[Retros] Solidarity chess (=SC)

[Kevin Begley]

Hi all,

Interesting variant idea...

In my experience, checkfree variants (win by capturing a royal) are quite
poor for problems.
As noted, not only does this corrupt directmates and helpmates, it also
tends to spoil retrograde analysys (legality of moves goes out the window).

It is still possible to define check as:
1) splitting the army,
2) threatening the royal unit (usually King), or
3) both.

Generally, fairy conditions do best when they do not alter the objective --
but instead only alter (generally by constraint) the set of legal moves, or
alter the outcome of legal moves (such as in the case of Circe rebirth).

The stipulation's AIM (e.g., #, =, capture, etc) best serves (as it was
intended, for problem chess) to change a solver's objective.

You could define a new aim -- (e.g., using the extended ascii symbol, let's
call it "÷" = irreparably divide the army), and then stipulate:
÷2 = white splits the black army in 2 moves.
h÷2 = help-split-in 2.
etc.
If you want to stipulate two ways to win:
(÷/#)2  = white checkmates the opposing King, or irreparably divides the
opposing army, in 2 moves.

Note, however, that the new aim would not change the rules of movement
(including with respect to check -- defined as an attack upon the enemy
royal unit --, and both checkmate and stalemate states -- either of which
would terminate play).
And, the situation gets more muddy, when you consider termination of play
by 'Dead Reckoning.'

So, not surprisingly, it tends to be a bad idea to introduce new aims.
But, more to the point, I don't think an AIM change is what you are after.
I think you might prefer a "fairy condition" which only constrains
movement.

I might be wrong... but, the alternative is wildly more complex to realize
in problem chess.

Best Regards,
Kevin





On Thu, Dec 20, 2012 at 8:48 AM, Noam Elkies <elkies at math.harvard.edu>wrote:


> Francois Labelle <flab at wismuth.com> writes:

>

> > Thanks. I managed to screw up my example (the WK should be put

> > elsewhere to prevent 1.Rxh1 gxh1=Q+),

>

> I was wondering about that...

>

> > but you get the idea: a mate in n can become a draw (just remove the

> > WR from my diagram),

>

> or keep it (as a superfluous piece, yes).

>

> > or it can become a win in n+1 if we feel like composing a bit.

>

> "n+1" is the normal situation; you probably meant either "n+2",

> for which

>

>   http://www.janko.at/Retros/d.php?ff=5brk/4p1pb/4P1p1/4N1K1/8/8/8/8

>

> is one simple setup, or "more than n+1", which can also exceed n+2

> if we feel like composing a bit more, e.g.

>

>   http://www.janko.at/Retros/d.php?ff=8/P1p5/2P5/5N2/6p1/8/4p1pp/4Kbrk

>

> Or we could amend the stalemate law...

>

> NDE

> _______________________________________________

> Retros mailing list

> Retros at janko.at

> http://www.pairlist.net/mailman/listinfo/retros

>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.pairlist.net/pipermail/retros/attachments/20121220/7e4fff10/attachment.htm>