Complete.Org: Mailing Lists: Archives: freeciv-dev: September 2001:
[Freeciv-Dev] Re: Profiles from huge game
Home

[Freeciv-Dev] Re: Profiles from huge game

[Top] [All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index] [Thread Index]
To: Trent Piepho <xyzzy@xxxxxxxxxxxxx>
Cc: Sebastian Bauer <sebauer@xxxxxxxxxxx>, <rf13@xxxxxxxxxxxxxxxxxxxxxx>, freeciv development list <freeciv-dev@xxxxxxxxxxx>
Subject: [Freeciv-Dev] Re: Profiles from huge game
From: Ben Webb <ben@xxxxxxxxxxxxxxxxxxxxxx>
Date: Fri, 7 Sep 2001 18:59:35 +0100 (BST)

On Thu, 6 Sep 2001, Trent Piepho wrote:

> On Thu, 6 Sep 2001, Ben Webb wrote:
> > I suspect the main problem with the client is that all effects
> > for all cities are updated every time a city packet is received. It is
>
> Do this means at the end of the turn, when a city packet for all cities is
> sent, that each city is recalculated each time.

The effects for each city are recalculated each time, yes.

> So you have made an O(n) algorithm into an O(n^2) algorithm?

        No; I have created an O(n^2) algorithm where no such algorithm
existed before. Obviously it would be nicer to make this an O(n)
algorithm, but that will take more time to do.

> I suppose that when a city packet is received the client will update
> each city's graphic and line in the city list too.

Nope. The code only updates the effects. All other behaviour is untouched.

> I can't believe a patch like this went in.

        I guess you should have complained about it at some point between
April and now, then, since the patch has been available for review since
then. As for CVS commit privileges, I have none.

> You might as well just go and announce that freeciv is not a game, it is
> a place for people to try and get their game ideas into someone elses
> code.

        Not my idea. See http://www.freeciv.org/roadmap.html, in
particular "Make rulesets for Improvements more flexible" and "Further
game styles: Civ:TOT, Civ:CTP, SMAC/Civ 3..." There are more than a few
comments in the code stating that this kind of thing is needed, too...

> In fact, why not just create a separate tree for research?

        I believe this has been discussed before, and rejected as too
unwieldy; however we do something similar with FreecivAC at
http://bellatrix.pcl.ox.ac.uk/~stewart/fcindex.html; it is separate from
the main Freeciv tree. The main problem is that such an approach reduces
the number of developers for each project.

        Ben
-- 
ben@xxxxxxxxxxxxxxxxxxxxxx           http://bellatrix.pcl.ox.ac.uk/~ben/
"Power tends to corrupt and absolute power corrupts absolutely."
        - Lord Acton




[Prev in Thread] Current Thread [Next in Thread]