Page 1 of 1

MateFinder 10

Posted: Tue Dec 04, 2018 10:50 pm
by jhellis3
For those who may find interest in such things: https://github.com/jhellis3/Stockfish/r ... /tag/MF_10

Changelog:
Increased max depth from 127 to 191.
Merged upstream SF changes in sync with Stockfish 10.

So not many changes since MF 9.

Re: MateFinder 10

Posted: Tue Dec 04, 2018 11:58 pm
by Dann Corbit
Thanks.
I think it is marvelous.

Re: MateFinder 10

Posted: Wed Dec 05, 2018 3:11 am
by JohnS
Thanks Joe. I like this version of Stockfish.

Re: MateFinder 10

Posted: Wed Dec 05, 2018 10:14 pm
by Jim Logan
Very strange - I cannot get Chessbase 14 or Fritz 16 to recognize the engine once installed. For those who know the interface, it will show up when I go to the engine management screen, but when I try to add it as a kibitzer, the engine is not an available choice - it does not appear in that list. Any suggestions?

This is one of two engines to exhibit this behavior (Mate 2.20 is another - but I can't recall when I installed that). I can add Sting, Stockfish 10, Stockfish Dev versions, etc. without incident. Matefinder 9 had no issues either. I have not tried other GUI's - at least not yet.

Re: MateFinder 10

Posted: Fri Dec 07, 2018 9:58 pm
by jhellis3
The peculiarities of GUIs is not an area of expertise for me, but I seem to remember there being some issue with Chessbase products and engines with certain names.

The first thing I would try is simply renaming the executable file (to something without the word mate in it). If that does not work, one might try renaming the engine in misc.cpp and making a new compile.

Re: MateFinder 10

Posted: Thu Jan 03, 2019 10:32 pm
by Jouni
I tried MateFinder 10 in my standard 400 position mate set. It was 20 positions worse than MateFinder 9. It was also worse than normal Stockfish 10 :o . I tested finding of shortest mate as usual.

Re: MateFinder 10

Posted: Sat Aug 21, 2021 10:52 am
by yurikvelo
Jouni wrote: Thu Jan 03, 2019 10:32 pm I tried MateFinder 10 in my standard 400 position mate set. It was 20 positions worse than MateFinder 9. It was also worse than normal Stockfish 10 :o . I tested finding of shortest mate as usual.
It is expected behaviour.
To find win/mate vs to resolve shortest possible DTM = different tasks.
Less pruning/reductions results in more CPU/time.
In other words for a limited amount of CPU/time it is expected that less win moves will be found.

Re: MateFinder 10

Posted: Sat Aug 21, 2021 11:18 pm
by Uri Blass
I download MateFinder and it does not find the shortest mate.
It find longer mate and later see shorter mates like normal playing engines.