scorekeepers with if and cond forms in the body

scorekeepers with if and cond forms in the body

[klaus schilling]

Are there any examples for the usage of the if- and cond- forms
in scorekeeper bodies?

Klaus Schilling

Re: scorekeepers with if and cond forms in the body

[Hans Ronne]

>Are there any examples for the usage of the if- and cond- forms
>in scorekeeper bodies?

The only stuff that really works is last-side-wins and last-alliance-wins.
You can define other scorekeepers, but then you will get a message "mplayer
does not understansd the scorekeeper". I guess some of them might still
work in a human player only game.

Hans

Re: scorekeepers with if and cond forms in the body

[klaus schilling]

Hans Ronne writes:
 > >Are there any examples for the usage of the if- and cond- forms
 > >in scorekeeper bodies?
 > 
 > The only stuff that really works is last-side-wins and last-alliance-wins.

so which files does one have to hack 
in order to make other scorekeeper bodies work?

Klaus Schilling

Re: scorekeepers with if and cond forms in the body

[Hans Ronne]

>Hans Ronne writes:
> > >Are there any examples for the usage of the if- and cond- forms
> > >in scorekeeper bodies?
> >
> > The only stuff that really works is last-side-wins and last-alliance-wins.
>
>so which files does one have to hack
>in order to make other scorekeeper bodies work?

You would have to add support for other scorekeepers in determine_subgoals
in mplayer.c. Possibly also hack some of the functions in score.c. Since
other scorekeepers have not been used a lot, I'm not sure how well they
would work, disregarding the AI problem.

Hans

Re: scorekeepers with if and cond forms in the body

[klaus schilling]

Hans Ronne writes:

 > You would have to add support for other scorekeepers in determine_subgoals
 > in mplayer.c.

Is it possible to implement AI scripting support,
such that one may use individual ai scripts for scenarios?
especially for RPG-style scenarioes hardcoded strategies
are not flexible enough, think of the variety of 
scenarioes in Age of Empires II.
So the AI behaviour of mplayers should be completely determined
from AI scripts, with the current last-side-winning support as the default
if no script is supplied? Then no recognition of scorekeepers would
be necessary.

Klaus Schilling

Re: scorekeepers with if and cond forms in the body

[Hans Ronne]

>Hans Ronne writes:
>
> > You would have to add support for other scorekeepers in determine_subgoals
> > in mplayer.c.
>
>Is it possible to implement AI scripting support,
>such that one may use individual ai scripts for scenarios?
>especially for RPG-style scenarioes hardcoded strategies
>are not flexible enough, think of the variety of
>scenarioes in Age of Empires II.
>So the AI behaviour of mplayers should be completely determined
>from AI scripts, with the current last-side-winning support as the default
>if no script is supplied? Then no recognition of scorekeepers would
>be necessary.

Some of the AI behaviour can be modifed on a side-by-side basis by using
different side doctrines. I guess this could be expanded into a more
general scripting support for different scenarios.

Another way to handle different scenarios is game-class specific code,
which is also supported in the AI.  This is how time.g works. However, I
don't think this is the way to go since it makes the AI code very
complicated.

Then there is the option of using completely different AIs for different
sides. This is not very useful right now since we only have two AIs: the
mplayer and its lobotomized brother the iplayer. I've been planning to
write a new AI for quite some while, but there has been to many other
things to do :-/.

Hans

Re: scorekeepers with if and cond forms in the body

[Eric McDonald]

klaus schilling wrote:
> Are there any examples for the usage of the if- and cond- forms
> in scorekeeper bodies?
> 
> Klaus Schilling

I know you asked the question more than a year ago, but here is a better 
answer than the one you got last time....

There are now examples of 'if' forms, now that the scorekeepers have 
been improved.

See:
   http://xconq.sourceforge.net/manuals/xcdesign_37.html#SEC175
and particularly:
   http://xconq.sourceforge.net/manuals/xcdesign_37.html#SEC179

   Regards,
     Eric