The Chess Variant Pages

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

Comments/Ratings for a Single Item

Later Reverse Order EarlierEarliest
Two Move Chess. Designed to alleviate the first move advantage for White using double moves, while retaining the tactics of international chess.[All Comments] [Add Comment or Rating]
Ted Larson Freeman wrote on 2021-09-25 UTC

I have posted Python code that validates games of Two Move Chess on Github at

This code leverages the python-chess module. In addition to validation, it was used to generate the board images used in the examples in these rules.

Ted Larson Freeman wrote on 2021-08-30 UTC

Thanks for the explanation of how ChessV works, Greg. I have rewritten the description of the rule for threefold and fivefold repetition to take into account the feedback below. I believe the way ChessV handles it will work, with just one thing to clarify. The move status in Two Move Chess is one of these:

  1. It is the beginning of a turn that is not a response move turn
  2. It is half way through a two move turn
  3. It is a response move turn

In ChessV, you might have to do something to ensure that otherwise identical board positions with these distinct possibilites for move status are not considered the same.

For simplicity, I would like to leave the 50 turn and 75 turn rules as they are. The 50 move rule in international chess considers 100 board positions, for example, and I think it would be unnecessarily confusing to have a "100 position rule" and a "150 position rule" in Two Move Chess. I agree that these rules will almost never actually be invoked in Two Move Chess.

I am also open to persuasion if this proves to be a stumbling block in ChessV.

Greg Strong wrote on 2021-08-29 UTC

Here's how the double-move variants in ChessV currently work.  (I don't think other variants have specified how 50-move and draw-by-repetition should work.)

Any position can be a repetition, including the first move of two, but will only match positions with the same move status.  So the first move of two will only match another position with the first move out of two (and same castling, en passant privs.)

For the 50-move rule, every individual move counts.  So a two-move turn could add two to the count.  (Internally, ChessV is built for individual moves.  The double-moves are accomplished by altering when the side-to-move flips.)  I'm not sure how hard this would be to change.

I also think, in general, that repetitions and 50-move draws aren't likely to happen in double-move variants.  Even with the one capture per turn limit, the game should still be pretty violent.

Ted Larson Freeman wrote on 2021-08-29 UTC

(Note: I have edited my answer below to improve the example.)

You make an excellent point, H.G. I will adjust the rule for threefold repetition to state that the positions are only considered equivalent if the entire state of the game is the same (e.g. if a position occurs half-way through a two move turn for Black and the same position is repeated later at the start of a turn for Black, those two positions would not be considered equivalent).

Your second point is thought provoking. The existence of the threefold repetition rule in international chess does provide a target for a player who is behind to still achieve a draw. My goal with Two Move Chess is to alter the game as little as possible while eliminating (or at least reducing) White's first move advantage, so I think we need to keep the rule.



Ted Larson Freeman wrote on 2021-08-29 UTC

Michael, you are nearly correct. If a player is reduced to a bare king and is not in check, the only way it would not be stalemate on his turn would be if the king could capture an opponent's piece, or if his opponent has just taken a single move turn and the king still has a legal move available.

See Example 4 in the Notes section. Also, consider the position at the end of the game in Example 5. If play were to continue it would lead to stalemate as soon as Black captures or blocks the advancing pawn (unless White is in a position to immediately capture Black's pawn or bishop).

Because of this, players need to carefully consider endgame positions to make sure not to inadvertently end a game in stalemate.

H. G. Muller wrote on 2021-08-29 UTC

I think it is a mistake to not take the turn situation fully into account for the repetition rule. For one, it is illogical: we do condider the same board position not a repetition of the other player is on move. Game-theoretically a position is different when the side to move can still make a double move, or must do his second move. So why consider them the same?

In general it is bad to declare draws through an artificial arbitrary rule in positions that could be won without that rule. It seems not far fetched at all that a position is vastly better when you can do a two-move turn. E.g. the first move could discover a Bishop attack on the Queen, which you then take with the second move.

Michael Nelson wrote on 2021-08-29 UTC

Correct me if I'm wrong, doesn't rule 3 imply that a player with a bare king is stalemated, even if he has legal moves, since he's not allowed to move his king twice in the same turn?

Ted Larson Freeman wrote on 2021-08-28 UTC

I've updated the rules, separating the rule for threefold repetition from the the one for a 50 turn and a 75 turn rule. I've also made the change I mentioned below for the rule on notation. (One could argue that a description of notation doesn't really belong in the rules section, but I'll leave it there for now.)


Ted Larson Freeman wrote on 2021-08-28 UTC

Thanks for your review, Greg. You have understood the design perfectly, and I would be delighted if you added Two Move Chess to ChessV!

After some thought, I see that the rule for three-fold repetition should be handled differently than the 50 move and 75 move rules. Let's take them one at a time.

For threefold repetition, let's consider the game to be a sequence of board positions. A two move turn results in two positions, while single move and response move turns result in only one. It doesn't matter what type of move produced each position--a player may claim a draw by threefold repetition simply by noting that a certain position has been reached for a third time (with the usual considerations of whose turn it is, what castling rights exist, etc.).

For the 50 move and 75 move rules, I think the simplest resolution would be to change them to a 50 turn rule and a 75 turn rule in Two Move Chess, with the reminder that a turn and a move are not equivalent.

I am open to other proposals, for example these rules could be replaced by a "25 turn rule" and a "38 turn rule", but I am inclined to think that would introduce unnecessary confusion.

Ted Larson Freeman wrote on 2021-08-28 UTC

Thanks, Fergus. I see your point about notation. Upon reading it through again, I think it would be more clear if the rule explained up front that it is just about notation. I'd like to update the first sentence to say, "When recording a game, a two move turn must be clearly distinguished from the moves of the other player."

Greg Strong wrote on 2021-08-27 UTCExcellent ★★★★★

This is an interesting and logical approach to tackling how to have a double-move variant addresses pesky rules like check and en passant.  They always require special-case rules to address, and how it is addressed here "feels" right to me.  Marseillais Chess handles the check thing fine, but falls down on how en passant is handled.  You seem to have neatly solved that, too.  I also like how you are limited to one capture per move and cannot move the same piece twice.  This also helps to preserve the strategical similarity to orthodox chess. I guess Marseillais is more of a "let's make double moves and we'll end up with an interesting but totally different game."  Originally, it wasn't even "balanced" (white started with two moves.)  This is an ambitious attempt to add the property of double moves games being "balanced" while changing as little else about the game as possible.

Extra Move Chess also provides similar benefits.  You can make a second move, but don't have to, as long as it doesn't capture or move a piece that just moved.  If you make a second move, it can be a two-space pawn move (which a first move can't, except for white's first move of the game.)  This also neatly solves check and en passant.

I'd like to add this to ChessV.  I think it's doable but I need to think some things through.  The thing I see that most concerns me is this:

Each position created by a two move turn is included in the count toward a draw by threefold repetition, or toward a draw by the Fifty move rule (or the Seventy-five move rule)

If I understand this, it would be difficult to implement and doesn't make a lot of sense to me.  Are you saying that any move in a single move turn or responsive move turn should not count towards the 50-move rule, nor should they be counted toward any potential repetition?

Fergus Duniho wrote on 2021-08-26 UTC

I modified the rule regarding notation, because the rules of a game cannot specify the notation to such detail unless they are for a specific tournament or something like that. Some games are recorded with different kinds of notation, and software like Game Courier or Zillions-of-Games will use their own style of notation for all games.

I unhid this page, but I did not examine the notes section in detail.

Ted Larson Freeman wrote on 2021-08-21 UTC

Thanks for updating the name of this variant in the database, Fergus. And thanks to everyone for the additional comments below. I have added a paragraph to the introduction to explain the motivation of the design. I hope this will make it easier to understand the reasoning behind each of the rules.

This submission is ready for another review. My hope is that it is now ready to go!



Fergus Duniho wrote on 2021-08-19 UTC

I have now updated the ItemID and the game's name in the database.

Joe Joyce wrote on 2021-08-19 UTC

Actually, I agree with the idea that one page more or less per person makes a noticeable difference. I meant literally 1 page, kept for the specific comments and discussion, done by the few people who do wind up with a separate revised version. And it's kept for the discussion.

It's a personal thing with me. I hate seeing information lost. I argued with everyone from John Smith to Derek Nalls about deleting games. I lost both those particular arguments, and lament it. Both had interesting stuff that they later decided didn't live up to their standards.

But I admit to being surprised at how many game courier settings files I have. Some of them can go, being early attempts at something I did better or gave up on. Some are non-chess prototype designs used for playtests of other people's games. Game Courier can handle a lot of abstracts besides chess variants. Should they go, too?

Greg Strong wrote on 2021-08-19 UTC

One page more or less won't make that much difference to this site.

I completely disagree with this.  Yes, there is a lot of clutter that should be cleaned up.  And, from time-to-time I do make progress on that.  But step 1 is not to make the situation worse.

Fergus Duniho wrote on 2021-08-19 UTC

Patience. I will get to changing the name.

Joe Joyce wrote on 2021-08-19 UTC

I haven't been following this conversation, but I was in the same situation. Create the revised game using the new name but put links into it to the original page, and edit the original page to link to the revised rules, with notes that the original is being kept for the history and comments. One page more or less won't make that much difference to this site. And you haven't disappeared all that work.

Ted Larson Freeman wrote on 2021-08-17 UTC

Actually, maybe it's better to let the comments go. I can create a new submission with the name "Two Move Chess" and delete this one. Earlier I was thinking we should preserve the comments to give credit to both of you for the clarifying discussion and for the suggested name change, but I also don't want to make extra work. What do you think?



Ted Larson Freeman wrote on 2021-08-14 UTC

After some thought, I would like to change the name of this variant to "Two Move Chess." If it's not too much trouble, it would be nice to rename it while preserving the comments.

I have gone over everything again and made additional minor edits that I hope clarify the rules. I have also added an example of stalemate in the Notes section, and added a note at the start of one of my earlier comments that was both unclear and incorrect, letting people know that they should disregard that particular comment.

Thanks again for all of the comments, questions, and suggestions below!


Fergus Duniho wrote on 2021-08-13 UTC

I am still pondering the question of the name. If I change the name, will the URL change? And if so, would we lose these most excellent comments?

The URL will change, but I can manually update the database to keep the comments with the page.

Ted Larson Freeman wrote on 2021-08-12 UTC

I have just edited the rules. I split them up into more numbered items, and (I hope) added clarifying language. Thank you for the comments and questions. Please take another look.

I am still pondering the question of the name. If I change the name, will the URL change? And if so, would we lose these most excellent comments?



Ted Larson Freeman wrote on 2021-08-12 UTC

That's a great suggestion! I might be more inclined to "Two Move Chess", but I like it either way. Thanks for the suggestion.

Ted Larson Freeman wrote on 2021-08-12 UTC

Okay, let me see if I can explain my thinking here. The terms "turn" and "move" are not interchangeable in this game. The rule for stalemate could be stated most simply as: the game ends in stalemate if at the start of a player's turn he is not in check and cannot complete a legal turn.

To break that down, assume that the player is not in check. At the start of his turn the game is in one of two states:

  1. His opponent has just made a single move. In this case he must make a response move. Since this is just one move, the possibility of stalemate is evaluated exactly as in international chess.
  2. His opponent has not just made a single move, so he may take a two-move turn (if possible) or a single move turn (if possible). Keep in mind that the set of moves that could comprise the first move of a two move turn is disjoint from the set of moves that could be played in a single move turn. If none of the moves in the first set can be followed by a second legal move, and the second set is empty, the game ends in stalemate.

When you ask above, "what if the first move is one that would normally prohibit a second move, such as a check, a promotion, a capture, or a double Pawn move subject to en passant?" you are asking about a possible single move turn. It is not stalemate if any such move is available.

I am trying to be succinct, and yet it seems I need to provide more detail here.

Fergus Duniho wrote on 2021-08-11 UTC

By rule #1, each move of a two-move turn must be individually legal for the position on the board at that moment--it does not matter whether a second move could remove any checks caused by the first move. So if there is no legal first move, it is stalemate.

I think you're saying that the first interpretation is incorrect. If there is no legal first move, it is stalemate.

Would this wording be more clear?

The game ends in stalemate if at the start of his turn a player is not in check and cannot complete either a legal two move turn or a legal single move turn.

No, it just applies De Morgan's theorem to what you originally wrote, leaving the original ambiguity intact. Since the third interpretation could be put succinctly and unambiguously as "The game ends in stalemate if at the start of his turn a player is not in check and has no legal first move," I think you favor the second interpretation, which I think is more clear when worded like this:

The game ends in stalemate if at the start of his turn a player is not in check and either has no legal move or, if he does have a legal move, cannot complete a second legal move after the first.

It seems that you favor the second interpretation, which is that the player must be able to make two legal moves to avoid stalemate. However, what if the first move is one that would normally prohibit a second move, such as a check, a promotion, a capture, or a double Pawn move subject to en passant? Would it still be checkmate in that case?

25 comments displayed

Later Reverse Order EarlierEarliest

Permalink to the exact comments currently displayed.