Nepo17 sent me this through e-mail:
https://www.youtube.com/watch?v=j5iy0VpwbSY&t=56s
Can someone check if it's true? Someone with 128 cores or so.
Mate in #130.
-
massimilianogoi
- Site Admin
- Posts: 393
- Joined: Thu Aug 04, 2022 1:42 pm
- Has thanked: 603 times
- Been thanked: 646 times
- Contact:
Re: Mate in #130.
It works out geometrically, and I think NM Lopez does a pretty good job explaining how it works in the video. Notice if you're looking at it with an engine, once you get to the point where you keep forcing the pattern to move the pawns forward, it will always find Qe1+ as its top choice, even if the evaluations of other moves also read 0.00. The engine is at least recognizing the point that it is the only way to start a continuation that doesn't force either a resetting of the pattern (takes two moves) which would increase the distance to zero (pawn push or capture to reset the 50-move rule) or a draw by perpetual check (like in the case that you choose to take the bishop). Up until move 125, it's just a bunch of forcing Black to avoid checkmate by blocking with the rook and moving pawns while keeping the white queen in position to catch the promotions and avoid getting checkmated themselves. When you've got it down to the last couple of pawns, Crystal will start finding the pattern quickly (nearly instantaneously in my case). Then the ending is just delivering a check to force Black into a zugzwang where the bishop is the only piece that can move (aside from Kc1, which runs into Qe1#), and you just repeat the pattern again, having exhausted all of Black's counterplay, finally capturing the bishop with checkmate (the bishop can move anywhere, but I like the finish where it is captured). Since the person who posted the video didn't include a PGN, I'll at least do that much.