GM
Garry Kasparov (2847) -
Deep_Junior (C)
(2633)
[E48]
X3D "Man vs. Machine" Match
New York City,
NY/U.S.A.
(Game # 5),
Feb. 05, 2003
[A.J.G.]
This game would be a very important
encounter for Garry. His last try with
the
White pieces ... he had to go
for a full point.
(In such a short match,
with the score now tied, this IS a really
big deal here.)
{
The Nimzo-Indian ... a very popular opening ... was played here.
If the players
had wanted to, they could have repeated the same line that was
played in the
first game of the Fischer-Spassky match! (Reykjavik, ICE;
1972.) }
1.d4 Nf6; {Diagram?}
A change of pace?
After getting pasted in the first game
and giving Garry a wonderful position
in the third game, (he definitely had
winning chances in that contest); I
can
only surmise the programming
team of Deep Junior decided to pass
on any
more Slav Defenses.
[
The machine played: 1...d5;
{Diagram?}
in games one and three.
(But got really bad positions.)
]
2.c4,
{Diagram?}
Statistically, the most standard
reply for White.
(But not the only
good move he can play.)
[
2.Nf3;
or 2.g3
]
2...e6; {Diagram?}
Heralding the Nimzo-Indian Defense,
some players choose to avoid this
line entirely, and play 3.Nf3 here.
(Of course Black does not have to
play the Nimzo, a transposition back
into the main lines of the Q.G.D. is
still possible.)
[
The move 2...g6!?;
{Diagram?} is more modern, and could lead
to a wide number of different
opening systems.
(The Benoni,
The Grunfeld, The King's Indian
Defense, ... or just
about any
line that features a fianchetto
of Black's King-Bishop.) ]
3.Nc3, {Diagram?}
Garry is not afraid of ...Bb4.
(He has even used it himself
as Black; i.e., see his game with
GM Jan Timman, from the
Euro Tel Cup Tourney, 1998.)
While Garry does not fear the Nimzo,
during the 1980's I was a spectator
at several U.S. Championships.
During this time, it seemed most
masters
(as White) were almost
universally reluctant to play 3.Nc3.
This meant that
3.Nf3 was almost
always played here, and that the
opening that saw more
action than
any other was, ... "The Queen's Indian Defense."
[
If White plays Nf3, we could wind
up in the Queen's Indian Defense.
E.g, 3.Nf3 b6;
4.g3 Ba6; 5.b3 Bb4+; 6.Bd2,
"+/=" {Diag?}
etc.
White maintains a small pull.
]
3...Bb4; {Diagram?}
This is ...
"The Nimzo-Indian
Defense." It was the brain-child
of Aaron
Nimzowitsch. (His
name is commonly spelled
'Nimzovich' in the English
language.)
Nimzovich was a very strong player,
easily at one time in the world's top
three or four players.
(He was the
leading contender for the position
of challenger. But because
Bogo
had better financial backing, he
played two matches for the {chess}
World's Championship.
And poor Nimzovich was denied any
opportunity
when he was at the
peak of his playing strength.)
He
(Nimzovich) was also one of the 'founding fathers'
of the whole
hyper-modern
movement.
(These players discovered that the
center could be controlled by pieces,
as well as by pawns. The hyper-modern players advocated
controlling the
middle of the board
primarily with their pieces, and
retaining a more flexible
pawn
structure. This is turn led to much
more fluid and dynamic systems
of opening play, especially for the
player who had to play the Black
pieces!)
***********
[
By playing the moves: 3...d5;
4.Bg5 Nbd7; 5.Nf3 c6; 6.e3,
{Diag?}
we transpose back into the
main lines of the classical
systems of ...
"The Queen's
Gambit Declined." {Q.G.D.} ]
4.e3, {Diagram?}
The Rubinstein System
... ... ...
still a viable system and much in
use, even today.
(Most books
consider this to be the main line
here.)
This line was worked out by the great
Akiba Rubinstein, who of the strongest
and most original players who ever
lived. (And had life been fairer he
would
have played for the World's
Championship - and probably won.)
Rubinstein worked out the best lines
of play in many different opening
systems.
(Most of these lines are
still considered viable and good,
even today.)
*****
[ Botvinnik - at least at one time
- strongly advocated the
'Samisch System'
by playing: 4.a3!? Bxc3+;
5.bxc3 c5; "~"
{Diag?}
but this line is not currently in
favor today. (out of fashion)
(I wish to point out that this
line is still perfectly playable.)
***
White can also play: 4.Qc2,
"+/=" {Diagram?}
which is the "Classical System."
(Invented by Rubinstein and
first championed by Capa.) ]
******
4...0-0;
{Diagram?}
Really this move and ...c5 are
nearly interchangeable at this
point in this line.
(In my opinion, 0-0 is a little more
flexible, and probably the main
line here
according to ECO & MCO.)
5.Bd3, {Diagram?}
The main line and probably
the most natural move here.
(The e4-square is influenced, a
piece has been developed, and
now White is one
move closer to
getting castled. Meanwhile, the
pawn at c4 is still guarded.)
[ Rubinstein also liked:
5.Nge2!?,
"~" {Diagram?}
as the more aggressive try.
(Basically an attempt to
refute Black's whole line of play.)
But this line is slower and not
guaranteed to give White an
advantage,
and it also blocks in
White's King-Bishop.
(I have at least 3 different books
on this one line.) ]
Black now naturally hits on the
center, which is probably the
most important
thing you can do
in the opening.
5...d5; ('!') {Diagram?}
This is probably the best move
here, and several books consider
it the main line.
[
5...c5!?
]
6.cxd5!?, {Diagram?}
I am not at all sure about this
move, one old book condemns
it as prematurely
freeing Black's
game.
My first instinct is that this is a
natural move designed to resolve
the central
conflict and avoid the
loss of tempo associated with the
main line here.
(Bd3, dxc4; etc.)
The drawback is that Black has
much greater freedom of play and
fewer problems
activating his QB
than he normally does.
***
[
MCO
gives the following line
in this variation:
6.Nf3 c5; 7.0-0 Nc6; 8.a3 Bxc3!?;
{Diagram?}
The move that is most often
played by Black in this position.
(8...dxc4!?; or 8...Ba5!?)
9.bxc3 dxc4; 10.Bxc4 Qc7;
11.Ba2!? e5!; 12.h3 e4; 13.Nh2!? Na5;
14.Bb2 b6;
5.c4 cxd4; 16.Bxd4 Qc6;
"=" {Diagram?}
This position is more or less
even. Since MCO gives no
game references
here, I must
assume that this is an analysis
line. (by GM Nick de Firmian.)
[ See MCO-14; page # 539,
column #1, and notes # (a)
through # (c). ] ]
6...exd5; 7.Nge2, {Diagram?}
I know f3 looks like the best square,
and this is normally true. But in this
line White avoids Nf3, as Black gains
play with the pin, ...Bg4.
White can now also play for a later
e3-e4, and try to dominate all the
key
central squares.
[
7.Nf3!? Bg4; "<=>" {Diagram?} ]
Now both sides continue to
develop.
7...Re8; 8.0-0 Bd6;
{See the diagram just below.} Both
sides seem to be doing OK here.
9.a3!?, {Diagram?}
This is a necessary precaution.
In some lines, White wishes to play
Qc2 but
does not want to have to
worry about a later ...N/c6-b4.
White can also
{possibly} consider
a later b2-b4, gaining valuable
space on the Queen-side.
[ White could also just play:
9.f3!?,
"~" {Diagram?}
in this position. ] 9...c6; 10.Qc2, {Diagram?}
To me, this looks like a completely
normal developing move, although
some on-line commentators roundly
condemned this here.
[ 10.b4!?
Ng4!?; "--->"
or 10.f3!? Qc7;
"<=>" ] 10...Bxh2+!?;
(Maybe - '?!')
{See
the diagram just below.} An incredible sacrifice, one
most computers would NOT make.
{Kasparov was obviously shocked by this move.}
In the final analysis, this move
could be unsound. But it does
lead
to some very interesting play.
I find it refreshing that a computer
could make a sacrifice of this type,
and basically play a move that is
based more on general considerations,
than on exact calculation.
The
position after Black sacrifices his Bishop.
[
One on-line commentator
highly recommended that
Black play:
10...b6!?;
"=" {Diagram?}
in this position. ***
Playable was: 10...Nbd7;
"~" {Diagram?}
with a fair game for Black.
] 11.Kxh2 Ng4+; 12.Kg3,
{See the diagram just
below.} This is forced.
The
position after White plays his King to g3.
[
Much worse would be: 12.Kg1? Qh4;
13.Rd1 Qxf2+; 14.Kh1 Rxe3;
15.Bxh7+ Kh8;
16.Bxe3, {Diagram?}
This is probably forced.
( 16.Qd3 Rxd3; "-/+" {Diagram?}
The main threat is: 16.b4? Rh3+!; 17.gxh3 Qh2#.
(This line also works against Bd3
or Bf5 by White on move 16.) )
16...Nxe3;
"-/+" {Diagram?}
and Black has a won game.
(Black threatens White's Queen
and a mate on g2 at the same
time here, there is no defense.)
] 12...Qg5!?; {Diagram?} Apparently the program felt this
was the best way to continue.
[
Probably not as accurate was: 12...Qd6+;
13.f4, "±" {Diag?}
and although White is very
clearly better, ("+/") the first player
has to play perfectly in order
not to get into trouble. ] ********** The next few moves look to be
fairly best, and ... forced. 13.f4, {Diagram?} This looks to be the most forcing,
maybe Garry thought this was safer
than Bxh7+, feeling this piece
could be "hung out to dry" in some
variations.
Several of my students saved the
transcripts of the on-line commentary,
as my computer was down.
Apparently several (on-line)
players and commentators gave the
opinion
that this move was incorrect.
Their 'snap judgment' was that
Bxh7+ was
superior, but this may
not be the case.
[
It appears to be a subtle offer of
peace negotiations after the
continuation of:
13.Bxh7+!? Kh8;
14.f4, {Diagram?}
White must take action, Black
was threatening a winning
discovered check.
( 14.Bd3?? Nxe3+; ("-/+") and mate follows next. (g2)
)
14...Qh5; 15.Bd3 Qh2+;
16.Kf3 Qh4!; 17.g3!? Qh5;
18.Kg2 Qh2+;
19.Kf3 Qh5;
"=" {Diagram?}
and the game is now drawn,
neither side can deviate from
this sequence of moves. ] 13...Qh5; 14.Bd2 Qh2+;
15.Kf3 Qh4; {Diagram?} The only way to continue.
[
15...Qh5?!; (Maybe - '?')
{Diagram?}
This is a mistake.
16.Rh1 Nxe3+; 17.Rxh5 Bg4+;
18.Kf2 Nxc2; 19.Rah1 Bxh5;
20.Rxh5 Na1;
21.Bxh7+ Kf8; 22.Nxd5!,
"±" (Maybe "+/-")
{D?}
and White should probably win. ] 16.Bxh7+!?, {Diagram?}
After this rather tame (lame?) move,
the game quickly resolves itself
into
a draw.
Garry said in the post-game interview
that he was in a state of shock
after
...Bxh2+, and could not think clearly.
[
Why didn't Garry try to win here? The best line - no doubt! - is the
continuation:
>/=
16.g3! Qh5; 17.Rh1 Nxe3+;
18.Rxh5 Bg4+; 19.Kf2 Nxc2;
20.Rah1 Bxh5;
21.Rxh5 Na1; 22.Bxh7+ Kf8; 23.Bd3,
"±" {Diag?}
and White has a very significant
edge from this position. ("+/")
{Two Bishops for a Rook.}
(Several programs evaluate this
line as basically a win ("+/-") for White.) ***
One student said Rh1 won on
White's sixteenth move, but I
think he
overlooked something! I.e.,
16.Rh1??
Qf2# ] 16...Kh8!;
{See the diagram just
below.} The machine finds the best
method here.
The
actual game position after DJ plays 16...Kh8.
(The game now quickly finishes up as a draw.)
[
A clearly inferior line would have
been: 16...Qxh7; 17.Qxh7+ Kxh7;
18.Rh1+ Nh6;
19.Rh4, "+/=" (Maybe - "±")
{Diagram?}
when White is
clearly better here. ("+/=" ... maybe
"+/")
(I am quite
sure Garry would have been
very happy if the box had
played
this line!) ***
Black could also try: 16...Kf8!?;
(Maybe - '?!/?') {Diagram?}
but White may gain the upper
hand with g3. ("+/") ]
Now close to 15 different analysis
engines verify the game is dead
drawn,
neither side can avoid the
repetition of the position.
17.Ng3 Nh2+; 18.Kf2 Ng4+;
19.Kf3 Nh2+; ("=") {Diagram?}
The game was agreed or declared
drawn at this point.
An interesting encounter that Garry
probably should have won.
Copyright (c) A.J. Goldsby I. Copyright (c) A.J.G;
2003.
(Code initially) Generated
with ChessBase
8.0
½ - ½
|