M2M Day 345: I’ve “officially” named this new style of chess
This post is part of Month to Master, a 12-month accelerated learning project. For October, my goal is to defeat world champion Magnus Carlsen at a game of chess.
Today, I started writing a few lines of code, which will be used to generate my chess algorithm. While doing so, I started daydreaming about how this algorithmic approach could change the game of chess if it actually works.
In particular, I tried to answer the question: “If this works, what’s the ideal outcome?”.
There are two parts to this answer…
1. Execution speed
The first part has to do with the execution speed of this method, which I believe can be reduced to less than 10 minutes per evaluation with the appropriate practice.
As a comparable, Alex Mullen, the #1-rated memory athlete in the world, memorized 520 random digits in five minutes. While, one full chess evaluation requires manipulating 816 digits if all the pieces are on the board and 560 digits if half the pieces are on the board.
While these tasks aren’t exactly the same, in my experience, writing numbers to memory takes just about the same times as reading numbers from memory and subsequently adding them.
To be conservative though, I’ve padded Alex’s five minute time, predicting that an expert practitioner at mental chess calculations could execute one chess evaluation in 10 minutes.
At 60 evaluations per game (which is what I predict for myself), an entire game could be completed in 10 hours, which, at least in my eyes, is completely reasonable. Many of the games in this World Chess Championship last for six, seven, or eight hours.
Of course, if a more experienced player can more intelligently implement the algorithm, I suspect she can cut down the number of evaluations per game to around 30, shortening the game to a fairly standard five hours for a classical game.
2. Widespread adoption
I’m not sure if this computational approach to chess will ever catch on as the exclusive strategy of elite players (although, I guess it’s possible), but I can imagine elite players augmenting their games with the occasionally mental algorithmic calculation.
In fact, I wonder if a strong player (2200–2600), who augments her game with this approach, could become a serious threat to the top players in the world.
I’d like to think that, if this works, it’s causes a bit of a stir in the chess world.
It might seem that this approach tarnishes the sanctity of the game, but chess players have been using machine computation for years as part of their training approach. (Chess players use computers to work out the best theoretical lines in certain positions, and then memorize these lines in a rote or intuitive fashion).
Anyway, if this approach to chess does catch on, I might as well give it a catchy name…
At first, I was thinking about something boring like “Human Computational Chess”, but I’m just going to commit to the more self-serving approach:
Introducing Max Chess: A style of chess play that uses a combination of intuitive moves and mental algorithmic computation. These mental computations are known as Max Calculations.
It’s probably cooler if someone else names a particular technique after you, but I think the cool ship has already sailed, so I’m just going to go with it.
Now, assuming Max Chess even works, let’s see if it catches on…
Read the next post. Read the previous post.
