Author: Drexel,Michael
Date: 13:28:18 04/16/05
Go up one level in this thread
On April 16, 2005 at 14:12:01, Roger D Davis wrote: >First you said... > >"No way, this can be fixed easily without any decrease in strength." > >Then when I ask "What would be an easy way to fix it without decreasing >strength?" you said... > >"You certainly can not solve the problem generally without decreasing strength >but it is at least possible to prevent the engines from playing this beyond >stupid move." You then said that preventing such moves might indirectly increase >strength. > >These statements appear to be mutually exclusive, or maybe you just changed your >mind? In my first statement I only referred to the stonewall setup. It shouldn´t hurt playing strength if you force the engines to go for b4-b5/b5-b4 in such positions and avoid a4-a5/a5-a4. The Haglund game was not a stonewall. It was a different pawn structure. I don`t think it is that easy to fix in general. If you did change your mind, I can certainly appreciate why...if it was >easy to fix without any decrease in strength, the Shredder or Hydra team would >probably have already done it. I do not think they are very concerned about this. > >Seems like the easiest way to deal with this would be by tailoring the book to >play ONLY open positions against human players. In other words, have the >computer use ANTI-HUMAN play before the human can use anti-computer strategies. >The operator should be able to choose what book the machine plays before the >game begins, just a human chooses his opening based on his opponent. Yes, but it is not that easy to avoid this positions. The human player who uses such a strategy will certainly try to lure the engine into closed positions by playing provocative moves. Michael > >However, if you're looking for a set of "chess general heuristics" that provide >the best move in any circumstance, then I guess you wouldn't want to exclude the >possibility of such positions. I guess your design depends on your >goals...desire to win the most games or desire to advance the art. > >Roger
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.