[MUD-Dev] Re: MUD-Dev digest, Vol 1 #142 - 4 msgs

Matthew Mihaly diablo at best.com
Tue Aug 17 20:14:31 New Zealand Standard Time 1999


On Tue, 17 Aug 1999, Par Winzell wrote:

> Matthew Mihaly writes:
> 
> As for the rest...
> 
> I don't -want- to investigate the bleeding edge of cellular automata,
> and the urge is -not- to mimic realistic ecologies. Would you consider
> SimCity self-sustaining? Where do the people come from? They appear as
> inevitable results of your town being a desirable place to live. That's
> a way to create an ecology too -- toss out breeding rates and predators
> and just assume nature will find a way [tm] to fill any empty niches.

Right, I fully agree that investigating the bleeding edge is not the way
to go with games. I guess my point was that doing something like SimCity
or whatnot in one player, and doing it massively multiplayer are very
different things. People make things difficult.

> 
> I picked the example of Gandalf precisely because it's about as far as
> you can get from bleeding edge, infact it has nothing to do with scalar
> fields or automata at all and so in your words cannot be interesting --
> yet it made the game so much more fun.

Right, that's fair enough. Perhaps it illustrates a way to do things too.
Maybe you start by making two simple, disparate background elements
collide (not necessarily 'physically'). Add a few more rules and you start
to get some pretty unpredictable interactions. Not a 'real' virtual
ecology, but unpreditable enough to be useful. 

> 
> Simulation can mean a lot of different things. If you want to be able to
> gaze at your ecology with eternal fascination, then perhaps you do need
> to blend just the right amounts of diffusion, propagation, sources and
> interactions and I'm sure that is a terrible complex field of science
> and I would perhaps go so far as to label this one of the pipe-dreams
> I want to avoid. However, to populate the wilderness with a sensible mix
> of critters that vary in concentration depending on where you are, that
> can be pretty simple and yet be called a simulation.

Yeah, I agree, but it seems that what you are doing then is basically
doing what Dr. Cat says he does. You're making the end, not the means, the
important bit, unless I misunderstand your example of populating the
wilderness. It sounds like you are putting the ends first, ie your goal is
to populate the wilderness so that things vary in concentration. You don't
need any sort of virtual ecology for that. Just have a limited number of
deer or whatever, spit out random rooms with whatever criteria for where
they go whenever it is you want them to appear, and then make there be a
chance of another deer appearing in a room adjacent to the room where a
deer appeared. Then do the same to the new deer room, and so on. You'll
get varying concentrations of deer that never concentrate in the same
place. But that has nothing to do with making a virtual ecology really.
It's chaos, which is just as uninteresting (in the sense that I use the
word) as order. I agree that it still might be fun, and frankly, I'm much
more likely to do something in the above way, than trying to create
cellular rules for the deer, hoping they work out correctly. I'd still
prefer to see something along the lines of virtual ecology, but I don't
think it can be done right now masively multiplayer (at least not without
a lot of constraints on it).
-


> 
> 
> All this said, though, I don't expect any of this to be trivial and I'm
> looking forward with great anticipation to digging into the details of
> it. If I sounded terribly arrogant I apologize. I'm not underestimating
> the problem, I think; I expect it to be tricky... but doable.

Well, it'd be very interesting, but I think if you let the players mess
around with it too much (and that's sort of the point of having things),
then they are going to push it to chaos or order, because people just tend
to seem to do that.
--matt




_______________________________________________
MUD-Dev maillist  -  MUD-Dev at kanga.nu
http://www.kanga.nu/lists/listinfo/mud-dev




More information about the MUD-Dev mailing list