Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: draw-score behaviour is in DeepFritz T28 too...

Author: Peter Berger

Date: 13:50:30 12/11/00

Go up one level in this thread


I can't understand your point of view .

You say Fritz doesn't understand a certain type of positions well . OK .

You probably will agree that still it remains at or near the top .

Isn't it clear that it understands some other positions better than the rest
then ?

If a program by "bean-counting" delivers a surprising mate in 12 outsearching
the opponent badly , why is this less valuable than judging some promising king
attack with +5 although not being able to see it to the end ?

If you look for data supporting my point of view : just look at any tactical
testsuite and see who is at the top of the solvers .

I think the Fritz way is only less popular because it is less similar to the
human style where a program like Tal seems more human as its strengths and
weaknesses correspond better with the ones you know from yourself.

So : if this is the case it is simply a matter of taste : do you like what you
get with Fritz :-) ?

Do you prefer Bach, Wagner or Dittersdorf :-) ?

Another question is : which program is best for analysis ; but again : cruel
tactics _is_ a part of the game . And if others with new paradigms will take
over I wouldn't be too surprised if the following Fritz version suddenly
displayed some most surprising new-paradigm behaviour .








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.