Sunday, May 22, 2005

Pondering. . .

First of course congratulations to Sir Pale Morning Dun, fear of the trouts, who just in a graceful way entered the Hall of Eternal Fame. He will be an example of endurance for youngsters everywhere in the world.
Further I decided to add Bahus to the Hall of Eternal Fame, because he did a DelaMazianlike program and he regretted he didn't join the Knights Errant earlier. We have to be benevolent to late converts.

I have done a lot of thinking and research lately, trying to find out the cause of my debacle last week. It's really an intriguing problem. I have experimented with the microdrills this week, but I'm now pretty sure this will not take the cause away.
Internet provides more good advice than one can handle. . .
For example the advice to manage your time well. Of course it is a good advice, as is the case with most advices.
Though it works against the symptoms, it doesn't take away the cause. And that is what really interest me.

So I have to find out what aspect makes me so slow.
First I thought it were positional patterns which I was lacking. But experimenting with Strategy 2.0 proved that that was not the case. What took me so much time during my games was different.
Then I thought it were the microdrills, or some sort of "extended microdrills" what I needed.
Experimenting with those convinced me that that will not help me either.

Being a former computerprogrammer I always look at a way to put knowledge into an algorithm. I can't help that, it's just second nature. Since a lot of you Knights have an ICT-background, you will know what I mean. With chess-tactics this is an extremely frustrating habit. The fuzzy knowledge of tactics and the complexity of combinations always tend to slip tru your fingers.
"Look to underprotected pieces" is such fuzzy advice that's not gonna help you in practice.
"Look at the characteristics of the position" is another one that drives me mad.

But now I believe I have found something. A starting point where to begin with when evaluating a position. The question you have to ask yourself in any given position, looking for tactics, sounds extreme trivial.
But I assure you, it will help you to focus.
The question is "which enemy piece bears the most responsibilities, has the most important function?"
I tested this question with 63 problems, and in 62 cases it lead me to the solution.
The one where the question did not work was a clearance problem.

Of course this need further testing, but this at least looks promising.
The 63 problems came from Intensive Course Tactics from George Renko and were of mixed themes.

Of course this is not a panacea, but it is a start.
Now I will show you a position, that took me 1,5 hour to solve, in spite of knowing where to look, thanks to the question.
It is a bit embarrassing, especially when you all see the solution within seconds, but what the heck, isn't chess about ego and elo?
It's a clear cut example what sort of problem hamper my play.

To make me humble I have to tell you that I have done this problem already 6 times, the last time was about two months ago. Though I recognized the problem as "this one I have seen before" I couldn't remember the solution at all.
Looking for the piece that bears the most responsibilities is easy, that's the black Queen. Defending f7, Bd6 and Nb6.

White to play.

Can you find it (without cheating)?
After I found the solution I didn't understand why it took me so long.
It's really a very simple problem!
Here you can see why an advice as "work on your timemanagement" is not gonna work for me.
If this simple problems take me so much time I have to work on the cause!

What kind of training is gonna help me to solve this sort of problems faster?
Since I did this problem already 6 times earlier, I am not working in the most efficient way, to say the least. I'm am going to experiment with this.
The approach of CD and King looks promising, so that is what I am gonna try first.
Which means two alternating minicircles of the same problems. One concentrating on understanding of the pattern and the other concentrating at speed.
To be continued. . .


  1. This is an example of my most hated type of tactics problem. There are a few lines that appear they might win in the long run, but only one that wins the fastest. The line I see begins with Rxf7. If the queen takes the rook, then the other rook takes, king takes the rook and then white's queen forks king and knight. If the queen moves away, then black gets into all sorts of trouble following the bishop sac 2. Bxg6! Rxg6 3. Qxh5 with a threat to the back rank as well as a threatened discovered check.

  2. I didn't calculate every line, but 1.Rxf7 stuck out right away. It should be pretty straightforward from there. Either the Queen commits suicide to prevent and eventual mating net with Bxg6 or she runs away.

    Usually with a Rook penetration to the 7th rank, it's goes bad for the other guy.

    Anyway Black goes, he either gets terrible position or material loss.

    I think your idea of looking for overloaded or lynchpin pieces is a good idea, but all the tactical software and books I have lists that as only one of many tactical themes.

    It's hard trying to keep all the themes in mind - and I still having problems recognizing for weak squares.

  3. Tempo, to tell you the truth, when I look at a position, I don't immediately look to see where the underprotected pieces are, or whether the queen has the most responsibility, etc. My tendency is to draw upon the patterns I have in my brain and simply begin calculating. My process is sort of like:

    "Hmmm...I see an open back rank with possibilities of exposing the queen...the castled position is a wreck...." After I do this initially, I get an idea of the theme of the problem. Then, I might look at loose pieces, etc. that can be exploited. My final step is to calculate.

  4. Jim, it's broader than just an overloaded piece. It's about THE most important piece of the enemy. If that piece wasn't there, things go awfully wrong. ANY combination to get that piece off place will do. The 62 problems I told you about were all totally different themes. But in the center of the solutions was ALWAYS the most important piece who played a keyrole. It's a pity the formulation of my central question is so trivial. Because of that it is hard to see the importance of what I have found.

  5. CD,
    Your solution is correct. It took me so long to see the Queenfork and to get the moves in the right order. I guess everybody has his own weak spots. In the past my approach was like you describe. But now I have a clue to take a more systematic approach. Especially in those cases where no particular pattern jumps on you.

  6. Hey TS,

    I really like the settings for your board. What chess program are you using and what are the settings?


  7. PS,
    I use Arena 1.1 (freeware, blue background for the fields and chess merida font.

  8. I know this blindness very well. Usually I look for the wrong thing, such as a mate when there is a material win.

  9. Have to respectfully disagree, Tempo.

    The Queen is not defending f7 - f7 is essentially undefended - unless you are willing to exchange a Queen for a Rook.

    To be honest, other than defending the Bishop, which technically is the only piece attacked at the moment, she is not really doing anything at all.

    In fact, if Black's Queen were off the board, it would still take 6 moves to mate.

    As it is, with the Queen on the board, she only forestalls the mating net by about 3 moves.

    Not sure if that makes her all that important here.

    Not arguing with your technique - it's probably a good question to ask, but I'm doubtful a single question can be the panacea to chess solutions.

    But, if it works for you, then go for it.

    By the way, conversations like these are why having forums are so much easier. I don't have to flip through a bunch of other blog comments to see if you responded to me - the forum would tell me instantly.

  10. Jim,
    The Queen is not defending f7 - f7 is essentially undefended - unless you are willing to exchange a Queen for a Rook.

    2 rooks that is. Yes, normally I would take happily 2 rook for a Queen. The problem lies in the Queenfork after that. Even now Arena says that the best move is Qf7:

    It seems to be difficult to explain my point. Does anybody understand what I am trying to explain?

  11. TS,

    Thanks. I'll check it out.


    We have had a forum since our first meeting but no one uses it. Most haven't even signed up for it yet.


  12. Tempo,

    I agree. 2 rooks for a Queen and then the fork.

    And Qx7 is the best move, I agree.

    I'm not arguing that the question you pose isn't a good one. It is.

    I'm suggesting sometimes it's not the right question to ask. Sometimes there just isn't a most important piece.

    But like I said, no big deal. If it works for you, go in peace. I have no quarrel.

  13. Hey, you get Queen and Pawn for 2 Rooks, that means 9.75 + 1 for 2 x 5 pawn units, so you win 0.75 p.u. even without the queen fork that follows!

  14. Munich:...I didnt find the solution. And I did train a lot forks and distractions.
    I think the reason why I missed it was because I found a promissing different line:

    and I still dont even know what is wrong with that?
    Probably it is wrong, and if I consult a program. But just for the fun of it: can anybody find how black shall defend?
    Many variations after 1.Rxd6 but I give some ideas:
    1.RxBd6 Qxd6
    2.Rf6 Qc5

    or 2...Qc7
    3.d6 Qd7
    4.Qf3 threatening Bc6 winning the bQd7 by pin.
    4...Rc5 Be3

    Well, because of these variations I didnt look further.
    "If you find a good move, put it in your pocket and find a better one" says Dan Heisman.