Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Elo of DeepBlue

Author: Dann Corbit

Date: 11:07:43 03/16/01

Go up one level in this thread


On March 15, 2001 at 23:13:20, Terry McCracken wrote:
[snip]
>Dan I agree in part but not in whole.
>I see it more in an opposite light. I don't believe this is a proper conclusion
>to a very very advanced  scientific project. It reminds me of the Apollo Project
>cut off as it was just getting interesting. It still will take many more years
>before we go back, let alone to Mars.

FWIW, I think the Hubble project is a trillion times more interesting than a bag
of rocks from the moon.  And the next generation space telescope even more so.
On the other hand, I do think space exploration is interesting.  Is it
interesting enough to send someone to Mars for another bag of rocks at a cost of
100 billion dollars?  I also think that a mission to Mars is something that
could possibly exterminate the human race, so I'm not entirely convinced it is a
great idea to go there and carry stuff back.  If there are Martian microbes, we
certainly would not have any immunity against them.

>Yes I agree to a point with many of your points!;) But much more needs to be
>done to bring chess to a true GM and WCC level. This was not accomplished and
>I'm not in a dream world I assure you.

I have no arguments with this point.  But creating 10 million dollar machines is
not the answer.  We are talking about a game, after all.  We're not solving one
of the deeper problems of mankind by coming up with an engine that can beat
Kasparov and Anand.

>Facts about Deep Blue and strength and how far it could be taken have been left
>to debate, which means far more work needs to be done to make this "Science"
>rather than fancy "Technology" with a large array of tricks to succeed in
>beating Kasparov.

That might even be the best part of it.  Here it is, 4 years after the fact and
it still comes up in conversation on a daily basis.  If we had a certain answer
to all the questions would is still be like that?  I doubt it.  On the other
hand, I want to know the answers, even if knowing them is not going to promote
interest in more research.

>Yes Kasparov behaved poorly and that truly was a shame, a " Dark Day " for
>Kasparov and the rest of the chess world. He's the main culprit! Sigh....
>He's suppose to be an example after all! But he's human and caved in to
>suspicions, conjecture etc.

I am harsh towards Kasparov, but (on the other hand) I suspect that many, many
great chess players might have reacted similarly.  I don't think that anyone on
the planet can imagine the tremendous pressure he was under.  It was a sort of
"John Henry" thing with literally billions of eyes watching.

>He lost due to psychological factors not that Deep Blue could out play him in
>_normal_ match settings.

I don't think anyone knows for sure why he lost.  Maybe purely outplayed.  Maybe
purely freaked out.  Maybe some wild combination of factors.  At any rate, both
Kasparov matches are without question the most famous chess matches ever to have
occurred on the planet.  They were the most interesting chess matches ever.

It *would* be wonderful to have Deep Blue to play with and do whatever
experiments we would like.  But, from a business standpoint, it really does not
make sense for IBM to do that.  Remember why they did it in the first place.
IBM is a business, and they are in the business of making money.  They have
stockholders who want to see a return on their investment.  Here at CCC, we
might not mind if we held $100,000 of IBM stock and they spent millions of
dollars fiddling with chess games.  But 99.99% of the world's population would
not think that way if they held IBM stock.  If IBM spends a big plug of money on
something, the stockholders (justifiably) want to see some kind of a return.  In
other words, the board of IBM owes it to those who pay the bills to react in
exactly the way that they did.



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.