Hacker News Reader

Comment by rfnslyr | Hacker News Reader

Chess is the one game I've tried dedicating time to yet I feel like I still make 0 progress.

I don't know what you mean by progress (you may already be at a level that is considered quite strong by a non-chess player), and I'm not a brilliant chess player myself, but my personal experience was that I very quickly progressed at the beginning, just by learning a decent selection of game openings. If you know enough of them to a decent depth, you first 5-6 moves can be made very quickly without sacrificing position. If you're playing against someone that hasn't studied openings, you have a good chance of creating an insurmountable lead in those first 5-6 moves, as well as saving time for thinking later in the game.

If you're playing against a good player, at least you won't immediately cede the game in the opening.

The next step that I found helpful was to play against a computer, but to have a real board set up, where I could play through different options to see where they lead. Here you have no time limit, so you can really take your time. I found that if I forced myself to take a minimum of five minutes of doing this, I could regularly beat the computer (in this respect it seems a bit binary - at a given difficulty setting, the computer beats you every time, until one day you skill edges past the computer, and now you beat it every time - there doesn't seem to be a phase where the outcome is unsure).

These two tactics advanced my game to a point where I could win about 50% of the time on blitz-chess sites - the good players would still thrash me, and I would completely thrash novices. Getting better from that point seems more difficult though, and seems to revolve around sinking massive amounts of time into the habit. Increasing your opening library knowledge gets exponentially expensive. Improving your mid-game analysis speed also seems to be very slow (although this seems to be linear, not like the opening library). I stopped developing chess at this point because I realized that I wasn't prepared to spend the time needed to become a competitive chess player - I'm good enough that I can beat most non-regular chess players, and that's good enough for me...

You must have been using a computer from 1960. There is no way that you would ever end up beating a modern chess program for the simple reason that they are now stronger than super GMs.

Key phrase there is "at a given difficulty setting."

I actually found playing against a computer to be extremely frustrating from the perspective of trying to learn & better my game. Computers simply do not play chess in a style that's comparable to human chess players. Human chess players that are far better than I am and will win 90% of games against me still occasionally miss tactics or fail to evaluate a position properly. That's absolutely not true of a computer by default.

Playing long games was fantastic for improving my chess. Most chess teachers also advocate long games as the best way to improve your game. The two ways I found to play long games, were to go to my local chess club and sign up for tournaments and to play in leagues on ICC, the internet chess club.

Chess is as much a psychological battle as it is a technical one. Playing against computers is really a different thing.

Agreed, a computer doesn't have intentions. Playing against a real human player you can disrupt their intentions and/or surprise them. That's obviously not possible with a computer.

Same here. I'm guessing the amount of time you need to practice to see any sort of progress is far greater than what I've put in.

You're right in that there's a mess of openings and endgame theory to know if you want to start competing seriously, but it's also possible to improve your tactical vision - which is the bulk of your practical playing strength - just with daily problem solving. http://chesstempo.com/ is a great site for it (not my site, just one I'm happy to recommend.)

Yes - it's a good site.

Reply on news.ycombinator.com