Author: Don Dailey
Date: 14:06:03 12/24/97
Go up one level in this thread
On December 24, 1997 at 08:23:05, Bela Andrew Evans wrote: >I am curious if anyone in this group has improved his/her play >greatly with the aid of a chess program/computer. If so, what >methods did you use? I have Levy's book on Computer Chess, which >gives some brief tips for using the computer to improve in the >openings, middlegames, and endgames, and I'm wondering if others >have their own techniques. > >One common idea is to play tactical positions against the computer >(play to the computer's strength), and I would like to also ask if >others know of particular positions are type of position that might >be especially helpful. For example, I've heard that certain queen >and knight endings are extremely tactical, to the point where humans >would have a hard time playing a computer -- or perhaps you would >suggest some specially composed 3 queens + 4 knights + 1 rook bonaza, >or some such thing. Any ideas welcome! > >Bela Hi Bela, I used to be tournament player although I never was any better than 1914 USCF. Once I got interested in computer chess my interest in tournament chess waned. But when I did play actively I thought a lot about a related problem, how best to improve. I believe it is possible to benefit from a computer chess program. The first thing I noticed was that the old timers at the club I attended regularly never improved. I wanted to know why because they were constantly exposed to chess, experimented all the time with openings and claimed to study chess occasionally. I eventually figured out that there was no real motivation for them. They simply loved chess as a hobby and this was what mattered to them. I noticed that when I studied chess (which was pretty rare but there were short spurts of intensity) I did not always concentrate that hard and often just casually READ the chess, not study it. So I started "playing games" with my study time to make it much more interesting. I challenged myself. The first thing I did was study books on tactics and drilled myself. I would do a full page (maybe 10 problems or so) and use a stopwatch. The goal was to solve the problems as quickly as possible with the correct key move. I would try to improve on my speed as I went along. This would give me a true performance number and goal which was a key motivating factor. I made up similar games for every type of study. I tried to quantify my progress and set goals that could somehow be measured. It's extremely important that you can see and feel your progress. The key factor I am convinced is hard work and concentration which is difficult to maintain for long periods of time without motivation. And motivation is hard to sustain. But the little "games" I played with study provided a lot of motivation for me. With computer chess programs you have a natural mechanism to improve but like the old timers it will be worthless without motivation and goals. If I wanted to seriously improve my game now, I would start by playing SERIOUS games with the programs. Not just fun games but serious games with no take backs, no cheating, strict clock and keeping a real score sheet just like in tournament play. And no quick restarts because you made a blunder on the 3rd move etc. Set the level as low as necessary to have a reasonable chance to win and keep a record of your results after every game. Play 10 game matches and be determined to win them. As soon as you do, move the level up. After each game, try to learn something about the opening if possible. Prepare for the next time you get the same opening and be determined to get a better game this time. When you win a game, allow yourself to win a second time the same way to reinforce your new knowledge (of how to beat this opponent) and then change the opening if the chess program allows you to change its book. You may find that even after the same opening it's not always easy to beat it the same way. Don't restart the game if you make an early blunder but play on, pretending that the result is crucial and the world title is at stake. Don't resign for the computer just because you won a piece either. Checkmate it. Try to understand the endings you get into and look them up in the books. Can you mate the program with Bishop and Knight? Do you know your endings? Set up endings from books to see if you can win them or draw them when required. Can you win even if the program does not follow the books line? Make everything a puzzle, a task a challenge a goal. I used to beat the Fidelity elite with the same opening over and over again. It was useful playing this out a few times because the computer had more than 1 move it would play at a couple of key points in the game. I learned a lot about the Ruy Lopez excahnge from these games and a couple of times won very similar games against people based on knowledge gained from these games. In another instance the Fidelty challenger 9 would play this non-book move and beat me every time. I suspected that the move was wrong but was completely miffed that I couldn't refute it. I looked hard at it, consulted a master friend of mine and learned why the move was wrong and how to refute it. It turned out a lot of players had trouble with the same move and I actually used it in tournaments (with success) but gave it up as soon as I learned why it was wrong. But the point is I learned something from this experience with my chess computer! I think there are lots of possibilities for you if you use your imagination and understand that real work will be involved, but that you can make it much more interesting. Depending on your current level, I do not believe it will require huge investments of time, just consistancy and intensity. Of course if you were a Master trying to become a Grandmaster this would be a different story! By the way, even though I only made it to about 1900 level, I was on the 10 most improved players list for Virginia at the time and it was a result of applying these study techniques. - Don
This page took 0 seconds to execute
Last modified: Thu, 15 Apr 21 08:11:13 -0700
Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.