IBM
Skip to main content
 
Search IBM Research
     Home  |  Products & services  |  Support & downloads  |  My account
 Select a country
 IBM Research Home
Deep Blue
Overview/Home
The Match
The Players
 ·Gary Kasparov
 ·Deep Blue
 ·The Deep Blue Team
 ·The Comparison
The Technology
The Community

Related Links
 Press room
 Chess conference
 Site guide
 Search Research
 Feedback
 
 


Deep Blue game 6: May 11 @ 3:00PM EDT | 19:00PM GMT        kasparov 2.5 deep blue 3.5
How Kasparov works
  

Kasparov vs. Karpov, Moscow, 1985
It is the 24th and final game of the World Championship match.Kasparov is ahead 4-3 and needs only a draw against his nemesis Anatoly Karpov to win his first world title. Karpov launches an aggressive attack against Kasparov’s king. Kasparov sits back, ponders the situation, then executes a brilliant pawn sacrifice (link to N.2) to halt Karpov’s advance. The attack fails, Kasparov wins the game and ends Karpov’s 10-year stranglehold on the World Championship.

As it turned out, Kasparov made the perfect move to defense Karpov’s attack. But how did he arrive at the decision to execute the pawn sacrifice? Why did he choose this move over others that may have seemed more advantageous? The answer lies within the way Kasparov uses his mind, his experience, and his psychological make-up to choose the ideal move nearly every time.

When Kasparov is presented with a chess position, a serious of biological processes is set in motion to help him determine the next move. The brain’s neurons, receptors and neurotransmitters work in concert to quickly formulate a picture of the best possible actions. Kasparov examines and evaluates the current position, compares it to the “database” of similar positions stored in his memory, then uses a combination of creativity and intuition to make what he feels is the correct move.

Most human chess players can search “maybe two positions per second,” says Deep Blue developer Murray Campbell, “but they have this talent, this intuition, [although] they aren’t very good at describing how it works.”

It is precisely this intuition that gives human players an added advantage over computer opponents. Kasparov must sometimes “guess” as to the best possible move -- albeit his guess is a highly educated one. He is able to call upon his years of experience at the chessboard and his own chess knowledge to determine the optimum move. Deep Blue, on the other hand, must simply draw upon vast resources of stored information make its moves.

Last year, Kasparov switched strategies mid-game to defeat Deep Blue. In essence, he used his ability to adapt to a given situation to provide himself an advantage that his computer opponent could not overcome. Adaptability and creativity have traditionally given human players advantages that chess machines cannot replicate.

Although Kasparov and Deep Blue play chess at a similar level, they employ different methods to arrive at similar conclusions. The human mind differs greatly from the processors, software and circuits that comprise a computer’s “brain”. And although both players utilize different “thought” processes, they still play a mean game of chess.

Said Garry Kasparov during last year’s match, “Here we are to discuss the ultimate question of whether one day the computer will beat the human world champion. Maybe it will happen in a week’s time. But I remain a cautious optimist in the progress of human brains as well. I still believe that there are some horizons it will be very difficult for computers to cross.”



  
Related Information

      Interview:Garry Kasparov's thoughts on the match, on the future of chess-playing computers and the psychology behind the game.

 
      Classic matches:The stories behind some of Kasparov's most engaging matches

 
      Kasparov FAQ:What you want to know about the greatest player in history

 
      1997 FIDE Rating List:How Kasparov ranks against the rest of the chess-playing world

 
      The Kasparov Team:The advisors in Kasparov's corner.

 
      Compare the Opponents:Kasparov can examine and evaluate three chess positions per second. Deep Blue can evaluate about 200 million per second.

 
      Chess Pieces
no. 9

The first chess game played by telephone was played by two gentlemen in Derbyshire, England, in 1878.

 
  About IBM  |  Privacy  |  Legal  |  Contact