[Freeciv-Dev] Re: for discussion: ics revisited

[Christian Knoke <chrisk@xxxxxxxx>]

On Mon, Mar 18, 2002 at 05:40:39PM +0000, Gregory Berkolaiko wrote:
> On Mon, 18 Mar 2002, Per I. Mathisen wrote:
> 
> > This patch limits the usefulness of _only_ playing with inifite city
> > sprawl (ics) as a player strategy, without imposing artificial limitations
> > like notradesize, citymindist and family.
> > 
> > It doubles research cost, and doubles the effect of all science
> > improvements.
> 
> This should be played with.  Hard to say how good it is from the 
> description.

I once had a thought (yes, really ;) that the tech cost increase
is unreasonable. At the moment we have

cost(n) = 10 * n

So, in the beginning there is a big increase, making it hard to
get new techs in midgame. At the end you have come up with developement
at its fairly easy to gain techs again.

So, why not

cost(n) = 30 + ( 5 * (n ^ m)) with m ~ 1.1 .. 1.3

Just as an example. I think this would change things pretty much,
and I *believe* it would be good for non-ics strategy.

Just my 0.001375 euro

Christian

> 
> > It also changes research cost calculation from civ1/2 standard, to
> > techstyle 1, which makes tech cost dependent on its depth placement in the
> > tech tree. This makes it easier to fill holes in your tech tree later on.
> > Tech leakage is further set to 1 to reduce cost of techs when other
> > players that you have an embassy with already have this tech. This
> > increases the value of embassies and makes it easier to catch up a tech
> > lag. But these changes are secondary.
> 
> Very reasonable changes.
> 
> > The AI at present doesn't understand the increased worth of science
> > improvements, and this would have to be fixed if the above suggested
> > changes should go in.
> 
> This is trivial to fix.  Less trivial is to teach AI to build diplomats 
> (to establish embassies).
> 
> G.

-- 
Christian Knoke     * * *      http://www.enter.de/~c.knoke/
* * * * * * * * *  Ceterum censeo Microsoft esse dividendum.