Huntsman and Stockfish matetrack

Discussion of anything and everything relating to chess playing software and machines.

Moderator: Ras

Jouni
Posts: 3821
Joined: Wed Mar 08, 2006 8:15 pm
Full name: Jouni Uski

Huntsman and Stockfish matetrack

Post by Jouni »

https://github.com/vondele/matetrack has 6554 mates from Chest database. I gave Huntsman 10 seconds and 6 cores to find mate (not necessarily the fastest). Score was 6294 = 96%. Stunning result. Best playing program SlowChess goes below 96% after 1000 positions, when I aborted.
Jouni
User avatar
Steve Maughan
Posts: 1315
Joined: Wed Mar 08, 2006 8:28 pm
Location: Florida, USA

Re: Huntsman and Stockfish matetrack

Post by Steve Maughan »

Huntsman is indeed impressive. Have you done the same analysis with a 1 million node limit per position?

— Steve
http://www.chessprogramming.net - Juggernaut & Maverick Chess Engine
Jouni
Posts: 3821
Joined: Wed Mar 08, 2006 8:15 pm
Full name: Jouni Uski

Re: Huntsman and Stockfish matetrack

Post by Jouni »

Yes. Score was 4650. But I find fixed nodes testing useless.
Jouni
Werewolf
Posts: 2082
Joined: Thu Sep 18, 2008 10:24 pm
Full name: Carl Bicknell

Re: Huntsman and Stockfish matetrack

Post by Werewolf »

Huntsman is old now....
User avatar
Steve Maughan
Posts: 1315
Joined: Wed Mar 08, 2006 8:28 pm
Location: Florida, USA

Re: Huntsman and Stockfish matetrack

Post by Steve Maughan »

Werewolf wrote: Thu Jan 22, 2026 8:39 am Huntsman is old now....
What is better?
http://www.chessprogramming.net - Juggernaut & Maverick Chess Engine
Jouni
Posts: 3821
Joined: Wed Mar 08, 2006 8:15 pm
Full name: Jouni Uski

Re: Huntsman and Stockfish matetrack

Post by Jouni »

When testing Huntsman I noticed , that using 1 or 6 core gave practically same number of solved positions in all suites! Only 6 core has slighly faster average solving time.
Jouni