I planned to redo complete parser thingy from scratch back in February, but before that a short detour to revisit book, more precisely, activation rules and do quick fixes here-and-there, then get back to coding.
Except, one thing leads to another, and here I am, two months in and still at quick detour. One reason is that I have to explain all emerging effects of rules, and new concepts in greater detail, and slower pace; so I have to redo quite a few examples. Another is that I wasn't happy with Miranda's Veil for quite some time now, so I took the opportunity, and rewrote the whole chapter.
I prefer to have a few simple rules from which all other emerging effects can be derived following logic. Sometimes, I get it wrong. For instance, last piece in a cascade can never check opponent's King, even if it has enough momentum and King is within range. This is so because activated piece, once it starts moving, can't change direction of its movement. So, to be checked, opponent's King would have to be positioned in a chosen direction before cascade even starts, which would mean King was checked before cascade even started. So, precheck and precheckmate can never happen. This is not the only thing I got wrong, or simply overlook, but it illustrates how nuanced things can become.
It's not all fixes and rewrites, either. I recalled that I read a paper on AlphaZero evaluating Classical Chess variants, so I recently added sideways moving Pawns into the fray, from Nineteen variant onward, to spicy things a little bit. I mean, c'mon, how can you say no to: "This is the most perplexing and “alien” of all variants of chess that we have considered. Even after having looked at how AlphaZero plays Pawn-side chess, the principles of play remain somewhat mysterious – it is not entirely clear what each side should aim for. The patterns are very different and this makes many moves visually appear very strange, as they would be mistakes in Classical chess. Lateral pawn moves change all stages of the game. Endgame theory changes entirely, given that the pawns can now “run away” laterally to the edge of the board, and it is hard to block them and pin them down." (assessment by Vladimir Kramnik).
It doesn't hurt that sideways Pawns are very roughly twice as valuable as ordinary cousins (which is very much welcomed for variants on an ever growing chessboards!), they offer more potential moves from the very beginning of a match, doesn't upset the balance of the game, doesn't slow down gameplay, ... all the good stuff! Would you be surprised if I tell you that I'm seriously thinking into doubling-down on sideways Pawns? I'm considering to introduce 2 new pieces, Scout and Grenadier, both remains in Pawns category (can be promoted, demoted-to, captured in Pawn-sacrifice, ...), but each with different movement. Issue is that now I'd have 3 different Pawns, and so I'd need to change en passant, and all other examples that involve Pawns. Also, introduction of Pawns category means now old Pawn piece has to be reintroduced as Private, to be able to distinguish between Pawns (category) and Pawn (old piece); this is huge undertaking. I'd like to postpone it, but I also very much like the idea, especially because thematically if fits so neatly into the book.
No comments:
Post a Comment