Friday, August 31, 2007

level 20, rep 3

I finished my 3rd rep of level 20 of CT-Art. I did this level at 95% -- hardly different than rep 2. Disappointing. I'll get 'em next time! By the way, level 20 has a handful of problems that appear twice in the level, quite odd for a level that has far more problems than any other level.

In positive news. I have located the chess club here in my new town. There is a decent size group of players who meet at a coffee house on Saturday afternoons. There are some good players there too. I got some good games last week -- and a couple good whoopings. It's great to have a place to play live chess, invigorating for the chess soul.

Unrelated to my personal chess improvement is the start of the US Chess League. The team from my old home town, Carolina Cobras, are off to a rough start. They have started blogging and as soon as they settle on the name of their site, I'll have a link to it in the sidebar.

CT-Art rep table:
| Level | Rep 1 | Rep 2 | Rep 3 |
| 10 | 97% | 99% | 99% |
| 20 | 92% | 94% | 95% |
| 30 | 86% | 86% | ?? |
| 40 | 77% | 77% | ?? |
See the last few posts for groaning about my lack of improvement and what I'm doing differently to counter act that problem.


Tuesday, August 28, 2007

CT-Art 267

It seems every time I go through the problems of CT-Art I come across new issues with the problem set. Problem 267 is a good problem, but happens to have a second solution to the one given.
Black to Move

The CT-Art solution is e2 gxf3 h1=Q+ Kxh1 exd1=Q+. Unfortunately I calculated the line h1=Q+ Kxh1 e2 where the knight can't stop the promotion. Here crafy plays Nf2 and black gets a queen vs. knight where it's easy to mop up the pawns and win. In fact, in the CT-Art line, white can throw in Nf2 and have essentially the same ending.

Maybe I'll play the "correct" line next time through.

Labels: , ,

Monday, August 27, 2007

New schedule, starting rep 3

I am not happy with my progress in tactical training. I spent about four and a half months doing difficult tactical problems (levels 40 - 90+ on CT-Art). I did the problems with effort and understood the solution. When I went back and re-solved levels 10 through 40, there was no significant improvement in my solving rate.

A lot of people have thought about and written about different approaches to the repetition of tactical excercizes. One thing I'm probably doing wrong is too much time between reps. Lots of other people have made study programs to better allow for repetition to help in learning. I'm hoping that will aid my success.

It took me about 3 weeks to do the second rep of levels 10 through 40. So I have created a new schedule for myself. I'll start at level 10 every 3 weeks. Hopefully I get more problems done every time I go through.

So I've re-re-done the first 250 problems. Once again I missed one problem on level 10 -- a different one this time, I don't know if that's bad or good. Through the first 140 problems of level 20, I'm at 97%, slightly better than the 95% last time through. The real test will be in levels 30 and 40.


Saturday, August 25, 2007

Level 40, redone

I've finished my second rep of level 40. As with previous levels, there is no significant change in performance. The bottom line is that I'm no better at tactics today than I was a year ago.

My plan is to go back to level 10 of CT-Art and try to really learn the ideas of the problems. I find that when I miss problems it's not because I can't calculate or visualize, it's because I don't see the idea, or can't come up with the right idea.

One of the benefits of the chess blogging community is that I've had a chance to read a lot of other people's blogs. I'm not the first one to struggle with this. Other people have written about how they deal with improving at tactics, so I've got some ideas of how to work on this.

CT-Art level 40 stats:
Rep 1: 77%
Rep 2: 78%


Tuesday, August 14, 2007

Level 40

I'm working through level 40 on CT-Art.

There are a lot of subtle problems on this level and some of them have alternate solutions. For example, problem 707. The diagram comes a few moves in:

White to move

The solution given is Rb8. The simple threat is c7, the rook on b8 allows Nxc7 to be met by Bxc7 simultaneously defending the b8 rook. Black can also try Nxd8 c7 Ne6 Rxe8 Nxc7 and white can easily pick up the a pawn to be ahead an exchange and a pawn.

A fine way to win, but white can also play c7 immediately from the diagram. CT-Art gives c7 a "?" due to the response Nxc7! (The "!" is their's as well). These are odd markings because Rxc7+ Ke6 (to trap the bishop) Bxg5 hxg5 Ra7 leaves white up three clear pawns in the endgame. Despite what people say about rook endings being drawn, I had no problem converting this against the built in crafy engine.

Labels: , ,

Monday, August 13, 2007

How did I get here?

This is fairly off topic for me as this blog is usually just about my chess study or chess games. But Blue Devil had a recent post that made me think a little bit this afternoon. I was about to leave him a second long comment when I figured this would come out smoother as a post on my own blog. His post was about what one can do to increase traffic to one's blog. Though he says himself that he's not that concerned with getting traffic to his blog and doesn't think other good bloggers are either. In any case, it started to make me wonder how I came to visit blogs and start a blog of my own.

By this point I am so used to reading chess blogs, I don't recall how I found them in the first place. I'm sure that I googled something like "chess improvement" or "chess study plan" or something like that. One of the first blogs that I ever read was Sancho Pawnza and sidebar links opened up a wealth of new territory.

Certainly people who read blogs will find new ones by following sidebar links and going to the blogs of commenters, but what about people who have yet to come in contact with chess blogs? How do they get here?

So just now I googled "chess improvement" and here are the blogs that come up:
7. Takchess
9. DK-transform
11. Temposchlucker
13. Fussy Lizard
14. The patzer's tale
16. Cratercat
17. Rook Van Winkle
etc. (Chess Confessions comes up 88, and I didn't have the patience to find The Back Rank)

Googling "chess study plan" turns up fewer blogs. So I am starting to feel at a loss for how I got on the chess blogging path. I also wonder how other people are doing it these days. What makes the chess blogosphere stand out of the background to your average web surfing chess geek? Maybe blogging in general is more popular than I realize and people are actually finding chess amongst the blogs rather than the other way around.

In any case, I've been enjoying chess blogs for over a year and I hope it continues.

Saturday, August 11, 2007

Level 30, no improvement

I'm plowing through my second rep of level 30 and if my form holds I will not improve at all on my first rep percentage. Given that my 'improvement' on levels 10 and 20 was pretty small, I'm inclined to think I'm not quite going about this the right way.

My current idea is that my first time through these problems was too long ago. It was 9-10 months ago that I was doing level 30 problems. In the DLM plan, the first rep takes 2 months and the second rep 1 month, etc. So the problems get repeated twice in the first 3 months. I don't know if this is an ideal timescale, but it has worked for other people. In any case, the long break that I had since last doing these problems was too long.

So I have two changes to make. One, I need to complete a program of learning these problems in one shot, starting and stopping and starting again didn't work. Two, if I don't get to the end of the problem set at the end of 2 months, I go back to the beginning no matter how far I've gotten.

CT-Art results for level 30:
Rep 1: 86%
Rep 2: 86% (162 done from 221)

Update: I've finished level 30 at 84%. Looks like I have a lot of work to do if I want to improve my percentages.


Tuesday, August 07, 2007

Level 20 done (rep 2)

I finished the problems of level 20 this morning. The percentage correct was only a little better than the first rep -- I don't know what this means. I did do this level in significantly fewer days, but this is mostly a function of having the time to do the problems and not working on anything else chess related.

CT-Art stats:
| Level | Rep 1 | Rep 2 |
| 10 | 97% | 99% |
| 20 | 92% | 94% |
CT-Art elo: 2122

Thanks to Blue Devil for pointing me to who has done a massive CT-Art set all in one day (in the comments of previous post). Even through the end of level 60 in a single day is quite impressive. Adding the last 3 levels would probably extend the time by 2-3 hours.

In the past I used to wonder if I can make it to expert or master level. These are somewhat abstract questions when we don't exactly what it means to play at that level. But something like doing a set of tactical problems is more concrete and hopefully attainable goal.


Sunday, August 05, 2007

Back in the saddle

I've recently moved and start my new job tomorrow. In the meantime I've gotten back in the CT-Art habit. I'm starting back at the beginning and doing a second repitition of the problems. In the last two days I've re-done the first 250 problems:

CT-Art stats:
| Level | Rep #1 | Rep #2 |
| 10 | 97% | 99% |
| 20 | 92% | 95% |
CT-Art elo: 2050

On level 10 I made a mistake on only 1 problem -- perfection will have to wait 'til the next rep. I'm through 140 of the 286 problems of level 20 with minor improvement in the percentage.

Knowing how hard these problems get in later levels, I'm still in awe that De La Maza did the whole problem set in a single day. In his article he says to gather 1000 problems -- does anyone know if he used the first 1000 problems of CT-Art or all 1209? Maybe someone who has read the book knows.

Edit: A quick trip to the book store reveals that De La Maza did all 1209 problems from CT-Art. It is an amazing feat to do them all in a single day and twice in three days. I'd be interested to know if there are other people who have done that.