Originally posted by Acolyte
I don't get this either. Here's a slightly different problem (just off the top of my head, it should work but I'm not sure):
On one chessboard, White has his king on a1, and Black has his rooks on e1 and h1, and a bishop on f5. Anot ...[text shortened]... e is a fundamental difference between these positions; what is it?
obviously, the bishop had to discover check to make it a legal position. since black's bishop could have moved from d1 to f3 to discover check in the second example, that position is legit if we put a black king on any legal square except e2 or anywhere on the first rank. to make it more of a problem, however, put white pawns on a2 and b2 (or black king on a3 or b3) to restrict white's moves (and thus produce mate) since otherwise you have essentially no difference.
the first example, however, looks
as if the black bishop could have started on b1 and discovered check by going to f5, but this is not
the case! if black's bishop would have been on b1, what was white's last move?
(of course, white could
have had something on f5 which black captured to discover check. to overcome this possibility, put a white pawn on c2 (or black king on d3) in both positions, making the bishop's travel from b1 to f5 impossible.)
another which appeared in chess life
many years ago was this: 40/7k/PPnP4/KRB5 (the position of the black king is irrelevant).
white stands smothered mated, right?
or does he?
these are examples of a promlem-solving technique called retrograde analysis
which is often useful and sometimes necessary.
here is another: 6k1/ 8/4r1q1/8/4Kp3/16. what was the position one and a half moves ago if all moves must conform to the rules of chess?