Support >26 piece types per variant

[TomFryers]

One for each letter of the alphabet defined in the ini file. I suppose this a pretty niche use-case, but is this currently possible, and if not is it feasible?

[ianfab]

Currently this is not possible, at least not unless some of the pieces only exist as promoted forms of pieces (like in shogi). However, since there currently are not that much more than 26 piece types anyway, I agree that this is a rather niche use-case. This might get more relevant when https://github.com/ianfab/Fairy-Stockfish/issues/58 is implemented. Is the question hypothetical, or do you have a concrete variant in mind?

[TomFryers]

The original reason this became a problem is that a simple GUI I’ve been making defines its own variant with 26 pieces. This is so that when a piece is placed the variant doesn’t need redefining, as doing that causes a small delay. I can work-around this by defining a new variant whenever a new piece type is placed, and other changes I’ve made mean it’s not now as big of a delay as it was when originally suggested it.

However, I realised that it would also be useful for dai shogi (and the even bigger variants), which has 29 different unpromoted pieces. The board for dai shogi is 15×15, so this would only really be useful after #38 is addressed. It would also depend on implementing the lion, which might be tricky.

[musketeerchess]

Hi Tom, Fabian Does your GUI Handles Musketeer Stockfish also?

I wish all of you a very nice and fruitful coming new year.

Zied

Le mar. 22 déc. 2020 à 13:01, Tom Fryers [email protected] a écrit :

The original reason this became a problem is that a simple GUI I’ve been making https://gitlab.com/Tom_Fryers/fairyboard defines its own variant with 26 pieces. This is so that when a piece is placed the variant doesn’t need redefining, as doing that causes a small delay. I can work-around this by defining a new variant whenever a new piece type is placed, and other changes I’ve made mean it’s not now as big of a delay as it was when originally suggested it.

However, I realised that it would also be useful for dai shogi (and the even bigger variants), which has 29 different unpromoted pieces. The board for dai shogi is 15×15, so this would only really be useful after #38 https://github.com/ianfab/Fairy-Stockfish/issues/38 is addressed. It would also depend on implementing the lion, which might be tricky.

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/ianfab/Fairy-Stockfish/issues/216#issuecomment-749506345, or unsubscribe https://github.com/notifications/unsubscribe-auth/AIIE4HJ6QHG63WXC3RJANO3SWCDBPANCNFSM4UZ2JKOA .

[atollk]

Hey. Since the question above was whether the proposal is hypothetical or an actual requirement: for me, I would like to have asymmetrical "armies" of 16 custom pieces, i.e. up to 32 pieces in total.