Strongest chesscomputer of the world (open source)

[NormanRKN]

if someone has a core or few cores free and like testing and maybe playing chess too and want to help the developers to optimize the code. (what ever ;) )
feel free to join the pioneers.

Overview

Fishtest is a distributed task queue for testing chess engines. The main instance for Stockfish is http://tests.stockfishchess.org

It is currently being used for testing changes on Stockfish, using tens of thousands of games, both on Linux and Windows. The following setup describes a step-by-step installation for a machine that will run test matches (a worker).
Setup Python on Windows

On Windows you will need to install Python 2.7.x for x86 (not 3.x series and not 64 bit) from http://www.python.org/ftp/python/2.7.4/python-2.7.4.msi

In case something is not clear please read windows-step-by-step-installation.txt
Setup fishtest

You can download fishtest directly from Github:

https://github.com/glinscott/fishtes...ive/master.zip

or, in case you have a git installation, you can clone it.

git clone https://github.com/glinscott/fishtest.git

Sign up

Create your username/password at http://tests.stockfishchess.org/signup
Launching the worker

To launch the worker open a console window in fishtest/worker directory and run the following command (after changing concurrency to the correct value for your system, see below), providing username and password you've been given.

python worker.py --concurrency 3 username password

or, after the first access, a simple

python worker.py

Option concurrency refers to the number of available cores in your system (not including Hyperthreaded cores!), leaving one core for the OS. For example, on my 4 core machine, I use --concurrency 3.

On Linux, you can use the nohup command to run the worker as a background task.

nohup python worker.py &

Override default make command

Once launched, fishtest will automatically connect to host, download the book, the cutechess-cli game manager and the engine sources that will be compiled according to the type of worker platform. If default make command is not suitable for you, for instance if you need to use some other compiler than gcc/mingw, then you can create a custom_make.txt file in fishtest/worker directory, containing a single line command that fishtest will run to compile the sources.
Running the website

This is only if you wish to run your own testing environment (ie. you are testing changes on another engine). As a pre-requisite, the website needs a mongodb instance. By default it assumes there is one running on localhost. You can set FISHTEST_HOST environment variable to connect to a different host. To launch a development version of the site, open a console window in fishtest/fishtest directory and do:

sudo python setup.py develop
./start.sh

on some machines you need: apt-get install PyQt4.x86_64
atm is only slow phyton but if someone can compile in C/C++ or somewhat else.
the optimium would be GPU-support :D

ready products to use:
http://abrok.eu/stockfish/

http://stockfishchess.org/download/

source: https://github.com/glinscott/fishtes...ster/README.md

Norman

[NormanRKN]

note: phyton is only the start environment not the optimized cores inside.
sorry for false connections ;)

[MooMoo2]

I did my small bit and contributed 10,836 CPU hours (~1.24 CPU years) and 685,030 games.

A total of 2,296.87 CPU years of work and 1,073,229,593 games were played by Fishtest users from 2019-09-28 to 2021-01-03 (463 days), which is an average rate of 4.96 CPU years/day and 2,317,990 games/day.

[Viliam Furik]

Originally Posted by ONeil

Hi here is my stats for the long game.

I don't think this has any use whatsoever to this thread...

[Dr Sardonicus]

Originally Posted by Viliam Furik

I don't think this has any use whatsoever to this thread...

That's good enough for me! Abracadabra, it's been moved to a thread in the poster's blog area