ShashChess

[amchess]

Nice! But even with GoldDigger option set no progress in test positions after my first tests.

Yes, there is room for other improvements!

[peter]

The main thread (id=0) is always A/B.
The MCTSThreads option specifies the number of threads (except the first main one) for MMCTS.

Hi Andrea!
Numbers of threads of CPU don't matter for this setting at all, correct?
I mean all CPU- threads set at UCI- option "Threads" are used for all MCTS- threads as well as for A-B-threads anyhow, no matter of setting of number of MCTS- threads, right?
Or is there any binding of MTCS- threads to certain CPU- threads?
Thanks again regards

[amchess]

The main thread (id=0) is always A/B.
The MCTSThreads option specifies the number of threads (except the first main one) for MMCTS.

Hi Andrea!
Numbers of threads of CPU don't matter for this setting at all, correct?
I mean all CPU- threads set at UCI- option "Threads" are used for all MCTS- threads as well as for A-B-threads anyhow, no matter of setting of number of MCTS- threads, right?
Or is there any binding of MTCS- threads to certain CPU- threads?
Thanks again regards

MCTSThreads must be less or equal than total Threads.
If greater, all threads except the first one, will be mcts.
Andrea

[amchess]

Fix release:

BrainLearn 16.1
https://github.com/amchess/BrainLearn/releases/tag/16.1

ShashChess 21.1
https://github.com/amchess/ShashChess/releases/tag/21.1

[amchess]

BrainLearn 17
https://github.com/amchess/BrainLearn/releases/tag/17

ShashChess 22
https://github.com/amchess/ShashChess/releases/tag/22
Tests
Match: https://github.com/amchess/ShashChess/wiki/Matches
Suite: https://github.com/amchess/ShashChess/w ... te-results

[peter]

BrainLearn 17
https://github.com/amchess/BrainLearn/releases/tag/17

ShashChess 22
https://github.com/amchess/ShashChess/releases/tag/22
Tests
Match: https://github.com/amchess/ShashChess/wiki/Matches
Suite: https://github.com/amchess/ShashChess/w ... te-results

Thanks for the new version, Andrea!
In an 128- position suite it scores best (of all tested engines, only Crystal is a little better as for Frank Schubert's EloSatTS- Elo having the edge as for time to solution of each position with 94/128) with combination of GoldDigger+Tal (95/128) followed by GoldDigger only and Tal only both 90. 20.1 GoldDigger (best setting then) had 94 too, e.g. CorChess3 230422 solves 92/128 regards

[criko]

Here the versions for Apple M1

BrainLearn 17 https://mega.nz/file/HNMwgToB#Q-U680cHM ... htgrHqPHgg
ShashChess 22 https://mega.nz/file/KIlRjQgK#fPUXEo_ig ... DH-5ACigjY

[amchess]

I have released a beta for BrainLearn and ShashChess experience file management.
Any bug reports or suggestions are welcome.
Enjoy.
https://gitlab.com/amchess/experiencema ... -/tags/1.0

[amchess]

New releases:
ShashChess 23
https://github.com/amchess/ShashChess/releases/tag/23
Matches result:
https://github.com/amchess/ShashChess/wiki/Matches
Hard positions result:
https://github.com/amchess/ShashChess/w ... te-results
Brainlearn 18
https://github.com/amchess/BrainLearn/releases/tag/18

[amchess]

New minor release
ShashChess 23.1
https://github.com/amchess/ShashChess/releases/tag/23.1
Refined Shashin theory, only at the engine bootstrap.
See the Readme file.
Non functional change in match play.