[Home] [Groups] - Message: [Prev in Group] [Next in Group]
10101: Re: [MUD-Dev] Blending graphics with text
[Full Header] [Plain Text]
From: J C Lawrence <claw@varesearch.com>
Newsgroups: nu.kanga.list.mud-dev
Date: Tue, 11 May 1999 17:53:42 -0700
References: [1]
Organization: Kanga.Nu
On Wed, 14 Apr 1999 07:32:16 -0400 (EDT)
u1391470 <u1391470@csi.uottawa.ca> wrote:
> Each position in the zone is associated with a room (for flags
> etc), because of memory constraints. Given the number of
> "coordinates" (ie. unique locations a person can move within the
> world), we found it to be too memory intensive for each to have
> its own structure.
The normal solution to this is to farm the structures out to disk
via a database or persistent store. Both ways have advantages. The
savings over having to keep everything in RAM (even with VM and
Linux's VM system ain't too wonderful) can be considerable.
> Of course, by removing all/most text (the more graphics we are
> able to create, the less text we would need), we are "moving away"
> from the whole concept of a mud, regardless of our base.
Is a MUD necessarily text based?
--
J C Lawrence Home: claw@kanga.nu
---------(*) Linux/IA64 - Work: claw@varesearch.com
... Beware of cromagnons wearing chewing gum and palm pilots ...
_______________________________________________
MUD-Dev maillist - MUD-Dev@kanga.nu
http://www.kanga.nu/lists/listinfo/mud-dev