This is still pretty rough ... HONEY-NN release

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

Moderators: hgm, Dann Corbit, Harvey Williamson

Forum rules
This textbox is used to restore diagrams posted with the [d] tag before the upgrade.
nabildanial
Posts: 126
Joined: Thu Jun 05, 2014 3:29 am
Location: Malaysia

Re: This is still pretty rough ... HONEY-NN release

Post by nabildanial » Fri Jul 24, 2020 10:38 pm

Raphexon wrote:
Fri Jul 24, 2020 10:24 pm
nnnnnnnn wrote:
Fri Jul 24, 2020 10:14 pm
What do you mean by "puzzle"? Study? Direct-mate?

And why would a neural net be good at these? I would have thought that nets would be weaker than pure engines when the solution was purely tactical, as is the case for most compositions, particularly when the winning moves are often ipso facto improbable - underpromotions, tempo losses, and so on.
It finds this sequence for example.

https://en.chessbase.com/post/solution- ... able-study

Normal SF never did.
Do you need to change anything to the config?

fishpov
Posts: 65
Joined: Sat Mar 07, 2015 5:05 pm

Re: This is still pretty rough ... HONEY-NN release

Post by fishpov » Sat Jul 25, 2020 12:03 am

Hi, Is this possible to get build for older cpu ?
Thanks

User avatar
Scally
Posts: 145
Joined: Thu Sep 28, 2017 7:34 pm
Location: Bermondsey, London
Full name: Alan Cooper
Contact:

Re: This is still pretty rough ... HONEY-NN release

Post by Scally » Sat Jul 25, 2020 5:19 pm

Hi all,

Mike helped me with a Raspberry Pi gcc-10 compile, I’ve just managed to compile it with clang-9 too, however it’s about half the speed of gcc-10

Both run well on the RPi 4 under Raspbian OS.


Cheers,

Al.
Alan Cooper
My Chess Computers

Post Reply