Author: Vincent Diepeveen
Date: 08:20:55 03/13/03
Go up one level in this thread
On March 13, 2003 at 10:25:31, Angrim wrote: >On March 13, 2003 at 01:52:34, Russell Reagan wrote: > >>I read that in the game of amazons, the initial position has 2176 legal moves! >>Is there a game with a higher branching factor? > >Leaving out the obvious non-chesslike games such as "Age of Empires", the >game with the highest branching factor that I have thought about writing >an engine for would be "progressive chess" in which white gets 1 move, then >black gets 2, then white gets 3, then black gets 4, then white gets 5, etc... >In theory the branching factor is infinite. In practice the games never last >long enough to get beyond a branching factor of 36^10 or so. > >more complex rules at: >http://www.chessvariants.com/multimove.dir/progressive.html > >Angrim This is not a realistic game. If someone doesn't win within 7 ply, then something went horribly wrong there. In netherlands it was called 'scottish chess' a few years ago. I played it a few times and always lost or won within 7 ply. Or 3.5 moves. 7 moves in a row for 1 strong player is just too much. I only remember first few games where major mistakes were made once the 9 ply was achieved. This game is therefore clearly not a serious games. The only serious game with a big practical branching factor is clearly backgammon. Go is not realistic to mention in that respect. My very weak Go program has a branching factor of 10.0 in the openings position and it obviously will not go up further in the game, only down. Also when searching deeper than the 6 ply it searches now in Go, it clearly means that the effect of the first move gets less, so the b.f. goes down considerably if ever 10 ply gets achieved. Best regards, Vincent
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.