[MUD-Dev] Hard Sci-fi muds was Character evolution

Brandon J. Rickman ashes at pc4.zennet.com
Tue Sep 16 02:40:00 New Zealand Standard Time 1997


On Mon, 8 Sep 97 21:53:33 MST, cg at ami-cg.GraySage.Edmonton.AB.CA
 (Chris Gray) wrote:
>I think we're preaching the same sermon here. My long-term goal for
>MUDs (from even before I'd heard of them) is a very detailed world,
>rendered in 3D. Since I'm not an artist, the idea of trying to draw
>pictures was never an option. So, an engine to produce the views has
>always been the goal.

Ah, the "not an artist".  If one could just qualify the title and
become a "programmer artist".  "Digital artist" is no good, it
conjures up images of richly-detailed renderings from Myst (or Riven,
I suppose).

>My (very hazy) view is this. Start with a terrain generator, seeded
>from the co-ordinates. Build a hierarchical data structure that can
>represent anything you need in your terrain, and in the features that
>you put on it. Store *only* those things that are different from what
>your pseudo-random terrain/scenery/town/building/room/container/item
>generators produce. The key is that these differences can be things
>that have been changed by players, or things that have been changed
>by builders. Thus, a builder is someone who can make deeper changes
>than your typical player can. For example, a builder can force some
>height values in your terrain generator, in order to produce the
>desired interesting features. This can vary in scale from the central
>China plateau to interesting looking bumps in your canyon floor.
>The same for generators of buildings/rooms, etc. Builders wander
>around the raw world, looking for something that interests them, and
>can then customize it, making just enough changes to get the effect
>they want.

I have done some thought experiments with this kind of idea.  Without
decaying the effects of players (as below), user changes will
require a larger and more complicated amount of data as your world
seed becomes more simplified.  It really is an entropy effect: when
the world is just an abstract function it is in a high state of order.
Added details to that world, which from an individual point of view
might seem to help organize the world, just add disorder.  But few
would disagree with me that these details are what make the world
interesting.  If only there was a way to identify which details were
slightly more interesting and get rid of the rest...

>There have been previous posts about this sort of thing, but as usual
>I don't recall what the conclusions were (if any!). Another aspect
>that has been discussed is that you can decay player-done changes >over
>time, so that eventually you can just delete the entries in the
>hierarchical structure for them.

Making things decay is tricky without introducing a weird kind of
physics.  I'm thinking about Wizardry VII where monsters vaporize when
you kill them.  Imagine how disgusting a mud would be if every room
in the newbie area looked like this:

Happy Forestland
You stand in a clearing in the middle of Happy Forestland.  The sun is
shining on the happy green trees.
Your movement is slightly hampered by the rotting corpses of several
thousand dead bunnies.
A fluffy bunny is here.

While this would make a mud quite novel it probably wouldn't be too
appealing to a large number of players, and the type of mud where such
a situation would be most likely to actually occur would probably 
want to have more than a few players.

So should we dismiss corpses as being relatively uninteresting details
(aside from special corpse-related activities (hey!) like looting and
sac'ing)?  I guess it depends on the situation.

_But_ imagine a scenario where for some obscure quest I need to fill
up a sack with rabbit bones.  If wholesale rabbit slaughter is a primary
focus of the newbie area, shouldn't I expect there to be at least enough
rabbit bones in the Happy Forestland to fill a sack?  Even if all the 
corpses had already decayed "to dust"?  

(Somehow I have gotten obsessed with the specific case of corpses as
opposed to the more general case of decaying the effects of players 
upon the world.  But perhaps the answer is hidden in the details?)

So corpses aren't as interesting as the components that corpses are
made of.  Those components are then just probabilistic combinations
of sub-components, and so on down to some quantum mud level.  If
we can easily generate and modify the, ah, probability curves for
various sub-component levels...

Are there rabbit bones in Happy Forestland?  First, check for a
rabbit corpse.  Next check the probability that there was a rabbit
corpse there in the recent past.  Then check the probability that
there was any kind of corpse there in the recent past modified by
the probabilty that rabbits were recently living in the area.
Finally check the improbability that the components of rabbit bones
(calcium, etc) exist in sufficient quantity in the area to justify
their spontaneous generation.

Some justification:

Check for a rabbit corpse: The direct benefit of persistence in
the world - observed quantities don't change.

Check the prob. of a rabbit corpse: The illusion, or simulation, of
persistence in the world.

Check the prob. of a corpse X rabbits: The illusion of scale - the
world seems to be impossibly large and complex.

Check for spontaneous generation: The illusion of history - unknown
actions in the past take effect in the present.

I will leave off for now and see if this makes any sense to me in
the morning.

- Brandon Rickman - ashes at zennet.com -
While I have never previously found a need for a .sig, this
may be considered one for the purposes of this list



More information about the MUD-Dev mailing list