gam@astrovax.UUCP (Gary Mamon) (03/16/85)
Regarding the following position:
....kBRK
....P.PP
........ White (uppercase)
......n. Black (lowercase) to move
........
........
........
........
We thank the 11 people who sent us mail or posted articles about our query.
Of these, 8 thought the position was a mate, 1 thought it was
a stalemate, and two did not care to give an answer.
Most of the answers were correct in that Article 12.1 of the FIDE rules
states that
"The game is drawn when the king of the player whose turn it is to move
is not in check and the player cannot make any legal move. The king is
then said to be 'stalemated'."
Article 10.2 of the FIDE rules states that
"Check must be parried by the move immediately following. If the check
cannot be parried, it is said to be 'mate'."
The note to Article 10.2 in these rules adds that
"Check may be parried:
a) by moving the king to a square which is not threatened by an enemy piece,
b) by capturing the opponent's piece which is checking the king and
c) by placing one of one's own pieces on one of the squares lying between
the king and the attacking enemy piece. This last means of defence is
evidently not possible when the check comes from the knight or in the case
of a double check."
In this position, after the black knight moves to f7, white is in check,
and as he has no legal moves at all, he cannot parry the check in any of
the three manners enunciated above.
Enough for the rules!
Don't think that this position is as boring as it may seem, because we
have discovered that some chess computers have difficulties playing it:
We used the Sargon II program on an Apple II. When the machine was playing
White, it responded to the check by printing "@/-E7", and DRAWING ON THE
CHESSBOARD AN EXTRA WHITE SQUARE between columns f and g and rows 7 and 8
which seemed to hover over the board!!! However, when the program played
Black, it was happy enough to check with the Knight and print out "MATE"!
We also tried Sargon II on a Commodore VIC-20. When it played White, it
printed "*" (the prompt) but refused to continue.
It is surprising that such a well-designed program would die when reaching
this perfectly legal position. Has anyone out there tried this position with
a different program (e.g. Sargon III or Chess Challenger)?
Is anyone in contact with the makers or distributors of Sargon II? They
surely would be 'happy' to learn of the existence of this bug.
Rich Gott and Gary Mamon
--
Gary Mamon Princeton University Astrophysics (609)-452-3579
{allegra,akgua,burl,cbosgd,decvax,ihnp4,noao,princeton,vax135}!astrovax!gam