Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Deep Blue and the

Author: Robert Hyatt

Date: 13:51:28 11/11/98

Go up one level in this thread


On November 11, 1998 at 15:44:32, Peter Hegger wrote:

>On November 10, 1998 at 17:27:43, Robert Hyatt wrote:
>
>>On November 10, 1998 at 10:56:40, Peter Hegger wrote:
>>
>>>On November 09, 1998 at 14:30:31, Robert Hyatt wrote:
>>>
>>>>On November 09, 1998 at 12:39:46, odell hall wrote:
>>>>
>>>>>
>>>>>On November 09, 1998 at 09:21:50, Peter Hegger wrote:
>>>>>
>>>>>>Hello, My name is Deep Blue
>>>>>>
>>>>>>Did I ever tell you about the ten game match I played a few years back?
>>>>>>Well, I played all the leading micros and made a perfect 10-0 score against
>>>>>>them. Wiped them all off the board easily.
>>>>>>What? You want to see the games?
>>>>>>Oh no sorry, I couldn't possibly do that.
>>>>>>You'll just have to take my word for it that it really happened.
>>>>>
>>>>>
>>>>>You are exactly right! If the match did in fact happen why haven't they
>>>>>published the games? The kasparov match is over so they can't fear that kasparov
>>>>>will see the games. What other possible motivation for not revealing the games
>>>>>other than that it never occured!!
>>>>
>>>>If you want to believe that pigs can fly, that there is a free lunch, and that
>>>>this match never happened, go right ahead.  P.T. Barnum had a comment that fits
>>>>really well.
>>>>
>>>>As to why won't they say anything?  After all the obnoxious posts here about
>>>>deep blue and how badly it plays and how lucky it was to beat kasparov and
>>>>how it can't hold a light to todays micros, maybe, just maybe, they don't give
>>>>a damn about what anyone here thinks?  Wouldn't blame them in the least...
>>>>
>>>>Never seen such a bunch of closed minds...  of course when something is closed
>>>>that tightly, it is *very dark* inside...
>>>
>>>Quite the contrary.
>>>It is because I am looking at the issue with an open mind that I pose the
>>>question about the games.
>>>For Hsu and Campbell to say that these games were not considered important
>>>enough to record strains credibility.
>>
>>
>>Why would this be true?  I played a long match vs genius several years ago and
>>beat it nearly every game (using a Cray) and with significant time odds to boot.
>>I expected to win, I did, and I didn't save a single game score because I didn't
>>consider those games particularly remarkable...
>
>So you're saying you tied up the resources of a multi-million dollar computer
>just to have a "for fun" match against genius? I didn't realize computer time on
>mainframes was that easy to come by.

For me, it was... if the "time" wasn't important.  I used to get all the
dedicated YMP or C90 time I wanted...  at say 2am-6am in the morning.  In
the case of the above match, the purpose was to test our search extensions
and eval changes, getting ready for an upcoming ACM event...

>
>>I'd suspect Hsu did the same.  They told me about the match months after it had
>>been done.  By the time I reported it the data was gone.  What was worth saving?
>>
>>IE I play new programs on ICC all the time, and they generally lose every game
>>for the first few months until they become more "complete."  I don't file those
>>games away... what would be the reason?
>
>No reason at all to save games like this, I can play the same programs at the
>kitchen table. But on the rare occasions when one has access to "big iron" I
>would think any game would be worth recording.
>Just my opinion
>Peter

That is the difference between someone that sees this every day, and someone
that hardly ever sees a supercomputer play chess.  What is mundane to me might
be earth-shaking to you..  and yes, that can be a problem...  but take the above
and think about things from Hsu's vantage point..



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.