From: Michael Vanier BBB CALTECH EDU> Date: 16 feb 1999 Subject: Re: shuffle chess(or shogi) [Manabu Terao] > Although shuffle chess seems accepted even by top level players, I > doubt shuffle shogi would be accepted in the shogi world. Three > things I would like to point out. Restriction of Lance's and kNight's > mobility in the original set-up in shogi, the difference in symetry > between shogi and chess, and exchangeability of Bishops. > > As you know, A Lance can move to only 8 squares in the file originally > located and a kNight can move to only 10 sqares(for example, the kNight > on 2i can move to 1g,3g,2e,4e,1c,3c,5c,2a,4a,6a) without either promotion > or drop, while a kNight in chess can move to all the squares. Judging > from that shuffle chess forces the Bishops to be on the opposite colour > which means there is no change in the movable squares of the Bishops, the > Lances and kNights in shuffle shogi should be located in the same squares > as original to keep the movable squares same. I think you're missing the point of shuffle-shogi. Unless there is some profound reason why the lances and/or knights MUST be located on the original squares I see no need for this restriction. In fact, allowing different placements of the minor pieces is what would make shuffle-shogi interesting. Larry's point about not putting the knights in the corner makes sense, but I'm not sure how much this would harm the game. Remember that pieces in hand can go anywhere, which takes away a lot of the force of your argument. Why should those particular 10 squares be the optimum ones for knights to be able to get to? What if the knight started one square closer to the king; how would this destroy the game? > > In the original set-up in Shogi, all the minor pieces and the Kings are > located symetrically by the line of fifth rank and at the same time all > the pieces are located symetrically by the point of the very center square > on 5e, while chess pieces are located symetrically by only the border line > between 4th and 5th rank. Since shuffle chess keeps the original symetry, > shuffle shogi should keep the original line and point symetry, too. It > means the Kings should be kept in the 5th file. I basically agree with this point (Kings on 5th file) but not with your reasoning. I think the reason the King should be kept on the 5th file is that it preserves the option of building a castle on either side. Of course, this will depend on the position of the generals too, so perhaps this restriction is unnecessary as well, or (alternatively) you could stipulate that on each side of the King you would have to have one of each of the minor pieces. Restrictions of this sort aren't done for shuffle-chess and it's not clear to me that they're needed for shuffle-shogi, although the lesser lateral mobility in shogi might make them desirable. With these restrictions and the bishop and rook on their original squares you would still have 24*24 = 576 possible opening setups, which is quite a bit already. Disallowing the knight from the corner reduces this to 18*18 = 324 setups, still quite a bit. And of course if the position of the rook and bishop are allowed to vary this goes up dramatically (by up to a factor of 72). What is important is that both players have equal chances. You can get that and preserve the symmetry of the relative arrangements of the pieces by having each player's setup be the same as the other player's, with the board rotated 180 degrees. > > Imagine removing all the Pawns, you can find the Bishops are originally > located to be exchanged quite easily in only one move. I think this easy > Bishop exchangeability is one of the essential and critical factors which > make shogi an attractive game. Shuffle shogi should keep this exchangeability. > Again, I think you're trying to make shuffle-shogi a clone of shogi. Is this really so important? > In consideration of keeping movable squares of each piece, symetry and > Bishop exchangeability at the same time, shufflable pieces and patterns are > very limited in shogi. This is why I think shuffle shogi would not be > considered seriously. In order to solve the problem of memorized opening > theory, I think starting a game at a specific position extracted from a > Joseki sequence or verbal agreement of what strategy both sides choose > before a game would serve. The latter could be an alternative to Lance > handicap, also. Hmm; I find that much more distasteful than shuffling the pieces. What is true is that it is not at all clear what restrictions, if any, on shuffling the pieces need to be imposed in order for the game to be playable. Mike