Wednesday, February 16, 2022
A friend introduced me to Wordle, a puzzle game on the New York Times web page. It is a guessing game where you try to guess a five-letter word. Any correct letter will show up in yellow, or green if it is in the correct position. You have six chances to guess the word.
The on-screen keyboard helps you because it shows which letters have succeeded or failed so far. In my second and third guesses, I tried to use new letters so that I could narrow down the answer.
Saturday, February 12, 2022
Funny thing about this game. I played it against a higher-rated player in a tournament in Bloomington around 1978. (This person is still around, and he has gone out of his way to be rude to me, both then and now.)
The game is very simple. My opponent made the mistake of letting me push the D pawn. He resigned after 8 moves.
I was so pleased with myself that I sent the game to the state chess magazine which published it. It was seen by some chess author who put it in a book of short chess games. I don't remember the title of the book. Over a dozen years ago I found the game in the database of games for one of the older versions of Fritz chess. It is not in Fritz 10.
Quite by accident, I discovered that the game is on chessgames.com. However, I had written them a couple of years ago to ask if they would accept games to publish, and they informed me that they only publish games played by chess masters.
Wednesday, February 9, 2022
Practicing with (emulated) old chess computers that I used to love to play with during the '80s in the early '90s.
I inadvertently may have found the Achilles Heel of the Super Constellation. It failed to find the drawing line because this would have involved a perpetual check. Instead, it makes a move that looks dumb.
It is not clear if the computer is smart enough to check for repetition of the position in its tree search because on an 8-bit computer this check would be very costly and slow it down The computer would not normally search deep enough to see a three-fold repetition anyway. Most chess engines today consider any repetition of the position to be a draw.
The Super Constellation only sees the draw after it reaches an 11 ply search. Back when I was using programs like Fritz 4 and Fritz 5 to analyze my chess games, I thought that an 11 ply was pretty deep, maybe equal to a low ranked Grandmaster from what I read. Today Stockfish quickly reaches 30 ply, and I use 35 to 40 ply to analyze my openings.