[QUOTE=axn;258753]If white pawn on c6 captures black pawn on b7?[/QUOTE]
Right!
The positions are legal after all. My objection was wrong, not taking account of the possibility that the pawn has just captured. And I suspect Mr. Christenson (who I hope is feeling better now) worked that out originally too but got temporarily bamboozled just now by the rubbish I wrote.
Thank goodness he didn't go ahead changing everything.
Sorry everyone.:redface:

[QUOTE=Brian-E;258754]Right!
The positions are legal after all. My objection was wrong, not taking account of the possibility that the pawn has just captured. And I suspect Mr. Christenson (who I hope is feeling better now) worked that out originally too but got temporarily bamboozled just now by the rubbish I wrote.
Thank goodness he didn't go ahead changing everything.
Sorry everyone.:redface:[/QUOTE]
Mr Christenson was actually feeling good that he had a serious criticism; remember the first one was perfectly on target...and the second one thought-provoking, good for demonstrating the kinds of traps these analyses can fall into. The bad feeling was sinuses.

If you want to see why it makes me feel better, consider whether the following is legal:
White king on a7, black bishop on b8, black king on c7. I'm pretty sure it isn't, and I'm sure I counted it and similar positions as legal earlier when trying to work out the total number of positions with three pieces on the board.

Consider also white bishop in the corner on h8, and black king on g7. (4 corners * 54 positions for the other king)

[QUOTE=Christenson;258760]If you want to see why it makes me feel better, consider whether the following is legal:
White king on a7, black bishop on b8, black king on c7. I'm pretty sure it isn't, and I'm sure I counted it and similar positions as legal earlier when trying to work out the total number of positions with three pieces on the board.[/QUOTE]
Yes, you've clearly investigated positions which are illegal because there was no possible previous move. I agree that this position is illegal, also the positions with queens and rooks which you described in post #185 (though the one there with queens seemed to me illegal simply because both kings are in check).

For what my muddled thinking is worth, after axn's correction I am now of the opinion that you have adequately covered the special illegal positions in your analysis of K+P vs K by observing that the opposing king cannot be in check if the pawn is still on the second rank. I don't know of any other special illegal positions. But others may discover them.

When I get time I'll plough through your document to verify the rest. As you say elsewhere, it's the type of work which needs different pairs of eyes to eliminate any errors.

[QUOTE=Christenson;258669]I've attached my cleaned-up, probably-suboptimal, enumeration, as promised.[/QUOTE]
OK, after going through the document properly and checking everything I have only come across [I]one[/I] more error apart from the previously flagged failure in section I) to include b8 for the opposing king coupled with I) F) being only 33 spots for opposing king, not 34:

Section VI) C) (pawn on h2, opposing king on g2).
I count 55 places for friendly king, not 59.

Apart from that there are only a few trivial errors here and there listing which squares you mean for certain categories, but it's always clear what you meant and it will not affect the final count.

Maybe you want to go ahead and calculate the total, or maybe someone else wants to check through first as well.

Whatever, your suggestion earlier of then doing the calculation with placing the 3 pieces in a different order and seeing if we end up with the same result will be a logical checking step. I can't say I have much stomach for that myself though, given the mind-numbing nature of the job.:smile:

[QUOTE=diep;258219]In this position you're not allowed to capture a single pawn en passant.

For an explanation on what en passant is see the great explanation at: [url]http://en.wikipedia.org/wiki/En_passant[/url]

:google:[/QUOTE]

If you're going to be insulting, you'd better be right, which, in this case, you aren't. I'm familiar with he e.p. rule. In the configuration whose diagram I posted above, any of white's pawns on the fourth rank (with black to play) or any of black's on the fifth (with white to play) [i]might[/i] be capturable e.p. by the imediately adjacent enemy pawn, if it was moved two squares the previous turn.

[QUOTE=Mr. P-1;258796]If you're going to be insulting, you'd better be right, which, in this case, you aren't. I'm familiar with he e.p. rule. In the configuration whose diagram I posted above, any of white's pawns on the fourth rank (with black to play) or any of black's on the fifth (with white to play) [i]might[/i] be capturable e.p. by the imediately adjacent enemy pawn, if it was moved two squares the previous turn.[/QUOTE]

only pawns I don't see moved by configuration a is the c pawn , d pawn, and maybe one of each pawn on one side only.

[QUOTE=Brian-E;258792]
Apart from that there are only a few trivial errors here and there listing which squares you mean for certain categories, but it's always clear what you meant and it will not affect the final count.

[/QUOTE]

Really I got totally lost!

[QUOTE=science_man_88;258798]only pawns I don't see moved by configuration a is the c pawn , d pawn, and maybe one of each pawn on one side only.[/QUOTE]

We are talking about possible captures [i]en passent[/i], not "normal" pawn captures.

In the configuration on the board attached to [url=http://www.mersenneforum.org/showpost.php?p=258211&postcount=164]this post[/url], which pawn (if any) can be captured e.p. depends upon the last move, i.e., the one which resulted in the position.

With white to play, if black's last move was a7-a5, white can play bXa. If black's last move was c7-c5, white can play bXc. If d7-d5, eXd. If f7-f5, eXf. If g7-g5, hXg. Any other move by black does not permit an e.p. capture.

Similarly, with black to play, if white's last move was a2-a4, black can play bXa. If black's last move was c2-c4, black can play bXc. If d2-d4, eXd. If f2-f4, eXf. If g2-g4, hXg. Any other move by white does not permit an e.p. capture.

Mr P, it's en passant...not en passent! (says the man as avoids picking out his own trivial errors!)

I know this is probably annoying but [url]ftp://ctan.cms.math.ca/info/symbols/comprehensive/symbols-a4.pdf[/url] page 94 symbols might come in handy ( chess pieces on different square types, in a package that allows drawing of chessboards).

[QUOTE=science_man_88;258832]I know this is probably annoying but [url]ftp://ctan.cms.math.ca/info/symbols/comprehensive/symbols-a4.pdf[/url] page 94 symbols might come in handy ( chess pieces on different square types, in a package that allows drawing of chessboards).[/QUOTE]

I guess we have:

[TEX]0-0-0 \rightarrow \text long castle[/TEX]
[TEX]0-0 \rightarrow \text short castle[/TEX]
[TEX]\times \rightarrow \text capture[/TEX]
[TEX]\mathfrak {p} \rightarrow \text pawn[/TEX]
[TEX]\mathfrak {k} \rightarrow \text king[/TEX]
[TEX]\mathfrak {q} \rightarrow \text queen[/TEX]
[TEX]\mathfrak {n} \rightarrow \text knight[/TEX]
[TEX]\mathfrak {r} \rightarrow \text rook[/TEX]
[TEX]\mathfrak {b} \rightarrow \text bishop[/TEX]

I could say use the alts to draw boards but that's hard.