ChessV 2.0 - open source GUI and engine for chess variants

[Greg Strong]

Thanks Jon. I understand how it works. The problem is that the way it works is very bad. In your diagram, the white king can escape by the other king moving to g1 and becoming royal. But in mirror-image position, black can not do the same. This is a critical flaw. The game has other problems too. Can't castle king side. It takes 2 bishops to cover the board but there is only one (should have removed a knight instead.)

If this idea is really worthwhile we should invent it properly, not implement random bad ideas. Why your interest in this particular game? Is it really that good?

[Nordlandia]

Thanks Jon. I understand how it works. The problem is that the way it works is very bad. In your diagram, the white king can escape by the other king moving to g1 and becoming royal. But in mirror-image position, black can not do the same. This is a critical flaw. The game has other problems too. Can't castle king side. It takes 2 bishops to cover the board but there is only one (should have removed a knight instead.)

If this idea is really worthwhile we should invent it properly, not implement random bad ideas. Why your interest in this particular game? Is it really that good?

Yes, i agree. something need to be done to readjust current rules.

I like this variant simply because of much similarity to regular chess. If both non-royal kings get captured then it's back to standard fide chess.

If you want, i'm willing to adjust current rules to allow kingside castling. Not sure though which piece that need to be removed to allow 0-0.

Alternative is to remove Queen so both kings get placed in the central files (d1 an e1 & d8 for e8). The compromise for queen removal is guaranteed queen promotion

[Greg Strong]

Yes, i agree. something need to be done to readjust current rules.

I like this variant simply because of much similarity to regular chess. If both non-royal kings get captured then it's back to standard fide chess.

If you want, i'm willing to adjust current rules to allow kingside castling. Not sure though which piece that need to be removed to allow 0-0.

Alternative is to remove Queen so both kings get placed in the central files (d1 an e1 & d8 for e8). The compromise for queen removal is guaranteed queen promotion

Ok. I think we can fix most problems and keep the start position (RNBQKKNR) with two small changes:

1. For white, royal king is the one closest to a1, but for black it is the one closest to a8.

2. The king on e1 can still castle even after the king on f1 has moved. This will allow king-side castling.

We would still have only one bishop, but I can live with that.

Or we could do something more ambitious by changing the opening position like putting the second king in place of the queen as you suggest. But some players will not want to be without a queen. What do you think?

[Nordlandia]

We can live without king's bishop

You're rules looks good.

Question: What symbol will be used for non-royal king?

Code: Select all

1. For white, royal king is the one closest to a1, but for black it is the one closest to a8. 

2. The king on e1 can still castle even after the king on f1 has moved. This will allow king-side castling. 

We would still have only one bishop, but I can live with that. 

Can you comfirm if my ranking order is correct, considering a8 for black.

Or do i have to update it?

It is now updated for mirror-image (exact duplicate)

Code: Select all

https://cdn.pbrd.co/images/MWGKW1mGU.png

[hgm]

Question: What symbol will be used for non-royal king?

Why would that need a different symbol? It is not a different piece anymore than a King in 3check that has been checked once is not a different piece from one that has not been checked, or a King that can still castle is not a different piece as one that cannot.

It is just that the winning condition depends on the location of the piece. There are several ways to treat multiple royalty; the most common are 'absolute royalty' (where capture of any King immediately wins) and 'extinction royalty' (only capture of all Kings wins). TwoKings is a version where it depends on the relative location. In Spartan Chess there is an extra 'duple check' rule, which specifies you already win when you have attacks on all remaining Kings, even though you can of course capture only one at the time. So it is like at the end of your own turn you can decide anew which of your Kings will be the royal one, to save the day when one of them is under attack.

Btw, WinBoard does allow castling with the e-file King after the f-file King has moved. Is that different on ICC?

I have severe doubts whether introducing a new variant of this variant merely to avoid the asymmetry would be a good thing.

[Nordlandia]

Btw, WinBoard does allow castling with the e-file King after the f-file King has moved. Is that different on ICC?

That's correct. Only Q-Side castling allowed.

What i'm trying to say is that the royal king and non-royal king should be easily detectable. Not hiding under any disguise.

[Nordlandia]

As hinted in my previous post --->

I've won all wild 9 games on ICC, except one which i lost.

The fact is that people do not know the rules and are confused once they realize they play with two kings. They do not know which of the two that is royal, unless if i say so in the chat.

So i suggest the symbol for the non-royal king should be different during the whole game.

Here is example using equivalent symbol for the piece on wikipedia.

Code: Select all

https://commons.wikimedia.org/wiki/File:Guard_(an_icon_of_the_chess_piece)_Classical_Version.png

Or upside-down king

[hgm]

Well, ICC players not knowing the rules is not a problem that can be solved by choosing a piece representation in ChessV.

In addition I think different represetation of the Kings is misleading, because the King might not stay of the indicated type when you move it. You cannot always step a 'non-royal' King into check, because the very move might 'promote' it to a royal one. And you can sometimes step your royal King in check, when it 'demotes' to a non-royal one in the new position. And some moves of the royal King to safe squares could be forbidden because they transfer the royalty to the other King, which is under attack. So I think it just adds to the confusion when the pictograms suggest that they are royal and non-royal Kings.

BTW, note that king-side castling seems ill advised in this game, because it tucks away the King that almost always would be the non-royal one, being so far from the a-file. While leaving the royal King out in the cold.

[Nordlandia]

Apropos piece representation.

Is piece exchange Mann for remaining Bishop considered equally good (fair trade). Thus leaving imbalance KK vs KB.

What side do you favour here?

[hgm]

A plain Commoner is worth somewhat less than a Knight, so also somewhat less than a lone Bishop. Bishop pair vs B + M is a significant advantage, though.

With extinction royalty, (as in Spartan Chess) the surplus King turns out to to be worth about 4.5, so that trading it away at the expense of a Rook or Queen is a bad deal. The delocalization of royalty apparently is worth a hefty bonus.

Another interesting observation is this: a second absolute royal is worth almost exactly zero! It is as much a liability as an asset. If you remove the Queens, and then give only one player an extra King on d1/d8, and the other nothing, they will score ~50% each. In the middle game the extra King just means that you have double the chance to get checkmated. But if you survive this, the extra King is a great asset in the end-game. Especially in Pawn endings, when checking is only possible by the not-so-dangerous Pawns, the extra King is a winning advantage, even against a Pawn majority.