Dec 28, 1995	Romania and Chess (Adrian German, Ariel Mazzarelli)
May  7, 1997	Kasparov vs Deep Blue (Ariel Mazzarelli, Eric Wang, et al)

Subject: England '96 preview: the Romanian team. 
From: "Adrian German" 
Date: Dec 28, 1995

It's been often said that a European Championship lacks some of the glamour of
a World Cup Tournament, since the South American teams are not coming. England
1996 is certainly no exception, but there is an amendment: from the eastern 
part of the continent Croatia, Bulgaria, and Romania will be carrying to the 
games a very distinct spirit, and style of play. 


It is impossible to make justice to the history of the Romanian soccer in such
a short space and knowing that, we will simply not deal with it here. In doing
so we are leaving out a large and significant part of the athletic phenomenon,
together with its social connotations. But we have no choice, and we will be
well justified, and our time well spent, if we concentrate on the present. 

The 16 teams for England '96 will start the competition on the 8th of June 1996
with a clear purpose: they are playing for the European crown. There is clearly
no other good reason for taking part to a final tournament if you are not going
there to actually win it. As such the Romanian team will fly to England in 1996
with a visible goal in mind: to win the continental title. 

1. Preliminary Round

Romania qualified atop Group 1, ahead of France, Slovakia, Poland, Israel, and
Azerbaijan. Romania - Azerbaijan (3-0) was the first game in Bucharest since 
the lost shoot-out in San Francisco (Petrescu and Belodedici). The game saw 
a beautiful goal from Romanian sweeper Miodrag Belodedici, and the Romanian 
team never left the Group's first place after that game. 

It is the great merit of the Romanian coach, Anghel Iordanescu, to have read
attentively all the fine print of these preliminaries. His team, that had given
its share in two of the most spectacular games in the USA '94, against Colombia
and Argentina, did not however put on an impressive show during the group's 
matches. Nonetheless, Romania drew only three games away (Israel 1-1, France 
and Poland, both 0-0) lost a game at home (which sealed the qualification for 
the team of France) and quietly won the rest of the games.

2. Anghel Iordanescu, and the Romanian style of play

Anghel Iordanescu was a gifted player. He started on the left wing, and 
finished his career (in Sevilla, May 7th 1986) as a playmaker, in the middle.
His coaching career (started around 1984, when he came back from Greece) saw
him at Steaua (1984-1990), in Cyprus (1990-1992, Anorthosis I think), at Steaua
again and then at the national team. 

For the reader who has seen the game against Sweden and has been left hanging 
waiting for a second round of display Hagi-Dumitrescu, Iordanescu requires a 
word of explanation. The pace of the game against Sweden was a deliberate, 
although unfortunate move. This important aspect helps us make the distinction
between the past and the present. 

Against Argentina and (especially) Colombia, Iordanescu played like the chess 
grandmaster - World Champion during 1894-1921 - Emanuel Lasker sometimes used
to: by constantly introducing unnecessary complications into the game. The 
purpose being to generate significant overhead in an opponent who's trying 
to make sense of your game, the strategy sometimes pays off. But combine this 
with the lessons that an attentive observer could have learned from the second 
Bearzot (1982) and you get all the ingredients of a rather strenuous game -- 
for player and spectator alike.

What Iordanescu knew, but had magnanimously been overlooked by others at 
large (teams and coaches alike), was that given the chance during the game 
he could fall back on one trump: this was the gift of the swift and classy 
touch out of which the spectacular Rose Bowl fireworks displays of Raducioiu, 
Hagi, and Dumitrescu have been created.  

For Iordanescu and his team this strategy was a way of concealing a certain
lack of confidence. It is the conviction of this writer that the preliminaries
for England have bridged the transition from a skilled, but often unfocused,
and shy Romanian team, to a team that can motivate itself to the utmost. And 
since these lines are written when we're still in 1995, we will leave it to 
the seasons between now and then to prove us right or wrong. 

[Wrong, unfortunately: France 1, Romania 0; Bulgaria 1, Romania 0; 
Spain 2, Romania 1]

From: (Ariel Mazzarelli)
Subject: Re: England '96 preview: the Romanian team. 
Date: Dec 29, 1995

You are confused. Lasker was better than Romania, he was Italia.

Romania was Steinitz (sorry Wilhelm). Look at the following game, specially
after move 17 (to be viewed with CBdemo). Annotations by Reinfeld (RR by me).

[Event "?"]
[Site "Hastings (09)"]
[Date "1895.??.??"]
[Round "?"]
[White "Lasker,Emanuel"]
[Black "Steinitz,Wilhelm"]
[Result "1-0"]
[ECO "C72"]
[Annotator "Reinfeld"]
[PlyCount "80"]

{RR This is perhaps the only game of this great tournament with which Lasker
was happy. Both players were in poor health, more so for Steinitz.}

1. e4 e5 2.Nf3 Nc6 3. Bb5 a6 4. Ba4 d6 5. 0-0 Nge7 

{One of the many defences which Steinitz invented, but which have not found favour with other masters. 
See Euwe-Keres, Hague 1948 however.} (5... Bd7 6. c3 g6) (5... f5) 

6. c3 Bd7 7. d4 Ng6 

(7... g6 {yields an easier game}) 

8. Re1 Be7 9. Nbd2 0-0 10. Nf1 

{This maneuver was invented by Steinitz himself.} 

10... Qe8 

{The critics are unanimous in applying to this move such epithets as 'bizarre', 
'typically Steinitzian', 'peculiar', and the like. In reality the text begins 
a profound manoeuvre. Black wishes to force his opponent to play Pd5. Once the 
centre is stabilised by this advance, Black can undertake a promising attack 
on the Qside. Unfortunately Black's Ng6 is not very well placed for the 
execution of this plan. But as we shall see, it is not on account of his 
'bizarre' moves that Steinitz loses this game.  @aAc6d4@cCd4} 

11. Bc2 Kh8 

{RR Weird stuff} 

12. Ng3 Bg4 {@cCd4} 13. d5 

{Black has finally forced the advance of the QP and should now continue Nd8 
(as he did, more successfully, in the 8th game of his second match against 
Lasker, Moscow 1896).} 

13... Nb8 (13... Nd8) 14. h3 Bc8 15. Nf5 Bd8 

{RR All we need is Ng6-e7-g8!} (15... Bxf5 $4 16. exf5 Nh4 (16... Nf4 17. Bxf4 
exf4 18. f6 gxf6 19. Qd3 {@cCh7}) (16... e4 17. fxg6 exf3 18. Qxf3 {@aAf3e4@cCe7} 
18... fxg6 19. Qe2 Rf7 20. Bg5 $18) 17. f6 Nxf3+ (17... Bxf6 18. Nxh4 Bxh4 
19. Qh5 {@cCh4h7}) 18. Qxf3 Bxf6 19. Qf5 $18 {@aCg7g6@cCh7}) 

16. g4 Ne7 17. Ng3 Ng8 

{The Deutsche Schachzeitung inclined to the opinion that Steinitz allowed his
sense of humour to get the better of him in bringing about the present 
position. This imputation of humourous intent is, however, merely a sign of 
the annotator's laziness. It is clear that after Pg6, etc., Black will
have a promising game.  @aAg7g6Ad8f6Af6g7Ag8e7Af7f5} 

18. Kg2 Nd7 

{Again Steinitz changes his plan. RR Reinfeld...} 

19. Be3 Nb6 

{The N is, of course, quite ineffective on this square, but it is to be brought 
to e7. @aAb6c8Ac8e7} 

20. b3 Bd7 21. c4 

{Now White has the initiative on both wings} 

21... Nc8 22. Qd2 Nce7 23. c5 

{RR  Black is getting stifled. It is hard to suggest a move that does nothing,
let alone one that does something.} 

23... g6 

{RR  This move of course does open up the K to attack, but Steinitz is sneaky. 
It would seem that he is hanging a pawn, and a pretty important one at that, 
but Lasker sees the real weakness.} 

24. Qc3 $1 

(24. cxd6 cxd6 25. Qb4 {@cCb7d6} 25... Bc7 26. Qxb7 Ba5 27. b4 Rb8 28. Qxa6 
Bxb4 29. Rec1 Bb5 30. Qa7 Ra8 {and Black has at least a draw by harassing 
White's Q}) 

24... f5 $2 

{Overlooking the threatened sacrifice.} (24... f6 {@aAe8f7Af7g7}) 

25. Nxe5 dxe5 26. Qxe5+ Nf6 27. Bd4 $1 

(27. g5 $2 Nexd5 $1 {@cCe5} 28. Qxe8 Nxe3+ 29. Rxe3 Nxe8) 

27... fxg4 

(27... Kg8 28. g5 Nexd5 29. exd5 Qxe5 30. Rxe5 $18) 

28. hxg4 Bxg4 

{RR  This gets rid of some pawn breaks, but it hangs the piece back with

29. Qg5 {@cCf6g4} 29... Qd7 30. Bxf6+ Kg8 31. Bd1 Bh3+ 32. Kg1 Nxd5 $2 

(32... Rxf6 33. Qxf6 Nxd5 {gives fighting chances}) 

33. Bxd8 Nf4 34. Bf6 Qd2 35. Re2 $1 

{The simplest way of putting an end to Black's demonstration.}

35... Nxe2+ 36. Bxe2 Qd7 37. Rd1 Qf7 38. Bc4 Be6 

{RR  This bishop on e6 does an obvious thing (break the pin) and a less 
obvious thing (hold f5). That is one too many.} 

39. e5 Bxc4 40. Nf5 

{@aAf5h6Af5e7@cCg8} 1-0

From: (Adrian German)
Subject: Re: England '96 preview: the Romanian team.
Date: Dec 30, 1995

Fine, let's assume it's as you said. First some background.

  "Lasker took 3rd place behind Pillsbury and Tchigorin at Hastings 1895. He 
  almost died before this event and was still recovering from typhoid fever."
  (from courtesy Palle Mathiasen)

Now the facts: 

   [...] "During the war, Lasker had invested his life savings in German war 
   bonds and lost it all. He tried to breed pigeons, but all the pigeons he 
   bought were male."

Such a great player with so little (should I say) common sense. I bet Steinitz 
wouldn't have made this mistake. 

   "After the war he won a strong tournament in Berlin in 1918. [...]"

Yet he recovered gracefully. Amazing. You were right. Lasker _was_ Italia. 
Capabalanca was Argentina, and Romania... Romania was Alekhine.

From: (Ariel Mazzarelli)
Subject: Re: England '96 preview: the Romanian team.
Date: Dec 30, 1995

Should we split the group with rss.chess? Anyway,

>  "Lasker took 3rd place behind Pillsbury and Tchigorin at Hastings 1895.

Before that tournament he gave some lectures in London that became a book,
Common Sense in Chess. It is now a $5 Dover, and it is really cool. He
analyzes other games against Steinitz as examples of general principles--and
his analysis is often an addition to what you find in, say, ECO.

> [Steinitz wouldn't buy all male pigeons or lose his savings]

It is hard to say, Steinitz never had a whole lot of savings to lose. Of
course it is very unlucky, money-wise, to be hit with a World War 1.0.

> Romania was Alekhine. 

You think Romania's tactics are that good? Could be. Romania has not won a big
cup yet, though.

You're right about Argentina, I'd add that Argentina '86 was like Tal '60.
Brasil is clearly Botvinnik, Germany is probably Alekhine, Uruguay is Fischer,
England is Staunton, and nobody touches Karpov or Kasparov.

I know that this is a ridiculous thread, but it has a certain logic.

From: (Ariel Mazzarelli)
Subject: Kasparov is no Diego
Date: 7 May 1997 02:35:02 -0700

First, from Reuter
Tuesday May 6 11:35 PM EDT 

Kasparov in Awe of Chess-Playing Computer

NEW YORK (Reuter) - World chess champion Garry Kasparov has called 
supercomputer Deep Blue an "alien opponent" but Tuesday he said it was 
playing like a god.

The best player in the history of the ancient game has suffered the 
double embarrassment of needlessly resigning to the IBM system on Sunday 
and then being held to a draw in Tuesday's third game of their six-game 
re-match despite the advantage of the white pieces. 

"The scientists are saying that Deep Blue is only calculating, but it has 
showed signs of intelligence," said Kasparov, who had no advance 
information on his opponent and has labeled it alien. 

The $1.1 million match is tied at 1 1/2 points each and Kasparov will 
have to play with the black pieces in two out of the three remaining 
games. One point is awarded for a win and a 1/2 point for a draw. Playing 
with white has the advantage of the first move, much like holding serve 
in tennis. 

The revelation that the Russian gave up on Sunday in what was in fact a 
drawn position, dominated and overshadowed the third game of the contest, 
a closely fought draw out of an English Opening that ended with Deep 
Blue's programmers accepting Kasparov's draw offer after almost 4 1/2 
hours at the board. 

"It reminds me of the famous goal that Maradona scored against England in 
86. He said it was 'the hand of God'," stated Kasparov, referring to a 
goal one of the world's greatest soccer players, Diego Maradona of 
Argentina, scored in the 1986 World Cup in Mexico. 

Maradona illegally used his hand to punch the ball into the net but the 
infraction was not spotted by the referee and the goal counted. 

"Suddenly you know it played like a god for one particular moment (in the 
second game)," an animated Kasparov told hundreds of spectators after 
Tuesday's game. It was unclear whether he was accusing the IBM team of 
cheating or just awed by the computer's performance. 

The Russian's remarkable and rare oversight was a reminder to human 
players that their emotions can be a handicap in clashes with machines. 

Several leading grandmasters admitted that they too had initially missed 
the continuation that would have saved the day for Kasparov. It was left 
to an untitled player in an Internet chat room to claim the credit for 
unlocking the problem. 

Deep Blue's programmers also said it had not calculated the moves 
correctly during the game, something Kasparov described as "very human 
from my point of view." 

Grandmasters intensely debated how the strongest player in chess history 
overlooked a sequence of moves that would have forced a draw Sunday and 
maintained his lead in the match. 

Instead, the 34-year-old Russian resigned the position after Deep Blue's 
45th move and the match was tied at one win each. Kasparov defeated the 
machine Saturday in the first game. 

Chess experts, almost without exception described Sunday's game as the 
best performance ever by a computer, likening it to the style of top 
human players. 

Kasparov was forced by Deep Blue to defend with his black pieces for 
almost four hours and looked tired and demoralized. 

"The computer has an advantage, it does not have this body of emotions. 
We humans get depressed," grandmaster Yasser Seirawan of the United 
States said. "The computer doesn't get depressed." 

Subsequent analysis showed that Kasparov could have played a series of 
moves to force what is known in chess as "perpetual check" -- one player 
repeatedly attacking his opponent's king, ensuring none of his other 
pieces can make further moves and thus a draw is the only outcome. 

The analysis began within hours of the game ending in "chat rooms" run by 
the Internet Chess Club. Surprisingly, it was an untitled player who 
first suggested the drawing sequence, according to Internet Chess Club 
director Gregory Belmont. 

Now that I've taken a look at it, the draw is pretty straightforward
once you decide to look for it. Here are the moves:

[Event "fics rated standard game"]
[Site "fics, Oklahoma City, OK USA"]
[Date "1997.05.04"]
[Time "13:45:35"]
[Round "-"]
[White "DeepBlue"]
[Black "GMKasparov"]
[WhiteElo "0"]
[BlackElo "2796"]
[TimeControl "7200+0"]
[Mode "ICS"]
[Result "1-0"]

1. e4 e5 2. Nf3 Nc6 3. Bb5 a6 4. Ba4 Nf6 5. O-O Be7 6. Re1 b5 7. Bb3 
d6 8. c3 O-O 9. h3 h6 10. d4 Re8 11. Nbd2 Bf8 12. Nf1 Bd7 13. Ng3 Na5 
14. Bc2 c5 15. b3 Nc6 16. d5 Ne7 17. Be3 Ng6 18. Qd2 Nh7 19. a4 Nh4 20. 
Nxh4 Qxh4 21. Qe2 Qd8 22. b4 Qc7 23. Rec1 c4 24. Ra3 Rec8 25. Rca1 Qd8 26. 
f4 Nf6 27. fxe5 dxe5 28. Qf1 Ne8 29. Qf2 Nd6 30. Bb6 Qe8 31. R3a2 Be7 
32. Bc5 Bf8 33. Nf5 Bxf5 34. exf5 f6 35. Bxd6 Bxd6 36. axb5 axb5 37. Be4 
Rxa2 38. Qxa2 Qd7 39. Qa7 Rc7 40. Qb6 Rb7 41. Ra8+ Kf7 42. Qa6 Qc7 43. Qc6 
Qb6+ 44. Kf1 Rb8 45. Ra6 
{Black resigned} 1-0

But the FOOL had a draw. Here is what happens if the computer plays like
an idiot:

 45.  Ra6     (3:32)     Qe3     (0:00)  
 46.  Qxd6    (0:00)     Qxe4    (0:00)  
 47.  Qe6+    (0:00)     Kf8     (0:00)  
 48.  Ra7     (0:00)     Qf4+    (0:00)  
 49.  Ke2     (0:00)     Qe4+    (0:00)  
 50.  Kd2     (0:00)     Qd3+    (0:00)  
 51.  Kc1     (0:00)     Qe3+    (0:00)  
 52.  Kb2     (0:00)     Qxa7    (0:00)  

and Garry wins (this would not happen). Here is a more reasonable and 
representative approach:

 45.  Ra6     (3:32)     Qe3     (0:00)  
 46.  Qxd6    (0:00)     Qxe4    (0:00)  
 47.  Qxb8    (0:00)     Qd3+    (0:00)  
 48.  Kg1     (0:00)     Qe3+    (0:00)  
 49.  Kh1     (0:00)     Qc1+    (0:00)  
 50.  Kh2     (0:00)     Qf4+    (0:00)  

This shit is NOT hard. There is a famous draw by Alekhine against Grau,
where the lowly Grau had the Champion on the ropes when a similar perpetual
check was found by the great Alekhine. ARGH.

From: (Chuck Pearson)
Subject: Re: Kasparov is no Diego
Date: 7 May 1997 08:54:59 -0400

kasparov don't know footy, either.  how do you get off comparing a
brilliant chess move that totally demoralizes you - to the point of giving
up a defeat you didn't have to - to the "hand of God" goal?

> This shit is NOT hard.

it's not hard, of course, unless you're the guy sitting on the wrong side
of the table and you have to find the perpetual draw with the world
watching you and analyzing your every move.

and you KNOW your opponent isn't going to choke, because it is physically
incapable of doing so.

i don't envy kasparov for a SECOND.

Subject: Re: Kasparov is no Diego
Date: Wed, 07 May 1997 10:17:47 -0500
From: (Doug Karpa-Wilson)

Neither do I, but I don't know about Deep Blue not choking.  If you've
ever done any programming, you have a pretty good idea that there is a bug
somewhere, particularly in something as complex as Deep Blue.  
Furthermore, Deep Blue only looks ahead fifteen moves (about fourteen more
than I can), so it had no way of noticing that possibility of the
perpetual check, no?   I wouldn't be surprised if Deep Blue choked.  In
fact, the NYTimes article, suggested that Kasparov's real error was in
assuming that a perpetual check was something that Deep Blue would
automatically prevent, when in fact it didn't.  Perhaps if Kasparov had been
thinking that Deep Blue could choke he have had his draw.....

Subject: Re: Kasparov is no Diego
Date: 7 May 1997 13:04:20 -0400
From: (Chuck Pearson)

: Neither do I, but I don't know about Deep Blue not choking.  If you've
: ever done any programming, you have a pretty good idea that there is a bug
: somewhere, particularly in something as complex as Deep Blue.  

but i don't see that as a choke.  i see that as a fundamental tactical
flaw.  a choke is when you should know that move x is the best possible
move, and you [for emotional reasons] pick move y instead.

: Perhaps if Kasparov had been thinking that Deep Blue could choke he 
: have had his draw.....

and that is a good ol' fashioned choke.  kasparov's thought process: 
"damn, this computer can't be beat!  i give!"  computers thought process: 
[imagine your favorite do-loop here...]

chuck.  machines are machines, and humans are humans, and ne'er in
        between shall the two meet...

Subject: Re: Kasparov is no Diego
Date: Wed, 07 May 1997 17:04:27 GMT
From: maf6@*nospam* (Max Attar Feingold)

 The chess program that Deep Blue runs on has been in existence for
seven or eight years and has been tested quite extensively by
grandmasters such as Miguel Illescas.  If it had such a serious bug,
it would probably have been located already.

 Also, most chess simulations use programming techniques that allow
the computer to "learn" from its initial mistakes and thus generate
its own parameters for the game engine to evaluatuate configurations.
 Again, it is doubtful that a bug as impottant as the one you are
postulating would have survive.

>Deep Blue only looks ahead fifteen moves [...]

 Since in a situation of perpetual check the configuration of the
board repeats itself after one or two moves, the computer should have
no difficulty detecting one if it has a look-ahead of fifteen.

 That's if the programmers remembered to implement a detection
scheme, of course.

Subject: Re: Kasparov is no Diego
Date: 8 May 1997 09:39:51 GMT
From: (Eric Wang)

Reuter writes:
>"It reminds me of the famous goal that Maradona scored against England in 
>86. He said it was 'the hand of God'," stated Kasparov [...]

This was in reference to Deep Blue's 37th move Be4, when it had
37.Qb6 available.  Qb6 forked an undefended Bishop and Pawn, and
Kasparov couldn't have saved both, so it appeared to win material,
and thus the game.  In the press room, nearly all of the 20 or so
chess grandmasters polled said they would have played 37.Qb6;
several of them, and Kasparov himself when later asked about the
position, said they would have played it "instantly, without
hesitation".  However, Deep Blue's analysis during the game revealed
that Kasparov could have met this move with a tricky two-pawn
sacrifice, conceding material to free his pieces and obtain
dangerous counterplay, and Kasparov did, in fact, have this
sacrifice worked out and ready to go.  DB's actual move 37.Be4 was
*materially* worse, since it won no material, but *positionally*
better, since it strangled Kasparov's position and slammed the door
shut on his only glimmer of a chance at counterplay.  It was this
positional awareness to play like an "anaconda" that has been
computers' traditional weakness, and obviously many human experts
lacked the foresight to see this move, too.  This is what had
Kasparov babbling and bubbling -- probably he hadn't seen a move of
such depth played against him since his last loss to Karpov.

BTW, Reuters are a bunch of dummies; Kasparov's words were meant in
awe and tribute, not to imply cheating.  The scuttlebutt is the
other way around: the conspiracists are suggesting that it's
Kasparov who's deliberately playing less than his best to make DB
look good and guarantee another lucrative rematch next year (and the
year after, and ...)  IMHO, that's a bunch of hooey.

>Several leading grandmasters admitted that they too had initially missed 
>the continuation that would have saved the day for Kasparov. It was left 
>to an untitled player in an Internet chat room to claim the credit for 
>unlocking the problem.

Yup.  Probably 50 grandmasters in attendance, and every one of them
thought Kasparov was kaput.  But this "mass hypnotism" was probably
the result of their collective awe in watching undoubtedly one of
the greatest, if not *the* greatest, games ever played by a computer
-- playing the positional style that computers are usually very bad
with, and playing it better than most humans could have.  Kasparov
himself was thoroughly demoralized by the end, and couldn't bring
himself to believe that the computer had missed anything.  (In fact,
Deep Blue *had* missed the critical drawing line, which is evidence
for its depth and subtlety.)

>[match details] 44. Kf1 

This was Deep Blue's mistake; at first, it didn't see the draw
either.  44. Kh1, running into the corner, would have deprived
Black's queen of the necessary first check later on, and thus sealed
the win for White.

>44 ... Rb8 45. Ra6 {Black resigned} 1-0

Ariel Mazzarelli writes:
>Here is a more reasonable and representative approach:

> 45.  Ra6     (3:32)     Qe3     (0:00)  
> 46.  Qxd6    (0:00)     Qxe4    (0:00)  

46 ... Re8! is the only move that saves the draw.  Otherwise, White
wins with the in-between move 47. Qc7+ Kg8, and now captures the
Rook _with check_, 48. Qxb8+ Kh7, after which White's own Rook can
drop back to help the defense, 49. Ra1, and now Black's queen cannot
go to the first row to deliver check, so White's King eventually

The position turned out to be extremely difficult, requiring more
than 10 hours of analysis by Kasparov's seconds before they felt
confident enough to break the news to him.  Computer analysis
remains incomplete to this moment, but after some 36 hours of
calculation, every line of play has been plotted out to move 60 or
beyond, and Black draws every one of them.

After 46 ... Re8, the best line is 47. h4! hanging the Bishop, 47
... h5!! declining the material to deprive the King of a critical
flight square, 48. Bf3, and Black has a laborious series of checks
that White cannot escape, and that eventually force three
repetitions of the same position.

Date: Sun, 11 May 1997 18:07:22 -0700
From: Ariel Mazzarelli 
Subject: corposilicon wins

Using a non-book sac in the opening (brought to ibm courtesy of some
$100/hour grandmaster, perhaps?), the machine beat Kasparov in the
last game and took the thing 3.5 - 2.5.

Serves Kasparov right for resigning that drawn game, I suppose. Maybe
it's time for a player that knows how to play these things to step up
to the board.

The move that took folks by surprise (although not if white had been
Tal, of course) was 8.Nxe6. A very good positional sacrifice, and
Kasparov probably defecated on himself at that point and it was
over pretty quickly. Maybe 11...b5 was reckless; I'm looking at the
variations starting with 11... b6 followed by a5 and Ba6 as possibly
better defenses, although this allows 12.c4 which keeps the black knight
from going to d5. One thing is clear, though--after 12. a4 black's king
won't be much safer on the queenside either.

After 19.c4 black resigns because he is just deep fried. 19...bxc4
20.Qxc4 and black loses at least the c6 bishop without fixing his
position--eg 20... Nb8 21.Bxb8 Rxb8 22.Qxc6+ Nc7 23.Rxa7 Bd8 24.Ne5 is
just sick--eg 24... Rxb2 25.Qd7+ Kb8 26.Nc6 mate.

btw what happens after 19.c4 dxc4 20.qxc4 nb4? I like white, of course, 
but I don't quite get the 'oh yeah' move. 21.ra4 nb6 22.qe6+ kd8 
23.Bc7+ kxc7 24.qxe7+ kc8 25.Rxb4 is the best I can do, and black is
certainly ugly, but... well, I guess Garry was a little tired and just 
wanted to go home. It's rough to get innoed by the computer.

[Event "fics rated standard game"]
[Site "fics, Oklahoma City, OK USA"]
[Date "1997.05.11"]
[Time "13:56:31"]
[Round "-"]
[White "DeepBlue"]
[Black "GMKasparov"]
[WhiteElo "2800"]
[BlackElo "2796"]
[TimeControl "10800+0"]
[Mode "ICS"]
[Result "1-0"]

1. e4 c6 2. d4 d5 3. Nc3 dxe4 4. Nxe4 Nd7 5. Ng5 Ngf6 6. Bd3 e6 7. N1f3 h6 
8. Nxe6 Qe7 9. O-O fxe6 10. Bg6+ Kd8 11. Bf4 b5 12. a4 Bb7 13. Re1 Nd5 
14. Bg3 Kc8 15. axb5 cxb5 16. Qd3 Bc6 17. Bf5 exf5 18. Rxe7 Bxe7 19. c4 
{Black resigned} 1-0