Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: So why *does* Fritz beat Crafty?

Author: Christopher R. Dorr

Date: 07:57:04 03/31/99

Go up one level in this thread


Hello!

I'm *not* trying to say that Crafty will beat Fritz, or that Crafty is the
'best'. It may or may not be. But the original post had devolved into some
questions about 'Best' or 'strongest', which I felt were a) smuggling in
implicit premises that should not have gone unstated, and b) were quite unclear
in what they meant.

I guess my problem is more with the way the questiosn were asked, and the
usefulness of the information generated in response. My points go more towards
methodology, and definitions of strength.

The original question, in some posts, had become "Why is Fritz stronger than
Crafty?" I'm saying that the evidence is not clear that it *is* stronger than
Crafty. Computer-computer tests indicate one thing, personal experience another
thing, and there exists human-crafty data (on ICC among other places) that can
support both points. If the original question is 'Why does Fritz beat Crafty on
an equal platform?', then *none* of my points are valid. If the responses had
stuck to this question, then I doubt I would have said anything. But the
question changed with several postings, with the point becoming more 'Why is
Fritz Stronger than Crafty?'

I'm simply trying to get a handle on the issue. To me, there been numerous
questions (with many different answers) posted by various posters, not at all
asking the same thing. Some of these are:

"Why does Fritz beat Crafty on equal hardware?"
"On optimal platforms, which is stronger?"
"Does superiority in computer-computer testing truly reflect greater strength?"
"Does superiority in computer-human testing better reflect greater strength?"

And so on. People would respond to one of these questions with evidence and
opinions that didn't support it, or supported another question.

There are some interesting points here. Will a SMP Fritz be better than a SMP
Crafty? (Don't know yet...will be fascinating to watch)

What is the objectively strongest (perhaps defined as capable of gathering the
highest rating on, say ICC, given equal formulae) program/system out there?
(probably Crafty on that hopped up Quad)

What's the strongest program available for the average user? (probably Fritz -
who knows?...an entirely different thread)

I'm not arguing with Bruce. I'm not saying that Crafty is necessarily any better
or worse than *anything*. I'm not advocating for Crafty as the strongest
program. I'm saying that I feel that we need to be more clear in what we're
discussing, before the information generated can really be useful.

To me, the question 'What's the strongest program?' is almost as unclear as
'What's the best program?'

I'm really not trying to start a flame war, or attack anyone.

Chris



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.