Flawed Strategy vs. Fierce Resistance banner WaveTop WWWDirectory
toolbar
May 13, 1997

Flawed Strategy vs. Fierce Resistance

By ROBERT BYRNE

Looking back over the match between Garry Kasparov and Deep Blue, it seems clear that the outcome was determined by a combination of two factors: a slightly flawed strategy by the world champion and tremendous, bone-wearying resistance by the marvelous machine.



Today in CyberTimes

Articles and Columns

Activism: Nonprofit Has Inner City 'Plugged In'
By Matt Richtel

Kasparov, Unbowed, Looks to Third Bout With Deep Blue
By Bruce Weber

Flawed Strategy vs. Fierce Resistance
By Robert Byrne

Apple Sets Out to Sell Developers on New Operating System
By Lawrence Fisher

Searching the Net, Continental Style
Eurobytes /
By Bruno Giussani

Brilliant Scanning, Dumb Design
Personal Computing / By Stephen Manes


Today's Section Front

Seven-Day Index

CyberTimes Forums

CyberTimes Navigator



Overalll, Kasparov's policy of slowmaneuvering, which helped him win his first match with Deep Blue last year, is exactly what most grandmasters would have tried.

But he carried it to extremes. And he also underestimated Deep Blue's ability to fathom what he had hoped to keep concealed.

These defects showed themselves in the first game. Kasparov did indeed obtain a small positional advantage, but to put down Deep Blue's ingenious counterattacks, he had to engage in the hand-to-hand fighting he wanted to avoid.

He was playing into the computer's strength, yet he overcame it for his only victory of the match.

Still, Kasparov seemed to lack confidence in his tactical excellence. In the second game, he played a purely passive opening, as though he expected Deep Blue to roll over and expire in confusion. What happened was that I.B.M.'s master plotter played the strongest purely positional game ever produced by computer. Incredible. It was Kasparov's turn to be beaten by his own weapons. Worse yet, he resigned in a position where he had overlooked a way to force a draw. It was little consolation that Deep Blue had overlooked the same thing.

Indeed, the machine overlooked several earlier moves that would have won.

One is tempted to say, "That's chess." It is not yet played perfectly.

The third game saw Kasparov again play a game of slow, meandering maneuvers. But when that failed to elicit awkward unpositional responses, he jumped into action and sacrificed a pawn for positional advantage. Deep Blue grasped what his opponent was up to and simply would not let him achieve anything tangible. The half-way point of the series was reached with a 1 1/2-1 1/2 tally.

The fourth game looked as though it would go Kasparov's way. He played a quiet positional defense but created an unclear situation where great deeds could be planned. And he soon gave Deep Blue a pawn for an impressive positional counterattack. But Deep Blue's defensive power was once more extraordinary: with great virtuosity, it fought through a grim rook-and-pawn endgame to a draw. Score another disappointment for Kasparov.

The fifth game followed a somewhat similar course. Deep Blue, with original play in the opening, had no trouble getting an equal game as Black. Kasparov threw all his concentration and energy into forging a victory here, but Deep Blue played as though virtuosity in difficult endgames was second nature. No one had foreseen its scintillating method of certifying the draw. It was another frustration for the champion.

The final game was a shocking disappointment, but on the part of Kasparov, not the computer.

Worn down by knocking his head against impenetrable defenses and seeing his hopes repeatedly dashed, he lost his concentration and made a fatal error as early as the seventh move. The chess literature has warned against that particular error since 1987, and everyone knows how to avoid it, but on Sunday Kasparov appeared too exhausted to think. When Deep Blue started to cut through to the black king, he threw in the towel.

If there is a rematch, Kasparov might do well to switch over to Bobby Fischer's uncompromising, all-purpose strategy: no matter who your opponent is, just play the best moves on the board.

Instead, Kasparov accommodated Deep Blue to the point where he muzzled his own brilliance.


Game 1

DEEP BLUE/BLACK

KASPAROV/WHITE
Position after 29 . . . e4

Facing the fatal 30 . . . ef, Kasparov counterattacks with 30 f4! Be2 (30 . . . Bf4? 31 gf Be2 32 Qg1!) 31 fg Ne5 (on 31 . . . hg?, 32 Nc4 either wins the queen or mates) 32 g6!, gaining irresistible pawns.



Game 2

KASPAROV/BLACK

DEEP BLUE/WHITE
Position after 45 Ra6

After resigning here, Kasparov learns 45 . . . Qe3! yields a draw: 46 Qd6 Re8! 47 h4! Qe4! 48 Ra7 Kg8 49 Qd7 Qf4 50 Kg1 Qe3 51 Kh2 Qf4 52 Kh3 Re7! 53 Qc8 Kh7 54 Re7 h5! 55 Rg7 Kg7 56 Qd7 Kh6.



Game 3

DEEP BLUE/BLACK

KASPAROV/WHITE
Position after 40 Kg4

Though a pawn ahead, Deep Blue is unpleasantly tied up and returns the material with the prudent 40 . . . Bc7 1 Nc7 Rc7 42 Ra5. After 42 . . . Rd8, it puts the d3 pawn under heavy pressure and gains a draw.



Game 4

KASPAROV/BLACK

DEEP BLUE/WHITE
Position after 36 . . . Ne6

With 37 . . . Rf2 or 37 . . . Re4 looming, Deep Blue dodges into an endgame: 37 Nd4! Nd4 38 ed Rd4 39 Rg1. And after 39 . . . Rc4 40 Rg6 Rc2 41 Rg7 Kb6 42 Rb3 Kc5 43 Ra7, it struggles to a draw.



Game 5

DEEP BLUE/BLACK

KASPAROV/WHITE
Position after 48 g6

As the g pawn races to become a queen, Deep Blue discovers 48 . . . Kb5 49 g7 Kb4! If 50 g8/Q, then 50 . . . Rd1 51 Kc2 Rd2 is perpetual check. If 50 Rb3, then 50 . . . Kb3 threatens mate. The result: a draw.



Game 6

KASPAROV/BLACK

DEEP BLUE/WHITE
Position after 19 c4

Kasparov has nowhere to hide. If . . . bc, then 20 Qc4 Kb7 falls into 21 Qa6 mate. If 19 . . . Nb4, then 20 Qf5 bc 21 Ne5 Bb5 22 Ng6. So he decides not to put himself through any further torture and resigns.



Home | Sections | Contents | Search | Forums | Help

Copyright 1997 The New York Times Company



WaveTop WWWDirectory