Author: Omid David Tabibi
Date: 21:14:38 07/15/03
Go up one level in this thread
On July 15, 2003 at 22:35:20, Uri Blass wrote: >On July 15, 2003 at 21:43:54, Omid David Tabibi wrote: > >>On July 15, 2003 at 20:13:13, Robert Hyatt wrote: >> >>>On July 15, 2003 at 17:13:47, Andrea Griffini wrote: >>> >>>>A question about repeated positions... is the en-passant square >>>>part of the "position" even when no en-passant pawn capture is >>>>possible (because there is no capturing pawn, or because the >>>>theoretically capturing pawn is pinned on the file or because >>>>just the en-passant capture is "pinned") ? >>> >>>I don't believe so. At least not by the rules of FIDE. If a move is >>>illegal, it is not possible to play it. For EP, if the pawn can't >>>be taken because (a) it wasn't moved two squares, (b) no pawn is on >>>an adjacent file to take it; (c) a pawn is on the adjacent file but >>>can't legally make the capture due to being pinned; then the move >>>simply is not possible and a position with every piece on the same >>>square, where one was reached by two pawn pushes and the other by >>>one two-square pawn push, would be considered identical. >>> >>>> >>>>If this is not the case then checking for repeated positions >>>>gets way more annoying as my position hash code doesn't tell >>>>the whole story ... :( >>> >>>I don't hash the ep status if (a) the pawn didn't advance two squares; >>>(b) there is no pawn on either of the adjacent files that _can_ make a >>>pseudo-legal EP capture. I don't screen out the pinned business, however, >>>and haven't had it cause a problem. >>> >> >>I also do the hasing exactly like this, but thinking about it, a problem can >>arise: > >1)I think that even if the price of fixing it is only being 0.01% slower then it >is still better not to fix it if you care about playing strength. > >2)Movei knows about pins but I cannot decide that a pinned pawn cannot do the en >passent capture because the pawn can be pinned by a bishop and in part of these >cases it can do the enpassent capture. > >Another point is that there are cases when enpassent capture is illegal inspite >of the fact that the pawn is not pinned because there is an enemy rook or enemy >queen in the same rank of the king. > This discussion remined me of an old puzzle: can you setup a position in which a king is simultaneously checked by two queens?! (answer down the page) [D]Q3Q3/3p4/8/4P3/4k3/8/8/4K3 b - - 0 1 Plack plays 1... d5 and white replies with 2. exd6 [D]Q3Q3/8/3P4/8/4k3/8/8/4K3 b - - 0 2
This page took 0 seconds to execute
Last modified: Thu, 15 Apr 21 08:11:13 -0700
Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.