Arimaa Forum - Print Page


    
      
        Arimaa Forum
        (http://arimaa.com/arimaa/forum/cgi/YaBB.cgi)
      

        Arimaa >> General Discussion >> A suggestion
        
(Message started by: DorianGaray on Nov 12^th, 2006, 5:10am)

Title: A suggestion
Post by DorianGaray on Nov 12^th, 2006, 5:10am

If we started out a game with a forty move time reserve (as is done in Chess) surely those timing out troubles would happen a lot less often. Of course someone could then use up their entire reserve in one move but I believe that that would be the lesser of two evils.

Any thoughts?

Title: Re: A suggestion
Post by PMertens on Nov 12^th, 2006, 5:33am

there have been some discussions about the timecontrol-issue and several people agree that managing the time should be totally up to the player ... several did not agree ...

Off Topic ... I would have read your post even if I had known by the subject what it was about ;-)

Title: Re: A suggestion
Post by Fritzlein on Nov 12^th, 2006, 10:07am

I have been opposed to giving a large block of time to the players to manage as they see fit, partly because I know it sucks to be a chess spectator when the players descend into thinks of twenty minutes or longer.

On the other hand, timeouts are really bad too. It wasn't such an issue last year, so I thought maybe we could ignore it this year, but I guess not. Larger reserve may indeed be the lesser of two evils.

One potentially relevant observation about a larger initial reserve: In the postal tournament this year, we realized that everyone had been pinched for time the year before. To compensate, we gave a huge initial reserve. Somehow, the number of timeouts went up! I fear that if we increase the players' control over their time in the World Championship, they will mismanage the time, and we will still have timeouts, just somewhat deeper into the game.

Title: Re: A suggestion
Post by RonWeasley on Nov 12^th, 2006, 6:26pm

I kind of like the shorter limit per move in that it makes decisions with uncertainty part of the game. If you think you're better in a complex position, you can steer the game that way to get an advantage.

IMHO, the breakthrough in computer performance will come when we figure out a better way to address the uncertainty due to the time (and therefore search) limits.