[MUD-Dev] non-'realistic' spatial models

Nathan F. Yospe yospe at kanga.nu
Fri Dec 21 23:55:21 New Zealand Daylight Time 2001


"Nicholas E. Walker" <new at gnu.org> said:

> There is a lot of discussion on rooms vs. coordinate systems and
> related topics.  This focuses on how best to model our reality in a
> game.  I am interested in different types of spaces.  I am less
> interested in 'fake reality' than I am in intuitive and interesting
> spaces to explore.

> As an open question to the list: has anybody done anything that does
> not attempt to be 'real'?  Surreal, abstract, fractal spaces, and
> anything else not 'real' qualifies.

> Isn't it more interesting to experience a type of space that we
> cannot experience in daily life, instead of getting frustrated with
> the limits on how we can experience the fake reality of the MUD?
> Maybe I'm looking for a MUD that accepts the fact that it is a MUD,
> a computer program--instead of some crude model of reality.

The difference between SF for scientists/philosophers/artists and
the drek they write for the general public is a good start on the
answer to this.  Books, not movies, here.  Look at the Star Trek,
or whatever other mass market, stuff, and contrast it against the
likes of Asimov's "Gods Themselves", Forward's "Dragon's Egg", or
any Greg Egan novel...

Actually, to be fair, the contrast holds true for the above stuff
against more literary fiction as well.  The stuff that really has
a penchant for stretching reality out of bounds is appealing to a
very small minority.  Of which I, and presumably you, and others,
maybe a good portion of the people on the list, are members.  The
extent that people remain comfortable with such things in a video
game generally limits itself to topological impossibilities, most
of which are restricted to offstage effects (a looped boundary on
a zone is a perfect example of this) and exageration of physical,
not to mention biological, capacities.

Not that surrealism isn't appreciated by many people... try doing
a setting like "What Dreams May Come" and you'll have fans.  Even
a setting like some of the realities of Gabriel García Márquez as
text might work.  (I don't know how you would do it as graphical)

BTW, if you've never heard of Marquez, he's a South American with
a unique surrealistic style (preserved with variable success when
translated... but a few of the short stories that friends who can
read it in the original have noted as perfectly translated are of
sufficient stimulative quality to motivate me to learn to read in
Spanish, as Giovanni Guareschi did with Itallian)

Fractal Space isn't really that meaningful a phrase.  Fractal map
imagery in astronomy (e.g. "Hubble Fractal Space Images") is just
a way of mapping particulate distribution as a function of scale,
which isn't exactly a world.  If you mean a universe where you're
mapped to the scale of your current coordinates, then this I will
lay first claim to.  It's not fractal space, per se, but the base
algorithms most certainly are a fractal mapping.  Not that a real
world simulation server would be mapping players to the size node
bounderies below their physical size, but conciousnesses could do
so with ease, and I could certainly create a server with this set
of parameters defined in the rules mapping.

I don't model the physics of my game at the core.  I host them as
algorithmic components.  Bindings exist on objects.  There's very
little involved in adding or modifying a force or binding or even
a dimensional framework on a server level.  Heck, there's even an
override on locality, where you could have a different law at the
macro/micro/nano/quantum scales (which you would, if I modeled at
that small a scale) or an entire planet where the laws of physics
were different.

Actually, for computational simplicity, planets are modeled as an
invariant plane mapped to a sphere, with uniform gravity.  When a
player is not in the locality of a uniform source, gravity uses a
more rigorous model.  Boundary transitions are handled without an
undue amount of pain... the physics of an object are inherited at
its local boundary, unless overridden.  I suppose a ship with the
local model of a spinning tube for gravity that entered a uniform
field region (planet) would want to stop simulating gravity after
it stops spinning; take off the gravity?  Make sure the model was
designed to factor in angular velocity, of course, but no need to
compute a zero value modifier on gravity...

So I suppose I could have an entire non-reality model in some VR,
or dream, universe with no common boundaries with the real world.
The player would map control to a different character, and so on.

I'm vaguely suspicious that I've done this before, with VR.  That
bothers me, because it's faded to a vague suspicion, not even the
murky memories I'm usually plagued with.

Fractals don't correspond to a physical model well.  They are the
graphical representation (or numerical) of an analysis of mapping
surfaces (or boundaries in any dimension or mathematical form) to
a nondiscreet "dimensional" model.  There are math books that you
can look at for a more lucid explanation.

In a way, puzzle games are alternate physical models presented as
symbolic graphics.  If you treat any self-consistant mathematical
model (tautology?) as a "physical model", you're simply left with
the question of how to present it.  Of course, interactions are a
problem in a mudlike game.  What interaction does a player (and a
character, perhaps a more important question) have with the game?
What interaction does a player have with another player?  Is that
limited to character-to-character interactions?

The list owner is a big fan of non-reality based realism.  JC, is
this getting into your realm of expertise?  I'm much more at home
in the realm of simulation than abstraction.

--

Nathan F. Yospe -  Programmer, Scientist, Artist, JOAT with a SAK
yospe#kanga.nu   Home: nathanfyospe#mac.com  Work: nyospe at a2i.com


_______________________________________________
MUD-Dev mailing list
MUD-Dev at kanga.nu
https://www.kanga.nu/lists/listinfo/mud-dev



More information about the MUD-Dev mailing list