[Freeciv-Dev] Re: Tech research (PR#1462)

[Gregory Berkolaiko <Gregory.Berkolaiko@xxxxxxxxxxxx>]

On Wed, 15 May 2002, Per I. Mathisen wrote:

> On Wed, 15 May 2002, Christian Knoke wrote:
> > Obviously the computation of points required for a new tech
> > has substantially changed. I can't see how it is done now.
> 
> The cost of technology is ((1 + num_parents) * researchcost), where
> num_parents is number of requirements for tech, counted recursively.

Re: tech cost style 1 analysis

obviously 
        ((1 + num_parents) * researchcost)
doesn't work very well.  You need something which grows faster that linear 
with num_parents.
        pow(2, num_parents) * researchcost  
is an overshot, with max number of prereqs around 70...  The square: 
        ((1 + num_parents) * (1 + num_parents) * researchcost)
is better, but still too much.  Maybe instead of num_parents we should use 
max_depth and then square?

> Yours,
> per
> 
> "As Israeli forces pursued militants, civilians
> continued getting in the way and dying as a
> result." -- New York Times, April 21

Did they get the militants though?