Matesolving test with SF18

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

Matesolving test with SF18

Post by Jouni »

Code: Select all

Huntsman       421
SlowChess      339
SF18           265
PlentyChess    261
Reckless       257
This is improved test with 470 long mates. No mates from Chest database. No need to find shortest mate.
Jouni
Werewolf
Posts: 2082
Joined: Thu Sep 18, 2008 10:24 pm
Full name: Carl Bicknell

Re: Matesolving test with SF18

Post by Werewolf »

Can you post 17.1 and 17 please? Also SF8 just for interests sake?!
Thank you.
Jouni
Posts: 3821
Joined: Wed Mar 08, 2006 8:15 pm
Full name: Jouni Uski

Re: Matesolving test with SF18

Post by Jouni »

SF 17.1 got 263. I have not tested SF11. It's clearly best Stockfish in solving. Simply because node speed is 3x faster!
Jouni
peter
Posts: 3534
Joined: Sat Feb 16, 2008 7:38 am
Full name: Peter Martan

Re: Matesolving test with SF18

Post by peter »

Jouni wrote: Mon Feb 02, 2026 5:08 pm No need to find shortest mate.
Well, Jouni, but then the results don't tell much more than but which one engine gives any DTM into output than the other ones do at certain positions, no matter how exact it is, isn't it?
I mean, I can say much earlier in many games but any engine does, that one side will checkmate the other one sooner or later, as long as I don't have to prove, how many moves that will take
:)
Peter.
chessica
Posts: 1059
Joined: Thu Aug 11, 2022 11:30 pm
Full name: Esmeralda Pinto

Re: Matesolving test with SF18

Post by chessica »

Jouni wrote: Mon Feb 02, 2026 5:08 pm

Code: Select all

Huntsman       421
SlowChess      339
SF18           265
PlentyChess    261
Reckless       257
This is improved test with 470 long mates. No mates from Chest database. No need to find shortest mate.
Hi Jouni, how did you perform the test?
Sopel
Posts: 397
Joined: Tue Oct 08, 2019 11:39 pm
Full name: Tomasz Sobczyk

Re: Matesolving test with SF18

Post by Sopel »

> No need to find shortest mate.

Do you accept truncated PVs? If so then I might have a pretty good entry for the list.
dangi12012 wrote:No one wants to touch anything you have posted. That proves you now have negative reputations since everyone knows already you are a forum troll.

Maybe you copied your stockfish commits from someone else too?
I will look into that.
Uri Blass
Posts: 11161
Joined: Thu Mar 09, 2006 12:37 am
Location: Tel-Aviv Israel

Re: Matesolving test with SF18

Post by Uri Blass »

peter wrote: Tue Feb 03, 2026 8:59 am
Jouni wrote: Mon Feb 02, 2026 5:08 pm No need to find shortest mate.
Well, Jouni, but then the results don't tell much more than but which one engine gives any DTM into output than the other ones do at certain positions, no matter how exact it is, isn't it?
I mean, I can say much earlier in many games but any engine does, that one side will checkmate the other one sooner or later, as long as I don't have to prove, how many moves that will take
:)
You can claim mate in at most 1000 moves when the engine evaluates the position as at least +40 maybe with no possible mistake because you are always going to mate in at most 1000 moves

I thought iniially to choose 100 moves but I was not sure that there is not some long tablebase win based on DTZ tablebases that is mate in more than 100 moves that the engine may evaluate as more than +40.