| |
Main
Date: 02 Jan 2005 23:57:19
From: Martin Euredjian
Subject: Calculation, how?
|
Looking for a tutorial/book/web link that deals with tips, techniques and in
general learning how to calculate moves and strategies. In studying master
games, the one thing that always stands out is the depth to which they can
analyze positions sometimes. I'm wondering what techniques can be used to
learn and hone skills in this regard.
For example, in a king vs. pawn race to promotion, if the king is within the
smallest rectangular area defined by the pawn and the board edges the king
can win the race. There is no need to mentally calculate all possible moves
step-by-step. I'm sure that there are many other tricks out there just like
this one.
Thanks,
-tin
|
|
| |
Date: 05 Jan 2005 19:09:04
From:
Subject: Re: Calculation, how?
|
On Sun, 02 Jan 2005 23:57:19 GMT, "tin Euredjian"
<[email protected] > wrote:
>Looking for a tutorial/book/web link that deals with tips, techniques and in
>general learning how to calculate moves and strategies. In studying master
>games, the one thing that always stands out is the depth to which they can
>analyze positions sometimes. I'm wondering what techniques can be used to
>learn and hone skills in this regard.
>
>For example, in a king vs. pawn race to promotion, if the king is within the
>smallest rectangular area defined by the pawn and the board edges the king
>can win the race. There is no need to mentally calculate all possible moves
>step-by-step. I'm sure that there are many other tricks out there just like
>this one.
>
>Thanks,
>
>-tin
>
I've been at this one for a while and I've concluded the best way to
calculate variations is to study chess puzzles. Get some nice chess
puzzle books like the recently released "365 Ways to Checkmate" by Joe
Gallagher from Gambit Press. Another one is "Excelling at
Combinational Play" by Jacob Aagaard by Everyman Chess that's chock
full of puzzles to solve. A juicy 500 chess puzzles, very
difficult...but helps alot. Yet another is "The Ultimate Chess Puzzle
Book" by John Emms by Gambit Press. This one is a monster...1000 chess
puzzles to solve. All of them from actual games played in top level
competition. And finally I bought this book the other day..."The
ChessCafe Puzzle Book" by Karsten Muller, that has another 500 or so
chess puzzles to solve.
So solving chess puzzles is one way to learn to calculate variations
better. Another way...and this is my way of calculating variations
better. I choose a game played at random from the 2 million games in
my Chessbase database files and randomly select a position that looks
interesting. From this point...I copy and paste the game position into
a chess playing program like Fritz or Junior...or whatever your fancy
and play against the computer from that point. I find this to be yet
another way of testing one's calculation skills. An added bonus using
this method proves even more interesting...I deliberately choose the
side that LOSES the game...so I can play from the loser's side and see
where he/she went wrong. This way I can choose a move that would have
allowed that side a better chance in the game. You'd be surprised at
how many games I've played this way makes me more alert to tactical
possibilities that were'nt found over the board during the actual
games played.
Finally, get Gary Kasparov's books..."My Great Predecessors" Why?
Because those books are chock full of variations you can go over with
a board and pieces. This makes you more aware of how to see the
positions that arise in various opening setups. From this you learn
how to think strategically and this allows you to plan your moves
accordingly...making the task of finding the variations easier to
implement.
Just my two cents.
|
| | |
Date: 07 Jan 2005 06:48:17
From: Martin Euredjian
Subject: Re: Calculation, how?
|
Thanks all for the input. Lots of good ideas. I am going to try and go
through all of the tutorials and analyzed games in Chessmaster with my son
first. There's a ton of material there. He just finished all of the basics
lessons and quizzes an is starting to work on the intermediate section
(first moves). Maybe I'll get a book or two (puzzles sound interesting) to
complement. Anything you'd recommend for a six year-old?
In general, I've watched him go from knowing nothing, to developing nicely
at the opening without getting too distracted with most unsupported threats.
Now I have to get him thinking about what to do after all your major pieces
are developed. He is starting to understand about thinking beyond the next
move and slowly realizing that taking that juicy pawn or exposed queen
doesn't always win a game. It's actually quite neat to see the game through
his eyes.
-tin
|
| | |
Date: 06 Jan 2005 15:49:32
From: Toni Lassila
Subject: Re: Calculation, how?
|
On 5 Jan 2005 19:09:04 -0600, [email protected] wrote:
These are all good suggestions, but might I add another method I've
used sometimes:
Pick a master-level player whose games have lots of tactics and
combinations. I've used Nezmethdinov, Nunn, Alekhine, shall and
others. Then I go through the game move by move, trying to guess what
the next move will be and calculate the variations in my head. Then I
see what the next move was and if I didn't guess it, I try to figure
out where I went wrong and recalculate all the variations. I also try
to spot and identify all tactical themes (pins, forks, discovered
attacks etc.) that lead to the tactics and combinations in the game.
Everything is written down. Once I've completed the game I fire up the
engine and verify my analysis to see where I missed something.
--
King's Gambit - http://kingsgambit.blogspot.com
Chess problems, tactics, analysis and more.
|
| | | |
Date: 06 Jan 2005 12:02:03
From:
Subject: Re: Calculation, how?
|
On Thu, 06 Jan 2005 15:49:32 +0200, Toni Lassila <[email protected] >
wrote:
>On 5 Jan 2005 19:09:04 -0600, [email protected] wrote:
>
>These are all good suggestions, but might I add another method I've
>used sometimes:
>
>Pick a master-level player whose games have lots of tactics and
>combinations. I've used Nezmethdinov, Nunn, Alekhine, shall and
>others. Then I go through the game move by move, trying to guess what
>the next move will be and calculate the variations in my head. Then I
>see what the next move was and if I didn't guess it, I try to figure
>out where I went wrong and recalculate all the variations. I also try
>to spot and identify all tactical themes (pins, forks, discovered
>attacks etc.) that lead to the tactics and combinations in the game.
>Everything is written down. Once I've completed the game I fire up the
>engine and verify my analysis to see where I missed something.
This approach is okay. The problem though is when you do this
method...is that you may not know if your move is truly incorrect if
yoiu didn't guess the master's actual move. After all, chess is a
nearly unlimited game with multiple possible moves and game
continuations.
You mentioned Nezmethdinov, a terrific player to study tactical themes
from! He was truely a great tactical genius. Now, using him as an
example, I take his games...then pick a position from any of his
games...and then play against the computer using that particular
position. The results are very interesting. This method, at least to
me, lets me see very rich tactical possibilities and it allows me to
test my theories on lines I won't normally get against opponents OTB.
There's a new book out by Batsford "The Princess of Chess" which
analyzes the games of Judit Polgar. Another good choice to study
tactics from. But I still think chess puzzles are the best way of
improving tactics and chess thinking altogether.
Final thought...endgame analysis...using Dvoretsky's Endgame Manual is
a real treasure find. This book beats all else on endgame analysis and
is a wonderful companion to use because endgame thinking also improves
tactical vision as well.
|
| |
Date: 04 Jan 2005 01:28:25
From: LSD
Subject: Re: Calculation, how?
|
"tin Euredjian" <[email protected] > wrote in message
news:zT%[email protected]...
> Looking for a tutorial/book/web link that deals with tips, techniques and in
> general learning how to calculate moves and strategies. In studying master
> games, the one thing that always stands out is the depth to which they can
> analyze positions sometimes. I'm wondering what techniques can be used to
> learn and hone skills in this regard.
>
> For example, in a king vs. pawn race to promotion, if the king is within the
> smallest rectangular area defined by the pawn and the board edges the king can
> win the race. There is no need to mentally calculate all possible moves
> step-by-step. I'm sure that there are many other tricks out there just like
> this one.
I guess I'll have to cozy up next to the 'shaved pictures' below, and 'BDSM
picture' above LOL. That's okay thought because I have an answer to contribute,
and I am interested in what others have to say on this topic.
There plenty of endgames that resolve with similar certainty as the King outside
the rectangle which you described (e.g., King and Rook vs. King, King and Queen
vs. King, etc. Then there are more complex endgames like Philidor's position
and the Lucena position, which IMO take it to a different level of understanding
and complexity. The better your command of the endgame, the easier it is to
conclude the middlegame with a position that you can win with certainty.
As for calculating more generally, well I don't think there's any way around the
fact that it takes practise, practise, practise. Study tactical positions,
practise in real games. That's all I can offer.
I am at a stage where I can calculate okay, and but I waste time REcalculating
moves: "Hmmm, was this a good move or wasn't it?" I can't remember what
horrible fate my calculations foresaw, and so I end up wasting time repeating
calculations, but not making forward progress in determining the best move.
So my advice to you, while you are still learning how to calculate: make sure
that you conclude each calculation with a quick mental note about what your
calculation says about that particular candidate move. Rank each move you
calculate in your mind, so when you are done considering all the candidate
moves, you will have an idea of what the best on is to play...hopefully, and
assuming you didn't miscalculate. ;-)
That's my two cents. What does anyone else have to say?
LSD
|
| | |
Date: 03 Jan 2005 23:01:45
From: Ron
Subject: Re: Calculation, how?
|
In article <[email protected] >,
"LSD" <[email protected] > wrote:
> As for calculating more generally, well I don't think there's any way around
> the
> fact that it takes practise, practise, practise. Study tactical positions,
> practise in real games. That's all I can offer.
Solitaire chess and analyzing complex positions on paper (and then
comparing your work to published analysis) is probably the best
practice. You will get better! Push yourself to see how far ahead you
can see.
Of course, sometimes you do this and you miss an easy response on move
two (see the thread "White to move and win" for an example of this
committed by yours truly) which busts your entire analysis, but the more
you do it, the less you'll make that (or any other) kind or error.
-Ron
|
| | | |
Date: 04 Jan 2005 13:31:24
From: Henri Arsenault
Subject: Re: Calculation, how?
|
In article <[email protected] >, Ron
<[email protected] > wrote:
>Solitaire chess and analyzing complex positions on paper (and then
>comparing your work to published analysis) is probably the best
>practice. You will get better! Push yourself to see how far ahead you
>can see.
>
Absolutely!
Tests have shown that grandmasters do not evaluate all possiblee moves,
but they quickly narrow the search to a few candidate moves (using pattern
recognition), then they analyze those candidates thoroughly.
IMHO the best training program to do this is Convekta<s CT-Art 3.0. Start
with the easiest levels, and see how many positions you can solve without
missing anything. Writing down your analysis is recommended by Kotov in
"Think Like a Grandmaster", you will be amazed at how much you miss.
What I like about the above program is that it allows me to see exactly
where my inadequacies are (superficial thinking, playing too fast, "hope
chess", cutting off analysis prematurely when a promising move is found,
not taking into account the opponent<s possible replies, difficulty of
visualizing, and defects of pattern recognition).
I haven t played any long games since I bought the program, whose inflated
ratings think I am a master, but if it did not improve my play, it
certainly improved my knowledge. And although the rating may be inflated,
it DOES keep increasing with time, so I seem to be getting better.
And I have nothing to do with Convekta or CT-Art 3.0, but I can<t
recommend it too highly. I try to spend a bit of time with it every day.
Henri
|
|
