Website founded by
Milan Velimirović
in 2006
17:48 UTC
| |
MatPlus.Net  Forum General Capture-Bound -- a new goal? |
| |
|
|
|
|
You can only view this page!
| | Page: [Previous] [Next] 1 2 3 4 5 6 7 8 | | (101) Posted by Kevin Begley [Sunday, Sep 19, 2010 14:19] |
@Juraj,
It seems to have been ammended:
The aim of CapZug has been met when the side on-move:
(1) has one or more legal capture(s); AND
(2) has no legal non-capture(s); AND
(3) is NOT in check
However, a question persists regarding stalemate.
 (= 8+5 )
h~x1?
Transmuting Kings + Rex Multiplex.
Perhaps Dan will clarify whether the following solves:
1.e3+ ...Kf4~x
This terminates, in a position where black is on the move, and
1) has at least one capturing move (namely 2.Rxg3+), and
2) has no non-capturing moves, and
3) is not in check.
But, according to the rules given, the problem might seem solved.
And this being an "aim," like checkmate, the problem does terminate upon achievement of the aim -- game over, correct?
Except, it might not be so simple...
If we look beyond the capturing move, which is not required to be played, 2.Rxg3+ forces 2...Kxh2+ (discovering check upon the bK at a8).
In turn, black has no defense (3.Rf3+ and 3.Rg3+, would result in self-check upon bK at h4, via the Transmuting Kings condition).
Thus, the bK at a8 would be checkmated.
Oh, but it is not legal to checkmate less than all Kings in the Rex Multiplex condition.
Therefore, 2...Kxh2+ constitutes an illegal move.
And, what becomes of 2.Rxg3+ (the move we never were required to play)?
After all, this too would have recursively checkmated the wK on g2 -- which, as already noted, is illegal.
So, is black stalemated at the end of the CapZug?
Yes, I think so... but, need we care, isn't the game already over?
Do recursive legalities (whether capturing, or otherwise) which may be dependent upon moves extending well beyond the truncated solution play a factor?
Furthermore, suppose we add bPa3h5, wPa2b2, and remove wSg3.
Is this a CapZug?
If so, it requires that I specify, well beyond the end of the CapZug game, why the non-capturing 2.Rg3+ is illegal.
And, having pushed beyond the barrier of CapZug's solution truncation, have I invited scrutiny as to whatever duals may lie beyond?
Or, will CapZug explicitly pardon me for such duals?
aside: it is unfortunate that capturing moves are truncated -- which might not always be entirely clear with complex fairies.
| | | (102) Posted by Dan Meinking [Sunday, Sep 19, 2010 16:50] |
"If so, it requires that I specify, well beyond the end of the CapZug game, why the non-capturing 2.Rg3+ is illegal."
If a capture-move is ILLEGAL, under whatever rules and conditions, then that move cannot be a valid element of CapZug. A solution being 'truncated' in notation does not change legality of play.
| | | (103) Posted by Kevin Begley [Sunday, Sep 19, 2010 21:40]; edited by Kevin Begley [10-09-19] |
@Dan,
OK, so last question... [the one I can never find an answer to in this thread!]
If the legality of a game, where capture bound is a new form of checkmate (the aim), depends upon the legalities of orthodox checkmates, which might occur well beyond the end of the capture bound game...
If the solution for a composition must pierce beyond the capture bound truncation, revealing what lies behind the curtain...
Should the composer suffer scrutiny for having revealed capturing duals, which are normally obscured by the capture bound truncation?
If, for some reason, I must reveal the captures following the truncation, what prevents a judge/Database from treating these as duals?
Can the author (somehow) depend, explicitly, upon capture bound's definition, to shield them from any such scrutiny? (Yes or No).
| | | (104) Posted by Dan Meinking [Sunday, Sep 19, 2010 22:05]; edited by Dan Meinking [10-09-19] |
"Can the author (somehow) depend, explicitly, upon capture bound's definition, to shield them from any such scrutiny? (Yes or No)."
Yes. We've covered this already.
 (= 2+1 )
#1
After 1.Rd8#, we do not consider 1...Ka7 2.Kxa7; 1...Kb7 2.Kxb7; 1...Kb8 2.Rxb8 to be "duals". In the same way, if by some cosmic fairy effect 1.Rd8 happened NOT to be checkmate (eg. due to illegality), then the composer must explain why -- even if that means going beyond the move itself.
The same rules apply to CapZug.
And that's my Final Answer.
| | | (105) Posted by Kevin Begley [Sunday, Sep 19, 2010 22:27]; edited by Kevin Begley [10-09-19] |
Your analogy does not apply to a problem which compels capture...
 (= 4+4 )
s#1
1.Rd8+ is considered to have a dual: 1...Bxd8#, and 1...Qxd8#
If my capture bound solution must reveal, beyond the truncation of 1.Rd8~#, that such a dual exists, will you provide no answer as to whether or not it should be deemed to be a cook?
If your definition does not shield a composer from the potential of such claims, against the very correctness of a problem, how can it be sufficient?
It has proven exceedingly difficult to obtain a proper answer to this simple question.
I am forced to assume the answer is: "no, you will not explicitly address duals in your definition."
| | | (106) Posted by Dan Meinking [Sunday, Sep 19, 2010 22:35]; edited by Dan Meinking [10-09-19] |
"Your analogy does not apply to a problem which compels capture."
Checkmate requires the ability to CAPTURE the opponent's king, although we don't examine the actual captures. Here's a checkmate equivalent of your (CapZug) example:
 (= 2+1 )
#1
After 1.Qc8#, we don't consider 1...Kb7 2.Kxb7/Qxb7 to be "duals", or some sort of defect.
| | | (107) Posted by Kevin Begley [Sunday, Sep 19, 2010 22:49]; edited by Kevin Begley [10-09-20] |
"The ABILITY to capture" is entirely different from "the COMPULSION to capture."
In selfmates (where the aim is to compel an opponent's action) multiple moves are considered duals.
[Note that selfmates do NOT truncate the moves compelled after achieving their "aim."]
What is the aim of selfmate? To have your own King captured? No, this can not be forced.
Selfmate is actually a direct-(compel mate) -- direct is the play, "compel mate" is the AIM.
[compel-mate = any legal move of the opponent would result in checkmate to the player's King.]
Does the solution stop after the aim is met? No, it does not -- it continues, with a very precise theory of what constitutes duals (read: correctness depends upon one, and only one, move having been compelled).
In checkmate aims [and stalemates, too] multiple methods to capture a King are not considered.
This is standard practice -- checkmate [and stalemate] are unique aims.
So, in your example, after 1.Qc8+, ...Kb7 is simply not considered to be a legal move.
[this may seem inconsistent (perhaps even unfair), but that's how it is -- I did not make these rules. Nevertheless, these unique aims are not without a unique remedy -- namely ideal-mate and ideal-stalemate.]
In all aims which compel an opponent's reaction (s#, hs=, semi-r+, rx, etc), that reaction is never truncated.
And, correctness always depends upon there being one -- and only one -- move compelled.
Imagine then, having a few dozen forms of your CapZug idea, all classified as aims.
Imagine having solvers puzzling over which aims allow truncation, and which require strict compliance with standard dual theory.
If this happens, it's not difficult to anticipate that a character would need to be reserved, to indicate truncation.
Short aims are much preferred, which constricts the characters available (as already discovered, in this thread).
To gobble a character for such limited, and inconsistent purpose, would not be good stewardship.
But, I have my answer... your definition deliberately refuses to sufficiently address these concerns.
| | | (108) Posted by Dan Meinking [Sunday, Sep 19, 2010 23:00]; edited by Dan Meinking [10-09-19] |
"But, I have my answer... your definition deliberately refuses to sufficiently address these concerns."
The definition is fine. You just refuse to listen.
If there are more than one moves invoking CapZug (or checkmate, or stalemate...), then THAT is considered a "dual".
Once a unique CapZug is invoked, if there are multiple captures available to the zugged side, that is OK (perhaps even desirable). Otherwise, it'd be the equivalent of saying that multiply-guarded flight squares in a checkmate or stalemate problem is a flaw.
| | | (109) Posted by Kevin Begley [Sunday, Sep 19, 2010 23:08]; edited by Kevin Begley [10-09-19] |
>You just refuse to listen.
No, I hear what you are saying... but I'm trying to explain to you that checkmate (and stalemate), which are orthodox events that end the game, are treated differently from other aims.
See my previous post, which has been edited somewhat...
I expect if you ask a selfmate expert, like Petko for example, they will say that selfmate is a direct-compel mate.
I know I read this in some book, in fact, but I can't remember where.
Maybe if you consider the aim of selfmate, you will appreciate why I view your aim [which similarly COMPELS an opponent reaction] to be inconsistent with standard practice.
If not, we can agree to disagree.
But, for the sake of completeness, why not simply state, explicitly, in the definition, that such duals are considered to be the equivalent of a single move?
Is this not a simple compromise (to provide remedy against any claim of dual -- whether by solver, judge, or database)?
| | | (110) Posted by Dan Meinking [Monday, Sep 20, 2010 02:42]; edited by Dan Meinking [10-09-20] |
Checkmate and Stalemate belong to a broader class which I will dub "Stifle": leaving one's opponent with no legal moves. So...
- checkmate = Stifle with check
- stalemate = Stifle without check
 (= 2+1 )
Stifle in 2 moves
1.Re7! (zz) 1...Kc8 2.Re8# or 1...Ka8 2.Rb7=
We can thus view Checkmate and Stalemate as "conditional" variants of the Stifle class. In the same way, CapZug is a "conditional" variant of the Self-Capture class, but it is closer in nature to the Stalemate. By contrast, a Selfmate utilizes both "conditions" (CHECK and ZUGZWANG).
Thus, comparing general Self-Captures or Selfmates (which include CHECK variants) to CapZug (which does not) is invalid. Apples to Oranges, as they say.
| | | (111) Posted by Kevin Begley [Monday, Sep 20, 2010 07:10]; edited by Kevin Begley [10-09-20] |
Stifle is already an aim (+/= in win chloe), and, your problem is anticipated.
Though, I find very few problems with this aim.
Erich Bartel
Diagramme und Figuren, 1969
 (= 2+1 )
+/=2
1.Rg5! zz
1…Kh1 2.Rg2#=
1…Kh3 2.Rh5#=
It is remarkable how often you find Erich's name when you search for unusual aims/stipulations.
| | | (112) Posted by Joost de Heer [Monday, Sep 20, 2010 08:22] |
QUOTE
s#1
1.Rd8+ is considered to have a dual: 1...Bxd8#, and 1...Qxd8#
Ah, the old 'Are multiple black mating moves in selfmates duals or variations?' discussion. It is NOT 'considered to have a dual'. It's considered by some to have a dual, and by some to have variations.
| | | (113) Posted by Kevin Begley [Monday, Sep 20, 2010 10:16]; edited by Kevin Begley [10-09-20] |
@Joost,
>"Ah, the old 'Are multiple black mating moves in selfmates duals or variations?' discussion. It is NOT 'considered to have a dual'. It's considered by some to have a dual, and by some to have variations."
I thought this matter was settled.
However, I just today stumbled upon:
P1092221 on PDB
Michel Caillaud
18th Place, 8. WCCT 2006-2008
 (= 13+9 )
s#3
1.Bd3! (>2.Rh5+ Kd4 3.Rd5+ Sxd5#)
1...Sxf7 2.e8=Q+ Sd6 3.Qh5+ Sd5#
1...Sxe6 2.e8=R+ Sxf8 3.Re5+ Sd5#
1...Sxc6 2.e8=B+ Se7 3.Ba4 S~#
This is a rare thing to find -- so, now I'm puzzled.
Is it merely an aesthetic flaw (if so, how severe is it?), or does it render the problem incorrect?
On the other extreme, I find this one -- marked C-!? -- in the Win Chloe database.
Uri Avner
Prize, Israel Ring Tourney, 1981
 (= 14+11 )
s#3*
1…Rxb3 2.Rd6+ Kf5 3.Se3+ Rxe3#
1…Rc2 2.Qc3+ bxc3 3.f8=Q+,f8=R+ (dual)
1…Sxb3 2.Se4+ Ke6 3.Sc5+ Sxc5#
1…Sc2 2.Ra6+ Ke7 3.Qxb4+ Rxb4,Sxb4,Bxb4#
1.Qc2! (> zz)
1…Rb3,Bd~ 2.Qç3+ bxc3 3.f8=Q+,f8=R+ (dual)
1…Rxc2 2.Se4+ Ke6 3.Sc5+ Rxc5#
1…Sb3 2.Ra6+ Ke7 3.Qc5+ Sxc5#
1…Sxc2 2.Rd6+ Kf5 3.Se3+ Sxe3#
I do not consider the promotion duals, on the final move, to be duals at all -- that's just what I have been lead to believe.
I am aware there are some holdouts on this matter.
I presume the duals in the mating move are what make this problem incorrect, and I am certain I have seen numerous other examples...
But there are too many C- selfmates to wade through in order to find a more clear case.
Even I wonder about the C- call here, given that the mating duals occur only in setplay.
In conclusion then, if this matter is not settled...
Perhaps even s# problems require a special character, to denote whether or not the final move should be truncated.
s# = must be a single mating move to be correct, and the solution continues unto mate.
s~# = truncate the final mating move out of the solution, and the number of mates in response are irrelevant (to the validity, thematic content, and aesthetic value of a problem).
This should be made perfectly clear -- such that everyone is plainly aware what constitutes a valid problem.
I have been operating under the advice that such duals constitute an incorrect problem.
Funny, this thread went so long here, and nobody disputed the point, until now.
| | | (114) Posted by Harry Fougiaxis [Monday, Sep 20, 2010 12:00] |
QUOTE
I do not consider the promotion duals, on the final move, to be duals at all -- that's just what I have been lead to believe.
I am aware there are some holdouts on this matter.
I presume the duals in the mating move are what make this problem incorrect, and I am certain I have seen numerous other examples...
But there are too many C- selfmates to wade through in order to find a more clear case.
Even I wonder about the C- call here, given that the mating duals occur only in setplay.
This amazing problem was (incorrectly, imho) marked as C-, because of the promotion duals on the final move. If you notice, there are also other directmates and selfmates (some among them are real masterpieces) marked as C- because of such "promotion duals" (which do not always occur on the last move). It seems that this is the standard policy followed by Christian nowadays; personally I do not agree.
The "mating duals" occurring in the set play of this problem are not considered duals by Christian (and they would not, even if they happened in the solution).
QUOTE
I have been operating under the advice that such duals constitute an incorrect problem.
Funny, this thread went so long here, and nobody disputed the point, until now.
There is a very old thread about "s# mating duals" in the forum and numerous counter-claims have already been made in various magazines (by Richter, Laue, Rehm, Vladimirov among others) answering to Petkov's article.
| | | (115) Posted by Bojan Basic [Monday, Sep 20, 2010 14:54] |
I believe that this comment is as official as it can be.
http://www.saunalahti.fi/~stniekat/pccc/8qa.htm:
QUOTE
a) Some problemists consider black dual in mating move as a serious weakness; others accept it as normal in S#. Do judges have right to lower their mark more than 0,5 points for such a subjective reason?
Reply: The so called "black dual" on the mating move SHOULD NOT automatically be considered a weakness. A "black dual" on the mating move might represent an interesting variety, like any variation in Black's play in reply to a white move. It is the judges' duty to evaluate in each particular case if (and to what extent) such a "black dual" adds to, decreases from, or has no influence on the problem's value.
| | | (116) Posted by Dan Meinking [Monday, Sep 20, 2010 16:47] |
Thanks all for some interesting reading! However, the discussion about "duals" in the CapZug context is moot, in my opinion:
 (= 3+6 )
h~x3½
After the solution...
1...Sa3! 2.Rh2 Sc2 3.Rg1 Se1 4.Bh1 Sg2~x
... the moves 5.Pxg2/5.R2xg2/5.Bxg2/5.R1xg2/5.Kxg2 are NOT duals. This is no different than post-mate or post-stalemate "continuations", which also are not duals.
Also, since Stifle is a recognized aim (making Checkmate/Stalemate "conditional" aims), CapZug should be treated as an AIM, not as a "fairy condition".
| | | (117) Posted by Kevin Begley [Tuesday, Sep 21, 2010 10:38] |
@Harry,
I still think Petko has a reasonable point... if the move is contained within the solution, it should be evaluated.
In the vast majority of compel problems -- especially selfmates -- the mating move is entirely unnecessary to the problem theme.
Therefore, if not to demonstrate uniqueness of solution, why are they not truncated (as they are in the CaptainZuggernaut thingie)?
I think Dan happens to be essentially correct about removing the duals from his solution.
I simply want consistency... across the board... tell me the rules, and don't disguise the truth... that's all I ask.
In that pursuit, I think the suggestion from my previous post (see "s~#" definition) might provide a suitable remedy, for all parties.
And, it could provide consistency for all compel-type problems (this phenomenon is not localized to a particular "invention").
At the heart of this dual-truncation, which has been pushed, there lies a MUCH bigger picture...
Obviously, PCCC should be setting such guidelines, in a manner which is clear, consistent, and intelligent...
Things can not be left to the inventors, or problem societies, or judges, or the compilers of databases, ... etc.
Somebody has to keep their eye on the bigger picture... which is easily ignored in the silly politics of problem chess.
I will respect that the aim of this thread is to focus on a particular creation (in the likeness of an aim.)
But, I would be pleased if this discussion reformulates elsewhere.
| | | (118) Posted by Kevin Begley [Wednesday, Sep 22, 2010 03:22]; edited by Kevin Begley [10-09-22] |
As I see it, Dan's house of CapZug ("aim") rests upon two cards:
1) CapZug terminates the play, like checkmate or stalemate.
2) CapZug does not in any way inhibit the rules of orthodox chess movement -- and is, therefore, not a fairy condition.
He has stated both, repeatedly, within this thread -- and upon both, the validity of this "aim" depends.
I will now demonstrate a direct contradiction (read: irrefutable proof that his "invention" is flawed as an aim).
"Smoking Gun"
 (= 2+4 )
~x1 ? Duplex. [note that either side may be on the move here.]
If, according to Dan's first claim, all moves are truncated (by termination -- akin to checkmate/stalemate) following the state of CapZug, then there can be no solution.
Both players are already in a state of CapZug -- thus no solution is possible (all moves following this state are truncated).
However, If -- according to Dan's second claim -- the rules of orthodox movement are in no way inhibited, there must be a legal CapZug for both sides: namely 1.Kxc2~x! and 1.Kxa2~x!
Both moves (1.Kxc2 and 1.Kxc1) are legally valid in orthodox chess.
Either way, both claims can not simultaneously be valid.
One of Dan's claims must fall -- either the solution must be inhibited (by CapZug's inherent fairy condition), or the solution must be truncated (as is done with the valid aims of checkmate/stalemate).
And, whichever card is removed, so goes the house upon which this "aim" rests.
I said, from the outset, that I "may yet prove 2=3 (with this aim)."
This delivers the contradiction promised.
Furthermore, consider how this "aim" would hold up, under the scrutiny of further stipulations: b) s~x1 ?, c) r~x1?, d) hs~x1? etc.
There are more plenty more contradictions where this came from...
I invite composers to have a field day with this, in the GoodZug tourney.
| | | (119) Posted by Dan Meinking [Wednesday, Sep 22, 2010 05:21]; edited by Dan Meinking [10-09-22] |
"Both players are already in a state of CapZug -- thus no solution is possible "
Incorrect. In a duplex, both sides don't play first simultaneously.
W: 1.Kxc2~x
B: 1.Kxa2~x
End of story. We could use the same logic here:
 (= 2+1 )
=1
1.Kg6!=
We don't say: "Because black is already stalemated, this has no solution."
| | | (120) Posted by Kevin Begley [Wednesday, Sep 22, 2010 05:42]; edited by Kevin Begley [10-09-22] |
I fixed my diagram -- took some time getting back online.
@Dan,
You are missing the larger point... I think deliberately.
If CapZug truncates the solution, and terminates (as in checkmate or stalemate), then the is no solution possible.
If CapZug does not truncate the solution, then the remaining moves must be shown (as they are for all other aims).
You said CapZug terminates, and truncates all further moves -- like stalemate and checkmate.
I can go back and quote you, if need be.
So, here neither player can move (both are already CapZugged).
There can be no solution, if your claim is correct -- just as there can be no further movement in =/==/#/## problems.
There would be no further movement in ~x~x (double CapZug).
So, if there are two solutions, as you suggest, then CapZug can not be an aim.
And, the truncation becomes a fairy condition.
Otherwise, answer this -- is there a r~x1? what about s~x1?
C'mon, admit it -- the gig is up.
Bring out Jammie Kennedy, Astin Kutcher, and Alan Fundt... so we can laugh along.
| |
Read more... | Page: [Previous] [Next] 1 2 3 4 5 6 7 8
MatPlus.Net Forum General Capture-Bound -- a new goal? |
|
|
|
ISC 2024
