Re: [Bug-gnubg] Doing a rollout as a background job

[Michael Petch]

On 14/06/09 6:39 PM, "Timothy Y. Chow" <address@hidden> wrote:

> Now that I have installed gnubg in my own directory on a shared Linux
> machine, I am wondering if I can do rollouts as background jobs.  That is,
> I would like to be able to prepare an input file with a given position,
> and then do something like
> 
>  /bin/nice -19 gnubg -rolloutoption <inputfile> <outputfile> &
> 
> and come back the next day to examine the <outputfile> for the rollout
> results.  Is there a way to do this?  I'm even having trouble figuring out
> how to input a position using only a line-terminal interface.
> 

This is not my area of expertise but I was able to do this:

Put this in a file rollout1

set player 0 name player1
set player 1 name player2
set gnubgid HN/ACAZiWzAMIw:sAFnAaAAEAAA
hint
cmark move set rollout 1 2
analyse rollout move
save position "rollout1.sgf"

And then did

nice -19 gnubg -t <rollout1 >rollout.log 2>&1


You can of course set all the rollout parameters after the player names. The
save position is a convenient way to save the rollout info into a file
automagically. Rollout.log will be raw gnubg output.

After setting the gnubgid you must do hint. That will allow the cmark
feature to mark the moves you wish to rollout for the current position.

Cmark move set rollout 1 2

Will mark the first two moves for current position.

analyse rollout move

Will analyze the currently marked moves for the position

If the position is a cube (before a roll) then you can analyze with (you
don't need to do hint first):

analyse rollout cube

You can in theory load an entire match, mark all the positions/moves with
cmark and then roll it out all at once.

The -t option prevents the gui from loading.