Marco's written some template+macro magic code that automates the generation of UCI option in SF to tune anything you want. Checkout the 'tune' branch of SF's git repo.Ferdy wrote:I am looking for uci engines that have more options especially evaluation parameters, options that are of spin type, example.I will try to optimize these engines as an experiment with my tuner.option name PawnStructure Bonus type spin default 30 min 0 max 99
I knew MadChess, and Amyan.
Wanted UCI engines with more options
Moderator: Ras
-
lucasart
- Posts: 3243
- Joined: Mon May 31, 2010 1:29 pm
- Full name: lucasart
Re: Wanted UCI engines with more options
Theory and practice sometimes clash. And when that happens, theory loses. Every single time.
-
Ferdy
- Posts: 4856
- Joined: Sun Aug 10, 2008 3:15 pm
- Location: Philippines
Re: Wanted UCI engines with more options
Thanks I will try that.lucasart wrote:Marco's written some template+macro magic code that automates the generation of UCI option in SF to tune anything you want. Checkout the 'tune' branch of SF's git repo.Ferdy wrote:I am looking for uci engines that have more options especially evaluation parameters, options that are of spin type, example.I will try to optimize these engines as an experiment with my tuner.option name PawnStructure Bonus type spin default 30 min 0 max 99
I knew MadChess, and Amyan.
-
tpetzke
- Posts: 686
- Joined: Thu Mar 03, 2011 4:57 pm
- Location: Germany
Re: Wanted UCI engines with more options
If you switch to file based parameter setting iCE will allow you to run a script file against it
iCE 3.0 v400 x64/popcnt [2015.6.9]
runcmdfile
usage: runCmdFile <filename> [-q]
In my case the script file is then mostly a collection of setweight commands followed by a reset command that resets the engine e.g clears the hashes so that no hash entries with evals from old evaluation parameters mess up search.
But if you look for exposed UCI options then iCE is not suitable.
iCE 3.0 v400 x64/popcnt [2015.6.9]
runcmdfile
usage: runCmdFile <filename> [-q]
In my case the script file is then mostly a collection of setweight commands followed by a reset command that resets the engine e.g clears the hashes so that no hash entries with evals from old evaluation parameters mess up search.
But if you look for exposed UCI options then iCE is not suitable.
-
Ferdy
- Posts: 4856
- Joined: Sun Aug 10, 2008 3:15 pm
- Location: Philippines
Re: Wanted UCI engines with more options
I am looking for exposed UCI options at this time, major eval components would be enough.tpetzke wrote:If you switch to file based parameter setting iCE will allow you to run a script file against it
iCE 3.0 v400 x64/popcnt [2015.6.9]
runcmdfile
usage: runCmdFile <filename> [-q]
In my case the script file is then mostly a collection of setweight commands followed by a reset command that resets the engine e.g clears the hashes so that no hash entries with evals from old evaluation parameters mess up search.
But if you look for exposed UCI options then iCE is not suitable.