Minimax

I am writing a game. If you assume I have the game tree, can you suggest a good minimaxing algorithm. I have one, but it doesnt always work brilliantly. Any suggestions.

I basicly find the smallest value of the opponent moves and group them into board positions. (Ie smallest value while at (i,j) when the counters before that are moving across possible positions). Then with the last AI piece I find the biggest. Find where opp was at when that was in that position. And then trace back to were the first piece needs to go to hope opp takes that position.

It seems to work at first, but as more positions are added in game it gives up :(

There is not one, single, all-encompassing algorithm for a minimax search- it's going to depend on many things that only you know. You haven't even told us what game or application this is for.

Developing a good minimax search algorithm isn't an exact science (unless you're talking about something very simple). Chances are you're just going to have to play around with it until you find something reasonable.

But what do you mean by "it gives up"?