game annotations

Main

Date: 03 Sep 2007 00:58:38
From: help bot
Subject: game annotations

I am still (yes, STILL) reading the August 2007 issue of Chess life,
and
have finally made it to GM Pal Benko's column (page 56). He starts
off
with a rapid event where GM Kramnik had White against GM Leko in
game 5. A rather unusual sequence of moves led to an early trade of
Queens, and I suspect, a heavily-analyzed (by GM Kramnik) position.

White starts off with the superior pawn structure, but at move 25 GM
Kramnik did some "free repair work" for his opponent -- a mistake just
glossed over by GM Benko, apparently because White wins eventually.

But the real problem came when GM Leko later mistimed the
liquidation of his a-pawn at move 35; this is quite simply a blunder,
but again it is glossed over, possibly because he had one opportunity
later on to escape with a draw (though this was due to other errors).
White failed to punish this mistake properly, and this somehow
resulted in zero question marks being doled out by GM Benko even
though the refutation was fairly obvious.

I sensed that the annotator was using computer help, as in some
cases his analysis is an exact copy of what appears on my chess
program's screen; but he does not make the same mistake as my
computer in the simple ending near the finish, recognizing a draw
as a draw, not a +1 point advantage! I had to look at that ending
closely, just to be sure the extra pawn was useless.

I find it interesting that this opening line -- which I had never
seen
before -- is walked into by 2700+ players with Black. It is obvious
that Black has not quite equalized, and that unless he can get
some free repair work out of his opponent (as in this game), he
stands worse. Maybe the thinking is that standing worse is okay,
so long as you are not down any material and the Queens have
come off.

GM Benko's comments attributed the win by GM Kramnik to his
endgame mastery, yet I felt this was not demonstrated and that it
was more due to GM Leko's unforced errors, some of which were
exploited and some of which went unpunished. For instance, at
one point GM Kramnik was well on top, but he then allowed his
opponent to walk his King half-way across the board(!) by messing
around, losing considerable time. Oh well, it was rapid chess.

In fact, in another article in this issue of Chess Life a different
annotator did something similar, crediting GM Kamsky when in
truth his opponent had simply refused to capture a hanging pawn
and let him keep it. Later on you see that he has this extra pawn
and think, wow, he must be good because he is still up a pawn!

I am mentally trying to compare these modern games against
classic games I replay on the Web, from many, many years ago.
In some cases, I find the old games to contain unbelievable
errors, like a back-rank mate delivered by GM Alekhine at Bled
in 1931. But in other cases, I see old games where both of the
players are ultra-precise, where just one tempo decides the
issue in favor of either player, and this sort of precision is, by
and large, nowhere to be found today, with little errors galore
in some cases. Perhaps the time controls were slower back
then, or maybe there is just less oxygen in the air today... .

-- help bot

Date: 06 Sep 2007 21:02:10
From: help bot
Subject: Re: game annotations

On 6, 9:59 pm, Sin...@webtv.net (SAT W-7) wrote:
> keep up the good work..
>
> what computer program do you have ?

I have Fritz 5.32.

Now I am looking at the new issue, tember 2007, of
Chess Life. It comes as no surprise to see a big focus
on (who else?) Bobby Fischer. Unfortunately, my program
is not strong enough to skewer the beaten-to-death analysis
of the first match game between GMs Spassky and
Fischer (though I suspect there must be something wrong
as no serious effort is made on White's behalf to find moves
of great finesse to match those for Black, tit for tat).

The program couldn't see the fortress idea, and got all
wrapped up in grabbing material (I too, plead guilty your
Honor; but I am just a weak player -- ask any-bot-y!
-- help bot).

Instead of seeking the truth, the authors (Suzan Polgar
and Paul Truong) trot out the opinion of Gary Kasparov,
suggesting "overconfidence" and a bizarre belief that
GM Fischer was trying to "make a statement" when he
miscalculated the Bxh2 line. My view is that GM Spassky
saw, but GM Fischer probably missed, the trapping move
B-d2 at the tail end of one key variation, which is just
careless analysis since he was not lacking in talent.

I found it interesting that a recent game had GM Spassky
as White against GM Portisch, and once again he went for
just such an ending and won it. This game is analyzed by
GM Benko in the August 2007 issue of Chess Life, page 57.

Typically, no special effort was made on GM Portisch's
behalf to proclaim that he had a "boring draw", or to show
some deep finesse which would have saved him after his
early mistakes. No, that sort of slant is reserved for GM
Fischer alone. This game is labeled as Heviz 2007, game
1, in case anyone is interested. The objective tone of that
article finds Black able to put up "resistance", and calls
a pawn hang a pawn hang, with no apologies for the loser.

Anyway, the final verdict will come when we have the
required endgame table-bases to settle this issue once
and for all. Then, I predict, the Fischer supporters will
shift to a new tack, bleating on about drawing "chances"
and match psychology all the more while never admitting
where they had been wrong.

The one thing I am learning from going over these
games on my computer -- aside from the fact that I can't
see tactics worth a darn compared to Fritz -- is that all
those exclams and other symbols tagged on to moves
in Chess Life (or elsewhere, for that matter) can be very
misleading. In some cases, a bad move is glossed over
by deliberately leaving off the appropriate symbol, and in
many cases a slightly inferior move is given an Academy
Award, while the best move is simply never mentioned
so as to enhance the overall perfect-white-teeth effect.

In many cases, now as in olden times, the winner's
choices and decisions will be routinely endorsed as
correct, since the eventual result was positive, while
the exact reverse applies to his unfortunate opponent.
Many, if not most, of these published annotations are
lacking in objectivity, not to mention quality. The new
style of choice is to run games through Fritz/Rybka,
and let the machine do all the work. But what is
really needed is for the ideas -- not just variations --
to be explained in a manner that most chess players
can grasp them. We need to allow for the fact that
humans can't calculate everything, and explain their
moves in that reality context.

Unfortunately, some of these articles are revealing
the fact that even top GMs do not always understand
what was going on in their own games, and when
asked why they did not play (insert computer's move
choice), they quite easily become irritated. So what
is the answer -- let weak players explain what is
going on in grandmaster games? Perhaps this could
work, if you have the right sort of weak player, and
arm him with the right tools (ChessBase/Rybka)... .

-- help bot

Date: 10 Sep 2007 10:40:43
From: help bot
Subject: Re: game annotations

On 7, 11:45 pm, Sin...@webtv.net (SAT W-7) wrote:

> what is the ELO on Fritz 5.32 ?
>
> You should get Fritz ten.

I don't really like any of the chess programs, because their
interfaces make it awkward to analyze games the way I want
to. Without any opening book installed, they all give absurd
variations as "best play" in the opening, and extra time makes
little difference. Were I to buy, say, Chessbase, I could look
up what moves had been played before and get a crystal-
clear idea of transpositions in the opening, but when you nix
analysis in favor of "what has been played before", you end up
with some real duds... which have indeed been played before!

I am also disappointed at how few ending I can resolve with
the tablebases; perhaps "Freezer" would be a useful tool for
someone with a lot of time to explore this area. I simply don't
trust books anymore; countless times I have discovered some
unexpected error in a supposedly authoritative book by a
supposed expert, who simply was too arrogant or lazy to do
the proper research before rendering his verdict.

Let me give an example of why I don't 'em much: right now,
I am running a few of my games through to uncover gross
mistakes, and verify if my "brilliant attacks" were sound or
if I just got lucky due to poor defense. With no opening book,
I set it to go over every move -- including move one! -- and as
a result I am bored to tears until the serious errors begin.
When I reach the most interesting points, I suspect that I
cannot always rely on the five-minute analysis to yield a
result that would not change at say, one hour per move, so in
effect, I may end up going back over some of these positions
manually. But I do not want to manually decide which of the
moves to analyze and which to simply accept as "book",
because I could (eventually) miss something that way. I also
like to see how the computer would play this type of position
with no opening book; sometimes it is stupid, but sometimes
it comes up with rather interesting ideas humans would never
have even considered.

-- help bot

Date: 12 Sep 2007 08:41:07
From: help bot
Subject: Re: game annotations

On 11, 11:49 pm, Sin...@webtv.net (SAT W-7) wrote:

> You do very detailed work..

In another thread, some-bot-y moans and groans that
the same game was annotated twice in the same Chess
Life issue, while another "explains" that the editor had
little choice because they were both paid for their work,
so it would be wasteful not to publish both articles.

But from my perspective, this can in fact be very
instructive -- especially if it is possible to compare the
annotations side-by-side and neither annotator had an
unfair advantage over the other, so to speak.

The problem here is that if both annotators were using
Fritz or Rybka to do the brunt of the work, we are in
effect cheated; all we will learn is which is superior at
(or spent more effort on) using computer help. In this
one case, it seems that GM Su. Polgar/P. Truong did
the far better job, but this may only be because of the
significant difference in strength between her and the
other annotator, who was not a GM.

In my "detailed work", I am finding so many errors in
my recent games that I can hardly begin to explore
them all properly. My brilliant sacrifices were not only
making Fritz yawn, but he also says that where I fell
victim to similar coups by an opponent, I could have
safely accepted the material, rather than decline and
"lose like a carrot", as Sanny might say.

Using the computer I am getting a better handle on
complex tactics, and beginning to comprehend such
things as the fact that a fork is not always as bad as
it looks. In one recent game, I declined a sacrificed
piece because immediately afterward, there would
have followed a nasty-looking fork; but the key here,
as Fritz showed me, is that no way can the forking
piece capture *both* forkees; no, it must choose
between one or the other, and I can help him choose
and thus (barely) survive the assault. I actually
declined and at one point was showing a deficit on
the graph of around four points before my opponent
began to let his winning advantage slip.

-- help bot

Date: 11 Sep 2007 21:49:05
From: SAT W-7
Subject: Re: game annotations

You do very detailed work..

Date: 08 Sep 2007 18:48:04
From: SAT W-7
Subject: Re: game annotations

Just wondering if you want to play Fritz 5.32 vs Ivan ?
Ivan was beat by Fritz 9 .......

5.32 mite be a good battle.....

Date: 07 Sep 2007 21:45:04
From: SAT W-7
Subject: Re: game annotations

what is the ELO on Fritz 5.32 ?

You should get Fritz ten.

Date: 06 Sep 2007 19:59:10
From: SAT W-7
Subject: Re: game annotations

keep up the good work..

what computer program do you have ?