Chessville - by chessplayers, for chessplayers!

Corus Diaries 2008 - The Caruana Kid
by Dr. Albert Alberts

Chessville is pleased to present notes and games from Corus by the noted author
Dr. Albert Alberts (See Chessville reviews of Dr. Alberts' book, How To Fool Fritz
- Explorations in Man Assisted Machine Chess
here and here.)  For Dr. Alberts'
observations and analysis covering the first twelve rounds of the event, click here.

CORUS 2008 Wijk aan Zee, The Netherlands.  The party is over.  The white demonstration tent can be dismantled and the five or six black horses can return to the large vacant green meadow in the centre of the village.  As far as I know Wijk aan Zee is the only beach village in the western hemisphere without ugly tourist exploitation.  And NO parking meters!!  Back to business as usual.

Feel free to apply for a job as a kitchen aid in hotel Sonnevanck, the kitchen closes as 10 pm and if you can play piano you can sit down behind the restored black Bluethner grand 80 years old, a magnificent instrument and entertain the guests.  If you are finished with the pots and pans at that time that is, which is a tough job, I can assure you.

Special guests can be granted lodging in the pyramid glass penthouse on top of the hotel with a view over the sand dunes on the sometimes rough North Sea.  You can write books up there and I think it has been done.

Carlsen and Aronian on top.  Computers say that Magnus missed a continuation to a good advantage and the tournament victory, while Judith Polgar (mrs.? but playing under her world famous maiden name) managed to draw a Ruy Lopez Marshall gambit.  The blackies played 4-5 of those in the final round.  Marshall's western chess invention by now is about 100 years old, a multitude of lines have been analyzed up to move 30, and in my view it is the most convincing proof so far that the noble game is a draw by nature if both players play the best move every step of the way.

We are still left with the question why computers play better then humans.  In particular in a tournament setting with head calculation in limited time.  Knowing that Kasparov and Kramnik went under, amateurs can forget about it.  The solution has to come from free style- advanced-man-assisted-machine chess or correspondence chess, which by now is all the same thing because it is hard to believe that correspondence players in their private attics and shacks (and villas, e.g. world champion millionaire Van Oosterom) do not resort to machine assistance.

Backed up by computers any amateur can play over Elo 2700, the theory is all in there, the endgames are exactly solved.  It just is not fair.  A complete chess library without royalties for the inventors for 50 eu or $.  Somebody is suffering.  Like the advent of the Hammond organ in movie theatres that put violinists out of business in the 1930s.

One does not need tournament books, no comments from professionals, the machines make the proceedings crystal clear.  Numerical rating says it all, over +2 (Fritz/Rybka/Junior, what have we, it makes no difference) the advantage for White is overwhelming, with -2 for Black.  The critical risk margin reaches up +3 resp. -3, which is considerably higher then the one operative in tournament games, usually fluctuating within +1 and -1.  Re-raising the risk a couple of times will ultimately drive HAL (the chess computer in Kubrick's Space Odysseia) crazy and bring about its collapse.

Best game I saw so far in that style on CORUS?  Carlsen-Anand, an open Sicilian, and with Anand I am still not convinced that somewhere the Carlsen plan going Qh4!? b3!? Rc1/c3!?, Bg2!? Rh3 Bf3!? sacrificing two pieces in the finale never mind the two pawns on White's queen side can not succeed.  A string of daring non-computer risk moves.  A work of art.

In my view the open Sicilian (Najdorf, Svesnikov, Dragon, Keres Attack, Sozin-Fischer, a.o.) is the prominent battlefield in computer chess due to the hideous asymmetry of the pawn structure.  And the best answer for Black on 1.e4, in particular with the e5-Svesnikov variation which in tournaments now is avoided via the older Rossolimo-line with White going Bb5.  The paradigm of Robert James Fischer.

Various examples can be retrieved from the literature in which very strong players get lost in the Sicilian jungle before move 20.  Classics are Ratmir Little Warrior Cholmov-Keres, Cholmov-Bronstein and Tisdall-Lee all to be found on www.howtofool.fritz.com.  100 best chess moves ever collected by Tim Krabbe and computer analyzed by the Fritz Slasher on that site.

Before I vanish in the night with a computer study of Karpov-Kasparov match games (two aliens playing chess taking the game to extreme limits) I collect a couple of open CORUS Sicilians.  First from the CARUANA KID (Italian-US double passport I believe) whose overwhelming win in GC-C is highly likely to promote him to GC-A next year.  We already highlighted Caruana-Carlsson in round 7, a Black-e5 Sicilian (hunting game is the Dutch term) responded with Nb3/f4.  Ruigrok avoided the open Sicilian against Caruana and lost in round 3.  Now we go Najdorf:

Caruana- van der Wiel, the latter a Sicilian expert par excellence

1.e4 c5 2.Nf3 Nc6 3.d4 cxd4 4.Nxd4 Nf6 5.Nc3 d6 6.Bg5 e6 7.Qd2 a6 8.0–0–0 Bd7 9.f3 b5 10.Kb1 h6 11.Be3 Ne5 12.Bd3 Rb8 (out of book) 13.g4 Nxd3 14.cxd3








A remarkable idea: Caruana tolerates an open c-file and the d-pawn will turn out to be a brave warrior going d4/d5 with Nc1/d3/b4/c6

14...a5 15.h4 g6 16.Nb3 a4 17.Nc1 b4 18.N3e2 Bg7 (about equal here) 19.d4 h5 20.g5 Ng8 21.d5 e5 22.Nd3








White is better at +1.

22...b3 23.a3 Ne7 24.f4 exf4 25.Bd4 0–0 26.Bf6 Bg4 27.Rdf1 Re8

The wrong one but how can you tell in these Sicilians with both pawn storms frozen?

28.Nd4 (1.88 for White) Qb6 (machine explores lines with f3) 29.Qxf4 Bf8 30.Nb4








The White offensive has switched from the king-side to the queen-side and very often the c6-square, weakened by the advance of ...d6 and ...b5, brings a decision in Sicilians while a king-side attack is still looming.  Compare Cholmov- Bronstein and/or Cholmov-Keres cited above.  Caruana s style definitely can conquer computers as well.

30...Rb7 31.Ndc6 Nxc6 32.dxc6? (over 3.13, Black lost) Ra7 33.Nd5 Qa6 34.Bc3 Re5 35.Nf6+ Kg7 36.Bxe5 dxe5 37.Qxe5 Qd3+ 38.Ka1 Re7 39.Nxh5+ Kg8 40.Nf6+ Kg7 41.Qb8 Be2 42.h5 gxh5 43.Rc1 Rxe4 44.c7 1-0
 

And although in this case Nijboer has to be credited for an intriguing novelty, we study:

Nijboer-Caruana
CORUS 2008 round 12

1.e4 c5 2.Nf3 e6 3.Nc3 a6 4.d4 cxd4 5.Nxd4 d6 6.Be2 b5 7.Bf3








A novelty on move 7, and a good one, in a thoroughly analyzed opening.  A rarity.  Classic chess is not dead RJF!

7...Bb7 8.e5 Bxf3 9.Qxf3 d5 10.Qg3 (White good) Ne7 11.Bg5 h6 12.Bxe7 Qxe7 13.f4 g6 14.0–0–0 Nd7 15.h4 h5 (about equal here) 16.f5 gxf5 17.Nxf5 Qc5 18.Nxd5








The culmination of White's idea.  True Sicilian style.  The black king is in a minefield in the Sicilian if you can prevent him from castling.  The risk for White is -0.5 which means compensation for the knight.

I think Nijboer's idea has a future, and is certainly worth analyzing in computer chess.

18...exd5 19.e6 fxe6 20.Qg6+ Kd8 21.Nd4 Kc8 22.Nxe6 Bh6+ (Rh6) 23.Kb1 0.00 dead equal here but White can play for a win Qc6 24.Qf5 0.28 (tilt positive!) Ra7 25.Rxd5 Re8 26.Nc5 Bf8 27.b4 Qf6 28.Qxh5 Rd8 29.Ne4 Qc6 30.a3 Rc7 31.Rc1 Kb8 32.Qd1 Rdc8 33.Nd6








-1.09 things start to look grim for White but he can still fight for a draw.

36...Nb6 34.Rd3 Bxd6 35.Rxd6 Qb7 36.c3?

A fatal human mistake probably due to time pressure.

36...Nc4 37.Rd4 Qxg2  Over -3, and with true gentleman's style White resigned.  Heavyweights like Kramnik will play on to try to swindle White out of a defeat at this level.

Great game with theoretical value.

See you next year!  - Albert Alberts

 
Chessville coverage / Official site

 

Chessville
logo by
ChessPrints

 

Follow Chessville1 on Twitter
 


Ranked #1 by Google
"largest chess database''




The
Chessville
Chess Store


The
Chessville
Weekly

Newsletter

Subscribe
Today -

It's Free!!

The
Chessville
Weekly
Archives

 

Advertise
with
Chessville!!

Advertise to
thousands
of chess
fans for
as little
as
$25.

Single insert:
$35
x4 insert:
@ $25 each


 

 

This site is best viewed with Java-Enabled MS Internet Explorer 6 and Netscape 6 browsers set at 800x600 screen size.

Copyright 2002-2009 Chessville.com unless otherwise noted.