Author: Chuck
Date: 10:10:35 05/29/99
Go up one level in this thread
On May 28, 1999 at 23:28:50, Joe Koss wrote: >Curiosity killed the cat.. > >This question is for both engine authors and chess players alike: > > How much mobility is a pawn worth? > >On one side of the equation, you can sac a pawn for increased mobility and on >the other side, you can 'save' a pawn at the expense of decreased mobility. > >Now, mobility isn't as straight forward a concept as I once thought.. The >effectiveness of my mobility is in direct relation to my opponents mobility. > >n = (my mobility) / (opponents mobility) > >how large does 'n' have to be before it is enough compensation for a pawn? First, I think this question should really be left for those with a lot of experience. But I think they would say that it depends on the position. The extra mobility is only worth something if you can do something with it. For a computer, this is more difficult, because you have to determine in code whether you can achieve a "promising" position from the pawn sac even if you can't see material gain. The ideal, imho, would be to recognize that with the extra tempo you can open lines for an attack or get your pieces lined up for a promising attack against the opponent's King. Because you are going a pawn down, I don't believe it's something you want to do unless you can see a strong attack or you want to exchange into what you KNOW would be a drawn or winning endgame. Of course, if you decide to play for a drawn ending, that would be because your opponent already has an edge. Anyone else with input on this question? Chuck
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.