Saturday, April 02, 2011

Blown away by the idea of Checks, Captures and Threats




















.
.
.
I used to think that CCT of Dan Heisman was some kind of blundercheck. Since I don't drop pieces very often, I didn't consider it to be very usefull. But in search for tasks to automate during calculation I stumbled upon it. In fact looking for CCT is a method for pruning the tree of analysis. In stead of learning to calculate branches very fast, it is much better to know when there is no reason to calculate a branch at all!
The idea behind it is fairly simple. If there is no CCT, there can't be no tactic. If there is no tactic, there is no need to calculate.

I have a tendency to end up in time trouble. I tried to manage that by avoiding complex openings like the King's gambit. The first tests with CCT during serious long games show that I calculate way too much!

Using CCT is not as straightforward as it sounds. So I'm first going to apply it to the >2000-rated tactical problems of Chess Tempo to get the hang of it. Once the thoughtprocess is clear, I will try to automate it. That should kill two birds with one stone: speeding up a sound thought process and pruning branches from the tree of analysis!

And of course continuing with automating tactical vision.

5 comments:

  1. a general question not related to this post. If speed is an important part why not blitz?

    ReplyDelete
  2. From a previous comment on blitz:

    Say in 1 blitzgame you encounter 1 knightfork. If 1 blitzgame = 5 minutes, you can play 264 games. That are 264 repetitions of a knightfork in 22 hours. That is way too little. Besides that there is too much noise in a game, so the brain can't self-organize a strategy for knightforks.

    It's like learning multiplication and starting with 2134 x 318. If you try that and speed up you enter a different game: gambling.

    With gambling on a move, your mind doesn't get correct feedback. So it can't adjust a strategy.

    ReplyDelete
  3. My coach says that you have to look at all the captures even if they look silly. Those that are indeed silly are quickly removed from the list of candidate moves.

    ReplyDelete
  4. I'm guessing using CCT should help analysis. The key thing in common with CCT is gain of initiative and gain of tempo moves usually end up as one of the best candidates.

    I think a great example is the April 3rd Sokolov - Mohr puzzle on Takchess's blog.

    ReplyDelete
  5. Hi Tempo,

    I am interested in your vision on how CCT relates to seeing moves by drilled skills and/or pattern recognition. In my opinion CCT and recognition are complementary but do not function at the same time. CCT demands concious effort while recognition demands (drilled) skills. In my practice I do not rely on CCT immediately, since i have the feeling that this would dehance the benefit of pattern recognition. So first I look to a position trying to spot patterns and only then I do a CCT. Also I think when you do tactical problems using CCT (from the start of looking at the position), your pattern recognition skills are hardly trained. I am very interested in your opinion on these matters.

    ReplyDelete