Complete.Org: Mailing Lists: Archives: freeciv-ai: May 2003:
[freeciv-ai] Re: patch: don't reevaluate ai govt so often
Home

[freeciv-ai] Re: patch: don't reevaluate ai govt so often

[Top] [All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index] [Thread Index]
To: "Per I. Mathisen" <per@xxxxxxxxxxx>
Cc: Freeciv AI development <freeciv-ai@xxxxxxxxxxx>
Subject: [freeciv-ai] Re: patch: don't reevaluate ai govt so often
From: Gregory Berkolaiko <Gregory.Berkolaiko@xxxxxxxxxxxx>
Date: Mon, 12 May 2003 22:33:27 +0100 (BST)

On Mon, 12 May 2003, Per I. Mathisen wrote:

> On Mon, 12 May 2003, Gregory Berkolaiko wrote:
> > 1. If we don't intend gov_reeval_turns to be variable, we should make it a
> > constant. Even if it is not a constant, we don't need to store it, just
> > decrease govt_reeval to zero rather than increase it.
> 
> I think this variable (and others like it) should not be constants, but
> variables that can be changed to increase AI speed. As you suggested, we
> might want to increase this variable as we get more cities (but setting
> it _to_ the number of cities would be gross!).

why is it gross?? :(
I am rather fond of the idea...

> I also think we should give each AI a different gov_reeval_turns so that
> they don't all do it the same turn.

They most likely will have different number of cities ;)

Anyway,whatever method we use, there is no need to have both the upper 
limit and a counter and then use % on top of that.

> > 2. When the first evaluation is done, AI might have no cities and the
> > results could be quite wrong. So we probably shouldn't have constant
> > reeval time.
> 
> We add some constant WAG values to the want to solve this. Yes, the result
> will be also be a WAG, but it doesn't matter - we're not likely to be able
> to choose another government in the first 10 turns anyway. The amount of
> research we do in those turns is also small. Finally, only after ~10 turns
> can we assume that the has AI managed to establish itself (settled its
> cities) and can properly be evaluated for government.

So these WAGs will only be added if turn is <10 ?

> And yes, it does think 'The Republic' is the best govt even in turn 0, I
> made sure of that :-)

Even in turn 0? ;)
In the game I observed the Monarchy was the favourite most of the time...

> > 4. How about making live_player_iterate macro? ;)
> 
> target_players_iterate(pplayer, aplayer) ?
> 
> Avoids all dead players as well as pplayer == aplayer and all allied
> players. That is the most useful one I think.

The above was a hint that you evaluate governments for dead players as 
well.  And they do have some preferences!

G.



[Prev in Thread] Current Thread [Next in Thread]