Author: Gordon Rattray
Date: 09:47:40 08/24/00
Go up one level in this thread
On August 24, 2000 at 12:06:57, Christian Pike wrote: >On August 24, 2000 at 11:36:56, Gordon Rattray wrote: > >>So, do you think that endgame table bases should be removed too? Where do you >>draw the line? Do we only have programs that figure everything out for >>themselves except for the rules of chess? Human chess players don't figure >>everything out for themselves - we use the advice of others in the opening as >>well as other phases of the game. It is up to us to decide whether to use >>advice as it may be good or bad, that's just part of the challenge of chess so >>it should not be removed. > >the problem is that the engine cannot decide which opening to chose, but the >programmer does, or his advisor. >IMO there should be an algorithm that the engine can decide too. >it can take an advise from book and correlate this advise with its own >ideas about the positio. Why do you think there should be an algorithm? To allow the program to play better? <snip> > >>Do you honestly believe that a program can win a "strong" championship by having >>a really good opening book but play badly afterwards. I very much doubt it. > >yes - i have experienced in many tournaments that the opening decides much. >i have seen programs, very strong programs lose because of opening >bugs they were not in charge but the advisor/programmer before the game. >strong programs giving away the chance to win the title because of dump >opening decisions. That was not my question. I asked if a weak engine could win with a good book, and not if a strong engine may lose because of a bad book. > >i can show you examples. but in the moment i am busy - like all - watching >the games. but it is IMO a pity that the engines cannot decide themselves. No need. I too have seen strong programs lose because of poor books. So, the programmers need to improve their program's opening play. Whether this involves a better book, the engine itself, or a combination is up to them, but they need to accept the outcome of their decision. > >>If they do, then they have weaknesses in their ability to play chess so they >>deserve to lose. > >nonsense IMO. >choose an opening YOU decide, throw the opponent out of book. >but have the initiative about the opening-decision. >not follow some stupid openings humans have played, there is no reason >to do so. >for a tournament you must have completely different tactics concerning >opening than for normal play. What if getting out of book involves accepting an inferior position? I'd rather have a program that plays the opening well due to using a book, than a program that plays the opening badly because it feels that it doesn't need to take advice. The majority of human players accept opening advice and apply it in their games, just like direct file access.
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.