No, black pawns can move to the second rank, but not the seventh. White pawns can move to the seventh rank but not the second. So "b2" and "b7" are both valid pawn moves, even if they are each only valid for one color.

Ahh -- gotcha.  I notice you also mention that kings can't check, but they can reveal a check (and also a mate) and so should have been included.

Good point. I corrected the orignial post. That adds 256 moves, bringing the total up to 45,586

There are 168024 possible legal positions with just a King and pawn vs. king endgame

Just do a Google or Wiki search for Shannon Number:

http://en.wikipedia.org/wiki/Shannon_number

Are you sure? I'm getting 334,496. If you assume the pawn is one color or the other (say, white), I get 167,248. So I guess you're doing that, and not accounting for the fact that a pawn that hasn't moved yet cannot cause check.

I think the two bishops are pretty different from each other.

Yes, but the point is that a bishop on 'square A' is the same as a bishop on 'square A'.

"Kasner and Newman estimate that the total of possible moves in a game of chess is 10 to the power of (10 to the 50th). This is a figure for which we have no name. It is composed first by taking a one with fifty zeros after it, thus: 100,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000 and then multiplying ten by itself that many times. An ordinary 200 page book accommodates 330,000 letters and numerals; three such volumes would provide space for about a million numerals. To write the number 10 to the 10th to the 50th power would take 30,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000 volumes: all this space being required merely to write down the numeral indicating all the different moves in a chess game."

oO yeah surprised me to.

but: in order for any pawn to reach a 'second rank'  e.g. b2 or b7, there must be something missing off the board - as every pawn is blocked by an opposing pawn in the initial position - and when something is captured, --- DEPENDING ON WHAT IS CAPTURED --- the number of possible moves remaining will be drastically reduced by some number - quite variable and impossible to quantify with a formula that actually accounts for the variability in a rational sense.

  You did not mention that reduction in your formula - making your formula invalid in the extreme because you have credited BOTH SIDES with potential to move all pawns to the enemy second rank (their own seventh rank) and that is physically impossible.

    Now that I mention it, that feature alone renders this an irrational exercise by those who have time on their hands.  the reduction in remaining possible moves or positions depends entirely on exactly what is captured and how many pawns remain after exchanges occur.

   applying rational thought to the irrational babblings.

I think the exercise was to enumerate the set of possible move notations irrespective of what preceded or followed the move.  In that context, the peice that previously occupied the spot the pawn is moving to (on the second or seventh rank) is irrelevant.

My calculations were not "invalid in the extreme" just because they don't answer the question you are interested in. I never stipulated or intended the count to be limited to one game. It was intended to be a count of the possible moves in all possible games, "move" being defined as a move with a distinct expression in algebraic notation. I think the first sentence of my post was pretty clear about that.

Please work on your "rational thought."

Thank you for the non sequitur.

hmmm, quite a lot of chat about this interesting subject, I will get my chess board out right now and work out every single possible position, I will post the answer as soon as humanly possible, then when everyone knows the correct answer, we can all relax!!! don't worry guys, I'm doing it for the team!!! 

A quick summary of the questions posed and the numbers supplied, for new-comers to the discussion ...

AWARDS FOR HIGHEST Number of responses:

FIRST/Second PLACE: Tie between:    artfizz  and ichabod801 - 7

third place to:  Eniamar - 4

honorable mention to:  cuendillar, GuyOnTheCouch, pvcmike, and last:  Sleeper.  with 2 each.

all others:  honorable mention,

and

       thanks for reading to the rest.

a gazillion!

I know I already plugged it earlier, but there is also some discussion here about the longest possible game:

I came to a ceiling of 5900 with some certainty that it is less than that.  Revan24 arrived at 5870.5, although I'm not sure exactly how as I've not seen his calculations, I arrived at 5897.5 (the means by which I discounted half-tempos from the ceiling leaving a lot of opportunity for error).

Any other thoughts on that one?

It appears the URL has gone AWOL. Could you possibly supply it again?