March 28, 2012 8:43

Review: Kaufman and his Komodo

Review: Kaufman and his Komodo

Nowadays opening books play a different role than before the database era. There is no longer a need for a complete survey of all possibilities. Instead, opening books may have different focuses, supporting or even avoiding the database knowledge. Quite familiar is an opening book written by an expert of high level, which supplies valuable explanation on a specific variation. A recent example of this is The Strategic Nimzo Indian by Ivan Sokolov. Another kind of opening book, aiming at somewhat lower leveled players, may give a structural, general overview and basic explanation of an opening (a recent example being The Nimzo-Indian, move by move by John Emms). In my opinion this is a very useful method to become acquainted with an opening before plunging into a database. Understandably, these days especially young players are tempted to start with the latter.

Next to Arne Moll, we've added a second reviewer to our team: Dutch IM Arthur van de Oudeweetering. This is his first review for ChessVibes.

Larry Kaufman’s repertoire for Black and White has recently appeared and does not fit in either category. I was already acquainted with its predecessor The Chess Advantage in Black and White, so the concept was not new to me. Of course, this book allows you to build up a complete repertoire with the help of one single volume, which will appeal to many. But above this, the book also has a unique approach, which may contain some unavoidable flaws but certainly is revolutionary and inspiring.

For starters Kaufman is a computer expert who has done work for Rybka, and is co-creator of the strong engine Komodo. Of course nowadays every chess author uses chess engines to check or back up his own ideas (by the way, in analysis you often run into a phrase like “my computer says” which seems to me either a sloppy use of language or a clever way to avoid mentioning the name of the engine you have been using). Kaufman has gone far beyond this: he has methodically used Komodo (and also Houdini) to evaluate virtually all positions in his book. This of course leaves little room for tactical mistakes, but also provides numerous improvements on grandmaster games. Or occasionally produces a stunning novelty, like this

PGN string

Position after (1.e4 e5 2.Nf3 Nc6 3.Bc4 Bc5 4.0–0 Nf6 5.d4 Bxd4 6.Nxd4 Nxd4 7.f4 d6 8.fxe5 dxe5 9.Bg5 Qe7 10.Na3) 10…Rg8!

A computer move if ever there was one

as Kaufman rightly says. He also mentions that this move has been published before, in Sosonko‘s column in Yearbook 91 on how Dzindzichashvili discovered this move when working with Rybka (the move now has also been played by Kosteniuk as I noted in the database). Still the book has many other – maybe less spectacular – recommendations which did not appear before.

Before dealing with the actual lines Kaufman devotes some space to the developments after the publication of the predecessor in 2003. He also elaborates on material values and the role of computers in the book. This pleasantly provides some clarification for the main part of the book. For example, I remember from The Chess Advantage for Black and White that I was struck by the number of times that an evaluation referred to the pair of bishops.  In his recent book Kaufman explains that he considers the bishop pair worth half a pawn. He apparently supported this with data in Chess Life already in 1999. In The Kaufman repertoire he states:

I think it is only a slight exaggeration to say that the main goal of the opening is to maintain the two-bishop advantage: certainly that is the main goal for many of the lines recommended for White in this book.

I cannot judge about the general validity of these ideas about the bishop pair, but is it useful food for inspiration as can be seen from the very first game in the book.

Saric,Sinisa (2448) - Miladinovic,Igor (2556)
SRB-ch Kragujevac (5) 04.03.2011 D06

1.d4 d5 2.c4 Bf5 3.cxd5 Bxb1 4.Qa4+ c6 5.Rxb1 Qxd5 6.Nf3 Nd7 7.Bd2 Ngf6 8.e3 Nb6 9.Qa5 Ne4 10.Qxd5 Nxd5

PGN string

and now instead of the game move 11.Bd3 Kaufman suggests 11.Ba5! This indeed seems a good move and may have something to do with the pair of bishops, but also with keeping pieces on the board to profit from a spatial advantage due to the mobile white pawn centre. Kaufman however does not refer to the latter.

So the engines have made a prominent contribution to this book, but Kaufman occasionally also notices shortcomings, like in the Botvinnik variation:

PGN string

Here Kaufman concludes: 

This whole variation with 16…Qa6 deserves more investigation as the resultant positions are too hard to evaluate even with the best computers.

The book itself is divided in two main parts, being, this may not surprise you, the repertoire for White, based on 1.d4 in 15 chapters, and the repertoires for Black (also 15 chapters), based on a 1.e4 e5 repertoire with the Breyer as main line and the Grünfeld against 1.d4. Many lines of this black repertoire are inspired by Carlsen’s choice of openings, Kaufman states.

The book has two “front covers”, one with a white background, the other with a black one. Both repertoires have their own page numbers, the black repertoire starting up-side- down from “the other cover”. Quite a nice and original lay-out, creating the illusion of two separate books (though you may get annoyed by having to turn the book time and again when looking for something in your other repertoire).

Kaufman has generally chosen sound main lines (no quick offbeat lines), with 2.Bg5 against the Dutch being the exception. The white repertoire also contains a chapter on the Pirc, Modern and Philidor (via transposition 1.d4 d6 2.e4) and an odd, not so relevant one on 1.Nf3 (which you may feel free to skip, in Kaufman’s own words).

Every chapter has a short introduction on the chosen variations. In general Kaufman avoids extremely complicated lines, and focuses on active, strategically sound variations. Kaufman seems to have done a good job in selecting all relevant lines to provide a more or less complete repertoire. I suppose this must be a hell of a job, what to include and what to omit. 

Kaufman occasionally works out some alternatives. like 4.Qc2 or 4.Qb3 against the Slav complex besides the mainlines that he provides. He also gives two different possibilities against the Queen’s Gambit accepted and covers both the Nimzo Indian and the Queen’s Indian. I’m a little skeptical about this because this may leave the reader in doubt, forcing him to make a choice instead of offering a clear cut solution in the theoretical jungle. On the other hand, now the reader will be able to form a repertoire more to his own liking or style.

The introduction mentions that the book is intended for a wide range of chess players:

The basic idea is that the chosen variations are suitable even for the strongest grandmasters, but also for the average tournament and club players. 

Of course this is always a good argument to sell more copies, but I do think Kaufman succeeded in this. Stronger players will be able to quickly scan an opening and pick up ideas from the high level analysis, while the openings are indeed also suitable for the average player to build up a comprehensible repertoire.  

One more thing that I quite like is that the complete games are provided, a kind of trend in opening books. In this way you can gain an impression how the game may develop after the particular opening. All in all the book is neatly organized and thus pleasant to read and easy to handle.

The trickiest part for the book’s concept is of course the point where the white repertoire meets the black one. Every chess player faces the same problem: sometimes playing White you have to face the system your prefer with Black. This is unavoidable, also in this book. The point where the two meet is in the Hungarian variation of the Grünfeld defence:

PGN string

Position after 1.d4 Nf6 2.c4 g6 3.Nc3 d5 4.Nf3 Bg7 5.Qb3 dxc4 6.Qxc4 0–0 7.e4 a6 8.Be2 b5 9.Qb3 c5 10.dxc5 Be6 11.Qc2 Nbd7 12.Be3 Rc8 13.Rd1 b4 14.Nd5 Bxd5 15.exd5 Nxc5 16.0–0

Here the White repertoire continues with 16….a5 adding that 16….Qd6 17.Rc1 Qb8 18.Bc4 (…) and Black doesn’t have enough for the bishop pair here.

The Black repertoire says: 16…Qd6! Previously 16…a5 was played 17.Rc1 Qb8 18.Bxc5 18.Bc4 Rfd8 White has the bishop pair but an isolated d-pawn, while Black has good squares for his knights on c5 and d6. Perhaps White is still slightly better after 19.Rfd1.

Here the author is clearly struggling like we all are when in the same sort of situation. I have never yet heard any clear practical (or psychological) recommendation how to deal with this. In any case, theory is continuously evolving, so a guaranteed advantage (or equality) just seems to be an illusion. 

Another similar point in case can be found in his introduction to the black repertoire:

(…) ” I would say that only the Italian, the Spanish with 6.d3 and the Spanish with 9.d4 lead to positions (with best play) where I would rather play with White than Black, and just marginally so.

Note he is here advocating 1.e4 e5 for Black! By the way recently the Spanish with 6.d3 was also given attention in Chess Evolution, a completely other kind of publication on openings. Both works cover the same mainline, Kaufman supplying a compact practical advice and Chess Evolution (Predojevic) the in depth analysis.

All in all I think this is a very original opening book, making excellent use of the contemporary means. It will be extremely useful for setting up a solid repertoire in a short period of time. It is also a very authentic book. Kaufman ends with fragments of his own recent games in which he has consistently tried all his recommended lines into practice. In the end he bemuses something which may sound familiar:

Now if only I could play the whole game as well as the opening. 



Csaba's picture

To clarify, when Kaufman said "(…) ” I would say that only the Italian, the Spanish with 6.d3 and the Spanish with 9.d4 lead to positions (with best play) where I would rather play with White than Black, and just marginally so." he was talking about white deviations before the main-line Breyer defence, which he also concedes is an excellent defence to the Spanish game, but it does not quite equalize (sorry, I don't have the exact quote!). Anyway, Kaufman clearly believes that with best play, White always has a tiny edge, and he writes more explicitly on this that the position where the two repertoires meet has White a small edge but Black can also play for a win.

Merlin's picture

Regarding the use of computers in Kaufman's book, I think that the analysis tools that he is using haven't been used before in writing a book. As Kaufman says in the "The Role of Computers in this Book": "My basic method was to use 'Aquarium' software 'IdeA' mode, which allowed me to input literally hundreds of thousands of positions to be analyzed one by one". There is more about this in that same section.

Mike van Rooyen's picture

Original layout with the Black repertoire 'upside down'.
However if it is from black's point of view,why not invert the diagrams?
In fact why not invert the diagrams for all opening books
that discuss the openings for Black.

jo's picture

I read a few years ago about blindfold simultaneous players back in the late 1800's almost to a man saying they always remembered the position from the white side of the board, for myself I can have the same position 12 moves into a game as black or white and they are two completely different games.
It would be interesting to find out if stronger players picture the board mentally differently than weaker players.

sligunner's picture

Just bought this (I also have The Chess Advantage in Black and White). The main things I don't like about the new book: I don't like having to turn the book upside down to switch from white openings to black openings; and a I HATE long algebraic notation. It makes it so much more difficult to visualise the moves. ALSO, in the chapter introductions, whenever he mentions a variation there should be a page-number cross-reference; I would have liked to have seen the ELO ratings of players in the games used (which he did in The Chess Advantage); lastly, I would have preferred the (short) notation NOT to have included the symbols of the chess pieces, but to have used the English lettering (K, Q, R, B, N, P etc.). Again, I think most readers would find this easier to 'visualise'.

Mindhunter's picture

Got to agree with these comments! Otherwise a nice book, but could and should have been easier to read!

sligunner's picture

I agree it's an excellent book – just poor typesetting (as usual in most chess books) and an ill-thought-out use of the chapter introductions (with no cross-references). I'll be writing them in by hand, which is a shame because the book is then no longer 'perfect' . . . but because I've had to, it clearly isn't perfect right out of the book store (Chess & Bridge on Baker Street, London). The irony of it!

GeneM's picture

I must agree with 'Mike van Rooyen' and 'sligunner'; that for books on a repertoire for Black, the diagrams should have row-8 at the bottom.

(GeneM , 2012-June-08 ,

I agree with 'sligunner' than LAN notation is hard to replay. However, LAN's problem is not that it gives the origin square. Rather the problem is that LAN gives the origin square *before* it gives the destination square.

Better is XSAN notation (eXtended SAN). When reading XSAN, the human eye can easily stop after reading the more essential destination square; whereas in LAN the eye is forced to process the less essential origin square:

[SAN] Nc3
[LAN] Nb5-c3
[XSAN] Nc3

GeneM's picture

(Weird automatic deletion of the trailing part of my XSAN move, so I will try again, by adding spaces around the LessThan sign: Nc3 LT b5.)

[XSAN] Nc3 < b5

LeonChess's picture

Hi...just one question...why do you prefer the letter than the figurine in the gama notation....I would say that it is easier to visualize...while I absolutely agree that long notation giving the origen square means a bigger effort to follow the game. Regards

rayiq's picture

the most ironic thing about the book is it claims to refute the KID and gives an equal position in the end without suggesting any improvements ...

S3-'s picture

Now they are writing chess books with the help of computers. What logic is this. If I wanted to know what a computer thinks is a good move. I would play a computer. What a waste of time in reading this book.

brabo's picture

The same logic as why chessbase has introduced the let's check feature.

Latest articles