Check out Symmetric Chess, our featured variant for March, 2024.


[ Help | Earliest Comments | Latest Comments ]
[ List All Subjects of Discussion | Create New Subject of Discussion ]
[ List Earliest Comments Only For Pages | Games | Rated Pages | Rated Games | Subjects of Discussion ]

Single Comment

Metamachy. Large game with a variety of regular fairy pieces.[All Comments] [Add Comment or Rating]
📝H. G. Muller wrote on Fri, May 1, 2020 01:28 PM UTC:

capture zone - n. For a given piece, those squares on which it has the ability to legally capture a piece should the opportunity arise. This may include squares with nothing on them that it may presently capture. What matters is that if an enemy piece did move to a square, the piece could capture it. See threatened.

This is still no good. If the piece was pinned, he could still not legally capture an enemy piece that moved to that square. Unless you consider 'opportunity arises' to also mean "when it doesn't happen to be pinned, and when the that player is not in check". But then it could mean anything, e.g. why not "if there had been no pieces in its path blocking it", or "if it had been on a more suitable square". Involving vague, undefined concepts like 'opportunity' just makes things worse.

The FIDE rules do define 'under attack' as follows:

3.1.2

A piece is said to attack an opponent’s piece if the piece could make a capture on that square according to Articles 3.2 to 3.8.

3.1.3 A piece is considered to attack a square even if this piece is constrained from moving to that square because it would then leave or place the king of its own colour under attack.

It seems the definitions in the glossary have removed the reference to articles 3.2 to 3.8 (which we have to generalize here) and replaced it by the the term 'legally'. This is where the trouble starts, because 'legal move' in the FIDE rules already has a specific meaning (involving check), while what the FIDE rule meant to say here was "would conform to the rules of motion for the individual pieces" (ignoring check, which is only mentioned in 3.9). And we appear to be missing the all-important qualification 3.1.3 in the definition of 'under attack'.

I am not familiar with the term 'capture zone', but it seems you want it to mean "the set of squares that the rules for moving the piece (given the board population elsewhere, but ignoring any check rule) would allow it to capture an enemy on.

Note that the concept 'attack' in the FIDE rules only serves the purpose of formulating the check and castling rule, for which purpose captures that expose their own King are also valid attacks. It is never used in connection with non-royal pieces. This definition of 'attacked' therefore can deviate from the colloquial meaning as used by chess players, who will say things like "my Queen is under attack by a Pawn". It is IMO questionable if they would say that when the Pawn was pinned. But apart from what I think, far more serious is that the FIDE concept of 'attacked' is at odds with the definition in Xiangqi. There the rules do apply the concept 'attacked' to non-royals, in connection with the definition of perpetual chasing. And in this context attack on (or protection of) a non-royal always means "by a fully legal capture (recapture)". 'Attacks' by pinned pieces are not recognized as attacks. (Attacks on the King by pinned pieces are recognized as checks, though.)

Since we run a website for chess variants, I think it would be ill advised adopting a terminology that was exclusively defined in the narrow context of the 'FIDE rules of Chess', but would fail to be useful in other chess variants, amongst which the world's most played variant. FIDE rules should obviously not be binding for us, or the whole concept of a chess variant would be outlawed.

So I think the definition "has a legal capture move to the square when it would have been occupied by an enemy" would be the best definition for 'attacked', with the note that when applied to a royal, every capture would be legal. (Note that the FIDE rules also explicitly point this out when they use the word  'attacked' in 3.9.1 for describing when castling is allowed, while in fact this is redundant, as 3.1.3 already defined 'attacked' in that way!)

It can seem awkward that 'attacked' means something else for a royal then for a non-royal, and that a King would be 'under attack' on the same square where another piece would not be under attack. But in chess variants we have to deal with that problem anyway, as there might be divergent pieces that capture royals in other ways than non-royals. Again Xiangqi comes to mind, where the king attacks the opponent king through a Rook move, but not other pieces. And of course the Ultima Chameleon. It would solve a lot of problems if we distinguished 'attacks a square' (non-royal) from 'checks a square' (royal).

Note that these complications are a result of applying the concept in anticipation of a move, rather than to an actual position. In the latter case it would be obvious whether a legal capture to a given square is possible or not. But for judging legality of castling, it requires several fictions: that it is the opponent's turn, and that you have moved your King to the square in question. Without those fictions, f1 would not be under attack by a Pawn on e2, as 3.7.3 (to which the definition of 'attacked' refers) clearly states that Pawns can only make diagonal moves to squares occupied by an opponent. As we all agree that white castling is not allowed with a black Pawn on e2, we apparently agree that the rule must not be applied to the position before castling, where the Pawn had no move to f1. The FIDE rules do not specify whether the intended fiction (for the "square the King must cross") is that we should imagine (1) the King having moved there, (2) a second royal piece having been placed on that square, (3) a new non-royal piece having been placed on that square. Because in FIDE this doesn't matter. But in other variants it could. E.g. when there is a non-jumping Dababba on d1, it could not capture to f1 after (2) or (3), but it could after (1). I think the natural generalization would be (1): you cannot castle to g1 when you cannot legally move your King to f1.

That brings me to another issue in Metamachy: what if there is a black Cannon on f1, a virgin King on g1 and h1 is empty. Can the King now move to i1? In interpretation (1) it could, as keeping in the direct line of sight of a Cannon never exposes you to capture.

Anyway, while awaiting for you to make a final decision of how or whether to fix the glossary to make it conform to the Metamachy rules, I repaired the damage you had done to the Metamachy article. I think it is a very bad idea anyway to refer to a glossary where the reader has to browse through several items with long and complicated definitions to learn something that could have been said in half a sentence.