Re: Re: Status of HAKMEM proposed programs?

Well maybe you can help clear something up for me, since I'm not too familiar with game theory. What exactly does it mean to "solve chess?" Does that mean to find a winning solution for say, white, given any board layout?

"As information travels faster in the modern age, as our days are crawling by so slowly." -- DCFC

Comment on Re: Re: Status of HAKMEM proposed programs?

Replies are listed 'Best First'.
Re: Re: Re: Status of HAKMEM proposed programs? by dragonchild (Archbishop) on Mar 13, 2002 at 16:42 UTC
"Solving" a game means to be able to state an algorithm by which the best move in any situation can be determined, solely by looking at the board (and possibly the moves that have happened already). For example, there is a very simple algorithm that solves Tic-Tac-Toe. ------ We are the carpenters and bricklayers of the Information Age. Don't go borrowing trouble. For programmers, this means Worry only about what you need to implement.	[reply]
Re(4): Status of HAKMEM proposed programs? by FoxtrotUniform (Prior) on Mar 13, 2002 at 17:35 UTC
Put a bit more simply, solving a game is playing it perfectly, all the time. (Hmm. Do you need a decision procedure for perfect play to call a game "solved"? Not AFAIK....) `-- :wq`	[reply]
Re: Re: Re: Status of HAKMEM proposed programs? by danichka (Hermit) on Mar 13, 2002 at 20:28 UTC
To put it another way chess would be solved if we knew what conclusion best play from the position would reach. Whether it be a win for white, a win for black, or a draw. We would also know the fastest way to reach that conclusion for the winning side and the longest way to hold that conclusion off for the losing side. Some endgames have already been solved. All 3 piece and all 4 piece endgames for example. Some of these are really not relevant however, because they are so lopsided it was rather useless to solve. If you want to play around with some of these EGTB visit Robert Hyatt's site.	[reply]