Author: Uri Blass
Date: 00:38:56 06/05/01
Go up one level in this thread
On June 05, 2001 at 02:38:06, Christophe Theron wrote:
>On June 04, 2001 at 22:05:16, Dann Corbit wrote:
>
>>On June 04, 2001 at 21:15:35, Christophe Theron wrote:
>>
>>>On June 04, 2001 at 20:56:50, Dann Corbit wrote:
>>>
>>>>On June 04, 2001 at 20:49:45, Christophe Theron wrote:
>>>>
>>>>>On June 04, 2001 at 18:15:03, Dann Corbit wrote:
>>>>>
>>>>>>On June 04, 2001 at 17:44:56, Christophe Theron wrote:
>>>>>>[snip]
>>>>>>>What a major mistake. Strong chess players are not the people of choice to
>>>>>>>create a strong chess computer.
>>>>>>
>>>>>>Berliner did OK.
>>>>>
>>>>>Frans Morsch
>>>>>Amir Ban
>>>>>Christophe Théron
>>>>>Christian Donninger
>>>>>Steen Suurballe
>>>>>Mark Uniacke
>>>>>Rudolf Huber
>>>>>Ed Schröder
>>>>>Bob Hyatt
>>>>>Johan de Koning
>>>>>Marty Hirsch
>>>>>Stefan Meyer-Kahlen
>>>>>Richard Lang
>>>>>
>>>>>
>>>>>These are the names of the programmers of the highest rated chess programs
>>>>>(order of the list).
>>>>>
>>>>>How many of these guys are "in the top flight of their national chess rankings"?
>>>>>
>>>>>How many of these top programmers are you going to reject by selecting people
>>>>>who are "in the top flight of their national chess rankings"?
>>>>>
>>>>>For your information I don't think I would get a 1800 FIDE elo rating if tried.
>>>>
>>>>I know for sure several are 2000 or better. Ed Schroder is a very good player
>>>>(IIRC -- at least I have seen games of his with brilliant moves). Robert Hyatt
>>>>was around 2000 [again IIRC].
>>>>
>>>>Be that as it may [and even if wrong], Berliner was a world champion and also a
>>>>GM.
>>>>It only takes a single counterexample to refute a statement {mathematically!}
>>>>;-)
>>>
>>>
>>>
>>>Then mathematically it is obvious that selecting programmers because they are
>>>good at chess is a sure way to reject people who are very good at chess
>>>programming.
>>>
>>>Mathematically the fact that some 1800 chess players are high in the above list
>>>is enough to refute their student selection system.
>>>
>>>BTW 2000 elo is not much nowadays. In most countries you need to be much better
>>>than that to be listed in the top 100.
>>
>>Yes. Actually, I misread your post. I thought you said that none of the
>>programmers could reach 1800.
>>
>>Anyway -- you are right that it is clear you do not need to be a GM to write one
>>of the world's best chess programs. But I doubt that it hurts. A combination
>>of GM and top level programmer (if you could find it) would be hard to beat.
>
>
>
>I'm not sure. In my opinion a very strong chess player has too many prejudices
>about how a program should think.
>
>From my own experience success in chess programming comes only when you admit
>that you have no idea how chess programming has to be done. Then you are ready
>to open your mind and discover new territories.
>
>It does not mean that a GM programmer could not understand that, but I believe
>that being a GM does not help to pass the difficult moment of "OK, I realize I
>understand nothing to the matter and I'm now going to think about it in a
>totally different way".
>
>
>
> Christophe
I believe that thinking like that can help people to become GM's.
Trying to think about the position in the chess game in a totally different way
is one of the things that help GM's to become GM's.
Uri
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.