Talk:Forsyth–Edwards Notation

Standards

Anyone knows if the letters for pieces (KQRBNP) are strictly defined by this standard? If, for example, I use FIDE approved letters, will it invalidate the FEN? --217.198.225.76 22:02, 16 Nov 2004 (UTC)

Well, considering that standards are used to provide valid communication between disparate programs, using different letter than provided in the standard will mean that your FEN will not work with programs designed to read FEN formatted inputs.

From the standard itself -

---

8.2.3.2: Piece identification

SAN identifies each piece by a single upper case letter. The standard English values: pawn = "P", knight = "N", bishop = "B", rook = "R", queen = "Q", and king = "K".

The letter code for a pawn is not used for SAN moves in PGN export format movetext. However, some PGN import software disambiguation code may allow for the appearance of pawn letter codes. Also, pawn and other piece letter codes are needed for use in some tag pair and annotation constructs.

It is admittedly a bit chauvinistic to select English piece letters over those from other languages. There is a slight justification in that English is a de facto universal second language among most chessplayers and program users. It is probably the best that can be done for now. A later section of this document gives alternative piece letters, but these should be used only for local presentation software and not for archival storage or for dynamic interchange among programs.

---

It appears that very few illustrated chess positions included in Wikipedia have a Forsyth-Edwards Notation (FEN) tag. Why? A FEN tag would be very handy for copy/paste entry into user's own chess database.

Because it can and does stretch the diagram display especially in Firefox. Looking at the situation at present to see if it can be resolved somehow. ChessCreator (talk) 17:53, 31 March 2008 (UTC)

That isn't a good reason. --Zzo38 (talk) 05:00, 24 December 2012 (UTC)

---

Re: "However, the FEN castling availability encoding (KQkq) is inadequate for Chess960 positions in which there are two Rooks on the same side of the King on the back rank." According to the Wikipedia article on Chess960, linked from this very article, a starting position requirement is that "The king is placed somewhere between the rooks". Therefore FEN castling availability encoding is NOT inadequate for Chess960, and the rest of the section falls apart (there is no ambiguity, no need for Shredder-FEN, etc.) —Preceding unsigned comment added by 76.213.240.115 (talk) 11:15, 12 February 2010 (UTC)

Help! Example please

Could someone please post a NUMBERED example of a chess diagram like I've seen in newspapers for forty years. I have still no idea what grid of the board F8 or anything else refers to, and that is extremely frustrating. Just one of the things that are fundamentally wrong with the world, and why everything sucks: the impossibility to find out the simplest things even with the internet, because they assume you already know?

En-passant square when no en-passant possible

Here is the FEN after the move 1. e4:

rnbqkbnr/pppppppp/8/8/4P3/8/PPPP1PPP/RNBQKBNR b KQkq e3 0 1

Is this correct? Although a pawn has moved 2 forward, no en-passant is actually possible. Evercat (talk) 16:37, 10 June 2010 (UTC)

I think that is correct. The en passant square is listed if the last move was a two-square pawn move, whether e.p. is possible or not. Of course, it would not really hurt anything to leave it off if no e.p. is possible. From the article: "This is recorded regardless of whether there is a pawn in position to make an en passant capture."Bubba73 ^{(You talkin' to me?)}, 17:16, 10 June 2010 (UTC)

Fifty-move rule

The second introductory paragraph states, "FEN does not represent sufficient information to decide on draws by [...] the fifty-move rule." However, the fifth item of FEN is apparently "used to determine if a draw can be claimed under the fifty-move rule." I'm not a frequent user of FEN, but it seems the half-move clock is sufficient for establishing a draw by this rule. Unless someone else has reason to believe it isn't, I'll remove the relevant statement from the introduction.--TrippingTroubadour (talk) 01:31, 2 July 2010 (UTC)

I believe you are right! Bubba73 ^{(You talkin' to me?)}, 03:18, 2 July 2010 (UTC)

Can you use "-" if you don't know the halfclock? --Zzo38 (talk) 05:00, 24 December 2012 (UTC)

FEN adjustment for chess variants like Chess960

Chess960 starting position rules do not allow both rooks to be on one side of the king. "This leaves three empty squares. Place the king on the middle empty square, and the rooks on the remaining two squares." The rules of Chess960 make the following statement superfluous. "However, the FEN castling availability encoding (KQkq) is inadequate for positions in which there are two Rooks on the same side of the King on the back rank. It is ambiguous which Rook is still available for castling without knowing their initial positions." — Preceding unsigned comment added by 151.193.220.28 (talk) 15:23, 5 September 2012 (UTC)

Came across the same question myself. It was well explained here http://www.dwheeler.com/essays/Fischer_Random_Chess.html. If in the game the other rook comes to the same side of the king as the rook with castling right, then it is ambiguous. Ketorin (talk) 16:56, 1 March 2015 (UTC)

Does anyone know, which chess engine understands which castling annotation ? Otherwise everyone has to check whether the engine undestands the castling info in the fen correctly or not.Wiki lofi (talk) 18:13, 20 March 2015 (UTC)

FEN overspecifies

Interesting that FEN is actually "overspecific", in that the same position can be represented by two different FEN strings, as detailed in the "COMMENTS" section here. In particular, the two FENs below are the same position, such that FEN alone cannot be used (e.g. with hashing) for counting unique positions. Possibly this should be mentioned in the article, and the word "particular" in the opening paragraph reconsidered?

rnb1kbnr/pppp1p1p/4pq2/5Pp1/8/8/PPPPPKPP/RNBQ1BNR w kq g6 0 1

rnb1kbnr/pppp1p1p/4pq2/5Pp1/8/8/PPPPPKPP/RNBQ1BNR w kq - 0 1

--Jientho (talk) 20:15, 23 December 2013 (UTC)

It HAS to be specific. Theoretically you can come to the same position by different moves, where in one position there is en passant possible, and in the other it isn't. The information in the FEN is necessary and not overspecific. In case of castling in Chess960 it is not specific enough. That's why there are different approaches to handle it. Chess GUIs and chess engines may or may not interpret it in the right way.

Wiki lofi (talk) 18:10, 20 March 2015 (UTC)

It sometimes has to be specific. When it doesn't have to be specific, there is no reason for it to be specific. Also, you seem confused what a position is: if an en passant capture is possible in one position but not the other, then they cannot possibly be the same position. ISaveNewspapers (talk) 05:01, 15 May 2022 (UTC)

Erroneous change "c5" to "c6" invalidated FEN

A recent edit has changed c5 to c6 in error. As the move was changed, but the FEN not, the FEN example became invalid. When I looked at the article, I was not aware of such a recent change, and adapted the FEN to be correct with c6. Also as such, I assumed the article has low attention, so made a few small corrections, which are not relevant. The change to c6 is clearly in error, it was not intended to introduce the move c6, the author assumed wrongly that the move meant is c6. As such I will roll-back my changes and this commit. I will do this edit by edit, this is just to document. Dlbbld (talk) 14:14, 20 July 2020 (UTC)

Performed as mentioned, in one edit. Dlbbld (talk) 23:07, 20 July 2020 (UTC)

Halfmoves?

This article mentions halfmoves, saying they're relevant to the fifty minute rule and linking to that page, but doesn't elaborate. The fifty minute rule article does not mention halfmoves. There is no other content related to halfmoves on Wikipedia.

I therefore suggest that this article requires an explanation of what halfmoves are. Anyone care to write something up? Birdsinthewindow (talk) 01:36, 13 January 2023 (UTC)