[Freeciv-Dev] Re: (PR#6595) Optimize/reimplement CM
[Top] [All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index] [Thread Index]
On Tue, 21 Oct 2003, Raimar Falke wrote:
> This issues is to track all efforts to optimize the existing CM
> (common/aicore/cm.h) and/or reimplement it. It is not about finding
> the best parameters nor about the best usage of CM.
Unless you call the thing very often during a turn, it seems to me you
might quite often get more than one worker's worth of change, for example
a unit moves in, displacing at least the worker on its final tile if not
those it passes through; and blows away a unit or two of population -
maybe even three units of population if armour can move and still get off
three shots? Maybeeven more units in modpacks that include units that can
get off even more shots than the armour in the default rules can?
However, is it possible that doing one worker at a time can arrive at the
best result? In the example of one citizen wanting a 3/3 tile whereas two
citizens might prefer a 5/0 and a 0/5, possibly if you first add the
second citizen then check whether moving one citizen can improve things
you might find that added the new citizen to either the 5/0 or the 0/5
tile and that having done that it suddenly makes sense to move the 3/3
chap to the other of the tiles that total five?
If so you could do just one citizen at a time but then also check each of
the remainder to see if now tthat you have moved one someone else might
also see an advantage in being moved. Which of course could rapidly get
back to being exponential...
-MarkM-
--
Got a website? Get 10,000+ hits a day FREE...
http://makemoney.knotwork.com/10000hits/
[Prev in Thread] |
Current Thread |
[Next in Thread] |
- [Freeciv-Dev] Re: (PR#6595) Optimize/reimplement CM,
Mark Metson <=
|
|