[freeciv-ai] Re: Measure the AI ability to harvest resources

[Gregory Berkolaiko <Gregory.Berkolaiko@xxxxxxxxxxxx>]

On Mon, 25 Nov 2002, Raimar Falke wrote:

> > Threefold increase in production but twofold slump in science!  Can CMA be 
> > adjusted so that both science and production are increased?
> 
> It can. But I don't think that this is a good idea. We are here
> speaking of the first 30 turns. IMHO it is better to produce settlers
> here to found cities. I have no facts to back this up.

Sounds reasonable.  But priorities have to change depending on the 
situation.  And the fact that you _can_ change them effectively is very 
good.

> > > To have a fairer ground for comparing (and also to see how bad/good my
> > > EA is) I now show all the players the map at the start of the game.
> > 
> > What does that mean???
> 
> show_map_to_all is called before the game starts.

Very very strange.  The only place I found where AI depends on knowing 
things is in find_a_direction...

Ah, wait a minute, here it is!
bool city_can_work_tile(struct city *pcity, int city_x, int city_y)

Okay, this is explained.

> > > I haven't searched why the server AIs perform badly. It is quite
> > > possible that they built warriors or barracks. Nevertheless this is an
> > > error. Everybody should built cities the first dozen rounds. And
> > > explore huts (no huts here to be more fair).
> > 
> > AI builds barracks by default, before building any attacking troops.
> 
> This is IMHO bad.

I am testing easy alternatives right now.  First conclusion: it's bad not 
to build barracks at all.

G.