The luxury of the salt mines
Since a few days I'm working on my thought process again. If I look at it, I waste most of my time due to a bad thought process. If I only could get rid of that spill, I would already be as fast as a master at tactics.
From that point of view, improving my board vision is a form of luxury. The usefulness of board vision is limited to the visualization of current and future positions. When I started with the salt mines, I was very well aware of that. The reason that I nevertheless spent a few months with the salt mines, which are supposed to work on board vision, is twofold. First I just want to know how things work, and second Aox has a plausible hypothesis about the improvement of board vision which I like to be proven or invalidated.
We narrowed the bandwidth within which subtasks can be improvable. Only the very chess "atoms" are possibly subject to improvement. Beyond them, subtasks become soon too complex to train salt mine style. Work on this will continue, mostly on the background probably, since it will take a while before we can draw definite conclusions in this area.
Back to the thought process
I'm doing a set of problems at CT, in order to get a series of positions that cost me a lot of time. Once I have a reasonable amount of positions, I will look for possible adaptations of my thought process.
Take for instance the following position.
4q3/2Q1nkP1/2p3r1/3p4/8/P5R1/KPP2P2/8 w - - 1 1
Solution
This position took me 8:48 minutes to solve, so it is ideal for optimizing my thought process. I made a new mind map of my yet to develop thought process:
The question "what are the attacking squares?" must be worked out, so that the square g8 ends up on top. How should I do that?
From that point of view, improving my board vision is a form of luxury. The usefulness of board vision is limited to the visualization of current and future positions. When I started with the salt mines, I was very well aware of that. The reason that I nevertheless spent a few months with the salt mines, which are supposed to work on board vision, is twofold. First I just want to know how things work, and second Aox has a plausible hypothesis about the improvement of board vision which I like to be proven or invalidated.
We narrowed the bandwidth within which subtasks can be improvable. Only the very chess "atoms" are possibly subject to improvement. Beyond them, subtasks become soon too complex to train salt mine style. Work on this will continue, mostly on the background probably, since it will take a while before we can draw definite conclusions in this area.
Back to the thought process
I'm doing a set of problems at CT, in order to get a series of positions that cost me a lot of time. Once I have a reasonable amount of positions, I will look for possible adaptations of my thought process.
Take for instance the following position.
 |
| White to move and win |
Solution
This position took me 8:48 minutes to solve, so it is ideal for optimizing my thought process. I made a new mind map of my yet to develop thought process:
The question "what are the attacking squares?" must be worked out, so that the square g8 ends up on top. How should I do that?
I solved this task in about 40-70 seconds. What I realized at this puzzle?
ReplyDelete1) N is pinned
2) passed pawn is ONE square from promotion
3) R attacks R, Q attacks N
4) Q, N and K defends promotion square
1. RxR! deflection of the King and removing the quard of the promotion square
1... KxR - Now the King threatens to capture the pawn and the N is not pinned anymore (it controls the promotion square).
2. QxN! deflection and removing the quard of the promotion square again! There is no N as it is captured and...
2...QxQ and now 3. g8Q and white is winning.
I tested just very shallow two options: 1) Rf3+ and Qf4+, but I did not find anything special.
I am not sure if it helps, but very often I start checking variations... with a move that captures something.
I am wondering what variations you have checks and looked at. It is HARD to imagine why you lost close to 9 minutes at this specific task. Could you reveal what was the reason? (and what was your thought process).
My TP isn't developed much yet, and even what is developed, is not entirely automated. Meaning that just random trial & error (T&E) plays a big role still. Sometimes I'm lucky and my T&E soon reveals what the position is about, and sometimes I'm not so lucky, like in this position, and dabble around for quite a while before I find the clue by accident. The average of this luck is called "rating". Of course you are invited to find your own "bad luck positions", since those indicate your own weaknesses.
DeleteLooking for targets, attackers and ways to attack the targets is pretty much automated. But when that reveals no clue, I'm kind of lost. I looked for ways to separate the rook from the king, for accidental mates, and for ways to force black to sac a piece for the promoting pawn.
Only when I fully realized the importance of g8, I started to look for ways to undermine it, and then the moves were found very fast.
So I have to find a way to systematically list the important squares, with g8 on top.
My thinking process
ReplyDelete( in real not quite that straight )
Count material : its about equal
What is the puzzle about
1.mate?
2.gaining wood?
3.Promotion?
1.mate?
no, a mate with rook and queen is possible but not likely. There are more defender 3 than attacker 2
2. Winning material
to win material we need not over protected pieces
Count on black queen +1
count on Knight +1
count on rook 1.5
so: no
3)remains promotion
the count on the promotion square is +3
To take at g6 would reduce the count at g8 by 2 and make the knight "critical"
Calculation with the hypothesei pawnpromotion and 1) Rxg6
I see the point. It is useless to decide which weakness to investigate based on the numbers. Since the numbers say nothing. What I need is the numbers after annihilation of the defenders.
Deletenot exactly, my point is that you need to have a hypothesis what the position is about.
DeleteA "strategy" you want to follow or a central weakness you like to play.
A winning taktik can only be : mate, material,promotion and promotion was the only "strategy" left.
Such decisions can be made on the base of numbers ( counts ) , win of material (count on material = 0) and mate (number of attacker>= number of defenders) is not ( easily ) possible so it was necessary to look intensive at the promotion aspect and the count on the squares in front of the pawn.
A correct counting will take into account potential removed defender
See the chuzhakin system ;)
At the end the black queen is overworked
Let's look at the position from two point of view like engine do (or worked in the past)
ReplyDelete1) moves that take pieces
2) moves that give check
Ad.1: RxR, QxN, Qxp
Ad.2: QxN+, g8Q+, Rf3+, Qf4+
This way we have just found the candidate moves (promising ones).
And IF we add the knowledge related to the passed pawn (it is just ONE square from promotion) we can THINK over if the promotion motif can be realized.
Analysis of the moves:
a) QxN+ we have destroyed the defender, but after QxQ the move KxQ - both the Rook and Queen protects the promotion square (and Q defends the R). After Re3+ Kd7 RxQ KxR g8Q+ is stopped because of RxQ
b) g8Q+ and Black has 3 answers: RxQ, KxQ or QxQ. After KxQ the R is defended and the N is unpinned. White did not gain an advantage
c) Rf3+ Kxp and White simply lost the pawn. After Re3 (double attack) Black simply Kf7 and after Rf3+ simply Rf6. It is won position again.
d) Qf4+ Kxp and White simply lost the pawn. After Qe5+ Kg8 Qe6+ (double attack) Black plays Qf7! and Black is fine again (all the exchanges fastens the end of the game)
e) Qxp does not make sense due to double protection (N and Q defends it well enough).
That leads to the conclusion (unless I miss something important): IF White has winning move (the move that preserve the advantage) it has to be the last one - RxR!
Why is that? Why RxR is the best one in the position? It is because it HELPS to realize the goal of pawn promotion. In addition it deflects the K from the last rank. The winning variation was mentioned in my previous post.
Let me know if such analysis (or at least "loud thinking protocol") gives you another view to see the position at fresh (new perspective).
What do you think about such description and explanations? Does it lead to any helpful suggestions or conclusions?
CCT is only a structured T&E method. You look for something, but have no idea what you are looking for. You investigate way too much irrelevant possibilities that way.
DeleteYou may use CCT to explore a position and it is a good method to gain control over your own thinking
DeleteCCT starts with checks first and the we have to calculate the tree as long as it is forcing till Quiescence. Thats a hard job at this example. Personaly i like to have a goal im calculating for.. if that dont work .. i look for an other goal
Did you get my email? I could generate more...
ReplyDeleteNo, I didn't get it. Something must have gone wrong. If you get me a fenarray, I make exercises of it with specific material.
DeleteI think this is where Tp might be shelved and David Bronstein fantasy could get you an answer. What do you want to do ? Trade pieces and get the pawn to promote without being captured on the next move. Begin with this end in mind and shuffle some pieces. Thats how I see it.
ReplyDelete