In today’s day and age, computers are obviously way better at chess than humans, and even the best players in the world have a hard time holding a candle to engines like Stockfish, etc. However, what I don’t understand is how is it possible to have them in different levels of strength. For example, on Chess.com, there are dozens of bots that you can play against depending on your own level. But for the weakest ones, how does this work? One would think that an engine either knows how to play chess efficiently, or it doesn’t. How can you “dumb down” a computer to the level of an intermediate player or even a beginner? Thanks!
In: 15
We can break bad chess bots down into three categories.
1. Ones that are a regular chess bot that knows how to crush any human, but simply doesn’t. Every bot on chess.com falls into this category. They know the best moves, know how to accurately evaluate. This means they can, instead of selecting the best move, search for moves that lower their evaluation and play one of those. These bots don’t feel authentic and you won’t learn anything playing against them.
2. Bots that are trying to play well, but are poorly coded. Not every chessbot is made by a team of people who know what they’re doing. Every once in a while someone who is learning how to program will try their hand at making a bot from scratch. Or as a fun challenge will code a bot with some kind of restriction. SEE [The Kilobyte’s Gambit](https://vole.wtf/kilobytes-gambit/) These kinds of bots often do play like people, but like beginners. They have limited calculation strength and holes in their fundamentals.
3. The last type are bots designed not to find the best move, but instead to find the move a player from a certain rating is most likely to play. This is probably the most difficult, and rarest, to make. Maia on Lichess is the only example I can think of.
Looks like even chess bots need to be babied just like humans. Welcome to the world of participation trophies, robots!
You know how parents will play games with their kids and make poor moves on purpose so that the kid can win? Yeah…. Most people can’t beat the computer, it’s letting us win because of our infant like intelligence level. The difficulty setting is telling it how many mistakes to make.
There are quite a few ways. You can limit its time to think into milliseconds. You can limit how deep it looks to evaluate. But for the weakest bots such as the ones you will find on chess.com, they intentionally play bad moves until their average playstyle equals a certain rating. Computers calculate many different lines. It’s not too hard to tell the computer to play the 15th best move which could be quite a poor move.
Latest Answers