Main
Date: 08 May 2005 14:42:04
From:
Subject: Chess problems don't work correctly anymore
I've purchased some very good books on chess for a while now...some of
them excellent material to study from. Unfortunately, thanks to the
infusion of powerful chess software via PC computers...they don't work
as advertised anymore. For instance...if any of you have the School of
Chess Excellence series by k Dvoretsky will know this is high
quality stuff. However, it's become apparent that many of his
questions via diagrams don't work when put to the test against
computers.

Here's a very simple example...taken from School of Chess Excellence
2: Tactical Play, page 10. Here's the FEN representation of the
diagram that he puts on that page:

1r2kb1r/1pRb1ppp/4p3/q2pP1B1/p2Q4/P2B4/1P3PPP/2R3K1 w - - 0 1

It's taken from Dvoretsky vs Zilberstein 1966 and it's White to move.
Now...I figured out the "idea" behind the tactical theme for the
diagram pretty easily enough... as White's first move SHOULD be
1.Rxd7! just as he essayed on that page. Unfortunately, when I played
the above diagram against Junior 9 on my Pentium 4...Junior comes up
with an incredible defence: 1...f6!! and I'm falling over my chair
when it chooses this move.

Now all of a sudden the sacrifice of the rook on d7 begins to look
silly as now there are two pieces hanging. Dvoretsky of course doesn't
even analyze this possibility in the notes. Why not? And that's a good
question because many of these books don't take the time to analyze
diagrams enough to justify inclusion in chess books. Clearly this
simple example shows that if such a diagrammed position can be
embarrassed by computer analysis means that many of those diagrams you
have in those very good chess books have to be re-evaluated in light
of such radical solutions given by the computer.

Another thing has to mentioned at this point...many of those tactical
analysis diagrams given for training are based on play from humans
against humans...meaning errors by the defence aren't given thorough
study and refutations aren't illustrated in many of these books. The
authors don't take the time to include such information because it's
"assumed" (my opinion) because these chess authors think these
examples are self explanatory in showing why tactical solutions to the
diagrams are fool proof. As my demonstration shows...this isn't the
case at all.

Does this mean you have to throw out all those great chess books
you've stored in your personal libarary? Of course not. What it means
is...thanks to computer software...you can now delve more deeply into
otherwise clearly labeled black and white positions and see different
ways of playing the offered positions and gives one a renewed interest
in finding ways to disprove or improve authors intentions.




 
Date: 09 May 2005 08:51:49
From: Ray Gordon
Subject: Re: Chess problems don't work correctly anymore
> Does this mean you have to throw out all those great chess books
> you've stored in your personal libarary?

Yep.


--
Ray Gordon, Author
http://www.cybersheet.com/library.html
Four FREE books on how to get laid by beautiful women

http://www.cybersheet.com/chess.html
Free Chess E-book: Train Like A Chess Champion

Don't buy anything from experts who won't debate on a free speech forum.




 
Date: 09 May 2005 00:12:16
From: =?ISO-8859-1?Q?Claus-J=FCrgen_Heigl?=
Subject: Re: Chess problems don't work correctly anymore
[email protected] wrote:
> 1r2kb1r/1pRb1ppp/4p3/q2pP1B1/p2Q4/P2B4/1P3PPP/2R3K1 w - - 0 1
>
> It's taken from Dvoretsky vs Zilberstein 1966 and it's White to move.
> as White's first move SHOULD be
> 1.Rxd7! just as he essayed on that page. Unfortunately, when I played
> the above diagram against Junior 9 on my Pentium 4...Junior comes up
> with an incredible defence: 1...f6!! and I'm falling over my chair
> when it chooses this move.

> Now all of a sudden the sacrifice of the rook on d7 begins to look
> silly as now there are two pieces hanging. Dvoretsky of course doesn't
> even analyze this possibility in the notes.

When reading older books, you have to live that not all content holds to
the tactical abilities of modern computer chess engines. When I reread
Nimzowich's "My System" a while ago armed with Shredder the engine
discovered several tactical inaccuracies. But that doesn't make the book
worthless as a whole.

I don't know Dvoretsky's book, but I bet most of the stuff is valid even
under computer engine scrutiny. As a student of chess the point of using
a computer engine isn't to prove or disprove a book's author's analysis.
The main advantage of using a computer is that you can check ideas of
yourself the author didn't think of. Without a chess engine you have no
evaluation how good your idea is, other than the book author didn't
think it worth mentioning. With the computer you can test your ideas.

In the position in question, 1...f6 is no refutation of 1. Rxd7. White
could simply play 2. Rdc7 with the threat of exf6. If Black balances the
material with 2...fxg5 White plays 3. Qg4 with the threats Qxe6 and Qh5+
which Black can't defend. If 2...f5 3. Qa7!! mates in 7.

Another excellent move in the starting position is 1. Qf4 with the idea
of 2. Rxd7 Kxd7 3. Qxf7+. Black doesn't have a defense. At best Black
can throw material at White to delay the outcome. 1...Be7 2. Bxe7 Bc6 3.
Bd6 Qxc7 4. Bxc7. Or 1...Bxa3 2. Rxd7 Rf8 3. Rcc7 Bb4 4. Qxb4 Qxb4 5.
Rd8+ Rxd8 6. Bb5+ Qxb5 7. Re7 mate, very nice!

Claus-Juergen


  
Date: 09 May 2005 08:50:48
From: Ray Gordon
Subject: Re: Chess problems don't work correctly anymore
>> Now all of a sudden the sacrifice of the rook on d7 begins to look
>> silly as now there are two pieces hanging. Dvoretsky of course doesn't
>> even analyze this possibility in the notes.
>
> When reading older books, you have to live that not all content holds to
> the tactical abilities of modern computer chess engines. When I reread
> Nimzowich's "My System" a while ago armed with Shredder the engine
> discovered several tactical inaccuracies. But that doesn't make the book
> worthless as a whole.

Wanna bet?


--
Ray Gordon, Author
http://www.cybersheet.com/library.html
Four FREE books on how to get laid by beautiful women

http://www.cybersheet.com/chess.html
Free Chess E-book: Train Like A Chess Champion

Don't buy anything from experts who won't debate on a free speech forum.




 
Date: 08 May 2005 21:35:57
From: Ed Gaillard
Subject: Re: Chess problems don't work correctly anymore
In article <[email protected] >,
<[email protected] > wrote:
>[snip] For instance...if any of you have the School of
>Chess Excellence series by k Dvoretsky will know this is high
>quality stuff. However, it's become apparent that many of his
>questions via diagrams don't work when put to the test against
>computers.

I'm sure that his books have an occasional error, but one should be
very careful about thinking you've found one; the odds are against
you.

>Here's a very simple example...taken from School of Chess Excellence
>2: Tactical Play, page 10. Here's the FEN representation of the
>diagram that he puts on that page:
>
>1r2kb1r/1pRb1ppp/4p3/q2pP1B1/p2Q4/P2B4/1P3PPP/2R3K1 w - - 0 1
>
>It's taken from Dvoretsky vs Zilberstein 1966 and it's White to move.
>Now...I figured out the "idea" behind the tactical theme for the
>diagram pretty easily enough... as White's first move SHOULD be
>1.Rxd7! just as he essayed on that page. Unfortunately, when I played
>the above diagram against Junior 9 on my Pentium 4...Junior comes up
>with an incredible defence: 1...f6!! and I'm falling over my chair
>when it chooses this move.

Assuming you've given the position correctly, Black is just busted
after 1.Rxd7 f6 2.exf6, for example 2...Kxd7 3.fxg7 Rg8 4.gxf8Q Rbxf8
5.Bd2, and after Black's Queen runs away (5...Qxd2 losing on the spot
to 6.Bb5+), White has a material advantage, and Black's King is still
stuck in the center and most of his remaining Pawns are weak.

>Now all of a sudden the sacrifice of the rook on d7 begins to look
>silly as now there are two pieces hanging. Dvoretsky of course doesn't
>even analyze this possibility in the notes. Why not?

Because he has the odd notion that students should actually do some
work instead of just passively absorbing material? Because if he
gave every defence in every sequence, the books would be hundreds of
pages longer?

>And that's a good
>question because many of these books don't take the time to analyze
>diagrams enough to justify inclusion in chess books. Clearly this
>simple example shows that if such a diagrammed position can be
>embarrassed by computer analysis means that many of those diagrams you
>have in those very good chess books have to be re-evaluated in light
>of such radical solutions given by the computer.

It's nice to check things yourself, but you really need to follow
tactical sequences out to the end before evaluating the results, nor
will you learn anything by just plugging positions into the computer and
passively watching it spin its wheels.

-ed g.
--
Caissa have mercy on a miserable patzer: http://altergoniff.blogspot.com


 
Date: 08 May 2005 23:21:25
From: Antonio Torrecillas
Subject: Re: Chess problems don't work correctly anymore
En/na [email protected] ha escrit:
> (...) However, it's become apparent that many of his questions
> via diagrams don't work when put to the test against computers.
>
> Here's a very simple example...taken from School of Chess Excellence
> 2: Tactical Play, page 10. Here's the FEN representation of the
> diagram that he puts on that page:
>
> 1r2kb1r/1pRb1ppp/4p3/q2pP1B1/p2Q4/P2B4/1P3PPP/2R3K1 w - - 0 1
>
> It's taken from Dvoretsky vs Zilberstein 1966 and it's White to move.
> Now...I figured out the "idea" behind the tactical theme for the
> diagram pretty easily enough... as White's first move SHOULD be
> 1.Rxd7! just as he essayed on that page. Unfortunately, when I played
> the above diagram against Junior 9 on my Pentium 4...Junior comes up
> with an incredible defence: 1...f6!! and I'm falling over my chair
> when it chooses this move.
>
> Now all of a sudden the sacrifice of the rook on d7 begins to look
> silly as now there are two pieces hanging. Dvoretsky of course doesn't
> even analyze this possibility in the notes. Why not? And that's a good
> question because many of these books don't take the time to analyze
> diagrams enough to justify inclusion in chess books.

In my opinion the difference between GM analysis and Engine analysis is
that Engine consider all moves and GM only good or natural ones.

That mean to play 1.Rxd7! you need to see that after 1...Kxd7! white has
2.Qa7! Qxa7 3.Bb5# once you have seen that you can consider if black has
moves whith more saving chances. Clearly 1... f6 is not one to consider
because white is easily winning no matter it can prolongate the game.

The natural answer to a rook sacrifice is acepting it. Once it has been
refuted you need to consider only moves who need refutation.

AT



 
Date: 08 May 2005 23:25:55
From: Toni Lassila
Subject: Re: Chess problems don't work correctly anymore
On 8 May 2005 14:42:04 -0500, [email protected] wrote:

>Another thing has to mentioned at this point...many of those tactical
>analysis diagrams given for training are based on play from humans
>against humans...meaning errors by the defence aren't given thorough
>study and refutations aren't illustrated in many of these books. The
>authors don't take the time to include such information because it's
>"assumed" (my opinion) because these chess authors think these
>examples are self explanatory in showing why tactical solutions to the
>diagrams are fool proof. As my demonstration shows...this isn't the
>case at all.

If this bothers you, use books written (or edited) after the
introduction of strong chess engines. The moves are then usually
checked by a computer. But perhaps your example of a cook in
Dvoretsky's problem has actually aided your study and not hindered it.
You're certain to remember this motif should you ever have it come up
again.

--
King's Gambit - http://kingsgambit.blogspot.com
Chess problems, tactics, analysis and more.