abulmo / edax-reversi Goto Github PK

Hello,

I wanted to create a bot that would play against people over the internet, but ideally by feeding him the board state and getting the next move, so that many people can play at their own pace. And wrapping the communication over HTTP. I have done this for chess using Stockfish.

Is there any tutorial on how to work with the program, ideally with some examples? After downloading (the linux version), all I get is 'Illegal instruction' and the help is not exactly helpful.

How does the position notation work? How do I tell edax to tell me his next move? Can it also validate move from input?

Thank you for doing this project, it looks awesome!

Line 576 a wrong pointer is tested after a malloc
stack->task should be stack->stack

There is no 7zip file for mac in Assets in the GitHub releases page.
I tried running the linux executable, which gave me an exec format error
On trying to build from the source, I did make release OS=osx COMP=clang ARCH=x64, (I have a 2016 macbook pro 13 inch without touchbar)
which gave me this log:
https://gist.github.com/ishanSrt/db03df64298e09b80a9d96a088174587
Can you help me out with this?

In hash.c line 677, the right test is likely:
if (hash->data.move[1] == move)
rather than
if (hash->data.move[0] == move)

otherwise the else if is useless.

That potential bug was already reported in okuhara/edax-reversi-AVX

And it would be great to know the details about how to generate the evaluation data, thanks!