Author: Stephen A. Boak
Date: 19:05:51 11/20/05
Go up one level in this thread
>On November 20, 2005 at 09:00:31, Nelson Hernandez wrote: >It seems to me that depriving engine developers the ability to >outwit their opponents in the opening is a lot like asking two boxers to fight >with one hand tied behind their backs. It may be a fair fight, but it would be >unnatural. As an improved example, it's like asking all boxers to fight with their left hands tied behind their back, so all fights are identically conducted with right hands only. Fully equal conditions, right? Yet some of the boxers are naturally left-handed, or right-handed but better with their left hands! There is a famous quote by the great American poet, Ralph Waldo Emerson: "A foolish consistency is the hobgoblin of little minds, adored by little statesmen and philosophers and divines. With consistency a great soul has simply nothing to do." Each program / hardware combination has its own advantages & disadvantages. Making a test foolishly consistent is full of hobgoblins. It will likely be 'fair' to some programs, but 'unfair' to others. With consistency, the great programmers & programs will have simply nothing to do. Why limit them in any way, from doing what they think is best. Let's get rid of the hobgoblins before they take over our minds! In professional checkers, I believe they sometimes randomly draw an opening and make the players play whatever is selected--not just varying the initial setup as in FRC (in checkers it's always the same), but the next move or two. I believe this is done because certain opening lines are very well known (maybe drawish), or overly favorable to one side. In chess, the variability of the opening gives room to players, programmers and programs to try to outbook, outwit, outplay the opponent. Nobody forces you to play the Sicilian, for example, in response to 1.e4. You may better understand, or better enjoy the Caro-Kann or Petroff Defense. You may have a greater chance of outplaying your opponent in something your program understands better. You alone decide which lines are suitable for your program, and how much risk you wish (or need) to take--to best attempt to achieve your game/tournament/rating/prize goals. --Steve
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.