Thursday, May 05, 2005

In the sweat of thy face shalt thou eat bread

A few weeks ago I did 6 circles with 1355 problems of Intensive Course Tactics from George Renko.
Because I started to memorize most of the moves instead of seeing the pattern in my mind I decided to postpone the 7th circle. In the meantime I did TCT step3,4 and 5.
Now I'm busy again with the 7th circle of Intensive Course Tactics.
Memorization of the variations decreased with about 90% by the delay.
I make an extra effort now to see the patterns in my mind before I play the solutions.
Allthough the pace isn't low it's far from playing a tempo.
So I think I'm gonna need more circles, maybe 8 or 9. I hope to finish that before the tourney at Witsuntide.

For some reason I have the feeling that I run tru things so easy. Compare this with the heroic battles of Don, Fussy, Pale (seduction!), J'adoube (frustration!) then my efforts look rather meager. I'm grown up with "In the sweat of thy face shalt thou eat bread" so when things aren't difficult then I have the feeling it's no work.
To make things difficult in an artificial manner I decided to couple my graduation at a gain of 168 ratingpoints instead of doing # problems 7 times.
So at least my feeling of heroism is done justice.
Besides I try to look at the total picture.
Cognitive science says that a grandmaster has 50.000 chesspatterns stored in his brainlobes.
Hence I assume that an average player of 1700 rating is about halfway.
So I have another 25.000 patterns to go.
Let's assume that 1000 problems a 2 variations per problem contain 2000 patterns, then some problemsets have to be done yet.
I trust that the sweat comes by itself. . .


  1. I read in Dan Heisman's Novice Nook (the article about de Groot's analysis...sorry I can't remember the number or date of the article) that GM's have about 100,000 patterns, but masters have only 10,000. I need to try the de Root exercise (analyze a position out loud and record yourself, and then compare your analysis to other strong players). That should be interesting...

  2. I have also heard that GM's "have" 100,000 positions.

    By the way, "having", in this context means memorized.

    Memorization is the key - you memorize openings, right? Kasparov is said to have memorized 1000 games.

    There's no harm in memorizing thousands of tactical problems - even if the ones you run across in a game aren't exactly the same as the ones you have memorized, there's a good chance that it will be close enough to the one's you have memorized that you will be able to see the first move and figure out the rest.

    That's what "pattern recognition" is - relating what you know to what you see, even if they aren't exactly the same.

  3. I agree that plain memorization is a drawback in the MDLM method: You recognize a whole position instead of a pattern, and you remember that the right move is a queen sacrifice, before you actually see why it works. My strategy against this is «just not to believe» and «try to forget everything» and to play the move only when I really see the pattern.

  4. Exactly, Mousetrapper! I don't move on to the next puzzle until I understand the solution. I don't care how long it takes, but I must understand it. That sometimes means running the solution by Fritz.

  5. You are right Mouse,
    if I memorized the whole pattern, then everything was ok and I'd followed Jims advise happily.
    But only one move is memorized, without any idea of the why or what else or context. And that is just useless. And that's why I had taken a rest between circle 6 and 7.
    The strange thing is that with TCT I had no problems at all with memorized moves. The difference is that that were more simple problems. So the whole line was easier to see in the mind.

  6. There are definitely CT-Art problems I know just by memorization. Once I see them, I know the move, and I have to admit, some of them I'm not sure why. This is not a good practice, I understand, but sometimes you gotta cut corners when you're trying to bag 300 problems in a day.