Author: Robert Hyatt
Date: 13:25:28 01/25/05
Go up one level in this thread
On January 25, 2005 at 16:13:01, Duncan Roberts wrote: >My understanding of dann corbit is that a 32 piece tablebase needs a 2.5 * 2.5 >cube of crystal. > >http://www.talkchess.com/forums/1/message.html?407489 > > >This is considerably less than a galaxy and perhaps could be slipped pass >greenpeace, although I think you were referring to all games. > >anyway is this calculation anywhere near right ? > > >duncan I honestly have no idea. First, I really don't know how big the actual game tree for chess will be. It could be _very_ big if the game goes on and on with every 50th move being a pawn push or capture to restart the 50 move rule. Or it might be that it is a forced win or draw or loss inside 50 move, with best play. Until we know how big the tree really is, it is hard to predict how much storage will be required to hold it. For the worst case, where some 5500 moves are possible before it becomes a forced draw (forced in that I assume one side will claim the draw if the other side will not, otherwise the game becomes infinite and all bets are off) that is so big that it is impossible to estimate anything about it. There are positions with a branching factor of one (one legal move) as well as positions with a branching factor of over 200. When talking about W^D where D is big and W can get big, I personally lose focus. :)
This page took 0 seconds to execute
Last modified: Thu, 15 Apr 21 08:11:13 -0700
Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.