Zulip Chat Archive

Stream: general

Topic: chess

Kenny Lau (Jun 11 2019 at 11:56):

I just had a strange idea: can we prove that certain positions in chess are "white to play, checkmate in 3" etc? Maybe this would be a fun project

Keeley Hoek (Jun 11 2019 at 11:56):

Sounds like a job for a tactic!

Keeley Hoek (Jun 11 2019 at 11:57):

(case exhaustion)

Chris Hughes (Jun 11 2019 at 11:57):

I was thinking about this the other day. I wouldn't mind seeing non-brute-forcy proofs of certain endgame positions.

Chris Hughes (Jun 11 2019 at 11:58):

It sounds like it would take a while to formalize the rules of chess. Maybe "chess is a draw" should be a Fabstracts theorem.

Chris Hughes (Jun 11 2019 at 12:06):

Prove that this position is a draw with white to move, and a win for white with black to move. (And assume the board is n squares long so you can't brute force it) Screenshot-2019-06-11-at-13.02.42.png

Kenny Lau (Jun 11 2019 at 12:12):

ah, the direct opposition

Kenny Lau (Jun 11 2019 at 12:13):

looks like if we do this then we should start setting up a game theory library

Chris Hughes (Jun 11 2019 at 12:45):

A doable example might be to prove the winning strategy for this game. https://chessplus.net/the-opposition-game/

Kenny Lau (Jun 11 2019 at 12:54):

wait am I being an idiot, how can you win if there are only the kings left?

Chris Hughes (Jun 11 2019 at 12:55):

The game isn't chess.

Chris Hughes (Jun 11 2019 at 12:55):

The aim is to get your king into the opposite corner.

Last updated: Dec 20 2023 at 11:08 UTC