Chaturanga/Shatranj: Bare-King rule and Statring Position Considerations

Sort:
Arma55

Hi everyone, I recently started to play shatranj here on chess.com and found I enjoy the game quite a bit. I was interested in getting better so I went looking for more resources on the Internet, however, I hadn't much luck. The only stuff I could get my hands on was the Wikipedia page and a couple of shatranj problems.

While reading the Wiki page I found out about the bare-king rule (for those who don't know, historically, capturing all enemy pieces leaving the king alone was considered a win, unless the opponent could do the same on the next move, which was a draw).

To be honest, the game made much more sense to me once I read that, considering checkmating with the relatively weak pieces in shatranj seems to be quite difficult and requires a considerable material advantage. If the rule isn't implemented and we play shatranj as it is on chess.com right now, wouldn't that mean that almost every game played between two highly skilled players would end in a draw? For this reason, I believe leaving the enemy king bare should be adjudicated as a win. It would make the game much more enjoyable and make possible ideas and combinations that you can't find in standard chess.

While trying to learn more about the variant I stated to use Winboard and ChessV and both implement the rule. I'm currently trying to make a shatranj variant without the rule, so that I may discover more about endgames.

On this same note, it seems like the "standard" starting position for the kings is the d file for shatranj (and this is also the setup used by Winboard and ChessV). However, the Wikipedia page claims that it wasn't uncommon at all to switch king and ferz. Does this variation in setup cause any change in gameplay? I don't think so but I was curious to hear your opinion on the matter. Even if, should we try to make the variant on the site as historically accurate as possible?

I'm curious about your take, let me know your opinions.

Arma55
MISTER_McCHESS ha scritto:

-1 I don't really think you could change the rules of a game that's been around for 1400 years

Exactly. The bare-king rule IS standard. Chess.com does not implement the rule however, making the game less enjoyable in my opinion.

tacticspotter

Eh the rule is in already...

tacticspotter

some people are wondering about bare king endgames being a win

Arma55
tacticspotter ha scritto:

Eh the rule is in already...

If I take my opponent's last piece I don't get a win though. How is the rule already in the game? Did I miss something?

tacticspotter
Arma55 写道:
tacticspotter ha scritto:

Eh the rule is in already...

If I take my opponent's last piece I don't get a win though. How is the rule already in the game? Did I miss something?

wait so it only work in some games?

Arma55
tacticspotter ha scritto:
Arma55 写道:
tacticspotter ha scritto:

Eh the rule is in already...

If I take my opponent's last piece I don't get a win though. How is the rule already in the game? Did I miss something?

wait so it only work in some games?

For what I've seen, it doesn't work at all. I'd really like some input on the matter from an admin since to me it looks like a detail that could make or break the game. @BabYagun maybe?

tacticspotter

https://www.chess.com/clubs/forum/view/chaturanga-bare-king

Arma55

I saw that thread but I think that's a bug or something, because in my experience if I couldn't mate or flag, even if the opponent's king was bare, the game would be declared a draw for timeout vs. insufficient material. Notice how the game isn't immediately a win after black captures white's last piece.

spacebar

What is the exact rule/condition? I will gladly add that

if one player has a lone K, and the other player has 2 or more pieces, and the lone K cannot make a capture, then it's a loss?

Arma55
spacebar ha scritto:

What is the exact rule/condition? I will gladly add that

if one player has a lone K, and the other player has 2 or more pieces, and the lone K cannot make a capture, then it's a loss?

@spacebar, thanks a lot for your help!

For what I know, the rule is quite simple: "If one player can capture all enemy pieces, leaving the king bare, that player wins the game. If, however, the opponent can do the same thing on the very next move, then the game is declared a draw".

To summarize my arguments in favor of this rule:
- It seems like this was an extremely common way of winning a game, probably the most common (after maybe winning by resignation?)

- It would make endgames much less confusing and way more enjoyable. I'm an endgame guy and I don't get any sort of enjoyment from shatranj when it comes to this department. The pieces are just too weak and don't cut it as a threat to the king, which is one of the most powerful pieces in the variant. Even if I have sufficient material to mate, the endgame is tedious, and boring since it takes forever for the pieces to move from one part of the board to the other.

-If there's no bare-king rule, almost every game between two skilled players (or really any two players of the same strength) would end as a draw. This is based on the fact that two players of the same strength will usually have roughly the same material, therefore making it impossible to reach the material advantage necessary to win.

-It gives shatranj a new level of complexity that is nowhere to be found in traditional chess or many other variants. As an example, rook vs 2 pawns or knight vs 2 pawns would be MUCH more difficult to win or draw for the side with the material advantage. Or in the middlegame, a player might choose to give up the exchange for numerical superiority.

- The variant would be much easier to study and enjoy thanks to shatranj software already available like WinBoard, ChessV, or FairyStockfish.

On the same note, both the current and "standard" shatranj initial setup are acceptable, historically speaking, but almost everywhere shatranj is played with kings on the d-file. I would understand if you preferred to keep them on the e-file, maybe to keep some resemblance to modern chess, but it might have some impact on the gameplay and it would be more complicated to use shatranj software (d-file setup is used everywhere basically).

I know almost nothing about writing code, so I'll leave that to someone who knows what they're doing, but I'll still write down an idea. So basically from a programming standpoint, my suggestion to implement the rule would probably be:
if (one player has a bare king) -> start countdown-to-win
countdown-to-win: 1 half-move
when the countdown reaches 0 -> if (1 bare king on the board) then (player with bare king loses) // if (2 bare kings on the board) then draw [this second condition is probably already coded under the insufficient material rule]

Thanks in advance for your help, I hope to hear soon from you!

spacebar

Any starting position can be played in variants, and I am interested mostly in general rules.
You'll notice that 'Stalemate is a Win' is not something than can be applied to just this game, but any game (currently it's only used for this game though).
So my first Q is, is this an option we want that can be applied to any game and i think, yes probably.
My next Q revolve around the many other possibilities we have in variants (what we've released so far is the tip of the iceberg, really ).
What if a player doesn't have a King? (like horde).
What if the royal piece isn't a King, but some other piece? (see for example https://www.chess.com/4-player-chess?g=4984742-0)


as for the position of the king, it's more of a "political debate", I'm all for it, what say you @BabYagun?

Arma55
spacebar ha scritto:

Any starting position can be played in variants, and I am interested mostly in general rules.
You'll notice that 'Stalemate is a Win' is not something than can be applied to just this game, but any game (currently it's only used for this game though).
So my first Q is, is this an option we want that can be applied to any game and i think, yes probably.
My next Q revolve around the many other possibilities we have in variants (what we've released so far is the tip of the iceberg, really ).
What if a player doesn't have a King? (like horde).
What if the royal piece isn't a King, but some other piece? (see for example https://www.chess.com/4-player-chess?g=4984742-0)


as for the position of the king, it's more of a "political debate", I'm all for it, what say you @BabYagun?

I forgot to address stalemate=win in my previous posts. This rule make it kinda easier to win (maybe?).
But really, it's not as big of a difference as someone would expect (doesn't affect a lot of pawns vs king endgames for example, since the bare king can just let the pawn/s promote without any consequence. You need to learn 2 Knights vs King right now, however, since now you can force a win. So that's pretty cool.

If you're interested in applying the bare king rules elsewhere, then I don't have much imput (if I understand correctly what you're trying to say, chess.com's staff wants to build a "toolbox of rules" that they then use to assemble any type of weird variant they'd like).

Replying to your questions, in my very modest opinion, I can't think of any other variant where the bare king rule would make much sense. It's really a shatranj thing as far as I know, I believe that if applied elsewhere, it might just interfere with the intended playstyle of the variant. However I am most certainly wrong on the counts that:
1) I don't know every chess variant in existence
2) Creating a variant where this rule is useful isn't that improbable

If a player doesn't doesn't have a king (or better, a royal piece), then the rule cannot be applied since the second part of the rule (mutual bare-king) cannot be satisfied. One exception might be an asymmetric variant where the whole premise of the game is "capture all enemy pieces (if you have the king) or bare the enemy king (if you are king-less)". That might be interesting to play, however I know of no such variant. Horde comes close, but is not an automatic win when baring the king.

If the royal piece is something else rather than a king, the the code might be programmed this way:
define (royal piece): (all the properties of a royal piece)
for shatranj: royal piece=king
if (one player has a bare RP) -> start countdown-to-win
countdown-to-win: 1 half-move
when the countdown reaches 0 -> if (1 bare RP on the board) then (player with bare RP loses) // if (2 bare RPs on the board) then draw

I hope BabYagun is favourable to the change of rules and initial setup as well. Since we're there maybe we could rename the variant Shatranj, since that's really what it is.

Also, I'd like to compliment the Chess.com staff for the work on variants. Made me much more interested in the site.

BabYagun

Thank you very much, @Arma55!

I hope BabYagun is favourable to the change of rules and initial setup as well.

I'll talk to him.

spacebar

2) Creating a variant where this rule is useful isn't that improbable

yep. i wonder if it could just be a "if someone has only one piece left he loses" rule, no matter what piece it is.

spacebar

I just noticed that wikipedia says the rule is: 'stalemated player wins'.
And I added the opposite, a stalemated player loses currently. Would you suggest this be changed? or left as a draw?

fyi, support for bare king rule is coming soon.

Arma55

spacebar ha scritto:

I just noticed that wikipedia says the rule is: 'stalemated player wins'.
And I added the opposite, a stalemated player loses currently. Would you suggest this be changed? or left as a draw?

fyi, support for bare king rule is coming soon.

I think you are mistaken. Stalemate is a win for whoever delivers it. Being stalemated is a loss. Anyway very happy for the bare-king rule, good job!

spacebar

ah. the source of my confusion:
https://en.wikipedia.org/wiki/Chaturanga

spacebar

The new version is now out, please let us know how you like it.

Arma55

Woaah! Awesome job! This will change the play of the variant completely! And for the better! Kudos to spacebar and the chess.com staff!