Website founded by
Milan Velimirović
in 2006

22:27 UTC
ISC 2024



Remember me

Forgot your
Click here!
to create your account if you don't already have one.

Rating lists


MatPlus.Net Forum General New (or already existing) Fairy Condition?
You can only view this page!
(1) Posted by Kevin Begley [Thursday, Mar 23, 2023 15:34]

New (or already existing) Fairy Condition?

I try to avoid inventing new fairy conditions, but I had to mention this idea...

The idea is so simple, it may already exist (if so, what is this condition called?), but if it doesn't exist, I would call it "Equipassant."

Why should a pawn have an exclusive right to capture by en passant?
When an adversarial pawn makes a double-step move, all attacking units should have the right to capture it by equipassant.

That's the condition -- that's the only rule change:
Pawns which move two squares forward are subject to capture on the first square by ANY hostile unit (as hostile pawns may capture them by en passant, so too may any hostile unit capture a pawn on the first-step of its double-step move -- and only on the following move).

Note: a locust makes an equipassant capture by hurdling the pawn on the 1st step of the double-step move (despite the fact no pawn occupies that square in the diagram).
This is in the spirit of en passant capture, as it was intended that the double-stepping pawn be moved back to the first step, and captured there.
This has no impact upon rules of rebirth following en passant capture (as may be instituted by other fairy conditions -- unless otherwise stated, assume the rules for rebirth follow those of a standard en passant capture; this condition only defines how an equipassant capture is executed).

So, the following (poorly constructed) #2 would have several duals after a lousy checking key.
1.Bh2+ ...e5 is not only met by 2.dxe6 ep#, it can also be met by 2.Kxe6 ep#, 2.Qxe6 ep#, 2.Rxe6 ep#, 2.Bxe6 ep#, and 2.Sxe6 ep#.

(= 9+5 )


The only point of this diagram is to illustrate that the pawn on d5 should not have exclusive rights to en passant capture on e6.
This is certainly not an original. :)
(Read Only)pid=24578
(2) Posted by shankar ram [Thursday, Mar 23, 2023 17:00]

WinChloe has something called "Prise en passant universelle".
Meaning ANY piece can capture ANY OTHER piece of the opposite colour, en passant, if the other piece moves through a capturable square.
There are some 12 problems in the Echecs database, but the condition is not programmed.
(Read Only)pid=24579
(3) Posted by Kevin Begley [Thursday, Mar 23, 2023 17:06]

Thank you, Shankar, but that's a different condition entirely.
There, you can capture ANY UNIT by en passant (not only pawns).

I'm not suggesting that all riders should be capturable by en passant.
That's an interesting condition, of course, and I appreciate your calling my attention to it, but it's not the same rule.
I'm suggesting that only pawns (not riders too) should be capturable by en passant, but that such captures should not be exclusive to hostile pawns.

In fact, that condition gets much weirder...
It seems you can capture two pieces in that condition -- a h#2 problem goes 1.Qa6-h6 followed by 1...Rb8xb4 (also capturing the bQh6 ep on b6). I already dislike the idea of capturing riders by en passant, but this gets quite absurd.

That does remind me: there is an interesting question about whether you can capture en passant with a neutral pawn which is reborn in Circe Parrain (or ContraParrain).

For example:

(= 1+3+1N )

h-ep-2? Circe Parrain

1.Kg4 Kxh7 2.g7-g5[+bPh5] h×g6 e.p.?

The idea of en passant was to capture a pawn which attempted to evade capture on the first square by jumping two squares forward, but in this instance, it's questionable whether a a capture was evaded when the reborn neutral pawn was reborn AFTER the move. If the bPawn g7 had stepped forward a single square, there would be no capture.
Yet, I expect the engines will all say this is a legal capture.

While the spirit of that ep reasoning (not allowing a pawn to evade capture) makes sense to me, I can not fathom why a pawn may evade capture from all enemy units except the enemy pawn.

That's the primary motivation for my idea.
I'm not entirely sure it has value as a problem form, and you'd probably play this variant a long, long time before you have an opportunity to take advantage of this rule change, plus, it doesn't fully simplify the rules for interactions with fairy units -- especially those based upon combinations with pawns (where en passant captures may become uncertain).

The simplest rule is to make it illegal to move two squares if the first square is attacked, but that's probably the least interesting resolution.
(Read Only)pid=24580
(4) Posted by Joose Norri [Thursday, Mar 23, 2023 17:49]

I claim predecessorship; I wrote about this in an article in Suomen Tehtäväniekat ca. 2002-04. But I would be extremely surprised if there's no anticipation.

I mentioned this only en passant (but I did); what I was really interested in was capturing the King en passant in circe rex inclusive: 1.0-0-0 Bxd1 [Ke1; Ra1]
(Read Only)pid=24581
(5) Posted by Kevin Begley [Thursday, Mar 23, 2023 17:58]


You wrote about capturing riders (and a castling King by en passant) in 2002?
You are (at least partially) anticipated by the "Prise en passant universelle" condition -- this condition dates back to at least 1983 (several problems from that year in the Win Chloe database from that date -- there's also a study by Albert Kniest which dates back to 1946, which clearly intended that a pawn could capture a rider by en passant).

However, your interpretation may be different (did you want to allow the capture of two units at once, when a rider takes a unit which also crosses an enemy rider's path?).
The "Prise en passant universelle" condition allows for that. I wonder if this is what Kniest had in mind, or was this interpretation added by others who took up the condition (in 1983)?

I can't tell you if that condition allowed for the ep capture of a Circe Rex Inclusive (or Rex Multiplex) King, as the condition is not programmed for Win Chloe (the problems are there in the database, but can not be solved). We'd need to dig up the rules for this old fairy condition.
I must admit, capturing a castling King by en passant is a very interesting idea!

So far, I see no evidence that my (more modest) idea is anticipated.
I would not allow the en passant capture of riders (or castling Kings) -- I only envision allowing other units to capture a pawn by en passant.
I must admit, my idea probably has less exciting potential for problem chess, but it makes more sense to me (we can all understand the reasoning for allowing a double-stepping pawn to be captured on the first step of its journey -- pawns are not like riders! -- but I see no logical reason to exclude other units from making the same capture that a pawn can make).

When I came up with this idea, I was thinking about en passant rules for Berolina Pawns in contact with Orthodox Pawns.
I am never certain about that rule. I always need to consult a problem database (or a solving tool).
That's when it occurred to me that all units should be able to capture en passant (if they have a legal capture with the pawn going forward a single square, all units should have the equal right to capture as a pawn may capture which attacks the same square).
(Read Only)pid=24582
(6) Posted by shankar ram [Thursday, Mar 23, 2023 18:09]

The oldest example in the WinChloe database is dated 1946!
By A.H.Kniest. Who else? ;-)

And as for capturing a King e.p using Circe Rex Inclusiv:

Thomas BRAND
Markus OTT
Theodor TAUBER
feenschach 1983
(= 9+8 )
Prise en passant universelle
Circe Rex Inclusiv

1.0-0-0! Zz
1…B×d1(Rh1),Bf1 2.Se1‡
1…B×d1(Rh1,Ke1) 2.0-0‡
1…B×f3(Sb1) 2.Rg1‡
1…g4 2.S×h4(Bf8)‡
1…K×f3(Sb1) 2.Bd5‡
(Read Only)pid=24583
(7) Posted by Kevin Begley [Thursday, Mar 23, 2023 18:32]

OK, wait...
I now understand that Joose is NOT saying he invented this condition (he wasn't around back in 1946, and I'd be surprised if he was composing in 1983), only that he wrote about this (old) condition (in 2002-2004).
I had misunderstood his remark when claiming "predecessorship."

But, if he's claiming "predecessorship" on behalf of somebody else, it's not against the condition I proposed -- in fact, it would appear he's claiming it against the very same condition (which Shankar mentioned).

I fully admit, somebody (almost certainly Kniest) has partially anticipated my idea; but this is no more than a very slight anticipation -- my idea is very, very different.

The condition I described is so much closer to orthodox, it's not immediately obvious how it makes a dent.
The advantage of my suggestion, as I see it, is primarily to simplify the rules for en passant capture of pawns when they come into contact with fairy units.

If a unit is partially made up of a Pawn (e.g., Dragon = Knight + Pawn), the en passant capturing rights are not immediately clear.
Why should a Berolina Pawn have the right to capture an Orthodox Pawn by en passant, but not a unit which is half Orthodox Pawn?

So, my fairy condition suggestion provides a way to simplify all those issues. While I admit, this resolves only half the uncertainty (which fairy units can make an ep capture?), I expect the other half of the question (which fairy units are capturable by ep?) is answerable by whether the unit can promote -- by default (unless otherwise stated), only a unit which can promote should be capturable by en passant.
(Read Only)pid=24584
(8) Posted by Hauke Reddmann [Friday, Mar 24, 2023 12:51]

Something similar(!) is known among Hamburg chess kiddies,
but I don't know the exact rules.
(Read Only)pid=24588
(9) Posted by Kevin Begley [Friday, Mar 24, 2023 16:23]

Curious. My idea is partially anticipated by Kniest, and is perhaps further anticipated by "Hamburg chess kiddies."

And in that good company, I may declare: "Ich bin ein Hans Berliner."
(Read Only)pid=24589
(10) Posted by seetharaman kalyan [Monday, Mar 27, 2023 04:54]

Castling is prohibited when the intermediate square is attacked. So I don't see the occasion for enpassant capture of a castling king.
So, the Thomas Brand..... problem has no solution
(Read Only)pid=24593
(11) Posted by Joost de Heer [Monday, Mar 27, 2023 08:04]

3.9.1 The king is said to be 'in check' if it is attacked by one or more of the opponent's pieces, even if such pieces are constrained from moving to the square occupied by the king because they would then leave or place their own king in check.

This means that in Circe RI, if we interpret the rules literally, in the position wKe1, bPe2 the move 1. Kd1 would be illegal, as this square is attacked by the pawn. In Circe RI of course, this is a perfectly legal move. So when dealing with Circe RI, 'attack' has a different meaning than in the orthodox sense.

So in the Circe RI interpretation, d1 is not attacked by the bishop.

And this also teaches us not to apply the orthodox rules literally to fairy conditions (or was that stipulations?).
(Read Only)pid=24594
(12) Posted by Nikola Predrag [Monday, Mar 27, 2023 13:43]

"So in the Circe RI interpretation, d1 is not attacked by the bishop."

This interpretation looks upside down. d1 IS attacked, as well as the respective castling King. Obviously, after 0-0-0, both wK and wR are captured due to that attack.
But the concept of check is altered in Circe RI, it requires both the attack and occupied rebirth square.
At least, that's what I understood from the play, not knowing the actual definition.
(Read Only)pid=24595
(13) Posted by shankar ram [Monday, Mar 27, 2023 16:56]

While on the subject of the illegality, or otherwise, of castling in fairy chess ... how many know that in madrasi, Popeye allows castling with a paralysed rook? ;-)
(Read Only)pid=24596
(14) Posted by Valery Gurov [Monday, Mar 27, 2023 21:17]

@ Shankar Ram: Рокировка - это ход короля, поэтому при парализованной ладь рокировка может быть возможна :-))
(Read Only)pid=24597
(15) Posted by shankar ram [Tuesday, Mar 28, 2023 01:56]

WinChloe этого не позволяет!
(Read Only)pid=24598
(16) Posted by Joost de Heer [Tuesday, Mar 28, 2023 07:54]

While in Einstein chess, castling turns a rook into a bishop, so there it's not a king move but a king+rook move...
(Read Only)pid=24599

No more posts

MatPlus.Net Forum General New (or already existing) Fairy Condition?