How long do you think this is? Not the longest practical game, but the longest possible. Any game will end up as a draw if each player plays 50 moves (so 100 half-moves altogether) and no captures or pawn moves are made. 15 captures are possible for each side, andeach pawn can move a maximum of 6 times before becoming a piece, which won't affect the 50 move rule. That gives a theoretical maximum of 96 pawn moves, but are they all possible? With captures, maybe.

If every pawn can make all its possible moves, and every piece is captured, then you have only the kings left, which is a draw under the rules (insufficient material), so the absolute theoretical maximum is 96 pawn moves, 15 captures for one side, 14 captures for the other side (making 125 50-move-clock-resetting-moves), 99 initial half moves before the first capture/pawn move (and this is probably the hardest time to avoid the threefold repetition rule, but it should be doable), 99 half-moves after each of the 125 clock resetting moves, and one last move, the final capture that just leaves the two kings (or if you prefer, the last move can just be any old move, and the 50 move rule applies).

Total of: 99+125+(125*99)+1=12500 half-moves. This is 6250 moves each. The longest games I've heard of have not gone for a tenth of this length.

The question is whether all the pawn moves and captures are possible. The captures would have to be done rather co-operatively, to get the pawns out of each other's way, but I think it can be done, partly because you can promote the pawns and use the newly promoted pieces for subsequent pawns.

Thoughts? Other than "OMG that would be a boring game", of course it would be, it wouldn't be chess, it would be the game of how to get the longest possible chess game, and it would neccesarily be co-operative. This is just an intellectual exercise.

so clearly not all pawn moves and all captures are possible (well, they are, but some of the pawn moves have to be captures too, so they take up double), and for black not to be doing all the work, it can't be 99 half moves all the time. the link (and what it later links to) says 5946 moves overall. (11892 half-moves)

Marvellosity: the pawns were initially counted as pieces that can be taken. Changing them to pieces after promotion doesn't actually add pieces, just changes them.

How long do you think this is? Not the longest practical game, but the longest possible. Any game will end up as a draw if each player plays 50 moves (so 100 half-moves altogether) and no captures or pawn moves are made. 15 captures are possible for each side, andeach pawn can move a maximum of 6 times before becoming a piece, which won't affect the 50 move rule. That gives a theoretical maximum of 96 pawn moves, but are they all possible? With captures, maybe.

If every pawn can make all its possible moves, and every piece is captured, then you have only the kings left, which is a draw under the rules (insufficient material), so the absolute theoretical maximum is 96 pawn moves, 15 captures for one side, 14 captures for the other side (making 125 50-move-clock-resetting-moves), 99 initial half moves before the first capture/pawn move (and this is probably the hardest time to avoid the threefold repetition rule, but it should be doable), 99 half-moves after each of the 125 clock resetting moves, and one last move, the final capture that just leaves the two kings (or if you prefer, the last move can just be any old move, and the 50 move rule applies).

Total of: 99+125+(125*99)+1=12500 half-moves. This is 6250 moves each. The longest games I've heard of have not gone for a tenth of this length.

The question is whether all the pawn moves and captures are possible. The captures would have to be done rather co-operatively, to get the pawns out of each other's way, but I think it can be done, partly because you can promote the pawns and use the newly promoted pieces for subsequent pawns.

Thoughts? Other than "OMG that would be a boring game", of course it would be, it wouldn't be chess, it would be the game of how to get the longest possible chess game, and it would neccesarily be co-operative. This is just an intellectual exercise.

http://blog.chess.com/kurtgodden/the-longest-possible-chess-game-revisited

Schu: also, if all the pawns can become pieces, this adds a further 16 pieces each to be taken.

yeah, again, thanks for the link :)

so clearly not all pawn moves and all captures are possible (well, they are, but some of the pawn moves have to be captures too, so they take up double), and for black not to be doing all the work, it can't be 99 half moves all the time. the link (and what it later links to) says 5946 moves overall. (11892 half-moves)

Marvellosity: the pawns were initially counted as pieces that can be taken. Changing them to pieces after promotion doesn't actually add pieces, just changes them.

ha!

Actually, the game can go on infinitley, someone has to claim a draw after 50 moves without a pawn move or capture.

chessy: well, yes, but firstly infinitley is not a word, and secondly, that leaves no interesting hypothetical debate ;)