mangue: Thank you ! It is true that these are very good stats, but Pythagoras is also right that it is too soon to tell what will happen later. I am still concerned about the possibility that there may be too many drawn endgames in the future.
For the moment, the game seems to work even better than I had thought

Pythagoras: Sorry I misread you ! I thought that you disagreed with my promotion rule because I missed the "only" in your first sentence. A more careful reading switched the meaning of your sentence completely :-)

wetware: Good old Compromise Chess is also a nice game, it is indeed the variant closest to Ambiguous Chess. It looks like it is usually played with the rule that if you have only one move to get out of check, your opponent has to accept it. However, I prefer the "wild" rule that in such a case you lose the game (because your opponent will capture your king).

Pythagoras: Cool about castling :-) I will stick with the promotion rule as it is unless it becomes clear that too many endgames are drawn, in which case having the promotion by the owner would be a solution. But it can require a three-step move instead of a two-step one (player A points the promotion square, player B chooses a pawn, player A chooses the promotion piece).

Pythagoras: Thank you ! Ambiguous Chess was born in January 2005, published on my web site, on chessvariants.com and some months later in the Variant Chess magazine.
Although it is a very simple idea, it came to my mind in a very roundabout way. By telling it I am a bit afraid to bore everyone with a long post, feel free to stop reading right here !

I read in the Encyclopedia of Chess Variants about a strange variant called "Unambiguous Three-symbol Chess", where the only moves allowed were those who could be written in three symbols in the English game notation. It gave me the idea of "Unambiguous Chess" where the only moves allowed are those who are determined by the arrival square alone (all ambiguous moves simply forbidden). This was probably a good idea for a problem stipulation (problem composers like strong constraints, which make it easier to avoid dual solutions), but not a good one for a playing variant (actually Unambiguous Chess is a forced win for White under the no-check rule).
Moving from Unambiguous Chess to Ambiguous Chess is not a very big step, still it took me something like five years and happened only when for some reason both Unambiguous Chess and the game of Quarto came to my mind in the same time - Quarto is an alignment game where you choose the piece which the opponent has to place on the board.
There were only two rules which were not obvious to settle, the castling one (argh!) and the promotion one. At first I had the "normal" promotion by the pawn's owner, but then a friend of mine rightly pointed out that it was more logical that the piece was chosen by the opponent, as the different promotions were different moves to the same square. I hope you will agree with this one :-)

Walter Montego, Pythagoras: I have already answered about castling somewhere lower on this page (May 16). It was the hardest choice and I am not 100% sure that I got it right, especially now that I have heard of four of you who are surprised that it isn't allowed.
What I am more sure about is that if one wants castling allowed, one should consider it as an two-piece unambiguous move that the opponent cannot replace by a rook move (because there is no point in having a special move if it is a disadvantage for the player at move), and that castling should be allowed only when it would be legal in normal chess (no castling under or through check). Like this, allowing castling would slightly complicate the rules but still be completely OK with me.
Maybe it is the opening theory (to be built) which should tell. If castling adds more variety to the games, it could be an excellent thing to allow it. If it turns out that castling is so strong that every games starts with 1.f4, 2.Nf3, 3.g3, 4.Bg2 and 5.0-0 then it is good that it is forbidden. Unfortunately, the game is still too new to tell.
For now, every other opinion on it is welcome !

Yes, there are examples, annotated games and tips here :
http://www.pion.ch/echecs/variante.php?jeu=ambigus&changer_langue=E
The tips should not be taken too seriously, everyone is still very much in learning phase.

Before the big July tournament I propose this one to you :
Ambiguous training

Play Ambiguous Chess for a 1-year rook membership !

tbart: No, it is the official rule for "Crazyhouse" (and Bughouse) but not for "Chessgi" (=Loop Chess) which is probably the oldest of the three games.
It was already discussed here, because all Bughouse players (me included) are acquainted to promoted pieces reverting to pawns, but there is no reason why it should be the same in Loop Chess.

Marfitalu: Good question, there is no outstanding reason for it, the main reason was simplicity.
I thought about three possibilities :
- The more natural one seems to consider castling as a king move. Then castling is ambiguous can always be replaced by a rook move, and the right to castle would in fact be a disadvantage for the player who can castle. For this reason I don't like this one.
- Consider castling as a two-piece move, hence unambiguous. This is now perfectly sound. But we would also have to state whether it is possible to castle under or through check. As the straightforward set of rules state that check does not exist and that the goal is to take the king, one would have to allow that. Personally I don't like at all the possibility to castle under or through check, like it is the case here in Atomic and Extinction. But this is probably a matter of taste.
- Banning castling is simple, clear-cut, easy to implement and can be phrased in very few words. That is what I like about it :-)

PS Someone composed a retrograde analysis problem of Unambiguous Chess, a variant I invented before Ambiguous Chess, where ambiguous moves are simply illegal (this variant is less playable but fun for problem composition). He asked me whether he could state that castling was an allowed unambiguous move, as his problem needed it, and I told him it was OK with me.

SMIRF Engine: To say the truth, I had initially also made the same request for 10x8 variants, but I have already received a lot more that could fit into the Encyclopedia.

WhisperzQ: Thank you !

SMIRF Engine: Link on this board or PM to me.

Good news for all of you who sent me games for the second edition of the Encyclopedia of Chess Variants : after the sad death of the author David Pritchard, it would have been a real pity that his huge work stayed unfinished, and indeed the task has been taken over by John Beasley.
He is the editor of the Variant Chess magazine and is undoubtly competent enough to continue David's work. The most interesting games might find there place in the magazine as well (on of them, with my annotations, has already been published there).
So, good-quality variant games are still welcome !

Pafl: Well your game is fun, doesn't look "pathetic" at all :-). The other one is stranger, with Black responding to a "check" by putting himself in a double check, and then winning the game !
But none of them really features play with KQRBN v KQRBN. As long as there is another piece on the board, this extra piece has unrestricted attacking possibilities and I would not call the position an endgame. However it is of course just a matter of words.

Pafl: Good question ! Of course a chess-like endgame is impossible in Extinction, but we could choose to call endgames all positions where both camps have only unique pieces except of pawns (that is, KQRBN v KQRBN). I didn't reach any here yet, but it doesn't seem an unlikely issue at all.
Now, many of the positions just after the last piece trade will probably be either simple forced wins, either dead draws with all pieces having to stay on their posts in order to avoid a breakthrough. Has anyone played an interesting "endgame" here ?

Anencephal: At least I've got White, so I don't have to look into books. Damn, it looks like my opponent already has a draw by Nb8-c6-b8 :-)

Pythagoras: I also hope that I won't get any Maharajah game. But what is wrong with Knight Relay ? I find it to be a lot of fun.

Fencer: I like the idea, this will be my first BK tournament !

I have sad news to report : David Pritchard, for whose second edition of the Encyclopedia of Chess Variants I collected games here, has died accidentally on December 11. A big part of the book was already written, but I have of course no idea whether someone will finish the huge work that he was doing. I will forward here the information I might get about that.

Among the great games collected here, a few of them may still be published in another way, either in the magazine "Variant Chess", either on my website. Again, if it happens I will inform the players who provided me the games.

Thank you one more time to all contributors.

Walter Montego: You are right, for the reader to get a right grasp over Dark Chess, at least one game position should be shown in all three views.

grenv: Thank you for that game, it looks really great, and I advice everybody to ask for the PGN !
Total agreement with you, to make a game private is IMHO a feature which should not exist.

grenv: Thank you ! I am all but expert in Dark Chess, but yes, I think that is what we are after. Unfortunately the Atomic game is a private one, so I would need the PGN score (here or by PM).

Walter Montego: Sure that the king promotion would be a definite diagram in the book ! I would like to see it. Good Dark Chess games are also called for, I doubt that David Pritchard has any of them yet except for the one published in Variant Chess by the inventor.

WhisperzQ: All variants played here ought to be covered. Variants where games are less likely to be included are the ones where it is most easy to get good games elsewhere, I would say Fischer Random, Xiangqi and Shogi. And in games where only the startposition is changed (like Fischer Random again), miniatures will probably be preferred, as long games will just look like normal chess.

Hi variants players,

I am writing this for David Pritchard, who is close to finishing the second edition of his famous book "The Encyclopedia of Chess Variants". He is always looking for good sample games, and when I offered him to make this call on Brainking, he found it was an excellent idea. There are many variants played here and the turn-based type of play makes for better games than blitz.

The present message is a call for "best games" by the top BK players - I have no precise idea about what "top" means, but probably it should be thought as something like having been in the top 5 BKR in the variant played. Best games can be won, but also drawn or lost games. If you are interested that one of your games shows in the Encyclopedia of Chess Variants, just submit to me your one or two best games in each variant. All I need is the game id (or the link to the game, which looks like http://brainking.com/en/ArchivedGame?g=123456), given either in a PM to me or on that board.

The time delay for this is about one month. After that, David (and me!) will still be interested in seeing your best games, but he will not be able to include them in the book any more.

Thank you by advance for your inputs !

02i: I am no specialist of the field so I could be wrong, but from what I read it looks like about all games of the type "one against all" are forced wins for the player with the army - the classical example being "Fox and the gooses". The reason for it seems to be that if one of them wasn't, it would probably be a straightforward win by direct attack for the lone piece.
The lone piece being much easier to conduct than a set of pieces, such a game is more playable if it is unbalanced in favour of the army.

Marfitalu: You can castle, but your king would stay on c8 and your rook would go to d8, as in the usual long castle. So you would need to move the rook currently blocking d8 first.
Do you remember the rule which was hard to translate ? Here the "longest range", which must be free from other pieces, is a8-d8.

andreas: Your proposal is a nice addition to randomization !

I also agree with Random Atomic Chess and Random Suicide Chess, the games as they are suffer a bit from a narrow non-losing path in the opening. But how to generate random position which will not be forced wins for White ?

If you have any other question on Anti-chess endgames, fear to be called stupid and can read a tiny bit of French, the endgame section of my webpage (http://www.pion.ch/Losing/endings.html) will probably answer it.
[Made link live ... WQ :]

rabbitoid: This actually has been called Recycle Chess and is a very cool variant : http://www.chessvariants.org/difftaking.dir/recyclechess.html . Unlike Loop it can be played over the board without problem. Try it !

I vote that Ra5-a5 is a perfectly legitimate Cylinder Chess move. Do we know whether BK allows it ?

WhisperzQ: Sure, no problem ! There was also a very fine endgame database server (up to 4 pieces) by Lenny Taelman, but this one went down recently :-(

Hmm, the links below should read without a dot at the end, so http://www.pion.ch/Losing/LCLIT3AR.html and http://catalin.francu.com/nilatac/book.php

Just to make some facts straight : the game of Anti Chess, more often called Losing Chess or Suicide Chess, is more than 100 years old, and the rules used here are the standard ones. For some more details, here is an historical research by John Beasley : http://www.pion.ch/Losing/LCLIT3AR.html.
The game has been studied a lot, by hand and by computer. Here is an opening book which should avoid you a lot of early trouble : http://catalin.francu.com/nilatac/book.php.
My personal homepage has tons of information too, but available only in French language.