Author: Christophe Theron
Date: 18:18:37 02/21/00
Go up one level in this thread
On February 21, 2000 at 16:17:22, blass uri wrote:
>On February 21, 2000 at 14:11:25, Christophe Theron wrote:
>
>>On February 21, 2000 at 08:00:13, blass uri wrote:
>>
>>>On February 21, 2000 at 07:18:23, leonid wrote:
>>>
>>>>On February 21, 2000 at 02:45:11, blass uri wrote:
>>>>
>>>>>On February 20, 2000 at 21:05:47, leonid wrote:
>>>>>
>>>>>>On February 20, 2000 at 19:25:10, blass uri wrote:
>>>>>>
>>>>>>>On February 20, 2000 at 14:39:24, Christophe Theron wrote:
>>>>>>>
>>>>>>>>On February 20, 2000 at 01:39:09, Drazen Marovic wrote:
>>>>>>>>
>>>>>>>>>What is Botvinnik's legacy to computer chess?
>>>>>>>>
>>>>>>>>
>>>>>>>>That to write a good chess program it's better not to be a strong chess player.
>>>>>>>
>>>>>>>I do not agree about it.
>>>>>>>You cannot teach your program things that you do not know.
>>>>>>
>>>>>>You don't teach your game to play but you depose exact logic to go after.
>>>>>
>>>>>I see teaching a program to play the same as deposing exact logic to go after.
>>>>>
>>>>>You cannot depose exact logic that you do not know about.
>>>>>
>>>>>One example:
>>>>>You cannot teach program that KRB vs KRP is usually a draw and that the
>>>>>evaluation should be close to 0.00 if you do not know it and your evaluation by
>>>>>only counting material may be +2 and you cannot see the 0 by search because you
>>>>>cannot search deep enough.
>>>>>
>>>>>Uri
>>>>
>>>>Maybe here it is possible to be lost somehow because of the nebulosity of
>>>>expression like "go after the logic". If the "go after the logic" represent some
>>>>transmission to the game certain human experiece about the game, this experience
>>>>is rather secondery. Such human experience could be recognition of certain pawn
>>>>structure, that make certain position vulnerable, or certain pieces too weak
>>>>when situated in certain board squares. Such a human know-how is not compulsory
>>>>for writing the good chess game. Base of the game can be all the time exact
>>>>calcualtion of the position based on material exchange. All human know-how is
>>>>nothing more but final touches for the game that is already written. And game
>>>>need it only because the hardware is too weak to reach the final solution on its
>>>>own.
>>>
>>>I use the word chess program instead of game and I understand that you mean
>>>chess program when you say game.
>>>The word game has nothing to do with programs and computer except the fact that
>>>programs can play a game.
>>>
>>>I agree that chess programs need need evaluation function only because the
>>>hardware is too weak but the fact is that practically the hardware is not going
>>>to be good enough in the next few years
>>
>>
>>You can even say that it will NEVER be good enough. This has been stated since
>>the very beginning of computer chess by Shannon himself.
>>
>>This definitely throws out the idea of "material only" evaluations.
>>
>>
>>
>>
>>> and it is important to have knowledge
>>>that the search cannot detect if you can do it without doing the program slower.
>>>
>>>I do not say that I am sure that adding the knowledge of KRB vs KRP is
>>>productive because if the program is slower because of this knowledge it can be
>>>weaker but programmers who do not know about this rule even do not think about
>>>adding this knowledge so they are at disadvantage relative to programmers who
>>>have more knowledge about chess.
>>
>>
>>Knowing this does not make you a champion. Any weak club player knows this.
>>
>>I did not say that knowing nothing about the game is an advantage.
>>
>>I said that being very strong in chess is not an advantage when you write a
>>chess program. Being an average player is more than enough.
>>
>>Either you try to distort my initial statement, or you did not understand it.
>
>I understand your initial statement and I only gave the example of KRB vs KRP as
>an example.
>I know that weak club players know this rule but there are other rules that they
>do not know and the point is that knowing more rules can be productive because
>you have more ideas.
>
>I understand your point that generalizations are important but I think that you
>can still use rules when to use the knowledge about special cases.
>
>You can do a slow searcher without using it most of the game but use it only in
>special cases.
>
>I think that it is possible to check at the root position the rules that may be
>important for the near future of the game and tell the program to forget about
>the other rules.
>
>Uri
Many programs do that already.
From my own experience, most rules used by human players do not help a chess
program to play better.
I don't mean that no rules coming from human experience will help, I mean that
most of them don't help.
So the job of the programmer is to carefully select which rules to use. In the
process, you can have to discard rules that is considered as very important by
human players.
Just an example: the concept of "tempo" is absolutely useless for a chess
program. Just my opinion of course. I don't see what to do with this concept,
and I don't think my program is lacking it.
If someone uses this concept in his chess program, I'm interested to hear about
it.
Christophe
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.