Recently I’ve gotten heavily into chess, and while I think it’s a great game, I feel that it has one flaw: if you want to be good, you have to do a lot of rote memorization of openings. The fun of chess comes from setting up tactics and positions, and when I was first starting out, I had a great deal of enjoyment right from move 1, since I was trying to calculate out all the different moves and not leave pieces hanging. Basically I was playing chess right from move 1.
Unfortunately, once you’ve played a few hundred games the game effectively starts later. This is because in any position in chess, there are often only a few valid moves. If that position is at move 20 of the game, it’s up to you to find them. If that position happens at move 5 of the game, tons of players have already gotten there, and you’re expected to know a very strong move in response to whatever your opponent has done. It makes the game quite boring, since I’m just making the moves that other people have told me to make. Even more, both my opponent and I are making very strong moves, which means that it is quite difficult to win or lose games quickly.
This is quite unfortunate, and, as luck would have it, lots of people agree with me. This has lead people to create two variants of chess, shuffle chess, and chess 960 (otherwise known as Fischer Chess). Shuffle chess is where the pieces on the back rand are entirely randomized. This has some problems, namely being that you can’t castle in shuffle chess. Furthermore, not being guaranteed one light and dark squared bishop can change the game far too much, and make it feel not like chess. Bobby Fischer, the former world champion, invented a version of random chess where the pieces on the back row are randomized, however, the players are guaranteed one light square bishop and one dark square bishop. Additionally, each player can castle long or short, as long as there are no pieces in the way.
Frankly I don’t love chess 960, because I feel it still changes the game too much. I feel it’s confusing for newcomers to castle when the king and rooks are totally out of their normal positions. That’s why I decided to make a version where the king and rooks stay in their classical positions. Each player gets one light square and dark square bishop, but aside from that the back row is randomized.
Doing the calculations, it turns out that there are 60 possible variations that can arise from those rules. For those wondering how I got there, there are 5 “available” slots for the moveable pieces to go (2 bishops, 2 knights, and the queen). 5! = 5x4x3x2x1=120. However, half of those positions will be invalid, since they will have two dark square, or two light square bishops. That leaves us with 60 positions, which is still easily enough variation that it destroys any opening preparation, and leaves the players to start playing again from move 1.
This also sidesteps the problem that chess 960 has, where some positions are simply unbalanced, and heavily favour white. In other words, it’s not really chess 960 because you have to throw away a lot of junk starting positions. Classical chess has a very well balanced board, which gives winning chances to both sides. It also gives plenty of tactics to both sides. Chess 60 (my version) gives a true 60 positions, since the limitations (king in the middle, rooks at the end) provide enough structure that (I believe) all 60 positions are valid for high level competitive play. Of course, I can’t really claim that without having a GM take a look at all the positions, and not just that but actually play them, but I believe my point remains.
Since I had better put this programming education to use, I wrote a small (129 line) program in c++ that generates a board, and then prints out that board in text. It took less than 2 hours from the first line of code to the finished product. And at the end, for the ultimate irony: the very first board it printed out, was classical chess.