MatPlus.Net

 Website founded by
Milan Velimirović
in 2006

9:41 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 General What is Fairy Chess?
 
You can only view this page!
Page: [Previous] [Next] 1 2 3
(21) Posted by seetharaman kalyan [Tuesday, May 8, 2012 03:13]

I think T.R.Dawson listed out the features of Orthodox chess thus: The 8x8 board, the six types of pieces, alternate moves of white/black, castling and enpassant capture, aim to mate the opposing King. Any change of any of these features is Fairy Chess. I believe this is still the generally agreed/understood definition of Fairy chess. Our Old FIDE Albums had correctly classified Helpmate and Selfmate also as Fairy Chess !

Having said that, I have no objection to a new classification based on other criterion: Defending play by black, Help-play by black, Mixed play etc.
 
   
(Read Only)pid=8470
(22) Posted by Kevin Begley [Tuesday, May 8, 2012 04:35]; edited by Kevin Begley [12-05-08]

Good point -- and we should certainly be mindful of T.R.Dawson's definition (along with other great predecessors).
But, this definition doesn't provide enough meaningful answers, today.

For example:
1) Is r#n is a stipulation, or is it actually a s#n + a fairy constraint (player must #1, when it is available).
2) Is it considered a dual, in a s#n, when the black player has multiple mating moves? (see related thread).
3) What fundamentally constitutes an aim/goal/stipulation versus a fairy condition?
4) Why are h#n and s#n considered orthodox, but not hs#n? Why not =n?
5) What are the default rules for pawns on the 1st rank?
etc.

The point is, we could benefit from a greater resolution of our divisions.
Already, many journals are adopting new genre divisions -- the genre lines are blurring...

I have no intent to impose my preferences, or dictate divisions to journals/TTs, or to alter any awards/titles processes -- that's all completely unrealistic (and far afield of my intent)!
If this discussion goes well, I'd like to present the delegates with a full set of the fundamental options -- each of which they can consider independently, for incorporation into our Codex (and I'll accept whatever their vote).

Our divisions are not completely meaningless (if they had been, TRD would not have helped establish them).
They serve an important function -- even if only to minimize redundancy of anticipation searches, it would provide a valuable service (but, we all know, the benefits go far beyond this)!

If our classification is not meaningful, why do the dividing terms (Orthodox/Fairy/Fairy Condition) appear in our Codex? If it is meaningful, why does the Codex leave these terms undefined?

I can think of an endless number of potential benefits... many of which can be easily illustrated (from unfortunate redundancies which could have been avoided).
But, I don't understand what harm is done by discussing how we *might* provide an improved framework, which fundamentally answers these questions for us (in a universal, standardized, logical fashion)?

The sky doesn't fall from such an improvement.
There is no reason to believe such improvement is beyond our abilities (or logically impossible).
Why not provide newcomers a solid foundation, from which they might have better opportunities to express themselves in this art form?
Whom would our effort harm?

Even if it proves impossible, we will have learned something (and, passed it down to the next generation).
 
   
(Read Only)pid=8472
(23) Posted by seetharaman kalyan [Tuesday, May 8, 2012 05:47]

Actually from TRD's point of view the answers are straighforward:

1. Orthodox is when both sides aim to mate the opposing King, and defend theirs. So obviously any reflex condition R# or R=, or reflex check/capture etc.. are all Fairy.

2.What is a Dual is not part of this discussion.

3. As in point 1, all goals other than mating opposing king are Fairy :)

4. H#, S#, =n are all fairy. They have only become more widely known/accepted.
 
   
(Read Only)pid=8473
(24) Posted by Nikola Predrag [Tuesday, May 8, 2012 09:51]

So an ending where white has enough material to mate and black has not, white plays orthodox and black fairy chess. Such study is orthodox or fairy?
 
   
(Read Only)pid=8477
(25) Posted by seetharaman kalyan [Tuesday, May 8, 2012 10:44]

Not clear. Material equality or inequality has nothing to do in this.
 
   
(Read Only)pid=8478
(26) Posted by Hauke Reddmann [Tuesday, May 8, 2012 14:10]

As a Science Fiction fan, I well know the "standard" definition of SF
(which also mocks the desperate attempts to clearly define the genre):
SF is what is labelled and sold as such! :-)

This is essentially a variant of item 4), "bug off", and I think it's
essentially the policy of many problem chess mags. (And in fact,
some fairys can become ortho just by habit. I remember a tournament
where only N and G were allowed as fairy pieces. From an axiomatic
standpoint this is utter balderdash.)

Hauke
 
   
(Read Only)pid=8483
(27) Posted by Nikola Predrag [Tuesday, May 8, 2012 19:17]

-...3. As in point 1, all goals other than mating opposing king are Fairy :)...
Any Rex solus is either a fairy problem or the lonely King is quite an optimist having a goal to mate the opposing King.
 
 
(Read Only)pid=8493
(28) Posted by Kevin Begley [Tuesday, May 8, 2012 19:19]; edited by Kevin Begley [12-05-08]

haha, Hauke, I love that SF def.

As to TRD's division... it has few adherents, today (if any).
Frankly, even in it's day, it really did not provide the most logical division.
I suppose strict chess players did enjoy this.

Today, Orthodox = h#n, s#n, even r#n (which makes little sense).
Tommorow, you may see a variety of additions: hs#n, =n, series-movers, grasshoppers, nightriders... who knows what?
Whatever you can sell the public, I suppose (hey, somebody sold them on reflexmates!).

My goal is not to reclassify Science Fiction into Drama/Comedy/etc...
The last thing I want is to argue about genre classification.

I'd like to see universal, logical divisions, and formalized stipulations, because I believe it would provide a better foundation for newcomers to appreciate a variety of chess problems; and, it would offer composers more opportunities for self-expression.

Orthodox/Fairy is analogous to separating music according to the preferences of two radio broadcasters.
There is no logically definable division anymore (if ever there was).
One station plays r#n, the other plays =n... nobody can explain why.

In fact, it's worse -- these artificial boundaries have become an impediment to the evolving progress of problem chess, because titles and awards have become so intertwined in these synthetic barriers.
It serves no purpose today, other than to pigeonhole art (and artists).

If you can't properly define "flotation device", you just might find yourself clinging to the anchors of a rusted tradition.
I don't come here looking for tourists -- we don't need another quartet playing, "Nearer My God To Thee."
I come here looking for a responsible crew, to whom I might scream, "ICEBERG!"
If I've come to the wrong place, please, return to your shuffleboard. :)
 
   
(Read Only)pid=8494
(29) Posted by Kevin Begley [Tuesday, May 8, 2012 19:39]; edited by Kevin Begley [12-05-08]

@Nikola,

> (quoting) "...all goals other than mating opposing king are Fairy..."

The present Codex shows that you can no longer apply TRD's classification system (rex solus or otherwise).
Dead Reckoning (a firm, codified Rule) effectively terminates a s#n problem, prior to achieving mate.
Therefore, the goal MUST BE something other than mate (in fact: the goal must be to achieve a s#0.5 position!).

S#n is NOT considered a Fairy Form -- virtually all journals and TTs (including the FIDE Album, and the WCCT) consider this stipulation to be an Orthodox (read: non-Fairy) Form.

There is (presently) no valid logical explanation (or definition) for the Orthodox/Fairy division.
Of all the problem experts in this forum, none have managed to provide anything close.
Nor is this the first thread to request such a definition.

It is an artificial construct, used to synthesize a specific preference in publications, awards, and titles.
There are only variants (the most recent FIDE Chess rulebook constitutes one single variant -- not a genre).
In fact, FIDE Chess has morphed into something few today would still recognize as Orthodox (mainly because the impact of Dead Reckoning is little understood).
 
   
(Read Only)pid=8495
(30) Posted by Nikola Predrag [Tuesday, May 8, 2012 20:53]

Orthodox chess is anything that might legally happen in a chess game, no matter how absurd it may look. Otherwise we should consider most (if not all) games ever played as fairy chess games. Show me a game won without help-play of the opponent. I've seen nice helpmates in chess games and nobody said that these were fairy games. Selfmates are rare but, as the name suggests, a reflexmate is a common thing in mutual Zeitnot.
 
   
(Read Only)pid=8497
(31) Posted by Kevin Begley [Wednesday, May 9, 2012 03:56]; edited by Kevin Begley [12-05-09]

>"Orthodox chess is anything that might legally happen in a chess game, no matter how absurd it may look."

Maximummer, Minimmummer, Bichrome, Duel, Ultraschachzwang, and Follow-My-Leader are but a few examples of what can legally happen in a chess game.
The shortest game I ever played was Monochrome.

These things probably happen (in interludes) more commonly than the strategies of s#n & r#n combined -- are they "Orthodox"?

In time scrambles, I've seen queens hurdle other units, and I've seen Knights leap like Camels -- but, it did not happen "legally" (though it sometimes goes unnoticed), so I guess Leapers, Chinese units & Grasshoppers are not "Orthodox"?

Still, I'm sure I've seen legal interludes of Isardam (perhaps even Madrasi paralysis) in Orthodox chess games.
Absurd things do happen... so, it's quite surprising what things may be considered Orthodox (by your definition).

Furthermore, there have been "orthodox" chess games where pawns promoted to pawns.
In fact, Steinitz famously argued that this rule should be adopted as Orthodox.
Time Krabbe showed, in an orthodox #n problem, that ultra-long castling was possible, according to Orthodox rules.
The orthodox rules have since evolved; and, they continue to evolve.

There are a variety of ways to discredit your definition of "Orthodox."
It's like the definition of "heat" -- try to define it.
Any child knows what it means (it's something we all feel); but, an adequate definition really requires some understanding of the statistics of small particle movement.

The difference is, Orthodox/Fairy are artificial -- there is no scientifically valid explanation for why r#n is orthodox, but =n is fairy.
[For the record: I once watched a strong, untitled player execute a =n upon a GM (who spurred all draw offers), in a major tournament. Surely, =n is much nearer to orthodox chess than any h#n/s#n/r#n.]
Same goes for hs#n -- it's components (h#n and s#n) are orthodox, but it is fairy.

You'll never construct a viable definition -- other than to artificially declare, element by element.

However, suppose you instead would consider the original FIDE Chess Rulebook, and each alteration which followed, to be descriptions of a unique set of chess variants (a kind of pseudo-set of temporally "orthodox" variants).
Furthermore, suppose you were to formalize aims & stipulations, such that they ONLY define the problem components (a primary objective, an ultimate aim, some deadline, and a motivation describing who helps/opposes your effort to achieve the primary objective).
That is, suppose no stipulation were allowed to redefine the rules (e.g., series-movers would not be allowed to redefine the rules of move alternation -- instead, changes in move alternation would be accomplished by adding a fairy condition, and the series-stipulation would be replaced with a valid stipulation).
Given such a logical implementation, you could define SOME particular FIDE rulebook (perhaps the present one) to be the base case (a complete, semi-permanent set of all default rules -- even for ortho-illegal possibilities).
[note: it is not absolutely essential to define a base variant, and explicitly state its default rules, but it certainly simplifies matters.]
Then, by filtering out all problems which have no fairy conditions (or fairy pieces, or fairy boards), you would arrive at some base set of "non-heterodox" problems -- which might still include all valid stipulations.

But, note that you'll not find r#n in a "non-heterdox" list -- because this stipulation concealed a fairy condition ("either player is obliged to play #1, if possible"), and after this condition was explicitly listed -- and we replaced r#n with a valid stipulation (in this case, to s#n) -- these problems would have been filtered out (for having a fairy condition).
It is possible that semi-r#n would appear as non-heterodox, but it depends whether you formalize stipulations using a recursive process of goal evaluation -- but, that's a long story (and somewhat tangential).

Furthermore, you would also have filtered out all other variants in the set described by some FIDE Rulebook (what somebody once called orthodox, has now become heterodox).
With each rulebook alteration (or if problemists decide upon alternative rules for the base variant), your orthodox compositions would have been transformed into fairies -- and this does happen (even the adoption of Dead Reckoning rules -- which seems insignificant -- seems to have had a considerable impact on some studies).

But, to continue...
Once you have a complete set of non-heterodox problems, you might apply sub-filters (e.g., by stipulations/aims/number of moves, etc), if you are only interested in a particular sub-set (e.g., #2, #3, #n, h#2, h#n, s#n, etc).
Thus, the majority of our traditional divisions might remain intact (save only r#n, and any studies impacted by DR).
However, an additional section would be required -- to cover all remaining non-heterodox problems (which should not be grouped with heterodox).

Furthermore, I would suggest that however you sub-divide all non-heterodox problems, so too should you sub-divide all heterodox problems.
The key to all mysteries, according to the Emerald Tablet: as above, so below.

The alternative would be to completely ignore the artificial Orthodox/Fairy boundary, and simply group all problems according by fundamental problem components -- such as: valid stipulations/valid aims/number of moves.
Then, each variant simply must define itself.
This is simpler, but the impact upon awards/titles would be considerable (and such a grouping would likely be unwelcome for those who strictly favor the pseudo-set of orthodox variants).

Whatever path you prefer, it is essential that we provide a universal, formal stipulation mechanism, which prevents the concealment of rule alterations (fairy conditions).
It should also provide composers a way to describe what they intend as their ultimate aim.
This would effectively provide a dual-suppression mechanism, which is necessary for problems intended to exhibit an alternative dual-theory, such as in s#n problems (and its derivatives), CapZug problems, and many other types (some of which are yet to be invented).
 
   
(Read Only)pid=8500
(32) Posted by Nikola Predrag [Wednesday, May 9, 2012 09:28]

Kevin, you try in each post to explain everything, to anticipate all questions and answer them in advance. I admire your energy and effort but it may be exhausting for the readers. We must find and establish some starting fundamental point. So far it looks as we are lost in the vast universe of rationalizing, having no central point of 'gravity'. I only try to awake the power of Intuition, millions around the globe know intuitively what the chess is. Chess composition is a specially organized way of presenting some features/possibilities of chess.
Madrasi allows illegal positions and moves, it can not be orthodox. Anyway, I did not offer a definition of orthodox chess, it is only a question about the fundaments.
 
   
(Read Only)pid=8506
(33) Posted by Kevin Begley [Wednesday, May 9, 2012 16:02]; edited by Kevin Begley [12-05-09]

@Nikola,

>"...it may be exhausting for the readers. We must find and establish some starting fundamental point. So far it looks as we are lost in the vast universe of rationalizing, having no central point of 'gravity'. "

OK, this is certainly a fair comment... a solid foundation is required, upon which we might build.
I am well aware that my wordy posts are better suited for a book... I'm sorry if it puts people off.
But, there is always much to say.

If I'm writing a book, chapter 1 (Weightlessness) established that there's no good definition for ortho/fairies.
So, let's come to the real point.... chapter 2 (Induced Gravity) aims to ground us with logical axioms.

I believe an axiomatic description of problems can provide all that we need.
Start with the task itself -- for any chess problem, what are the essential components?

At first, it seems there is no universal set of elements -- some problems provide no board, some show no pieces in the diagram, and some offer no stipulation.
But, many of these elements are actually implied -- if you're provided no stipulation, the obvious curiosities dictate an implied stipulation: what happened prior? what is happening now? who has the move? what will happen next?

So, what are the fundamental elements necessary to describe any formal chess problem?
1) Some primary goal must be provided (or implied).
2) Any deadline for achieving this goal must be provided (or implied).
3) The motivation (play) of any virtual opponent, with respect to the primary goal (help/oppose it), must be provided (or implied).
4) The form of a complete solution must be provided (or implied).

5) Some board (or set of possible boards) must be provided (or implied).
6) Some pieces (or set of possible pieces) must be provided (or implied), in some relation to the board.
7) The rules of play (or some set of possible rules) must be provided (or implied).

Beyond that, it may be important to establish:
8) A means to determine the soundness of the problem must be provided (or implied).
9) What constitutes a constructional defect?
10) What are the established problem themes (and what constitutes a legitimate realization of these forms)?
11) How do we resolve anticipations?
etc ... however, these are not considered essential elements to convey the problem's task (as are the first 8).


The first 3 elements are generally provided by a formal stipulation.
There is much to be said for the 4th element.

First, it should also be included in the formal stipulation.
You can not correctly identify a dual, without first understanding threat-notation.
Threat-notation has generally been presumed the default of all direct-problems (otherwise, a basic game-notation is required).
However, in some conditions (where a player may idle), threat-notation breaks down.
Unless we want to stipulate the notation form of a solution (threat- or game-), threat-notation must account for idle opponents.

Second, the 4th element requires some indication of what constitutes completeness.

In general, we had always assumed that a problem is not complete until it reaches some final, indivisible aim (#, =, x, ep, 00, etc) -- what I call the ultimate objective.
That need not always be true -- there are legitimate cases where a problem's ultimate objective may be to realize a stipulation (which is not intended to be recursively resolved).
For example, in s#n problems (and derivatives), the author may have no intent to include black's mating moves.
Further, by eliminating the mating moves from the solution, the "duals" (which some perceive from black having multiple mating moves), are effectively suppressed (if it's not found in the solution, it's no dual).
This dual-suppression mechanism is also useful for some self-type fairy inventions.
For example, CapZug is essentially a sxn (self-capture n) stipulation, (plus some fairy condition) -- except that these problems are intended to end prior to the actual capture.
It's axiomatically improper to obscure this conglomerate invention into a faux-stipulation (and again, neglect the redundancy which accrues); however, if the inventor is provided no formal mechanism for dual-suppression, we've left him no good alternative (our fault, not his).
Finally, note that this dual-suppression is not a fairy condition (per se) -- it does not alter (or constrain) any rules of movement.
It is the composer's definition of the task, and it should be formally indicated in the stipulation.

Elements 5 thru 7 are generally provided by the diagram (and any conditions listed).
Except, unfortunately, we have allowed many game rules to creep into our formal stipulations (and aims).
I don't want to dwell on these, or focus on any particular fairy conditions that have caused redundancies.
Let's just create a good set of axioms; and, tangible improvements will accrue.

It is worth noting that element 7 (the rules), are generally conveyed by way of a base variant (orthodox chess), plus a list of any modifying conditions.
From an axiomatic standpoint, we should establish a complete description of the base variant.
Further, we should strive for a fundamental set of modifying conditions (that is, ideally, each should strive to make only a single, non-redundant rule change).
Obviously, there are variants which require several modifiers (and, result in a multitude of special cases).
Therefore, it may be worth distinguishing the fundamental conditions from complex variants.
Furthermore, when a modifying condition offers a variety of alternative interpretations, it should provide a new base variant for its entire sub-variant family (e.g., Circe Family, Madrasi Family, etc).

From this list of axioms (which I concede may not be complete), I we can begin to define some fundamental terms: aims, goals, stipulation, fairy conditions...
We can work our way up to some set of options to define Orthodox, and Fairy (strictly according to an axiomatic logic -- rather than by a nonsensical favoritism).

That's my only goal... and, the progression I've outlined seems the most intelligent path to it.
I don't have anything against the stipulations which may require the addition of a fairy condition.
You'll find many reflexmates in my short list of favorite problems; and, they'll still be there, even after we make explicit r#'s underlying fairy condition.
I'm only interested in consistency.

I hope, if nothing else, this inspires others to care more about it.
I believe we would benefit significantly, if we respect our inherited classification system, and its potential.
 
 
(Read Only)pid=8508
(34) Posted by Neal Turner [Wednesday, May 9, 2012 16:05]

The Orthodox/Fairy boundary is not artificial.
The Game is the baseline, and in presenting an orthodox problem there's always the implication that what you're showing is interesting because it couldn't happen in the Game.
But what of Fairy problems? Most fairy chess forms don't have a game to compare with, so how can we appreciate them?
The answer is that the baseline for fairy problems is the orthodox problem!
A fairy problem is interesting because it presents something that couldn't happen in an orthodox problem (in practice of course most fairy problems don't do this and are therefore not very interesting).
It is for this reason that the Orthodox/Fairy boundary exists and should be retained.

There's no point in trying to rigidly define what is meant by 'Orthodox' or 'Fairy' because, as has been pointed out above, notions are changing all the time.
And there's always going to be a grey area, but it's not something to lose sleep over as most problems reside outside it and it's very clear whether they're Orthodox or Fairy.
 
   
(Read Only)pid=8509
(35) Posted by Kevin Begley [Wednesday, May 9, 2012 16:36]; edited by Kevin Begley [12-05-09]

Neal, if you can't clearly define what is Orthodox, how can you have a baseline?
And, if you can't clearly define what is Fairy, how can you know if a particular problem should be demonstrating something beyond the Orthodox baseline?

Before we all settle upon a path of despair, let me just say that my primary goal in this thread was to establish that we don't presently have good definitions for these terms; Having now achieved that, I aim to establish that a clear, logical, and axiomatic definition -- for both the Orthodox baseline, and the Fairy beyond -- is easily in our grasp.

To do this, we have only to reestablish the fundamentals (the essential elements of any problem).
Define aim, goal, stipulation, fairy condition, etc.... build from a simple, solid foundation!
Once we establish a full set of rule options, we let a vote of delegates decide a complete Orthodox baseline.
And, we would have defined ortho/fairies axiomatically -- rather than incoherently, inconsistently, and based largely upon select favoritism.
 
   
(Read Only)pid=8510
(36) Posted by Per Olin [Wednesday, May 9, 2012 18:11]

Is there a need for strict classification?

In post 18 is asked the question, is a clear-cut division needed, is there a need to have an exact definition of orthodox/fairy chess? If problem chess is an art, then no classification system is needed. But, as this has been turned into a sport, then there must be clear rules. There are World Champions in Fairy Chess composing; there might come the day when somebody in this forum misses becoming World Champion only due to lousy wordings and interpretations of the Codex for Chess Composition. In no sport is there a World Championship with such incomplete and vague rules as in problem chess.

The Dawson definition / stalemates

In post 21 is a definition attributed to Dawson of orthodox/fairy chess. In this is stated that fairy chess is anything, where, among other things, the aim is not to mate the opposing king. This then categorizes stalemates (direct, helpstalemates etc) to be fairy chess. This could be the answer to a question that has bothered and bothers many of us: why are stalemates regarded as fairy chess? This is, however, not mentioned in the Codex; how well this suites the situation of today deserves attention.

Objective of this discussion

The Codex is the property of WFCC, inherited from PCCC. WFCC decides what they do and what they don’t do with the Codex. In this tread has been indicated that a new classification system will be drawn up. Drawing up a new classification system is ideally led by the property owner WFCC to ensure credibility; this is in my mind step 2. Step 1 is to convince WFCC and the problem chess community that an update of the classification system in the Codex is needed. Then hopefully actions will be taken for step 2. If making step 2 without an assignment serves that step 1 will be accomplished, then just go ahead!
 
 
(Read Only)pid=8511
(37) Posted by Nikola Predrag [Wednesday, May 9, 2012 18:54]

@Kevin
I certainly would not connect a threat with any fundamentals. In a seriesmover it matches the solution and in an alternate play it is absurd as it is illegal. The idea of a threat surely simplifies a description of the real content and may add some virtual content. I have to accept it as a subordinate of the chess problem community but in private, I disagree with imposing a threat as a basic rule. Absurdity in the basis undermines the logic of the whole structure.

Threat is a move that solves a problem after the specific class of the opponents moves. There's no threat in a 'Zugzwang' although everything is the same. The only (virtual) difference emerges out of the illegal series of two or more moves by one side (if the opponent plays no move, there is a threat). Such an illegal discrimination justifies duals in 'threat-problems' but claims them in 'Zugzwang-problems'
 
 
(Read Only)pid=8512
(38) Posted by seetharaman kalyan [Wednesday, May 9, 2012 22:13]; edited by seetharaman kalyan [12-05-09]

I am sorry that I misquoted TRD from memory in my Post 21.. and hence the valid objection by Nikola in Post 24. Here is the reproduction from his book "Caissa's Wild Roses"

"Normal chess may be regarded as consisting of
1. playing space-the 8x8 plane board
2. six men KQRBSP with specific movement, and
3. various limitations on move freedon, such as the Principle of opposition to the adversary, the alternation of White and Black moves, capture, Pawn one or two P-promotion, PxP e.p., castling, check, checkmate, draw by repetition of moves or positions or 50 move rule, and draw by reduction of force."
He continues:
"From this point of view, normal chess is evidently an arbitrary group of elements selected from an infinity of analogous geometrical conceptions. Fairy chess comprises the study of all such elements taken in arbitrary groups at will"

Anybody will agree on this wholistic view.

It is clear that he did consider Helpmates are Fairy chess. That many chess games are won by help from the opponent is not relevant, as it is the INTENT that counts.

From the above definition of his some may feel that he did not consider Direct Stalemates as Fairy (as he mentions Stalemate in item 3). But in a separate chapter of this book he quotes Direct Stalemate problems under the category NEW AIMS. He considered MATE the only orthodox AIM.

He wrote while discussing Stalemate as a new AIM : "It is probably not generally realised that the familiar ending 'White to play and draw' is essentially a branch of this Fairy Chess subject".
 
   
(Read Only)pid=8514
(39) Posted by Nikola Predrag [Wednesday, May 9, 2012 23:44]; edited by Nikola Predrag [12-05-09]

I'm afraid that the intent is not a firm ground for fundamental definitions. A player may wish to draw and his moves sometimes show it as a clear intention. A player may wish to win, so he wishes to mate in the end but you don't see his intent to mate from the first move, and often never in the whole game. Impatient intentions to mate usually don't lead to that goal. Beginners nicely play and capture the opponents pieces but usually they don't know how to mate. Sometimes the experienced players can't mate with B+S within 50 moves, not to mention the more delicate positions. We may believe that there is the intent to mate but a fundamental definition asks for more.

The rules of chess are mandatory but the goal is not. Why would a chess composer proclaim a mate of the opponent's King as a mandatory goal in orthodox problems when it is not mandatory in orthodox chess game?
 
   
(Read Only)pid=8516
(40) Posted by Kevin Begley [Thursday, May 10, 2012 02:30]; edited by Kevin Begley [12-05-10]

@Nikola,

>"I certainly would not connect a threat with any fundamentals."

I'm not saying threats are fundamental element of a problem description.
However, without a threat-notation, you can not define a dual.

Simple example (which illustrates why threat-notation is fundamental to the solution of a direct-play problem):

Fred Lazard
L'Eclaireur du Soir, 1914
(= 4+3 )

#2

Unless you format the solution according to threat notation, you will incorrectly perceive duals.
After 1.Rb8! ...Sa3, white would have two mates: 2.Qb7#, and 2.Qd6# (a dual).
Only after formatting the solution by threat-notation, do we see that the key threatens (> 2.Qb7#).
Threat-notation shows that 1...Sa3 provides no valid defense against the threat, thus it's not a legitimate part of the solution (and constitutes no dual).

Furthermore, the solution's format can become particularly important, when evaluating themes (especially alphabetical themes).
I believe the format of the solution is something that the composer implicitly requires of the solver.
Same is true when you are "tasked" to submit a document, in a specific format -- the format becomes part of the task!
To the extent that threats are a fundamental element of a significant format, they must be considered essential.

Even if chess is a highly evolved game of tag, threats remain a key component.

>"...in private, I disagree with imposing a threat as a basic rule. Absurdity in the basis undermines the logic of the whole structure."

I've made no claim that a threat should be imposed as a rule.
I'm only saying that the format of a solution is an essential element (albeit, a generally implicit element) of the task.

In direct-play problems, the composer expects the solution to be formatted by threat-notation.
In all other problems, the composer expects the solution to be formatted in a basic game-notation.
However, there is an exception to this rule -- which occurs in fairy conditions where a player may idle.

To resolve this matter, either we need threat-notation to make some exception for an idle player, or we may need to provide the composer some formal mechanism to express what format the solution is intended to follow.
I prefer the former option, if it is possible.

>"There's no threat in a 'Zugzwang' although everything is the same."

Zugzwang is already well covered by threat-notation -- all legal moves are considered a defense to a ZZ threat.

>"The only (virtual) difference emerges out of the illegal series of two or more moves by one side (if the opponent plays no move, there is a threat). Such an illegal discrimination justifies duals in 'threat-problems' but claims them in 'Zugzwang-problems'"

I don't exactly follow you here... are you attempting to fix threat-notation to handle ZZ threats in idle-mover type direct-problems?
I honestly haven't given the matter much thought -- I just know the solution's format often fails in direct-play idle-movers.

Take the simple case: series-movers.
Take a ser-#n, filter out the implicit fairy condition (which changes the move alternation), and you're left with a #n stipulation.
According to the axioms, this is the proper way to define series-movers.
However, at no point in the course of the solution does the composer intend for a threat to creep into the solution.
In the more complex case (e.g., apply the same procedure to a pser-#10), however, there may be points in the course of the solution when it becomes important to incorporate threats into the notation.

Zugzwang may present further challenges for threat-notation in idle-mover fairies -- I really haven't given it much thought.
But, I know that the idle-mover family of fairy conditions could often be significantly improved by their ideal expression in a more intelligent threat-notation.
And, this would help us return to an axiomatic system -- where altered rules (fairy conditions) are always explicit (no more conglomeration & concealment in the stipulation).
 
   
(Read Only)pid=8518

Read more...
Page: [Previous] [Next] 1 2 3

MatPlus.Net Forum General What is Fairy Chess?