Surprises in Calculating Variations
The following endgame is chapter six from the famous book, Dvoretsky's Analytical Manual.
This is what Dvoretsky has to say about this position:
The diagrammed position, taken from a game plaed in the English League, has already been published twice on ChessCafe.com, in Tony Miles' column, in 2001. The grandmaster got it from a friend of his, who played White (it is a shame that the names of the players have remained anonymous).
Miles' commentary seemed insufficient to me, so I soon published my own treatment in the Russian - language newspaper, Shakhnatnaya Nedelya. Since then, I have often employed this exercise in workshops, while touching up and filling out my analysis. In 2006, I published a new and considerably wider-ranging version of my article on various chess websites.
For those unfaimiliar with this position, I recommend using it as a training exercise - and ideally, following the approach I have used more than once in my own lessons, thus:
Imagine that you are playing the fame yourself; and that, as white, you have to make your last move before time-constrol, with only three minutes left on the clock. You will not be able to calculate everything. It is more important to sense correctly the ideas in the position, to see a few of the possibilities, and then to make an intuitive decision. Write down the move you would have played, and perhaps a few of the ideas underlying your choice (so as not to froget what they were).
And then, follow the same procedure - but this time, give yourself ten minutes. That is hardly time-pressure. Now you will have time for a bit of calculation. Write down your new choice (it might be the same decision you made earlier, or something different).
And finally, on the third try, add twenty more minutes to study the position. Perhaps you will not needs this much time (or only a part of it) if you selected a quiet continuation, one that does not require, analysis, and you have no intention of second-guessing your choice (I remind you that we are imitating your behaviour in a practical game, under the given conditions). Otherwsie, if you were unable to caclulate fully the complex lines you enter into, now you can accurately support or refute the line that interested you.
Once the results of this training have been fixed either in your memory or on paper, you may begin to acquaint yourself with the analysis itself. Comparing the course of your own thinking (with three minutes to think, and then with the longer time allotment) with the variations and assessments offered here, you can assess the quality of your intuitive guesses, your ability to find all the resources contained in the position (or "candidate-moves"), the depth and accruacy of your calculation, and the rationality of your time-expenditure (whether or not you wasted precious minutes on variations that really did not need to be examined at all), etc.
The main outcome of all this will be either a small step or a giant leap forward in the process of self-understanding, without which any truly effective work on self-realization is unthinkable.
Solution - Part one
Since the solution offered by Dvoretsky is quite lengthy, I have decided to split it in two parts. First, let us take a look at the variation revolving around the idea of the queen capture.
Solution - Part two
Now when we have discarded the lengthy combination analyzed above, let us look at the alternatives on White's first move.