Hydra Chimera (C) - Nickel,Arno (2586) [C07]
Corr. Chess Match Abu Dhabi/Berlin, 16.09.2004

Semi-"live" comments and analysis.

1.e4 e6

I have the impression, that the French Defence has been played rather seldom in
man/machine events, although the high strategical impact of such games could be an
incentive for human players. This is why in contrast to my further habits and preferences
this time I chose 1...e6, especially as we had an unofficial test game before, where Hydra
could see what I was going to play in the Caro-Kann Advance Variation. Before
getting surprised in the opening I preferred to surprise Hydra by myself.

2.d4 d5 3.Nd2

The Tarrasch System is a good choice in order to avoid the strategical more
comlicated positions after 3.Nc3 Bb4 4.e5 c5 5.a3. I suppose this choice goes back to
the book author GM Christopher Lutz. 3...c5 Player who prefer a closed centre - especially against computers - would rather play 3...Nf6 ziehen, in order to provoke 4.e5 and
afterwards attack White's central pawns by c7-c5 or f7-f6. On the other hand many
players of the French Defence prefer 3...c5 against the Tarrasch System as for instance
Wolfgang Uhlmann. I thought this move to be more suitable in order to secure a
draw with Black, and the respect to Hydra and her team was big enough, so that I
avoided to strive automatically for a closed position. That Hydra in case of doubt
would be able to play a closed opening position could be seen most impressive in the
first of her two games with Ex-FIDE-Champ Ruslan Ponomariov in Bilbao 2004,
where the Ukrainian GM played to carelessly with the white pieces.

4.exd5

Also possible is to keep the tention for a while with 4.Ngf3.

4...Qxd5

The classical choice is 4...exd5, taking the risk of an isolated pawn at d5
(after d4xc5 or c5xd4), but enjoying active play with his pieces. By recapturing
with his queen Black avoids the isolani at the cost of a slower development, as his
queen loses one or two tempi. Now the pawn-structure is more likely the Sicilian
than the French, whereas this time Black has already managed to advance with the d-pawn.

5.Ngf3 cxd4 6.Bc4 Qd6
 

7.0-0

White could also do without recapturing the d-pawn and play more actively:
7.Qe2!? Nf6 8.Nb3 Nc6 9.Bg5 a6 10.0-0-0 b5 11.Bd3
see Adams-Buhmann 2002/ 03 and Glek-Rustemow 2003/04,
German Bundesliga, both games won by White.

7...Nf6 8.Nb3 Nc6 9.Nbxd4 Nxd4 10.Nxd4

Sometimes 10.Qxd4 and trading queens is to be seen. However, I did not suppose
Hydra to play like this, in particular as her endgame skills
(like most other computers too) are not that outstanding.

10...a6 11.Re1 Qc7

One of many possible variations for both sides. Black prepares the development of his
bishop to d6 or c5 (on e7 it would be too passive). The neuralgic point e6 isn't in danger so far, but Black has to keep an eye to Bc1-g5 (and Bxf6 after 0-0) or Nd4-f5.

12.Qe2 Bc5

12...Bd6 The main move. 13.Bg5!? 0-0 14.Bxf6 gxf6 15.Bd3 (‹15.Qg4+) 15...Qc5!?~~
(Suspicious seems to be 15...Bxh2+ see the duel of two Ex-World champions of
correspondence chess Oim-Sloth (jubilee tournament 50 years ICCF 2001, 1-0 after 55 moves). 16.Kf1 Bf4 17.Qh5 f5 18.g4)

3.c3 h6

GM Glek put a "!" to this move in Informator 88/250 (Kotronias-Glek, Silivri 2003, drawn
after 34 moves) This assessment could have been questioned pracically, if Hydra would
have played 14.Nf5 threatening a piece sacrifice on h6 (after ...0-0), as proposed in the CSS(Computer)-Forum. Without to comment and analyse those variations in detail, I can say
I would most probably have played 14...Kf8. Giving up the right to castle does not seem
such important, if Black is able to activate the h8-rook by pushing forward the h-pawn, while
on the other side he gets an active bishop on the diagonal a8/h1. Finally, White's knight on f5 loses time and has to retreat. 13...0-0 the normal move which is considered as reliable. 14.Bg5 Bxd4 15.cxd4 Nd5 16.Rac1!? (16.Bxd5 exd5 Tal-Kortchnoi, Brussel 1988, darwn after 21 moves) 16...Qb6?! in Speelman-Nogueiras, Barcelona 1989, 1-0 after 39 moves. (16...Qd6!?=)

14.Bd3

14.Nf5

14...Kf8 (14...0-0 15.Nxh6+ gxh6 16.Bxh6 Be7 (16...Rd8? 17.Qf3->)
17.Bb3 (17.Rad1 Rd8 18.Rxd8+ Bxd8 19.Bd3 Bd7 20.Qf3~~; 17.Bxf8 Bxf8~~)
17...Qc5 18.Qd3 Qh5 19.Bxf8 Kxf8 20.Bd1 Qh4 21.Bf3 Kg8

(21...Ng4 Shredder 8 in a search depth of 21 moves after 16.Bxh6: 22.Bxg4 Qxg4
23.Rad1 b5 24.h3 evaluated 0.63 for White (A. Strangmuller).)

22.Rad1 Ng4 Shredder 8 in depth 24/63, +0.34 (J. Rang).) 15.Ng3 b5 16.Bd3 h5 17.Be3 Bd6 (17...Bb7 18.Bxc5+ Qxc5 19.Rad1 h4 20.Ne4 Qc6) 18.Rac1 h4 19.Nf1 h3 20.g3 Bb7

Black's countre-play on the diagonal h1/a8 becomes evident, though the position
remains unclear and very complicated (A.N.).

21.Nd2 Rd8 22.Bg5 Bc5 23.Ne4 Bxe4 +/= (0.67) depth: 23/48 (A. Strangmuller).]

14...Bd7

The bishop's fianchetto by b7-b6 or b7-b5 is not good her, firstly because
White is able to attack the queen-side pawns, and secondlx because e6 still needs to be protected. 15.Be3 If now 15.Sf5 king-side castling would not be safe,
but Black could castle to the queen-side.

15...Bd6 16.g3 0-0
 

17.Qd2?!

If White has nothing better as preparing Bd3-f1-g2, then Black has already euqalized the game. Also Hydra's book author GM Christopher Lutz was not impressed by the outcome
of the opening and considered 7.Qe2!? as a more active choice for the future. - What can
be said more to this position? Black frees his play by Nf6-g4 and changes knight versus
bishop e3. 17.Rad1 looks more flexible as 17.Qd2. Also advancing with the pawns on the
queen-side came into consideration. Even the sophisticated manoeuvre Qe2-f1-h3
would appear more lively than the powerless move Qd2.

17...Ng4= 18.Rad1 Rfe8 19.Nb3 Nxe3 20.Qxe3 Rad8 21.Qe4 g6

Black is without any problems. The bishop's fianchetto Bd6-f8-g7 is now a further option.

22.Bf1 Bc6 23.Qe3 Kh7

Oder 23...Lf8.

24.Bg2 e5

Not an easy decision. Black has to be careful, as White could use the square d5 as turntable
for his pieces. On the other side, as an endgame bishop versus knight becomes more and
more probable, the pawn chain g6/f5/e4 should secure Black good counter play at the
king-side, so that he has not to restrict himself to passive defence at the queen-side.

25.Bxc6 Qxc6 26.Qe4
 

A quiet draw offer? Black is right trading the queens now, otherwise he
might get some problems when struggling for the initiative.

26...Qxe4

26...Qc7 27.Rd5 f5 28.Qe2 e4 29.Red1 would be too committing,
though the game is still equal.

27.Rxe4 Bc7

As his king is still far away Black decides for a.reliable though
somewhat passive defence at the queen-side.

28.Ree1 b6 29.Na1
 

Well, Hydra hasn't completely lost her humour. Indeed this is the only and best way to bring the knight to d5 (via a1-c2-e3 or -b4), without to disturb the d1-rook.

29...Kg7 30.Nc2 Rxd1 31.Rxd1 Rd8

A so-called easy andgame position, which does not put real problems, but is to be played with proper endgame technics by both sides.

32.Rxd8 Bxd8

A human player with the white pieces would now try to centralize his king and put his pawns
at the king-side on white squares (opposite to the black coloured bishop) . Hydra however as probably other chess programs too at first centralizes her knight at d5 and mobilizes her
queen-side pawns. This is still playable but becomes dangereous at some point later on...

33.Nb4 a5 34.Nd5 f5 35.c4 Kf7 36.a4 Ke6 37.b3 g5 38.h3 h5 39.Kf1

Here Hydra offered a draw, which was completely okay, but I refused it, because in an exhibition game people would prefer to see how both sides solve the remaining problems.

39...g4 40.h4?!

White gives himself a hard time. 40.hxg4 was still a clear way to draw.

40...f4 41.Ke2

[41.gxf4 came into consideration, in order to reach a study-like draw. It may be a matter of taste, if one prefers 41.Ke2, whereafter White also has to fight hard for the draw.

41...Bxh4 42.Nxb6 Be7 43.Nd5 Bd6 44.fxe5 Kxe5 45.f3 g3 (45...gxf3 46.Kf2) 46.Kg2 Kd4
47.Nf6 h4 48.Ne4 Bf4 49.Nf6 Kd3 50.Nh5 Bb8 (50...Be5= 51.c5 Kd4 52.f4 Bb8 53.f5 Kxc5
54.f6 Kd6 55.Nf4 Kd7 56.Ng6 Ke6 57.Nxh4 Kxf6 58.Nf3 Ke6 59.Nd2 Bc7 60.Nc4 Kd5
61.Nxa5 Bxa5 62.Kxg3=) 51.Nf6 Be5 52.Ne4 Ke3 53.Ng5[] (53.c5? Kd4-/+) 53...Bd6
54.Ne4 Bf4 55.Nc5 h3+ 56.Kxh3 Kxf3 57.Nd3 Bb8 58.Ne1+ Ke3 59.Kg2 Kd2
60.Nf3+ Kc3 61.Kh3 Kxb3 62.Nd4+ Kxa4 63.Nc6=]

41...Kf5
 

42.Kf1?

42.Kd3[] was without alternative, but astonishingly sufficient to reach a study-like draw.

42...e4+ (42...Bf6 leads to a different move order) 43.Ke2 fxg3 44.fxg3 Bf6 45.b4!?
(45.Ke3 Be5 46.Ne7+ Ke6 47.Kxe4 Bc7-/+) 45...Be5!? (45...axb4 46.Nxb4 Be5 47.c5!
(47.Nd5? Bxg3 48.Nxb6 Be5 (48...Bd6? 49.Nd5 g3 50.Kf1 Kg4 51.Kg2 Kxh4 52.a5 Bc5
53.Nc3 e3 54.Ne2 Kg4 55.Nxg3 h4 56.Ne2 h3+ 57.Kh2 Bd6+ 58.Kh1 Bc7 59.a6 Bb6
60.Ng1 Ba7 61.Kh2 Bb8+ 62.Kh1 Kf5 63.c5 Ke6 64.Nxh3=) 49.Nd5 Bd4-+

and White is without counter play against the Black's passed pawns.)

47...bxc5 48.Nc6 Bxg3 49.a5 Bxh4 50.a6 g3 51.Ne7+ Ke6 52.Ng6 Kd7 53.Nxh4 Kc7=)
46.c5! Ke6 (46...axb4?! 47.Nxb4 bxc5 48.Nc6 Bxg3= see variation A2; different move order.)
47.Nf4+ Bxf4 48.gxf4 axb4 49.f5+ Kd7 50.cxb6 b3 51.b7 Kc7 52.f6 b2 53.b8Q+ Kxb8
54.f7 b1Q 55.f8Q+=; 42.Kd2 Ke4 43.Ke2 different move order

 42...Ke4-+ 43.Ke2 fxg3 44.fxg3 Kd4 45.Kd2 e4 46.Ke2 e3 47.Ke1 Kd3
48.Kd1 e2+ 49.Ke1 Kc2 50.Kxe2 Kxb3 51.Kd3 Kxa4 52.Kc2 Bxh4 53.gxh4 g3 54.Nf4 Kb4
 

 White lost by time on 4.th of May 2005.

0-1

Please see the games here :