Comments by NeodymiumPhyte
Short castling seems to break the interactive diagram.
In the rules, it states, "If it captures both Saint (Priest) and Ghost (Pirate) in one turn, it promotes to the one that was farthest away from it." in the section about the Wolf and Saint, but the diagram seems to always just promote a Wolf corresponding to the one it captured second.
Also, the Elephant doesn't have the option to defer promotion when capturing the Capricorner. I don't know why anyone would want to defer promotion in this particular case, but it seems to be allowed in the rules.
In my play against the diagram, I did come across a different situation in which one might want to promote an Elephant to a King instead of a Saint or Ghost. This occurs when one wants to deny the opponent the opportunity to get the Saint or Ghost by re-capturing. The diagram used this ability of the Elephant to prevent me from getting a Saint, and that actually saved it from almost certain defeat, though I doubt the diagram would have let that situation arise if Elephants didn't have this ability. Still, I completely didn't see it coming and went on to lose the game.
* | * | * | * | * | * | * | * | * | * |
* | ::: | ::: | ::: | ::: | * | ||||
* | ::: | ::: | ::: | ::: | * | ||||
* | ::: | k | ::: | K | ::: | ::: | * | ||
* | ::: | ::: | ::: | ::: | * | ||||
* | ::: | ::: | ::: | ::: | * | ||||
* | ::: | ::: | ::: | ::: | * | ||||
* | ::: | ::: | ::: | ::: | * | ||||
* | :c: | L | :P: | ::: | ::: | * | |||
* | * | * | * | * | * | * | * | * | * |
If a Chameleon is to Capture a Cannon Pawn, need the mount be adjacent to both the Chameleon's original Square and the Captured Piece's Square? May the Mount be a Long Leaper, and if so, in the situation shown somewhat crudely in the table above, would the Chameleon be allowed to Capture both the Long Leaper and the Cannon Pawn by going to c1? The section on the Long Leaper states, "A Long Leaper may never jump over a friendly piece, jump over two or more pieces in a row without any empty spaces between, or move to an occupied square.", but are these rules actually so intrinsic to the Long Leaper's ability that they also apply to the Chameleon when it tries to capture a Long Leaper, or are they simply observations about the Long Leaper's abilities? I'm sure there are other ambiguities like these regarding the Chameleon, but I haven't thought of any more explicitly.
edit: corrected notation
While playing a game with an internet friend of mine, she revealed that she had misinterpreted the rules regarding edge squares. After the game, we both agreed her misinterpretation was better than the actual rule. She thought that the rules were roughly "you need to capture to get on the edge, and you need to capture to move while on the edge, either to stay [on the edge but move to a different square on the edge] or leave". I think that this is much simpler than the rather complicated current rules and serves the purpose of restricting pieces from edge squares almost as well. No one would want to put a piece on the edge if it meant they couldn't easily get it back. She also pointed out that the Swapper could be powerful with her rules because if it were on the edge, it could threaten to swap with an Immobilizer, and that Immobilizer would be stuck. I think that these rules should be considered. Also, a version where pieces could move to the edge without capturing but had to capture to move within or off the edge might be worth considering.
edit: The Immobilizer would be stuck after being swapped onto the edge, not just from the threat of being swapped onto the edge.
edit 2: removed unnecessary information
You write
And then an Alice move is legal (as usual) when it does not expose the King to pseudo-legal Alice capture.
. As far as I can tell, this isn't quite true. It must also not expose the King during the intermediate time between the piece making its move and transferring to the other board. This can be seen in this quote from Vernon Parton's work:
Fools Mate in Alician style.
(1)P - K4, P - Q4, (2)B - K2,PxP, (3)B - Q Kt5 and the black monarch is checkmated.
Here it will be seen that the move Q - Q2 (as well as B - Q2) fails to intervene as the
Q (or B) would be transferred to the other board, still leaving their King in check to the White
Bishop.Naturally, the move K - Q2 is forbidden, because the King would break the Alician
rule that he must make a legal orthodox move before being transferred. (This quick mate was
given by Mr. C. H. O. Alexander on radio.)
.
Shouldn't move option 2) include a lame dabbaba move?
I have a couple of questions about the rules.
- Can you travel back to a move in which you are in check? I'd assume not because you'd still be in check after dropping the king.
- If a man is scheduled to arrive on your move, can you go back in time instead of dropping the man on the board? What if that man happens to be a King?
- If a pawn travels forward in time, may it reappear behind the square from which it left? I know it can't reappear on the first rank, but can it reappear on the second rank if it left from the third? I ask because the bishops can only reappear on the squares they could have reached through on-board moves.
- If a man from move 5 is scheduled to appear on move 7, but a player on move 6 decides to go back to move 4, will that man still arrive on move 7 in the new timeline?
I think it makes more sense if all types of a piece's compulsion must be simultaneously satisfied for that piece to count as satisfying your compulsion.
I do agree, though, that different types of pieces shouldn't be "collapsed" just because they only differ for repetitions.
Apparently, in China, the vao is called "弩", meaning "crossbow". So I think in English, it should be called "crossbow". I am getting this information from https://discord.com/channels/634298688663191582/894530740820262932/1209401956297351258.
Oddly, if you click on a piece and then a pawn, the pawn is shown to have knight movement. I'm reporting this in case it is an issue with the diagram code more generally rather than just in this specific instance.
It appears from the Game Courier Preset that the answer is that the Long Leaper is not Captured in this situation and that only the Cannon Pawn is Captured. I suppose that I will accept this as the correct rule for now even though I don't know if this is intentional or a bug.
I don't think it's bugged. "P a6-a8;L-a8" seems to allow the Cannon Pawn to promote.
can you castle through attacked square if you still have another king
I think the given move of the Deva here is incorrect, though it could be the Maka Dai Dai Shogi page that's wrong.
The kings can jump in the diagram, which seems to be a bug.
rules not listed here (from Looney Labs webpage https://www.looneylabs.com/sites/default/files/literature/MartianChess_Rules5.pdf):
"If you have no Queens, you can create one by moving a Drone into a Pawn’s space (or vice versa) and merging them. Similarly, if you have no Drones, you can make one by merging two of your Pawns. (The pieces being merged should be removed from the board immediately and swapped for a piece of the combined size.)"
It seems both the pieces you are combining must be on your own board region.
"Your opponent may not 'reject' your move; if a player moves a piece across the canal, the other can’t move it back to the same square it was in."
"During the final stages of the game, the action slows down and may become stagnated. If any player feels the game has reached a state of deadlock, during their turn they may “call the clock.” From that point on, if seven moves go by in which neither player captures a piece, the game ends and the final score is computed. If someone makes a capture, the counter is reset."
"If the final score is tied, and one Zone is empty, the player who ended the game is the winner. But if the game ends in a deadlock with a tie, the result is simply a tie. Play again!"
If black has 1 Persian King and 1 Spartan King, is there duple-check for them?
It appears that the Black Buffalo promotes to a White Fire Demon in the interactive diagram. This should probably be fixed.
I had Fairy-Stockfish play against itself with this, and White won. So perhaps it's not as good of an example as imagined. But I think that may have been influenced by Black not being able to Promote to Queen.
I configured it myself. It is possible to train an NNUE for a custom variant for Fairy-Stockfish, so if someone were to do that, we could probably have much better version that should understand the value of the pieces and king safety properly. Unfortunately, I do not have the hardware to train it right now.
25 comments displayed
Permalink to the exact comments currently displayed.
1 thing I don't like (from an aesthetic perspective) about the above ruleset is the advisers being able to move diagonally within a single 3-by-3 layer without advancing or retreating. Logically (if the palace is made of a plus sign extended upwards, which it would be if you think about pieces being on 6-way intersections rather than in cubes), the adviser would be outside the palace when it is halfway through the movement, so I don't think that should be allowed. Also, the official xiangqi rules say "Advisers may advance or treat", so they would not be allowed to move without advancing or retreating. This quote could be interpreted as just a description of what happens in the 2-dimensional game and not a separate rule, but even then, my previous point stands about the adviser being outside the palace halfway through the movement (assuming it follows a straight line). I guess you could see the palace as a (45-degrees-turned) square prism in order to refute that argument, but seeing it that way seems a bit strange.