AI for 2D logic game

Hi,

I'm currently working on my first game,. It's a logic based game. I didn't plan it in detail so I won't really explain it in detail. I'll just explain the mechanic which is important to know for my question.

Basically you have a playing field which consists o 2x2 fields. On each field there is one of 8 elements. The goal is to combine these elements in a good way. Also there are bonus elements which do stuff like double the received points, making more combos possible etc. If you chose an element, it'll randomly get replaced by a new one. If a bonus element is used, it also gets replaced randomly. The bonus elements are player specific, like the card sets on magic the gathering.

In IRC someone told me about miniMax and Alpha-Beta Pruning, I read a bit about it. If I understod it correctly, it bases on knowing full information and precalculating every possible end situation of the game. My problem is, that I don't have that much of information. Of course, since my elements are limited I can easily guess the possible states. Also there are way more than 8 bonus elements, there will be like 80 or something. A player can only have like 10 bonus elements on each game.

Anyway, of course, I can predict every possible way but since I have to predict that much, the whole thing will explode like chess. So, do you think miniMax and alpha-beta is the correct way to go?

Thanks.

 

push