Julia's Fairies

No.670 (KM)

Karol Mlynka


Original Problems, Julia’s Fairies – 2014 (III): September – December

?Previous?Next ; ?List 2014(III)

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

No.670 by Karol Mlynka – AUW in a problem with Imitator and Pressburger King!  (JV)


Imitator(I): Every time a piece moves an Imitator (or a set of Imitators) moves simultaneously in an identical manner. An Imitatorcannot move of itself. If an Imitator cannot imitate the move of a piece, the move is illegal. An Imitator may only pass through or enter an unoccupied square and cannot move off the board. Castling is imitated by decomposing into a King move followed by a Rook move. 

Pressburger King (Super-Transmuting King): King which definitively takes up the nature of the checking piece (and thus loses his royal status).

No.670 Karol Mlynka

original – 20.12.2014

Solutions: (click to show/hide)

White Kh1 Pe7 Pc5 Pg4 Pe3 Black Kh4 Rf7 Pc6 Ph2 Neutral Ib7

hs#4                                      (5+4+1i)
Black Super-Transmuting King
Imitator b7
b) Kh4?f1; c) Kh4?f3; d) Kh4?Pe3

Notify of

Newest Most Voted
Inline Feedbacks
View all comments
December 20, 2014 15:07

Very ingenious and complicated solutions! I just cant imagine how the composer conceived such an idea. Just great !

Kjell Widlert
Kjell Widlert
December 21, 2014 01:57

Am I missing something? Cannot Black in a) play (4.Kxh2(Ib8)+) 4. – Rg7(Ia8)! ?

The move seems legal, as neither Bc8xh3 nor Kh2xh3 can be imitated by Ia8. And why would the wK be in check now if it wasn’t one move earlier (which must be because Kh3 is now moving like a B)?

So I don’t think the solution works (so no solution, and a bug in Py) or am I missing something?

December 21, 2014 12:24
Reply to  Kjell Widlert

Indeed you are right! I also missed this black defence.
But having looked at the position again, I have one more basic question. Cannot black simply play 4….Rh6 (Ib7) or any other rook move down the h-file? Since after the check by white bishop there is no royal black piece why should black try to prevent white playing 5.Kh2xh3?

Georgy Evseev
Georgy Evseev
December 21, 2014 12:58
Reply to  Kjell Widlert

As far as I know Super-Transmuting King has to respond to check by moving. It is not allowed to ignore the check (even if it loses royal status) or defend it by moving any other piece.

This is one of the reasons I do not like this fairy element.

Karol Mlynka
December 21, 2014 13:33

I’m sorry. Excuse me, please. Too I trusted computer program, and in addition, the desire to achieve something so complex and share it with experts has powered the rapid publication. And so this is only my mistake and the error of the program.

December 21, 2014 15:07

True. Moreover the checked King is moving as the checking piece:

Stipulation ~1
Pieces white Ka1 Bb1 black Ra3
Cond whitesupertransmutingking

1.Ka1*a3=R !
1.Ka1-a2=R !

1.Ba2 is forbidden to parry the check.

Btw the notation Ka1*a3=R looks in wrong order. The other possibility Ka1=R*a3 seems to better explain what is going on here.

December 21, 2014 15:30

Super Transmuting King is not programmed in WinChloe, but for ordinary Transmuting King the above uncheck 1.Ba2 is allowed.

I don’t know if the reason of the forbidden 1.Ba2 for Popeye is a bug or comes from a specific STK rule.

Karol Mlynka
December 21, 2014 17:10
Reply to  dupont

Complete definition: Pressburger King (SuperTransmutingKing) = King which definitively takes up the nature of the checking piece (and thus loses his royal status). After being checked, the Supertransmuting King must move. If the checked King can not move, it is a mate.

Kjell Widlert
Kjell Widlert
December 21, 2014 22:43
Reply to  Karol Mlynka

So with this addition to the rules, the problem must be sound. But like Georgy, I don’t very much like such “illogical” rules: it should be legal to parry a check by any possible move that stops the capture of the king.

(Another strange consequence of the supertransmuting rules is that there can be at most one check (from each side) in each line, including the mate; for after that, there are no more kings on the board. Unless you start with several kings on the board, that is.)

December 22, 2014 20:37

Thanks all for explaining clearly the rules of Supertransmuting King. Karol Mlynka’s problem is so good, it makes one excuse the illogicality of the Fairy condition. Congrats again !

peter harris
peter harris
December 23, 2014 15:49

Re the Definition: Karol writes:

“After being checked the Supertransmuting King must move. If the checked King can not move it is mate”.

This is not according to Popeye, otherwise in (d), 4. Qe2-a6 [Ia5] would be mate – i.e. Popeye would not move 4. …….Rf7-f3 [Ia1].

December 23, 2014 16:32

In this situation the bK is checked by the wK, hence is not supertransmuted – bK remains a King. So probably Popeye is considering this situation as “normal” – in particular black may undo the check without moving its King.

peter harris
peter harris
December 23, 2014 18:28

That is not the explanation.

As per the definition:

the bK “takes the nature of the checking piece” – King or not..

Further: The definition of SuperTransmutingKings does not include a provision [and never has] that the King HAS to move otherwise mate*. Popeye would be not be programmed with that intention. Normal defences are allowable – which makes sense all round.

* It is an incorrect explanation of why the defence Kjell mentioned in (a) is not allowed.

There is a BUG in Popeye.

[Popeye sometimes goes awry when combined with other conditions].

peter harris
peter harris
December 23, 2014 19:20

whi kg2 rf8
bla ka3 bb7
stip +1
cond whitesupertrans

I was expecting to see a wRf3+ but it did not appear.

So much for my presumption about Popeye.

We are in want of a Definition.

Perhaps Popeye could give us the one they used.

December 23, 2014 19:53

The scheme 21/12 15:07 (which is not combined with other conditions) demontrates that, in this situation, Popeye allows only to parry the check with a move from the checked ST King.

I made another test, with the setup:

Pieces white Kh1 Pa6 Bc8 black Kh2
Cond Imitator b8
Cond whitesupertransmutingking

The only legal move is 1.Kh1-g1[Ia8]=K. But if we add a wS on g1, the only legal move is then 1.Bc8-b7[Ia7] +

It shows that Popeye is following the rule: a checked ST King must move in priority, but without any such legal move, the checked side may defend differently.

It explains why Rg7(Ia8) is forbidden in solution a), and why the checked ST King is moving in solution d).

December 23, 2014 20:02

Oups crossed messages Peter. With my above explanation of which rule Popeye is using, your Rf3+ is illegal as the white ST King may parry the check by moving, and hence it must parry it that way.

peter harris
peter harris
December 23, 2014 20:43

Incidentally, if all leading magazines use a particular definition – a definition accepted by the fraternity at large – then programs should follow the definition and write their programs accordingly and should not alter the definition according to their ideas – and whims.

The definition of PressburgerKings and the matter relating to making a King move compulsory is a case in point.

But perhaps Popeye divergence was unintentional.

[The divergence spoils the PressburgerKings condition].

peter harris
peter harris
December 23, 2014 21:05

Your explanation about Rf3 being illegal is not correct. Rf3 does not appear even with the wK on h1.

Also: you are bringing in something new namely PRIORITY. This is nonsense.

I have no idea how you would define PressburgerKings.

The plain fact is that the PressburgerKings Definition as shown above the diagram is correct and FULL.

Popeye is going its own way and confusing everybody – including you.

peter harris
peter harris
December 23, 2014 21:25

Having the wKh1 still allows the K to capture the bBb7 [even if defended by a bQa8] and does not, as I suggested, alter the position from having wKg2.

Nevertheless Rf3 should be allowed to check and not be disallowed by Popeye – which is the main issue at hand.

December 23, 2014 22:00

I’m not confused at all, as I just want to understand which rules Popeye is using (if some), and not to claim that those rules are the true ones which define ST Kings!

And my above suggestion (priority to a move by the checked ST King) seems to fully explain each examples – Karol’s problem, my 2 schemes and yours.

December 23, 2014 22:20

If you replace, in your scheme, the bBb7 by a black Locust a8, then 1.Rf8-f3 + is now accepted by Popeye, as well as
1.Rf8*a8 + (but no other white move), because in this new situation the wKg2 has no legal Locust move but each of the 2 Rook moves parry the check.

peter harris
peter harris
December 24, 2014 02:08

Unlike you I am confused.

You simply do not know what is going on and are trying to figure it all out.

If I was an Editor I would say to an Author:

Dear Author, I cannot publish your problem giving solutions that do not follow the Definition I use. It is ridiculous to do so. So please let me have your Definition or the Definition your program uses. I will then publish your problem. [I am not forcing you to use my Definition]. Thank you.

[Whether major magazines would agree to alter a generally accepted Definition to that used by Popeye is another matter. They may ask Popeye to rewrite its program].

December 24, 2014 08:21

There is an interesting small detail – the notation Ka1*a3=R used by Popeye in my first scheme. It suggests that the ST King doesn’t become a Rook just after the check, but only after having moved.

I verified this suggestion with the addition of the Madrasi condition to the same scheme. Then Ka1*a3=R is still allowed, which would have been illegal if the King were a Rook before its move (Madrasi paralysis).

In conclusion, Popeye fulfills the “classic” ST King definition, modulo a precision: the ST King must move BEFORE taking up the nature of the checking piece.

So maybe it is worth to rephrase the definition as follows:
“When a ST King is checked, it must move like the checking piece and then definitely takes up its nature. When no such move is available, the ST King side parries the check normally. ”

I don’t know if it was the author’s intention – maybe he didn’t think at all about the possibility for a ST King to be “stalemated” – indeed this is impossible if everything except ST King is orthodox. But this slightly modified rule is the one followed by Popeye, and seems consistent.

peter harris
peter harris
December 24, 2014 17:04


It would not have been apparent from my Comments but I have admired your efforts to understand and explain Popeye’s solutions! – that is to say to find out what DEFINITION Popeye is using – in which regard perhaps you could obtain the information from Popeye and so make your detective work unnecessary.

I have made many Transmuting Kings problems and as I wrote in a Comment above, have found that when the condition is combined with other conditions [perhaps particularly with Isardam and PatrolChess] Popeye sometimes has difficulties.

So you should not draw too many conclusions from what you have seen when you introduced.Madrasi.


whi kh1 re8
bla ka4 ba8b7
stip ~1
cond whitesupertrans madrasi
opt duplex
cond whitesupertrans isardam


1.Kh1*b7=B !
1.Kh1-c6=B !
1.Kh1-d5=B !
1.Kh1-e4=B !
1.Kh1-f3=B !
1.Kh1-g2=B !

Duplex: No moves given.


1.Re8-e4 !
1.Re8*a8 !
1.Kh1-h2 !
1.Kh1-g1 !

Duplex: “The side to play can capture the king”. [i.e. the wK is in check].

Many of the answers given are inexplicable.

Nicolas Dupont
Nicolas Dupont
December 24, 2014 21:30

In the Madrasi case, everything looks ok. Indeed the ST King is checked by a Bishop, hence must move like a Bishop (and will take up the Bishop nature only after its move).

In the Isardam case (asked separately, not as a twin), Popeye 4.68 gives only the possibility 1.Re8-e4 + and once again it seems correct. Indeed the ST King is also checked by a Bishop but each Bishop-like move from the ST King would let it on the diagonal and then take up the Bishop nature, hence would lead to a Madrasi paralysis, which is Isardam-illegal. As a consequence the ST King cannot move and Re4+ is the only way to parry the check.

NB: my “detective” work may have been necessary as programmers themselves may not always fully understand their own creature, just as Dr. Frankenstein!

peter harris
peter harris
December 25, 2014 10:13

[The bugs in my earlier Popeye version have evidently been removed].

Re your: “my detective work may have been necessary”:

If we were given the Definition used by Popeye programmers, efforts to find out what the Definition is would NOT be necessary.

[You would be doing a service if you disseminated the Definition Popeye uses, to all Fairy editors so that they can use the Definition when Popeye solutions are given in a problem – in replacement of the Definition they are currently using – thus avoiding a mismatch between Definition and Solution – which mismatch makes it impossible for a solver to solve].

Kjell Widlert
Kjell Widlert
December 25, 2014 13:34

Hmmm, the exact definition of supertransmuting kings (Pressburg kings) seems elusive. I see now that I don’t understand why the mate in part a) is a mate – why not 5.Kh1=R (Ia6) or 5.Kh3=R (Ia8) ?

The wK moves like the checking rook, and then takes up its nature permanently, losing its royal status – becoming a normal white rook. The moves cannot be self-check, as a normal rook cannot be checked (only royal pieces can).

So does a supertransmuting K in check actually keep its royal status a little while longer than we have assumed?

The mystery deepens…

peter harris
peter harris
December 25, 2014 15:59

Hello there Kjell.

Only the bK is SuperTransmuting.


Kjell Widlert
Kjell Widlert
December 25, 2014 20:32
Reply to  peter harris

Thanks – I had forgotten that, with this long conversation between the actual problem and my post 🙂

Thomas Maeder
Thomas Maeder
February 1, 2015 16:49

> the exact definition of supertransmuting kings (Pressburg kings) seems elusive.

Can somebody from Pressburg (i.e. Bratislava) or somebody else come up with the original publication?

The way it is programmed in Popeye seems really weird, but before fixing, I’d like to know what target to aim at.

Juraj Lörinc
February 1, 2015 22:35
Reply to  Thomas Maeder

For me is supertransmuting king elusive as well – and I do live in Bratislava. However, the Marianka Cup 2014 was dedicated to supertransmuting king and there was a considerable effort by inventor Karol Mlynka and judge Juraj Brabec to clarify any confusion. The English translation of the award is available in the Marianka bulletin starting from page 91. It includes some theoretical discussion of supertransmuting king properties as well. (With enough time, see also the rest of the bulletin, there is a lot of interesting material there.)

February 1, 2015 23:54

Thanks Juraj, it seems everything is now clear:

– A checked super-transmuted King must move, using the mobility of the checking piece. If no such move is available, it is checkmated.
– Problem 670 is cooked, as the author recognized himself (December 21, 13:33).
– Popeye is buggy, but not deeply: the only bug I can see is that, when a checked super-transmuted King can’t move, Popeye allows the other possibilities to uncheck rather than declaring a checkmate.

Kjell Widlert
Kjell Widlert
February 2, 2015 21:55
Reply to  dupont

The composer referred on Dec. 21 to my question if part a) isn’t unsoluble because of 4. – Rg7(Ia8), but we now know that this move is illegal because a checked ST-king must move.

The problem must be unsound anyway, however: the same argument implies that part d) has no solution as 4. – Rf3(Ia1)#? is illegal.
(I assume it doesn’t make a difference that the piece checking the bK happens to be the wK.)

February 2, 2015 22:46

Yes. It is also interesting to note that the problem is cooked precisely the way Popeye is buggy – it allows an illegal unckeck at the end of solution for part d).

This phenomenon can be described with only 3 pieces – wSTKb7 wPe4 bLd5. The white ST-King is checkmated by the black Locust, although Popeye considers e4xLd5 as a legal way to parry the check.

Thomas Maeder
Thomas Maeder
February 11, 2015 19:30

Fixed for Popeye 4.71.

Would love your thoughts, please comment.x