Druss Blog

An account of my attempts to try and improve my chess.

Monday, October 31, 2005

Progressing with 3rd circle

I've now finished level 30s and started the level 40s - I've completed 50 of them, or about 25%.

The 40s are definitely harder than the 30s, but not a lot so. I find myself either solving them quickly, or being stumped. But I felt that with the 30s as well.

Wednesday, October 26, 2005

Thoughts on more difficult problems

Some interesting feedback on my last post. Phorku described level 30 as 'rearranging the pieces' to make the tactics work, and that there aren't really any new ideas. I think this is spot on.

Pale Morning Dun said that the jump between level 20 and level 30 is one of the more dramatic, and that don't rush the early problems. Again, I agree with this.

Both these made me think that following the plan of stopping early and not completing all the difficult problems above 50+ is good. Also, repeating the erroneous problems on levels 10 and 20 until I got them right is worth while. When I first read about these two approaches I was sceptical. Shouldn't tactical training be about solving all the problems in CT Art? I thought this approach of stopping early was wimping out. Now I'm glad I've switched. Yes, it is about solving all of them eventually, but the easier problems feel like the foundation.

Monday, October 24, 2005

Chugging through level 30

I completed 155 out of 221 level 30 problems last week. I did at one stage have hopes of finishing them all, but it was not to be.

It has been difficult transitioning from level 20 to level 30. I had got into the 'got to score 100%' mindset by repeating level 20 until I got them all right, this made trying the level 30s very disappointing.

Still, getting through them more quickly than in my first or second circle.

Monday, October 17, 2005

2 weeks into my 3rd circle - finished level 20

A nice week last week - I finished off the level 20s. Also, I went back over the ones I got wrong using the 'erroneous' filter and redid them until I got them all right. I tried this approach with the 10s and it works well. You get to repeat all the ones you didn't get the first time round ... or in some cases the second or the third time round!

My new plan is to progress this each circle - on circle 2 I repeated level 10 until I got them right. Circle 3 will repeat levels 10 and 20. And so on. Also, I'll progress the maximum level I try, so I stopped at level 40 on circle 2 and I'll aim to stop at level 50 on circle 3. At least that's the new plan this week. I'll see how I get on with this as this circles progress.

It is an interesting exercise to repeat the problems you got wrong - it highlights areas of weakness in your thinking. I think mine fall into two areas:
(1) blindness to certain themes
(2) not enough breadth of analysis.

(1) blindness is easy to spot. Here's a good example (problem #149):

I thought through the problem to this stage (problem #149 a couple of moves later):

But then thought, "So what?". I couldn't see that taking either rook would open up the back rank mate.

This is an example of (2) - not enough breadth of analysis (problem #259):

I kept wanting to play h4 instead of f4. I am not thinking through all the options thoroughly. In fact the only way I managed to finally solve this problem was by learning f4. Am I just learning some of the specific problems off by heart rather than understanding them? Is this bad? Or will I remember the pattern somehow?

There are other examples where I get carried away with quite a complicated line of analysis, only for it to not draw any definite conclusions. My initial thought is that I'm just not seeing deep enough, and there is in fact a solution possibly a couple of moves more ahead. What I'm really doing though is missing an easy 2 or 3 move combination! I can get distracted focusing just on one line and miss a number of other (simpler) ones.

Monday, October 10, 2005

OK - definitely started my third circle

Bit of a strange week last week. For the first 4 days I struggled with level 50 in my second circle. I managed to complete about 5 a day, but it was really hard going and I wasn't really fully understanding the problems.

Then I decided to have a bash at the level 10 ones again, as I detailed in my last post. These were much easier, and I rattled through all 110 in just over an hour.

Spurred on, I tried some level 20s. Over the last couple of days I've completed 146 of these, or about half. So that's over 250 easy problems in the second half of the week vs 19 tricky ones in the first half. That's telling me something!

Much as I want to think I can solve all the problems in CT Art - I can't. I have tried hard to struggle through the more difficult ones, but have realised that it will take ages. Reluctant as I am to give up on level 50 plus, I feel I get more benefit from solving the easier ones.

So my revised plan is much like a number of other knights have already concluded - circle round the easy levels in CT Art. I can always come back to the more tricky ones later, and indeed this might be a better way to solve them. The more difficult problems tend to rely on a more simple theme, but with added complications. If I'm not seeing the simple theme all the time, then I'm never going to really understand the complications.

I'm going to get more benefit understanding all the themes well in levels 10, 20 and 30.

Thursday, October 06, 2005

Had another crack at doing level 10

I've been getting a bit bogged down in the level 50 ones. I occasionally get one right, but it is slow going. However, I am very emotionally attached to finishing them - I feel that I should try to reach the stage where I can solve every puzzle in CT Art, even though this might take a very long time.

Just out of interest, and without really thinking that I was giving up on the 50s, I thought I might have another go at running through the 10s.

They were easy! Some of them I definitely remembered, but even with the ones I didn't the answer just jumped out. I felt really on top of them - that's just another easy knight fork. It felt good solving them, and it is much better practicing success than failure.

I completed all 110 problems in 1 hour 15 minutes and scored 92%. The problem with level 10s is that you lose loads of points as a percentage if you get one wrong - one slip and half the points are gone. So I got 11 wrong, but then tried these ones straight away and got them all right a second time. Except one. Problem 86.

Also, I think I found an alternative, equally valid solution for problem 83. This is the first time I have encountered a minor error in CT Art. This is problem 83:

Ct Art says Qf4 Qxf4= . Isn't Qxg6 Kxg6 equally a stalemate?

Maybe I'll start my 3rd circle after all ...

Wednesday, October 05, 2005

Completed level 40

I have finally finished level 40, which is pleasing. I ended up with a score of 58%. So my scores for the first 4 levels are:
level 10 - 92%
level 20 - 80%
level 30 - 61%
level 40 - 58%

I've fallen below the recommended 60% threshold! So I'm wondering about continuing on, especially given the comments on my last post. But I think I'll have a bit of a go at some of the 50s, until I get frustrated.

Some of the 50s are easy, I solved this one (#832) straight off with no mistakes:

But then had my confidence dented with the following 50 where I didn't have a clue (913). Even now I still have problems remembering the solution.

There is definitely a pattern here - I struggle with the problems with a lot of pieces on the board. I am distracted by lots of sidelines which seem interesting, but don't lead to anything.