[Home] [Groups] - Message: [Prev in Group] [Next in Group]
871: Re: [MUD-Dev] Population container
[Full Header] [Plain Text]
From: clawrenc@cup.hp.com
Newsgroups: nu.kanga.list.mud-dev
Date: Tue, 24 Jun 97 16:40:05 -0700
References: [1]
Organization: Kanga.Nu
In <Pine.SOL.3.94.970619203725.20110E-100000@eduserv1.rug.ac.be>, on
06/19/97
at 08:39 PM, Wout Mertens <Wout.Mertens@rug.ac.be> said:
>What is it? Well, you take a kind of monster that you want to
>populate a zone with. Now make a population model, something like
>Fibonacci's row or so (rabbits!), and update the model every once in
>a while. Now, if a player is about, and he 'spots' one (or more) of
>the population, decrease you population by one (or more) and make the
>appropriate monster appear.
>This way, you can have zillions of flies, musquito's, dwarfs, etc
>etc, you could even describe the critters from the container (some
>dwarfs are lurking about) (tho I think that's a bit farfetched), and
>all with virtually no memory or cpu overhead.
A good idea. Problem:
Bubba does a speed run thru the zone and manages to "see" every
single mosquito (he's exceptionally shard sighted). You container is
empty and you have lots of live, animated mosquito objects.
You also need a way of folding the "dead" (ie unseen) objects back
into the container at some point. The problem is guaranteeing that
that folding is not detectable by players:
> l
You are in a corridor that runs east west.
There is a T-Rexx here.
> w
You are at the end of a corridor that runs east west.
The only exit is to the east.
> e
You are in a corridor that runs east west.
There is a T-Rexx here.
The T-Rexx leaves west.
> wait a really long time.
> w
...no T-Rexx...
--
J C Lawrence Internet: claw@null.net
(Contractor) Internet: coder@ibm.net
---------------(*) Internet: clawrenc@cup.hp.com
...Honorary Member Clan McFUD -- Teamer's Avenging Monolith...