Author: Ricardo Gibert
Date: 08:14:42 08/02/01
Go up one level in this thread
On August 02, 2001 at 10:53:02, Ed Panek wrote: >Lets say 2 programs are playing a standard game. the opening books lead to >instant replies to move 20. then at move 20 your chess engine begins pondering >its move for 7 minutes and then it moves. The game ends in a win for your >program. Is there a danger to adding this 7 minute pondered move into an >extended type opening book file? So next time it plays this opening it will see >that for 7 minutes it saw this as the movelast time...so why waste 7 minutes >again reaching the same conclusion? play the move right away. > > >Ed This sort of thing is possible, but you need to take other variables into consideration such as: (1) The hardware, since a minute on a 500mhz PII is not the same as a minute on a 1.4 Mhz Athlon. (2) The new move may have led to a poor position, but the program won anyways due to a blunder. (3) The time controls, since the new move may have been from a bullet game and only a few seconds were used to generate it. (4) The rating of the opponent, since what works well against a weak player may do poorly against a strong player. This is related to (2). I'm sure there are other issues to take into consideration, but the above list is sufficient to show it is not so simple, but there is no reason why this type of idea should not be workable with a little care.
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.