Greetings,

As has been already noted above, by both yourself and others, you see the basic motifs and moves - the problem is that you're not trying them in different orders. Without meaning to sound facetious, that's the point of "combinations"!  

You need to list all the possible checks and captures (and threats to do so) in the position - then try them in different orders. This needs to be done for both sides to ensure that the opponent doesn't have a "killer" reply. perhaps most importantly, this needs to be done at each move for either player.

As swingdoc points out, you need to practice on complex positions as a means of honing your combining skills.

Peter Griffiths, in his book Better Chess for Club Players, makes a number of interesting comments whilst annotating games played by club players:

Quote:


Quote:


Quote:


In these and other comments he makes throughout the book, it is driven home that if you wish to pursue an attack (whether on the king's position or in general), you must train yourself to calculate accurately - there is no other solution to a successful attack (or defence).

Bearing the above comments in mind, you may like to have another go at both diagrams and give as full an analysis as you can.

Kindest regards,

Dragan Glas

Dink Heckler wrote on 02/18/09 at 23:24:32:


Thanks for your input.

The idea with 2.Nf6+ is to remove the threat of gxh6 by black and so that 3.Rxe8 pins the bishop at the same time as attacking the rook on d2. 

2.Nf6+ gxf6 3.Rxe8 Rxg2+ 4.Kxg2 Qxe8 for example. And 4...Qg4+ 5.Kh2 is just bad.

This is a much more difficult problem for me than just being more careful and taking more time.

Here is another problem which I had very very similar problems with

I found 1.Nxe4 fxe4 2.Bxe4 Ne7 reasonably easy. In this position (again) I saw the idea that if I could move my bishop with tempo I might be able to play Ra8 to pin, attack and take on e8 with my rooks. I tried 3.Bxc6 Nxc6 4.Ra8 and saw that it didn't work. Then I tried something with 3.b5 Rc7 4.Bxb7 Rxb7 5.Ra8 but realised that the knight on e7 was in the way. So I concluded that 1.Nxe4 is ok but not winning strongly. And trust me, I looked at the position after 2...Ne7 for a long time to see if I had missed some other candidate move. My problem however was completely different, I didn't understand that I could try 3.Ra8 emmediately instead which is winning.

Do you see the similarity between those two mistakes, in this post and the one above? I am pulling my hair about this because I know the solution but I just can't calculate for some reason.

I seem to have a lot of problems with my calculation. Let me explain to you what I mean.

This position was shown to me by a friend. I found 1.Qxh6 pretty fast and declared that was my answer. He told me I had to give all the lvariations so I gave it another effort and naturally saw that black could go 1...Kg8. So far easy.At this point however I could not get it to work anymore. I thought for perhaps 5 and saw the ideas with a discovery against the rook on d2 with my queen but I could not get it to work since for example 2.Rxe8 Rxg2+ 3.Kxg2 Qxg4

Now, if I had said to myself that "what is the problem, why doesn't it work? Answer: because he can take my queen with the g-pawn" I might have come to the conclusion that black g-pawn is overloaded and consequently found 2.Nf6+

How can I learn how to think in this manner instead of just relying on my fast but inaccurate pattern recognition? There is no logic in the way that I think and in order to take the next step in my improvement I need to change this. I can not just simply calculate every forcing variation, I need to learn how to ask myself questions about the position.

Before someone suggests me to read Kotov or Aagard, I have already read those books but as you can see, it does not seem to have helped me much.

Any suggestions from own experiences of how you improved at calculation dramatically?