Author: Don Dailey
Date: 09:40:05 03/03/99
Go up one level in this thread
On March 03, 1999 at 09:59:51, Robert Hyatt wrote:
>On March 03, 1999 at 04:36:43, Lin Harper wrote:
>
>>Kaspy should never have agreed to such a short match. He would surely have
>>caught up and passed the computer if he had time.
>
>I think Kasparov might be one of the best two or three players of all time.
>But in _that_ match with deep blue, I personally believe that if the match
>had gone two more games, it would have been two more won games for DB. Kasparov
>"lost it" somewhere in the first 3-4 games and never recovered. And I don't
>believe he would have recovered unless there had been a month break in the
>match which would not have happened.
>
>The last two games of the match had him looking 'beaten' before he ever made
>the first move. I think it would be very difficult for him to 'come back'
>with additional games, given the way he appeared to feel...
This is pretty speculative but it's certainly not implausible. I
personally thought Kasparov was tougher than that but now I'm not
sure. Up until the last game I really feel as if he had outplayed
Deep Blue for the most part, but let the fact that he couldn't quite
"close" psyche him out and lose confidence.
Having said that, I believe that credit should be given where credit
is due. Kasparov's own psychology is one of the variables here and
is built into his rating and his performance. So for me to say
Kasparov "should" have won this match is not really an accurate
statement, because he didn't. The fact that Deep Blue DID hold
on proves it was able to take advantage of a weakness in Kasparov,
also a variable in the equation. Whoever faces the computers of
tomorrow will have to deal with this too if they want to claim
superiority.
I'm not sure it would have mattered given Kasparov's phychology,
but at the time I thought is was very foolish of Kasparov to
play such a short match. Perhaps he knows nothing about math
or was more interested in the money, but if it were me and I
was in a position to make lot's of demands before I would accept
a match, I would have insisted on a reasonable number of games,
unless of course I percieved myself as being the weaker player.
I'm tired of speculating however. The truth of the matter is
that Deep Blue won. The only scientifically valid conclusion
you can draw from this is that it is slightly more likely that
Deep Blue is the better match player. Anything more than this
is a very fallible judgement call and speculation.
I have also been speculating a lot about how good Deep Blue is
compared to todays Micro's by comparing Deep Thought's results
at the last ICCA tournament. This is a horrible comparison
however because of at least 3 reasons:
1. Only 5 games were played by each, that means almost nothing.
2. Todays micro's are completely different programs.
3. Deep Blue is a completely different program than the Hong Kong
program they called "Deep Blue prototype" which was actually
not Deep Blue at all. More different than even the micro's.
Speculation is a lot of fun but there is frustratingly little data
to do a good job of it. I also believe it is also a good thing to
accept the results we get. We can alway speculate till we are blue
in the face but a small amount of "real" information and results in
a controlled and well documented setting means more to me than
all the anecdotes and second hand stories (and opinions) in the
world.
- Don
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.