Board Game Arena

I just started playing Oxono in the last few days. Its a delightful little game. I have not yet come up with an opening strategy, though with the symmetry of the board, its seems some preferred ones will appear.

My thought is on machine learning with this game. The number of opening combinations does not appear oppressive to me for CPUs to calculate with the symmetry of the board. Additionally, because pawns remain in their place the entire game, the moves become forced pretty quickly. I have often found myself knowing I will win or lose after my fifth or sixth move.

I assume Oxono is easily solvable by ML- somewhere closer to Connect Four than Checkers. I will enjoy it until someone unleashes AI upon it, but I suspect my days of enjoying this game are short lived.

Given Tic-Tac-Toe, Quarto and Connect 4 are solved games and they all share similitarities with the game in discussion, I believe that Oxono is probably solvable as well.

Still it is very compelling and it's one of my favorite games ever, even though it has just been released.

Every non-random game is solvable eventually.

Yeah, more or less axiomatically, any turn-based game of perfect information is necessarily solvable, though whether it will be solved is a separate matter. But without any ambiguity or doubt, it can be solved, at least.

Is there enough depth for the human mind? For me that is the ultimate question. Is it solvable by the human mind? I am assuming it has more depth than connect four, and it would be very rare for a human to play a perfect game of connect four, So I don't see any real problem. It is truly a great game. Where would others place the depth of the game? I might be inclined to place it somewhere between connect four and Checkers.

McCoo wrote: ↑25 March 2024, 12:38 Every non-random game is solvable eventually.

Chess and Go have yet to be solved and likely never will, at least not in our lifetimes.