Beat Your Computer
I've read a lot lately regarding chess players and their chess software, namely attempts to beat their computers. I decided to do a bit of research and present a very basic atricle on some strategies useful for beginners. As always, I remind everyone that I am simply an average player who enjoys the game and also enjoys writing (so don't look for GM level analysis). Therefore, the concepts I present are very simple in structure. I have stayed away from the mathematical complexities of programming stucture because it would require too much writing and, truth be told, probably put half of you into an intense coma for days after reading it. This is some basic theory I have researched in a nutshell. Enjoy this or complain about it, your choice (for those of you choosing to send semi-angry notes to me, try to use clever analogies and metaphors rather than statements such as "you suck." It doesnt bode well in demonstrating your maturity level!
Because my schedule as a musician ties up a lot of my time, chess computer programs have become a good friends. I approach chess and most things in life the only way I know how: the way one would study music theory or a scientific problem (because of my science background, although applying scientific reasoning to love usually fails). Advances in technology have allowed us to purchase a number of sophisticated programs at a very reasonable price. While the chess software program offers a great training partner, it is no substitute for human interaction. Programs such as Fritz, Rybka and Chessmaster offer a broad range of levels suited for most players. One of the things I've noticed via comments on various chess forums is an alarming lack of common sense from novice players. That's what I'll be covering here.
The most obvious piece of advice is don't play your software program on it's highest setting. Even if you're the luckiest person on the face of the earth, your luck will run out here! Sofware programs come with a number of different playing levels designed to level the playing field. Use those levels. You should probably set the levels a bit lower that your rating. The reason is simple: The computer doesn't get tired and doesn't have a mind apt to wander. Also, getting beat by your computer 26 games in a row can often deflate and numb one's ego a bit. Of course don't set the computer's level on the "I was dropped on my head as a small child" setting because you'll learn nothing (even though you might feel like a minor chess god for perpetual wins!
This is the point where you want to consider your style. It's never too early to start developing a style of play. This style will become important in your later games against the computer. These types of programs are not "knowledge based systems." In other words, computers do not learn from their mistakes. Chess programs have improved over the last 25 or so years but these improvements have more to do with advances in hardware issues such as CPU speeds, system architecture, etc. The improvement is in the computer's ability to analyze large fields of possibilities (running 25,000 possible replies, for example, to your single move). If the computer can analyze your single move in nano seconds and analyze a huge number of responses, what hope do we have of beating the all powerful PC or MAC? To answer this question, keep in mind that I am using relatively average playing levels rather than the highest setting your software has.
The first concept to consider when trying to beat your computer is tactics. This is so simple that it is often overlooked. The computer's ability to calculate positions and possible outcomes of those positions at such a high speed give it a huge advantage from the onset of the game. Therefore, you want to avoid complicated tactics. You need to use tactics that you have a grasp of and tactics in which the outcome is more limited. It's easier for you to play a move with 16 possible outcomes that a move with 1600.
Of course, I am presenting an overview that is extremely simplified. You can further you studies through a number of articles and books related to the subject. With the above thought in mind, is it possible to battle your software in a tactical game? There is hope! Most chess software is not absolutely brilliant when it comes to seeing far into the game. This is called the "horizon effect." Therefore, you want to try to develop a long range strategy. The strength of most basic chess programs is in their ability to analyze the next few moves in your game in great detail. If a program were to analyze your move, say 20 moves in advance, you'd grow old waiting for the computer to move. Here's where you want to capitalize on things like pattern recognition and experience based on human game play. Being human can work to your advantage when playing the computer. However, don't try the rookie mistake of throwing the computer an "off the wall" move in an attempt to throw it's game. Most programmers have worked their design around this trick.
This brings us the the age old rule, keep it simple! Your want to make moves that require you not having to calculate more that three or four moves in advance. It's tough when your past the opening and into the beginning of the middle game where combinations can seem to have endless possibilities. However, you want to stick to this concept until you approach the end game. Most chess programs tend to play their worst in the endgame. The reason for this endgame deficiency has to do with the program's "tablebase." When the computer gets to the point in the game where material has been reduced to three or four men, it goes into it's "tablebase" mode. This is where reduced material positions have been completely pre-analyzed by the computer program when it was first designed and the best moves programmed into it's data base. In other words, you make your move in the end game and the computer replies with a preprogrammed response. If the computer finds your move in it's database, it stops calculating it's move in the normal fashion ( a huge number of moves calculated per second), checks the database response, and makes it's counter move. This means the odds start to adjust more toward you.
While none of this gives you sure fire ways to always beat your computer. It should provide some intellectual food for thought regarding beginner's strategies. I am going to do another blog with game illustrations from a GM and Fritz. I will try to post it at the end of this weekend. I have a big show this weekend and have to go into my studio and practice some guitar solo changes I want to try. Like chess, it's all a matter of practice. If our next CD hits and I get more money, I am retiring, staying home and doing nothing by playing chess, using my new microscope and telescopes, and reading. Yeah, there's misspellings in the above but I'm not an English Major and I hate giving things over to the editor for changes.