Re: [Glob2-ideas] non-flat terrain

[Kai Antweiler]

> Just compute the cost of a hill to be the average cost of going up and going
> down, since ultimately on average all the units will go up the hills on the
> map approximately as many times as they will go down the hills, when you
> look at the big picture. So the individual costs doesn't matter and they can
> be considered equal because ulimately they will balance out.

I think this would be the way to go.

Given the costs that Steph suggested, it would result in a penalty for
using slopes in pathfinding.

If we use symmetrical costs, our gradient algorithm could stay as it is.



Though hight based speeds would be strategically more relevant.
-- 
Kai Antweiler