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
This item is a game information page
It belongs to categories: Two dimensional
It was last modified on: 2005-05-01
 By Joe  Joyce. Modern Shatranj. A bridge between modern chess and the historic game of Shatranj. (8x8, Cells: 64) [All Comments] [Add Comment or Rating]
H. G. Muller wrote on 2021-05-07 UTC
satellite=motern promoZone=1 promoChoice=G graphicsDir=http://www.chessvariants.com/graphics.dir/small/ whitePrefix=W blackPrefix=B graphicsType=gif squareSize=35 darkShade=#FFFFFF symmetry=mirror stalemate=win baring=0 Pawn::fmWfcF:Pawn:a2-h2 General::K::e1 Elephant:B:FA:Elephant:c1,f1 Knight:N:::b1,g1 Rook::::a1,h1 King::K::d1

Modern Shatranj


Kevin Pacey wrote on 2021-04-08 UTC

Question for H.G. (or any who might know):

Are two opposite-coloured FA (ferfil) elephants on average worth more than one FA plus N (or worth more than 2 Ns) on 8x8, according to computer studies?


Kevin Pacey wrote on 2018-01-02 UTC

Hi H.G.

I thought about trying to find a position with a stalemate by bared K created by a series of forced moves (such as in a problem), but it seemed unnecessary at the time, as even though the 'stalemate' can be prevented in the initial position I gave, there's still the possibility of the 'error' I selected as the only line of play, which at least shows that such a final 'stalemate' could possibly arise by legal moves in a real game (and I've seen some pretty bad legal moves played over the years...).


H. G. Muller wrote on 2018-01-01 UTC

Note that, despite the clever construction of the position, this is basically a non-issue: irrespective of the rule precedence the given position is always a win for black, as he can also bare by Kxa2.


Jose Carrillo wrote on 2017-12-31 UTC

The additional move for the lone King will still be required for this preset, but the Bare King rule precedence over Stalemate will assign the win to the correct player.

I would expect the White player in the center position to resign, rather than to play, but if he plays the preset logic will now award the win to Black (no matter what White plays), and formally end the game.


Jose Carrillo wrote on 2017-12-31 UTC

I fixed the preset. Now it will correctly give preference to a win by Bare King over Stalemate, and will correctly give the win to Player B in the position that Kevin had sugested below.


Greg Strong wrote on 2017-12-29 UTC

Agreed.  The Bare King rule is clear and it only permits another move if the opponent can bare the other's king on the next move.  Stalemating would require allowing another move where there is no indication it should be allowed.  This is how ChessV is coded and I've confirmed that when 1...Pb3xa2 is played it immediately decalres black the winner.  So all is good :)


Jose Carrillo wrote on 2017-12-29 UTC

>>Kevin Pacey wrote on 2017-01-25 EST
>> My own suggestion would be that for the diagrammed example, for Modern Shatranj, let the stalemate=win rule
>> override the bare king consideration - the 'logic' being that the stalemated king will perish if the stalemated side
>> attempts to move, whereas the bare king has freedom still.
>> In any case, I don't know how Jose's rules enforcing preset for Modern Shatranj currently would handle the
>> diagrammed example, after the final move is made. [edit: the preset's rules say a lone bare king is an Automatic
>> Loss (if the other side's king isn't immediately bared), so I think I ought to take that at face value, even for the
>> example situation I gave.]

Forgive me for my late response... Better late than never! At least I'm still responding in the same year as the question! :-)

As currently programmed, for the given (unlikely) position, the result of my preset would be a win for the bare King.

I need to check the logic to account for this strange position.

I agree with Joe, it should be an automatic win for Player B, as Player A was bared prior to the stalemate, and the rules only allow for the disadvantaged bared King to play, if he would bare the other king in the next move. Nothing is provided for the case where the bare King could stalemate the opponent on his next move.

I guess this position must have had a lot to do with both stalemate and bare kings (by insuficient material) being draws in modern Chess, to avoid this dilema.


Kevin Pacey wrote on 2017-04-15 UTC

I've noticed that in Game Courier play of the reputedly drawish Shatranj there's been 2 draws in 55 games thus far, while there's been 7 draws in 33 games of Modern Shatranj so far, if I've counted correctly. It's possible the opponents were more evenly matched in more of the Modern Shatranj games, but I'm wondering if the Generals in the setup position for Modern Shatranj might make for even greater defensive resources than may be available in the game of Shatranj. A case where more test playing of Modern Shatranj is needed to reach a firm conclusion? In my case, I take a variant that may be tougher to avoid drawing (i.e. win) at as something of a challenge, if I'm not so happy with the idea of drawing before a game begins. Fwiw, chess played at a high level still has a far greater drawing rate on average than 7 out of 33 games.


Joe Joyce wrote on 2017-01-27 UTC

Since stalemate was sometimes a win for the stalemated side, and "bare king" is immediately obvious to everyone, I lean strongly toward the bare king rule winning out. Your argument, Kevin, cements my position.

Now, who's got stuff hanging? Contact me.

 


Glenn Nicholls wrote on 2017-01-26 UTC

I, too, have a CVP reference page submission awaiting clearance.


Kevin Pacey wrote on 2017-01-26 UTC

I've edited my previous post.


Kevin Pacey wrote on 2017-01-25 UTC

My own suggestion would be that for the diagrammed example, for Modern Shatranj, let the stalemate=win rule override the bare king consideration - the 'logic' being that the stalemated king will perish if the stalemated side attempts to move, whereas the bare king has freedom still. In any case, I don't know how Jose's rules enforcing preset for Modern Shatranj currently would handle the diagrammed example, after the final move is made. [edit: the preset's rules say a lone bare king is an Automatic Loss (if the other side's king isn't immediately bared), so I think I ought to take that at face value, even for the example situation I gave.]

P.S.: Alert again to CVP editors: I have an aging CVP submission awaiting review.


Joe Joyce wrote on 2017-01-25 UTC

Good question, Kevin, and as far as I can see or find (so far), the answer is undetermined. There was little standardization of stalemate rules until a couple hundred years ago. Different areas did different things, and I suspect that in the situation diagrammed, all 3 possible outcomes were, at some place and time, accepted.

With rook-level pieces, in the spirit of shatranj, I find the knightrider stylistically wrong. Of course, I find the NN an awkward piece, and I am terrible with awkward pieces. To me, ches should be fighting with your opponent, not fighting with your own pieces. That being said, the NN is a limited piece, far more in keeping with the limited pieces of ancient chess than the modern versions of strong pieces. I see it as a sort of limited missile, able to strike across the board, but with restricted targeting. It seems like a piece for a very large board with lots of pieces worth a range of values.

At this point, the hero ands shaman pieces (D+W) & (A+F), or the bent versions (D+/-W) & (A+/-F), while more powerful, or possibly the Oliphaunt (AF+AF) or Lightningwarmachine (DW+DW) seem to me to be the best fits, being very roughly worth around 5 or so pawns and short ranged. Pieces like the half duck (HFD) or scout or other such pieces seem more awkward, to me. ... Hm, I guess there's a giant shatranj variant lurking somewhere in my head.


Kevin Pacey wrote on 2017-01-23 UTC

[comment edited to include diagram] I have a question about the rules of this game, based on an unlikely situation. If a player A's king has just been bared, but the following move he stalemates his opponent (Player B) by using his own king to do so, what should the result of the game be? Below is an example, i.e. if 1...Pb3xa2 (baring White's king) 2.Kc3-c2 (stalemating Black immediately after White's king was bared), what is the result of the game?

P.S.: To write again of examples of what a rook level fairy chess piece may be, I seem to recall others before me have valued a nightrider as worth about 5 pawn units. That's besides the superbishop (aka promoted bishop in Shogi), which also happens to be placed in the same piece type class as a rook by the inventor of the popular 8x8 variant Pocket Mutation Chess.

[edit: Note to editors of CVP: I have a submission I gave to CVP last week that's to be reviewed.]


Kevin Pacey wrote on 2017-01-17 UTC

Hi Joe.

My own calculations for my invented game of Sac Chess place a 'Missionary' (aka a promoted bishop, in Shogi) as very close to a rook on that game's 10x10 board. On the other hand, what I called a 'Judge' (aka a Centaur, i.e. a piece that's a Knight+General compound, in wikipedia's & many other people's usage) I rate as being worth a pawn better than a rook, in that game's fairly large 10x10 board. So, I'd say either long- or short- range pieces can be roughly in rook class as far as value goes. However, I don't know enough fairy chess piece types yet by heart, in order to name several already invented rook class pieces. Regarding the values of the knight & elephant in Modern Shatranj, for a knight I used the value that the late Dutch player Euwe (former chess world champion) gave (i.e. 3.5 pawns), on chess' 8x8 board (to try to be consistent with my placing a knight at 3 pawns on a 10x10 board). Using his value proved to be a good thing since I believe like you, at least for now, that an elephant is worth a shade less than a knight in Modern Shatranj, and by my own rough methods of calculating estimated piece values for chess-like games, if I set a knight to 3 pawns on an 8x8 board, an Elephant I would work out to then be worth a knight exactly. :)

P.S.: Fwiw, for those interested, on chess' 8x8 board, I'm agreeing with Euwe's having a Rook=5.5 pawns, too, which I also gave it for on Sac Chess' 10x10 board, since for that game I have the short-ranged piece (knight) reduced to 3 pawns in value (in my estimation), at least as a way to take into account that game's slightly bigger board.


Joe Joyce wrote on 2017-01-17 UTC

Thank you for the comment and rating, Kevin. Regardless of the exact value of the "minor" pieces, they are all within a point of each other, allowing fairly free exchanges among the pieces, and sometimes giving the end of the game a "different armies" feel, where a pair of elephants face a knight and general, for example.

In shatranj, there are 3 "levels" of even exchange, between pawns, between minor pieces, and between rooks. Modern chess adds the queen exchange for a 4th level. There are several "queen-level" pieces, from the R+N minister to the B+K dragon bishop of shogi. What are decent rook-level pieces? DO they need to be short range, more "area-effect" pieces to keep them rook level?


Kevin Pacey wrote on 2017-01-16 UTCExcellent ★★★★★

I'd tentatively estimate the relative piece values in Modern Shatranj (current version) as: Pawn=1, Knight=3.5, Rook=5.5, King's fighting value (noting it cannot be traded)=4, General=4 (noting it can be traded or put what be 'in check', unlike a K, but I've judged their value in action to be similar enough), with the Elephant=3.125.


Joe Joyce wrote on 2016-09-12 UTC

Aurelian Florea Verified as Aurelian Florea wrote on None

If you don't mind modern shatranj is an inspiration allong with shatraj kamil for my own 15x10 I shall complete and publish in a few months or so, actually I'm thinking on the name great modern shatranj. Are you ok with that, Joe?

Full circle! A decade ago, I emailed Christian Freeling 2 game write-ups, asking if I could use the name "Grand Shatranj" and copy his setup, as his variant Grand Chess had inspired me to create 2 games where I'd only seen one muddled game before. He was very courteous and friendly, and thus Great and Grand Shatranj were posted together here, direct outgrowths of Modern Shatranj. Grin, so if Christian Freeling approves of your game, I'd be happy to have it named great modern shatranj. ;)

Seriously, thank you for the compliment. I certainly have no objections to your use of the name with the caveats that your game should bear some resemblance to mine and further not be significantly offensive to the social mores. Without actually seeing what you're going to post first, that's about as close as I can come to saying I'd be honored. It's always nice to hear that someone appreciates your efforts. And I wish you the best in yours.

I see a game design that's intended to be played as a collaboration between the designer and the player(s). I've been lucky enough to see a couple of my games become what passes for moderately popular on this site. That people modify the games to suit themselves is a good sign, in my opinion, that the games have some merit. But the games I post here are public property. Anybody may do whatever they want with them. So I truly appreciate, in more than one way, that you asked. Take your ideas and run with them. Enjoy!


Aurelian Florea wrote on 2016-09-11 UTC

If you don't mind modern shatranj is an inspiration allong with shatraj kamil for my own 15x10 I shall complete and publish in a few months or so, actually I'm thinking on the name great modern shatranj. Are you ok with that, Joe?


Joe Joyce wrote on 2016-09-10 UTC

Thank you for the comment, HG. It seems to be unanimous that promotion should only be to general. Vox populi, vox Dei! It is changed. ;)


H. G. Muller wrote on 2016-09-08 UTC

Promotion to General already solves the problem with promotion in Shatranj, namely that the promotion piece would be virtually worthless, especially if you get them on the same square shade as wehere you already have some of them. The General in Modern Shatanj is a Commoner, which does have mating potential.

Promoting to stronger pieces is actually of little help only, as almost always the opponent can sacrifice one of his minors to prevent such a promotion. So in practice a promotion gains you a minor.

I therefore agree that there is little need to sacrifice the simplicity and elegance of the game by introducing complex promotion rules.


Joe Joyce wrote on 2016-09-07 UTC

Hi, Greg, thanks for the comment and rating. And it's nice to see some of the old gang around. To fully answer your comment, I think we'd need the 3rd shatranjeer, David P. He was the one who said it was a variant if I added promotion rules!

I've always considered games to be a collaboration between at least 2 people, the designer and the player(s). While I designed it to be a bridge between shatranj and modern chess, and so split the difference between no promo to lost pieces and unlimited pieces of any non-royal type, I certainly have no objections if people play it with promotions only to general. The only real effect it would have on the games is to extend the games with promotions a bit, because the generals are slow compared to the other pieces. 

Finally, grin, how do you tell from the current state of the board if a pawn can be captured en passant? ;)


Greg Strong wrote on 2016-09-05 UTCGood ★★★★

A promising game that might be worthy of upgrade to Excellent pending play-testing, which I will now try with Jose's new preset.

Reading through the comments, the promotion rules seem to provoke the most disagreement.  I must admit that I don't like the promotion rules as written.  I can see both promotion only to general, or promotion to general or to any lost piece as reasonable options, both leading to good although different games.  For myself, the part I find troubling is this:

At most, only 3 lost pieces may be regained: 1 rook, 1 knight, and 1 elephant, even if the player has lost both of any type.

The problem with this is that it is no longer possible to look at a board and know what moves are legal. You'd have to also know about all past promotions.  This makes the game much more difficult to program.  Chess has this issue too with castling - you have to know which rooks/kings have moved, although when the game has progressed enough that these pieces are no longer on their original squares it becomes a non-issue.  Also, Chess has established standards for how the castling information is preserved in the FEN game notation.  If we wanted to notate positions of Modern Shatranj with FEN notation, (certainly a worthy goal), new notation standards would need to be invented.  I would question whether the value of this particular rule justifies the significant added complexity.


Joe Joyce wrote on 2016-09-05 UTC

Thank you, Jose, for the comment, rating, and especially the preset. Modern Shatranj is my simplest and in many ways most successful design. Grin, there's probably a lesson there. As for the shift in promotion rules, I consider games to be a collaboration between at least 2 people, the designer and the player(s), so "adjusting" a rule to suit the player(s) is okay with me. Just means someone is interested enough to try a game. Thanks again.


25 comments displayed

Later Reverse Order EarlierEarliest

Permalink to the exact comments currently displayed.