]; edited by Linden Lyons [14-03-02]

Hi Kevin,

To the best of my ability I’ll address your questions.

I consider this idea to be defined as a ‘movement aim’. In other words, the game isn’t over once the aim is achieved. Targeting data will be necessary in the stipulation, e.g. kiss-(bRc5)-2. I regard it as perfectly legal for black to move the targeted piece into a kiss position, but the composer may wish to avoid this from arising (as in my example). However, I don’t know if there are good zugzwang possibilities or if, as you suggest, they’ll be displeasing. I’ll give this some thought.

In relation to your three diagrams:

(1) Help-kiss-1
(2) Help-kiss-1 (2 solutions)
(3) Help-kiss-1 (2 solutions)

In a nutshell, all the normal rules of chess apply. Both kings would be present in the position. Only the aim is different, i.e. not checkmate or stalemate, but kiss of a target piece.

I welcome any further thoughts.
 
 
(Read Only)pid=11737
(6) Posted by Kevin Begley [Sunday, Mar 2, 2014 15:49]; edited by Kevin Begley [14-03-02]

Wise choice...
Best to avoid the hybrid options... I had prepared a brief synopsis explaining why, but it seems you have smartly managed to avoid the need for this comment.

Therefore, I will expunge all but the one quote which best explains the most pertinent danger:

"Ph'nglui mglw'nafh Cthulhu R'lyeh wgah'nagl fhtagn" -- H.P. Lovecraft.
 
   
(Read Only)pid=11739
(7) Posted by Kevin Begley [Sunday, Mar 2, 2014 16:03]; edited by Kevin Begley [14-03-02]

One thing you may want to consider...

You may need to provide multiple target units (for white and black).
This is especially true, if you desire a truly reflex-form.

I'm not certain, but this might be an issue for some solving tools.
It is wise to consult with somebody on the popeye team, if you have the opportunity.

Oh, and final thought: you might want to consider finding a character representation (like "x" = capture).
I don't know if there are good characters left, but who knows -- maybe some are moving to Unicode.
 
   
(Read Only)pid=11740
(8) Posted by Frank Richter [Sunday, Mar 2, 2014 16:21]

"...which itself has no adjacent black pieces."
Wouldn't it be a defence to move a piece to an adjacent field of the black target piece?
 
   
(Read Only)pid=11741
(9) Posted by Kevin Begley [Sunday, Mar 2, 2014 16:35]; edited by Kevin Begley [14-03-02]

@Frank,

No, I don't think that's a problem -- remember, this is only a move-aim.
Much like a capture (or a check), there is no defense after the capture is made (or check is given).
So long as there's no adjacent defender, at the instant of move completion.

@Linden,
Here's maybe a better idea for targeting (because you might even wish to allow multiple targets!).

Popeye would probably be smart to implement the targets by way to ATTRIBUTES (on a unit).
And, editors could similarly represent targets in the diagram (e.g., using highlighted, or upside-down units).
This makes your stipulation easier (eliminates all target info), and gives you wider targeting options.
(e.g., you may stipulate a kiss against any of multiple targets -- much like a capture problem would allow)!

I don't foresee any difficulties in full-reflex-kiss, either...

(= 1+1 )
reflex-kiss-1? (suppose both units shown are targets -- and non-royal, or place Kings someplace).

Does 1.h4 force ...Sg5 (kiss)?
I presume so, because 1...Sg5 should be possible (the kiss is not a state, thus the move is considered legal).
And, since the kiss is a move, white has not kissed the black Knight (but black Knight was compelled to kiss the white Pawn).

So, full-reflex seems to work.
 
 
(Read Only)pid=11742
(10) Posted by seetharaman kalyan [Tuesday, Mar 4, 2014 09:17]

Nice idea Lyndon.

@Frank
<<<"...which itself has no adjacent black pieces."
Wouldn't it be a defence to move a piece to an adjacent field of the black target piece?>>>

I see your point. Yes. Black would be able to defend by moving a piece adjacent to the threatened piece.... BEFORE white executes the kissing move.
 
 
(Read Only)pid=11754
(11) Posted by Thomas Maeder [Wednesday, Mar 5, 2014 15:17]

So first we had mate, then Face to Face (and Cheek to Cheek), and now Kiss - and still there are only very few problemists around!

 QUOTE 
My intended unpin key is 1 Qd2! (threat 2 Qxd5).

Popeye starting at 4.67 will support kiss as a new goal. Your direct twomover is found to be correct, including your analysis of the tries.

 QUOTE 
In relation to your three diagrams:

(1) Help-kiss-1
(2) Help-kiss-1 (2 solutions)
(3) Help-kiss-1 (2 solutions)

Total agreement here as well.

No support for the two target idea, which I haven't understood.

One can simply specify
stipulation kc52
meaning "kiss in 2 moves of the piece starting at c5", and of course self-kiss, reflex-kiss, help-kiss, etc. etc..
 
 
(Read Only)pid=11766
(12) Posted by Linden Lyons [Wednesday, Mar 5, 2014 16:08]

@Thomas

That's fantastic that Popeye will support kiss as a goal. Thank you!

I'm glad that my kiss-2 is correct!

I think Kevin's idea for multiple targets is that only one of these targets needs to be kissed in any variation, e.g. if there are two targets (A and B), the threat might kiss A, some variations will kiss A, while the remaining variations will kiss B. I don't have an example at this stage, but it's worth exploring.

@Kevin

Is my interpretation of your multiple-target idea correct?

@Seetharaman

I'm glad you like this idea.

@Frank

Black could indeed defend by moving a piece next to the target piece. In this case I think white's only reply could be to capture the former.
 
   
(Read Only)pid=11767
(13) Posted by Geir Sune Tallaksen Østmoe [Wednesday, Mar 5, 2014 19:20]

 QUOTE 
Black could indeed defend by moving a piece next to the target piece. In this case I think white's only reply could be to capture the former.


Or in some cases, perhaps a zugzwang would force Black to move the pieces apart again?
 
   
(Read Only)pid=11768
(14) Posted by Kevin Begley [Wednesday, Mar 5, 2014 22:11]; edited by Kevin Begley [14-03-05]

@Linden,

Yes, that's a good characterization behind the idea to allow for multiple targets.
Another is the reflex-kiss idea (which you had already expressed), which requires one target for each of the two armies.

Your invention inherently offers these extended targets (beyond one per side), which can dramatically extend complexity (for those interested).
It is rare that an invention offers rich opportunities for exploration of settings both elegant, and complex.
This seems a likely accomplishment of your invention; and, because it is so clear, I'm glad to see popeye offer support.

My suggestion was really only an improved methodology for presenting this information (placing targeting information outside of the stipulation, and into the diagram).
This suggestion should make life much easier for editors, and hopefully this extends opportunities for composers (and solving tools) to elegantly express extended complexity.

note: it seems logical for programmers to shift ALL targeting information to the diagram (since there's little value supporting two methodologies).
 
   
(Read Only)pid=11770
(15) Posted by Kevin Begley [Wednesday, Mar 5, 2014 23:09]; edited by Kevin Begley [14-03-06]

In fact, given multiple targets, you could also extend any stipulation involving a Kiss-aim, such that the problem objective is defined to achieve two (or more) Kisses, simultaneously.

e.g.,
(= 1+2 )
help-kiss-1 (2)
= Targets
b) h7->a1

a) 1...Se3 (1...Sd4?) 2.Bd3
b) 1...Sd5 (1...Se4?) 2.Be5

So, once again, this invention yields remarkable versatility; and, logical alternatives seem to naturally progress, from its clear, fundamental definition.
You could give n number of targets, and stipulate help-kiss-1 (m) {where m <= n}.
I'll leave it as an exercise for the reader to determine how large m can be, in the orthodox case.

Aside: just imagine if our classification system were arranged (much like the universal symbolic language of mathematics and logic) to help guide inventors toward a similar unfolding of possibilities... Maybe this can not be more than a wildly hopeful dream; but, we'll never be certain, unless we actively pursue the collective good.

ps:
I'm not certain how best to express the objective of X simultaneous kisses, but not more than Y (where Y >= X).
Maybe it should be expressed as an inclusive range ( e.g., help-kiss-1 (2,3) ) ??
Similarly, maybe a summation symbol could be used to stipulate goals based upon an accumulation of Kiss-aims.
 
 
(Read Only)pid=11771

No more posts


MatPlus.Net Forum Fairies New stipulation: kiss in 2 moves