Author: K. Burcham
Date: 08:46:09 09/08/01
I find the deep blue-kasparov games very interesting. lots of people do.
Two of the most respected people here are Uri Blass and Robert Hyatt.
here is a comment Uri made the other day:
"I prefer that other people do the analysis with Deep Fritz
because my hardware is not the fastest but unfortunately people
are not interested to try to prove that I am wrong so I am afraid
that I will have to try to prove that my opinion about Deeper blue
is wrong without help from other people".
Uri
we all have read enough of their opinions on this comp-human match,
that we know where they stand. but---i agree with Uri's approach.
so for the last year i have been analysing the kasporov-deep blue
games. i have not posted these results because i have not found any
"mysterious" moves. or moves like the shredder vs deep junior7
example where shredder said "whoops i see it now, let me
adjust my eval". and then shredders eval at +5 was adjusted
to 0 in the world tournament. i have not found
any moves similiar to this. i did find that some of the
moves (very few) were 5 to 8 minutes deep on my 1500 mhz using
deep fritz, tiger 14, and shredder5. there were several moves
that deep blue chose that deep fritz would have the same eval
for another move and it would choose the other move.
i also have not posted any of my results because i know the type of
comments that Robert makes when refering to the IBM machine and its
specialized processors and its software. Robert will make a technical
comment about the hardware etc., but not position evals. although
i did see a follow up this week where he posted a crafty eval
on a deep blue position that Uri had.
we all have tremendous respect for what these technicians accomplished
with deep blue. it was so amazing that if you just mention deep blue
the threads here go crazy. i cant say enough about what they did.
but----what i would like someone to post here are the moves
i missed. i downloaded all the games in word, very easy to access.
i dont think i missed any moves, if you know of any positions
that you think any of todays top programs on 1500+ mhz cant find
please post them, or some of them,,,,i cant find any.
so i have concluded after lots of time analyzing deep blue positions
that todays programs seem to be very close or equivelant to deep
blue in playing strength.
if you disagree with this, could you post a position, or
positions, that you have found, or you might just think is, or
someone told you it was, or a position you have analyzed
but not with a pc. in other words i am looking for any positions
that my system will not choose deep blues next move. or does
not see deep blues next move as an equivelant eval.
also by the way, it was very interesting to see how many
moves the programs chose of kasparovs moves.
kburcham
This page took 0.21 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.