Micah 1.0

[flok]

Now on the web: Micah 1.0

Get it from: https://vanheusden.com/Micah/

- when the pv starts repeating, stop generating that string (causes crashes in certain uci clients)

- replaced cpw king-shield code by own implementation (from embla)

- proper stalemate detection

- verify null move

- eval distance of N/R/Q to king

This is the "I'm hoping CCRL will test it" release

[tsoj]

nice will this version also play in the lichess tournament?

[Graham Banks]

This is the "I'm hoping CCRL will test it" release

Will play in my Amateur Series Division 9, which starts in 2 weeks time.

[Gabor Szots]

This is the "I'm hoping CCRL will test it" release

I don't think I'll be able to resist such a nice request. But it will have to wait a couple of days so I don't think it will reach this week's update.

EDIT: It does not run, complaining that cyggomp-1.dlll is missing.

BTW, is the tune file required?

[flok]

nice will this version also play in the lichess tournament?

Possibly!

[flok]

This is the "I'm hoping CCRL will test it" release

I don't think I'll be able to resist such a nice request. But it will have to wait a couple of days so I don't think it will reach this week's update.

EDIT: It does not run, complaining that cyggomp-1.dlll is missing.

BTW, is the tune file required?

Oops! Ok I've added the missing dll and re-uploaded the zip.

The tune.dat is not entirely required but it gives +30 elo so...

[flok]

The tune.dat is not entirely required but it gives +30 elo so...

Oh sorry, in this release it is required (defaults are '0').

[thunderstruck]

Hi Folkert,

I/O buffering is still missing in Linux.
The whole PV output is only send after making the move.
In Micah.cpp line 196 I inserted
setvbuf(stdin,NULL,_IONBF,0);
setvbuf(stdout,NULL,_IONBF,0);
This is for C but it also works in C++.

Best regards,

Antonio,