Julia's Fairies

No.420 (PH)

Peter Harris (South Africa)


Original Problems, Julia’s Fairies – 2013 (III): September- December

  →Previous ; →Next ; →List 2013(III)

Please send your original fairy problems to: julia@juliasfairies.com

No.420 by Peter Harris – This nice Miniature with a combination of Isardam and Chameleon Chess has opened up some questions. And finally I’ve decided to present it to you with some comments about the interpretation of Chameleon Chess condition in WinChloe and in Popeye, as well as comments about possible bugs in Popeye on the example of this problem. (JV)


Isardam: Any move, including capture of the King, is Isardam illegal if a Madrasi-type paralysis would result from it.

Madrasi: Units, other than Kings, are paralysed when they attack each other. Paralysed units cannot move, capture or give check, their only power being that of causing paralysis.
Madrasi RI (rex inclusive): the rule applies to Kings as well, so the two Kings may stand next to each other.

Chameleon Chess: All pieces on the board which are displayed as orthodox Q, R, B, S, are Chameleons. A Pawn can promote only in Chameleon-pieces.
Here I’d like to mention that this interpretation worked in Popeye‘s version till 4.61, but doesn’t work in the last version, 4.63. I’m not sure either it’s a bug or some new interpretation.
Also, WinChloe has a different interpretation of Chameleon Chess, where orthodox pieces on the board don’t turn into Chameleons, but a pawn promotes into Chameleon piece. I’d like to offer you a little explanation with examples about WinChloe’s interpretation of Chameleon Chess and Promotion into Chameleon Allowed kindly provided by Christian Poisson.

Chameleon: On completing a move, a Chameleon (from classical standard type) changes into another piece, in the sequence Q-S-B-R-Q… Promotion may be to a Chameleon at any stage in the cycle if any Chameleon piece is presented on the board.

No.420 Peter Harris
South Africa
original – 13.11.2013
hs#3            b) wKe6→f1           (3+2)
Chameleon Chess
Solutions: (click to show/hide)

Notify of

Newest Most Voted
Inline Feedbacks
View all comments
Kjell Widlert
Kjell Widlert
November 14, 2013 00:06

The Popeye result when only e8+f8 are Chameleons is correct, not a bug – at least according to Petkov’s own rules for Chameleon play. In the cook, Black promotes to an orthodox Queen, and this of course cannot observe the bK as an S. The point is that orthodox pawns may promote to orthodox OR chameleon pieces (an unfortunate rule in my view, but that’s another matter).

Kjell Widlert
Kjell Widlert
November 14, 2013 00:20
Reply to  Kjell Widlert

I might add that this fact is probably behind the bug in Py 4.63 producing the “strange looking” extra solution. Popeye seems to differentiate between “moving a chameleon, turning into a different chameleon” and “moving a pawn, turning into a chameleon by promotion”. When the pawn is already a chameleon, as in Chameleon Chess, there is no difference between the two so there is only one solution.

Kjell Widlert
Kjell Widlert
November 14, 2013 00:24
Reply to  Kjell Widlert

I should say something about the problem too… While it’s true that the two parts have essentially no strategy in common, both are quite pointed and obviously very economical. I like this a lot!

Peter Harris
Peter Harris
November 14, 2013 13:04

I would agree with Petko: just because there is a Chameleon piece on the board should not mean that pawns HAVE to promote to a Chameleon piece. I forgot about a choice/option when I said there was a bug.

I also failed to notice that in the quoted Popeye output [solutions] the Q is not followed by = c .

I think I may have been distracted by the WinChoe solution! – where there is no choice and is therefore is not following Petko or Popeye.

So if compulsion is required a pawn should be designated Chameleon or the [Popeye] ChameleonChess condition used.

Kjell Widlert
Kjell Widlert
November 14, 2013 22:51
Reply to  Peter Harris

I don’t want to start a big debate on this detail: the rules have been declared by the inventor and they should stay. But I should perhaps explain why I used the term “unfortunate”.
In theory:
Some fairy forms are additional properties of certain pieces, properties that you cannot acquire or lose by pawn promotion. Neutral pawns promote to neutral pieces only, and orthodox pawns don’t. The same goes for paralysing pawns. It seems to me most natural to regard “being chameleon” as such an additional property, in which case chameleon pawns would always promote to chameleon pieces and orthodox pawns would always promote to orthodox pieces (or any other kind of fairy piece that is present).
In practice:
The rule means that with at least one chameleon on the board, a normal pawn can promote to eight different pieces (not only the type of the chameleon that is on the board at the moment – which also proves that a chameleon queen, for example, is not just an ordinary fairy piece with respect to promotion). It is often impossible to differentiate all these promotions, and it will always require at least one more move by the promoted piece. The dualized version of no. 420 with chameleons only on e8+f8 is an example of this.

Diyan Kostadinov
Diyan Kostadinov
November 16, 2013 05:33

First of all the most important is that Chameleons are different programmed in Popeye and WinChloe.
In WinChloe Chameleons are defined as clear fairy pieces which mean that for example Chameleon Bishop do not paralize normal Bishop in Madrasi; also in Circe the captured white Chameleon Queen do not reborn on d1 but o the file where is captured (as a fairy piece) etc.
In Popeye chameleons are used in the nature of piece which phase they are in the moment – Chameleon Bishop can paralize normal Bishop in Madrasi; captured white Chameleon Queen reborn on d1 in Circe etc.
This is the reason because of which both programs give different results on this problem.
Julia, please note that in version with screenshot of WinChloe the program do not present the same solution in b) as the authors solution! The promotion on c1 is to Chameleon Rook, not normal Rook as in authors problem, because if on c1 is regular Rook, the white move cRf3=cQ?? do not avoid the check (the black Rc1 still can capture the wKf1 because the chameleon on f3 can not paralize the regular Rook).

Diyan Kostadinov
Diyan Kostadinov
November 16, 2013 05:41

I forgot to add that in my memories Petko Petkov told me that the most correct is when in Circe the captured Chameleon piece reborns acordind its phase in that moment; in Madrasi Chameleon pieces can paralizing normal pieces when they are in the same phase etc.
P.S. In Chameleon Chess ALL PIECES are Chameleons and change their phase on each move! So after pawn promote some piece – this piece also change its phase when moves.

peter harris
peter harris
November 17, 2013 16:36

After reading the article “WinChloe’s interpretation of Chameleon Chess and Promotion into Chameleon Allowed “ and Diyan Kostadinov comment about Madrasi, I have learnt that I was wrong about why WinChloe only gives one (b) solution and not two like Popeye. It is NOT because WinChloe disallows Orthodox promotions – which I assumed was the [only possible] explanation. It is because with WinChloe there is no paralysis if the one piece is Orthodox while the other like-piece is Chameleon. What a surprise!

The 4 examples given in the above cited article wherein there are various combinations of Chameleon, pieces marked Chameleon, with or without other unmarked pieces and “promotion into Chameleon allowed” are quite bizarre and I would say confusing.

One can only sympathize with an editor who when presenting a [Winchloe based] problem with Chameleons would have the task of making Definitions for all the possibilities. In the absence of [full] definitions solvers would not know where they are. Also with Madrasi: the definition would have to be expanded.

So [at present] there can be no single definition for many conditions with the same name. There has to be Madrasi I , Madrasi II and so forth.

The only way around this is to have a WFCC approved Guide of Definitions. Surely everyone would approve an Authority [an imposition of discipline] to bring some sort of order.

It is not that there are BUGS in programs that is the trouble. It is that the programs are using different DEFINITIONS. Different definitions are being used almost at the whim of various Programmers – causing confusion and making life very difficult for editors – like Julia.

November 17, 2013 19:08
Reply to  peter harris

Quite right Mr.Harris. I wish there is an universally agreed definition covering centrally accepted by WFCC.

Diyan Kostadinov
Diyan Kostadinov
November 17, 2013 20:49
Reply to  seetharaman

Practically the “universal fairy definitions guide” is not possible because even if all fairy conditions and pieces are finally clear defined (which will be difficult, because of many opinions) it will be not possible all changes to be programmed in all chess solving programs. So it is not possible to (re)define all when after that will be not possible to be used because of not possible computer check.
Even if all fairy conditions and pieces are programmed in same way in all programs (which is impossible) again different programs will give different results sometime because of different priority of the fairy conditions if they are used in combination (which is considered first).
So practically – we shoud accept what we have now and try to make some improvements, but the problems and mistakes will occur always…

peter harris
peter harris
November 18, 2013 10:32

On the one side there is the Fairy chess world and the WFCC and on the other computer Programs and programmers.

Who is in charge here?

In considering the concept of a WFCC Guide to Fairy Definitions forget about computer programs. [They are all over the place].

The WFCC [representing all of us] should not be hamstrung by or dictated to by Computer programs and programmers – who are more concerned with their own interests than with the general good – and so resist any form of a universal dispensation.

If a WFCC Guide was brought into existence computer programs could choose whether to participate or not. It is as simple as that.

To say regarding present computer programs that this, that or the other thing is “not possible” even if true, is hard luck. And with regard to “many opinions”: a democratic process would prevail.

[The issue of which condition is given priority over another in a multi-condition problem is a quite separate and altogether secondary issue – and has no bearing whatsoever on the issue at hand – a WFCC Guide to Fairy terms].

A WFCC Guide could give rise to a computer program specifically programmed to conform to the Guide. Both the Guide and the program could become universally used – by magazines, websites and for Album judging purposes.

Of course everyone in the Fairy problem chess world [except vested interests] would welcome a Guide if it brought some order to the present disorder.

The creation of the Guide could in fact be the most important issue / task facing the WFCC.

Juraj Lörinc
Juraj Lörinc
November 19, 2013 00:27
Reply to  peter harris

The creation of the Guide could in fact be the most important issue / task facing the WFCC.

Considering the importance of fairy chess in the eyes of many composers as well as WFCC delegates, it is clearly doubtful assertion. Not very important actually.

Moreover given the variability of already existing fairy elements, numerous possibilities of their interaction and contant inflow of new elements, the task of Guide creation would be gargantuan, if not Sisyphean.

peter harris
peter harris
November 18, 2013 15:46

Regarding a WFCC Guide: directly related to the subject is the following.

[There are some questions that for some reason are not asked. It may be similar to the story about the Emperor’s clothes. As he rode by in his carriage people cheered and said how wonderful his clothes were – when in fact he was naked].

In the Popeye’s HELP is listed Pieces and Conditions [and other things].

Now one would think it would be normal and natural for the DEFINITIONS Popeye used when programming to be given – in the same HELP.

The two: the name of the Piece or Condition and the Definition go hand in hand.

Is it not extraordinary that the definitions are not given? [One could almost ask how they have got away with the omission all these years!].

What is the explanation?

As the foremost program in the world is there not a duty or an obligation involved here?

But perhaps I am missing something in all this.

Diyan Kostadinov
Diyan Kostadinov
November 18, 2013 16:26

Dear Peter and Julia,
Wishes are one thing but reality is another…
Many of delegates in WFCC are not fairy composers, many are even against fairy compositions… So how you explect to work for such a big project!??
A Fairy sub commitee is needed, but even there – who will participate? All members should be with big fairy knowledge, but many of them don’t want to participate in it – the reasons are a lot. Such a sub comitee should include also a computer member (representing chess programmers) and literally english member (for the definitions).
And let accept that all is done as should be – how you expect the programmers to be agree to hard work (for free!) to change their programs and all to be universal?? Probably you don’t understand how important these computer check is – many top composers even don’t look fairy conditions or pieces if they can not check it.
So why the Fairy sub comitee members will work hard for such a Fairy Guide when after that almost all will be the same? Even now almost all fairy pieces and definitions are well defined but they continue to differ… For example – this year I do not participate in World Cup fairies because my problem (with two very well defined fairy conditions) was impossible to be checked – one program was with wrong programmed 1st condition and another program was with wrong programmed 2nd condition… And I know that these conditions can not be corrected because both inventors was discussed it with the programmers. This is only one example how important are the chess programs.
In my opinion WFCC SHOULD PAY to some group skillful programmers which can guarantee the correct program results and then we can start with correct definitions etc. But who can guarantee something???
Wishes are one thing but reality is another…

November 18, 2013 19:50

I generally agree with what Diyan wants to convey. Programs are indeed important tools nowadays for composition. But I dont think that WFCC should first set up a group of paid programmers and THEN start with correct definitions. Actually agreeing on correct definitions should be the FIRST task. Correcting the existing programs may not be necessary for most of the Fairy terms. If needed WFCC can think and decide LATER about paid programmers. Of course the opinion of the existing programmers would be an important input in framing any such Code of Fairy Terms.

Nikola Predrag
Nikola Predrag
November 19, 2013 01:56

Interesting that noone says: “I’ll begin that project”, but “They should do it”.

November 19, 2013 08:16
Reply to  Nikola Predrag

My knowledge is limited, but I am offering my services if the future Fairy Codex Committee needs any help.

Diyan Kostadinov
Diyan Kostadinov
November 19, 2013 04:58

Nikola, nobody said himself “I will do it” basically in two reasons: 1. this project is extremely huge and difficult and no one can do it himself, and 2. he should be composer with big experience and big knowledge in fairy field and everyone to trust him as an expert.

So in my opinion if our fairy community want to put things in order we should do it as follow:
1. WFCC should vote for creating “Fairy Sub-comitee” (because only team of composers which work in fairy field can understand where are the problems in the genre and what is needed to be done). I had been planned to propose this in last 2-3 congresses but still do not do it because was not sure is this proposal will be supported. Maybe we should do it and discuss it in Bern;
2. The members should be carefully selected (in my opinion there should be GM Petko Petkov plus more 2-3 top fairy composers, plus 1-2 programmers, plus 1 native english speaking);
3. Special site/forum should be created where all fairy composers can send their questions, proposal etc. cases which should be discussed and solved by the Fairy sub-comitee;
4. The sub-comitee should prepare the “Code of fairy terms”, including definitions, schemes with the specific demonstrations and some short selection of examples (which actually means thousands of special prepared schemes and selected compositions!!);
5. If is neccessary WFCC should pay to the programmers team which have to prepare solving program (new version of existed program or completelly new if the corrections are not possible – the programmers should say which is better).

Only if all these 5 points can be realized the effect will deserve all hard work. Otherwise the final result will be again debatable.

peter harris
peter harris
November 19, 2013 12:07

I am going to change the name!

********* By doing this I hope most of the objections and what has been written in Comments will fall away.

The new name is

The WFCC Fairy Registration Bureau!

The Bureau would contain a Name of a Piece or Condition and its Definition.

That is all!

Does this not put a totally different complexion on the matter?

No need for perturbation!

[So for example I could submit the following for Registration:

Name of Condition: ForwardKings.
Definition: Kings can only move forwards.
Example: wKe4, wBa2, bKe6 is #

The Bureau will register this if there is no similar condition and will not register
the Name again.

The WFCC and magazines may not accept problems with the ForwardKings condition unless C+ with Popeye or WinChloe].

Concerning the Gargantuan task:

******** If the Definitions in Popeye and WinChloe were/could be simply Copied into the Bureau 95% of the job would be done. (!)

******** There would be no need to exclude from the Register any definition supplied by either of the two programs. All that would be necessary would to alter a Name given by one of them if the one Name had different Definitions.

[Concerning Juraj’s comment: Yes, I was imprecise. It is important only in a Fairy context. Yes, the Register would never be complete – always growing – with ForwardKings!].

[Concerning the absence of Definitions in Popeye’s HELP: the point is for example: it may have taken Thomas [Maeder] hours to program the new Condition FTF and input it into Popeye. It would take him 5 minutes to input the Definition. So the Popeye program should provide for this. I do not know whether HELP is the right or best place to put the definition but it should be put somewhere].

A sophistication in the construct of the Register could be that it could indicate for each Definition which computer program supports the Definition – by initials e.g. P, W, A after a Name. [Or it could indicate which program does NOT support the Definition – perhaps the simpler option] These initials could also be shown in Definitions given for a problem in a magazine.

[With regard to priority of Conditions: does there HAVE to be priority state – can it not always be that the side on the move can CHOOSE [to enable the stipulation to be achieved]? I am not sure that what I am suggesting here makes sense!]

Regarding the practical creation of a WFCC Fairy Registration Bureau: there is of course a common denominator in the management of two particular websites! I wonder whether the WFCC could subcontract a site to do something on its behalf.

Would love your thoughts, please comment.x