Reply
 
LinkBack Thread Tools Display Modes
  #1   Report Post  
Old October 10th 05, 06:32 AM
LSD
 
Posts: n/a
Default Game Over: Kasparov and Deep Blue

I watch this movie tonight, sort of, while I was working on something.

I found it interesting..and was shocked to hear Kasparov say he was cheated.

Anyone got strong (or better still, informed) opinions about this. Did the
Deep Blue team cheat and have grandmasters available to kill off errant
candidate moves for DB when Kasparov got tricky, or is Kasparov a whiner who
needs to make excuses when he loses?

Personally I like Kasparov (he will make a good politician), and I dislike
cheaters...but I can't believe those computer scientists would cheat.


  #2   Report Post  
Old October 10th 05, 09:05 AM
Martin Brown
 
Posts: n/a
Default

LSD wrote:
I watch this movie tonight, sort of, while I was working on something.

I found it interesting..and was shocked to hear Kasparov say he was cheated.

Anyone got strong (or better still, informed) opinions about this. Did the
Deep Blue team cheat and have grandmasters available to kill off errant
candidate moves for DB when Kasparov got tricky, or is Kasparov a whiner who
needs to make excuses when he loses?

Personally I like Kasparov (he will make a good politician), and I dislike
cheaters...but I can't believe those computer scientists would cheat.


I don't think they cheated at all. They played all out to win the match
- different matter entirely.

There is a pretty good book by the chip designer Feng-Hsiung Hsu called
"Behind Deep Blue" that describes the trials and tribulations of the
Deep Blue design team from their very earliest efforts at MIT.

Regards,
Martin Brown
  #3   Report Post  
Old October 10th 05, 10:25 AM
Ruud
 
Posts: n/a
Default

Kasparov, brilliant at chess as he was, prob. was a whiner.
Why else would someone who believed he was the best player be intimidated by
such a possibility of computers being supported by other players.
It's a contradiction.
The move b3 (do I recall this correctly?) was unexpected back then, but many
engines nowadays will consider it the best move.
At some point engines were bound to overtake humans in logica.
"Martin Brown" schreef in bericht
...
LSD wrote:
I watch this movie tonight, sort of, while I was working on something.

I found it interesting..and was shocked to hear Kasparov say he was
cheated.

Anyone got strong (or better still, informed) opinions about this. Did
the Deep Blue team cheat and have grandmasters available to kill off
errant candidate moves for DB when Kasparov got tricky, or is Kasparov a
whiner who needs to make excuses when he loses?

Personally I like Kasparov (he will make a good politician), and I
dislike cheaters...but I can't believe those computer scientists would
cheat.


I don't think they cheated at all. They played all out to win the match -
different matter entirely.

There is a pretty good book by the chip designer Feng-Hsiung Hsu called
"Behind Deep Blue" that describes the trials and tribulations of the Deep
Blue design team from their very earliest efforts at MIT.

Regards,
Martin Brown



  #4   Report Post  
Old October 10th 05, 11:44 AM
David Richerby
 
Posts: n/a
Default

Ruud wrote:
Kasparov, brilliant at chess as he was, prob. was a whiner.
Why else would someone who believed he was the best player be
intimidated by such a possibility of computers being supported by other
players.
It's a contradiction.


No it isn't. A strong player in tandem with a strong computer is much
stronger than either on their own. See, for example,

http://www.chessbase.com/newsdetail.asp?newsid=2461

Summary: a recent tournament for people with computer assistance was won
by a team of two people rated USCF1685 and USCF1398 using three computers.
Other players in the tournament included two Hydra machines, one running
unattended and one running with minimal human input: neither made the
quarter-finals. Third- and fourth-placed overall were grandmasters
playing with computer assistance so the field was by no means weak.


Dave.

--
David Richerby Salted Apple (TM): it's like a tasty
www.chiark.greenend.org.uk/~davidr/ fruit but it's covered in salt!
  #5   Report Post  
Old October 10th 05, 11:52 AM
Bateman
 
Posts: n/a
Default

No you dont recall correctly. The move was bd4. It was not a move a
computer would have chosen for the simple reason that it allowed
Kasparov to draw with a perpetual check. Kasparov overlooked this
forced draw at the time, but Deep Blue must have been aware of it. Deep
Blue isn't programmed to make moves which Negatively affect it's
evaluation, but at the time it played bd4 DB already had itself better,
so why would it have played a move it would have seen as creating a
0.00 evaluation? When Fischer accused the Soviets of rigging the
candidates tournaments everyone said he was whining too. Why do you
refer to Kasparov as brilliant in the past tense? Is he past it now? Do
you think he'd be a sucker for your Kings Indian down the club? Dont
tell me, you beat him already, right?



  #6   Report Post  
Old October 10th 05, 12:09 PM
Gian-Carlo Pascutto
 
Posts: n/a
Default

Bateman wrote:
No you dont recall correctly. The move was bd4. It was not a move a
computer would have chosen for the simple reason that it allowed
Kasparov to draw with a perpetual check. Kasparov overlooked this
forced draw at the time, but Deep Blue must have been aware of it. Deep
Blue isn't programmed to make moves which Negatively affect it's
evaluation, but at the time it played bd4 DB already had itself better,
so why would it have played a move it would have seen as creating a
0.00 evaluation?


The draw was simply outside the search horizon.

IBM published the logfiles later on, and furthermore it can be easily
verified that modern programs will also play Bd4 (and miss the draw
until a lot later).

--
GCP
  #7   Report Post  
Old October 10th 05, 12:42 PM
Taylor Kingston
 
Posts: n/a
Default


LSD wrote:
I watch this movie tonight, sort of, while I was working on something.
I found it interesting..and was shocked to hear Kasparov say he was cheated.
Anyone got strong (or better still, informed) opinions about this.


For the most informed opinion, I would second Mr. Brown's
recommendation of "Behind Deep Blue" by Feng-Hsiung Hsu, a very good
book.
Hsu was an important member of the Deep Blue team. Kasparov was not
cheated, he was just out-worked (and that takes some doing). Hsu has
some very harsh words for Mr. Kasparov, whose allegations are
groundless. The "Game Over" filmmakers seem to have been somewhat
misleading and irresponsible in this regard.

  #8   Report Post  
Old October 10th 05, 12:48 PM
bruno
 
Posts: n/a
Default

The move wasn't even Bd4 as a lot of people think for some reason, but the
big fuzz is about the move axb5 (with Bd4 next) instead of the move Qb6
(which wins a pawn).
Still to this day none of the common engines (shredder,fritz,nimzo,junior)
play this move; they all go for Qb6.


Bateman" schreef in bericht
oups.com...
No you dont recall correctly. The move was bd4. It was not a move a
computer would have chosen for the simple reason that it allowed
Kasparov to draw with a perpetual check. Kasparov overlooked this
forced draw at the time, but Deep Blue must have been aware of it. Deep
Blue isn't programmed to make moves which Negatively affect it's
evaluation, but at the time it played bd4 DB already had itself better,
so why would it have played a move it would have seen as creating a
0.00 evaluation? When Fischer accused the Soviets of rigging the
candidates tournaments everyone said he was whining too. Why do you
refer to Kasparov as brilliant in the past tense? Is he past it now? Do
you think he'd be a sucker for your Kings Indian down the club? Dont
tell me, you beat him already, right?



  #9   Report Post  
Old October 10th 05, 01:13 PM
Bateman
 
Posts: n/a
Default

That's right, It's a long time since I played through the game. The
perpetual check wasn't outside D.B.'S search horizon either. And
anyway, all computer chess programs will always selectively search a
sequence of checks, so even if the draw was 15-20 ply deep (not outside
D.B.'s brute force) most computers will find it. That's the
significance of bd4, axb5 doesn't allow the perpetual, bd4 does.

  #10   Report Post  
Old October 10th 05, 01:20 PM
Martin Brown
 
Posts: n/a
Default

Bateman wrote:

No you dont recall correctly. The move was bd4. It was not a move a
computer would have chosen for the simple reason that it allowed
Kasparov to draw with a perpetual check. Kasparov overlooked this
forced draw at the time, but Deep Blue must have been aware of it.


Do you mean in game 2? ISTR it was Be4. But I thought it was the last
two Deep Blue moves Kf1 and Ra6 that allowed the potential for a draw.

Deep
Blue isn't programmed to make moves which Negatively affect it's
evaluation, but at the time it played bd4 DB already had itself better,
so why would it have played a move it would have seen as creating a
0.00 evaluation?


One suggestion was that the transposition table had hidden a draw by
repetition at the available depth of search.

Regards,
Martin Brown
Reply
Thread Tools
Display Modes

Posting Rules

Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are On
Pingbacks are On
Refbacks are On



All times are GMT +1. The time now is 01:42 PM.

Powered by vBulletin® Copyright ©2000 - 2019, Jelsoft Enterprises Ltd.
Copyright 2004-2019 ChessBanter.
The comments are property of their posters.
 

About Us

"It's about Chess"

 

Copyright © 2017