[Fwd: Re: Grids and curvature of a sphere]
KaVir at dial.pipex.com
Sun Apr 19 06:54:43 New Zealand Standard Time 1998
This is a multi-part message in MIME format.
Content-Type: text/plain; charset=us-ascii
I'm going to invite Mr Black to join this list.
From: Hal Black <hablack at vt.edu>
Subject: Re: Grids and curvature of a sphere
Date: Sun, 19 Apr 1998 05:20:40 +0000
Message-ID: <353989A8.E77BF87A at vt.edu>
References: <3533D491.E23BBC07 at vt.edu> <35345540.6769 at dial.pipex.comNOSPAM> <3534B72F.ACC82D8C at stny.lrun.com> <6h2jii$c95$1 at news.cc.ucf.edu> <35379261.77989FF1 at vt.edu> <353943FB.5533 at dial.pipex.comNOSPAM>
Content-Type: text/plain; charset=us-ascii
X-Mailer: Mozilla 4.05 [en] (X11; I; Linux 2.1.78 i686)
Xref: plug.news.pipex.net rec.games.mud.admin:33129
Richard Woolcock wrote:
> Hal Black wrote:
> > I am not making most muds. Most muds have already been made, so I don't
> I wish more people had this attitude!
Heh, thanks, so do I.
> > need to make them. I am making a specific mud that will be modeled
> > after a sphere which will certainly have more than one continent or land
> > mass. The idea here is to do something different, which may introduce
> > some new possibilities not seen before on most muds.
> Perhaps some sort of 3D coordinate-based system could work - but you would
> have to drop the concept of 'rooms' if you did that.
I fear that you may be right about this. I would prefer to avoid
dropping the room construct because it allows for such convenient
abstraction. Without rooms, I would essentially have to build a 3d
engine. While this would be introduce some truly fascinating elements
of play, a 3d engine is not only nontrivial to implement, but also
resource-intensive to process. It is a difficult enough problem in a
Cartesian coordinate system, I suspect it would be a total nightmare in
a spherical one.
Even if there is no 3d object rasterizing involved (for display on the
screen), things like collisions and line-of-sight beg consideration.
For instance, when something happens, who witnesses it? And when
someone speaks, who hears it? With many players online at the same
time, this could consume a lot of processing power (and memory) better
used for other things.
> Do you want the world
> to be a 3D object or a 2D plane which is spherical? Will the mud be the
> entire game world, or just a single planet floating in space?
Hmm, I hadn't thought of multiple planets. I guess at this point I'd
like to keep it simple. I don't think there is much loss of generality
in defining direction relative to some point that corresponds to the
planet. A separate system could be used to deal with interplanetary
interaction. I think that having an intraplanet coordinate system would
keep things a lot simpler. For instance, it doesn't directly matter to
the player that he's moving very fast in orbit relative to the sun.
Plus it would be a royal pain to keep "absolute coordinates" of objects
on a moving planet. Then there are relativistic effects that start to
come into play. Oh, I don't even want to think about it 8')
So yes, consider the planet to be the entire game world. Of course,
the sun and perhaps some moons will exist that can be simply modeled in
a geocentric manner. This is important because those heavenly bodies
produce key atmospheric effects resulting from having a spherical
planet. I think it would be very cool to have an xearth-style display
of the sun's light on the mud planet (for my own viewing, not
necessarily for joe player). 8')
One of the things planned for the story line is a vast network of
underground caverns. Also, the world should not preclude buildings that
have more than one story. So yes, it would be a 3d model.
I was planning on worrying about the 3d portion after finding a
solution to the sphere problem. While it is true that the sphere's
surface area increases with radius, I think that the range of vertical
interest is sufficiently small to ignore this (this is analogous to the
cylindrical approximation of the sphere sufficiently close to the
equator that I spoke about in an earlier article.)
To implement height, I was thinking of using either an onion skin
model or absolute coordinates. I don't want to prematurely choose a
height solution lest it impose limitations resulting in a sub-optimal
solution to the more significant sphere problem.
> Then you really will have to change movement from the typical n/s/e/w
> approach - although you could still have such movement possible, eg:
> ]north 10
> You glance at your compass.
> You spin 37 degrees to your left and take 10 paces forward.
I was planning on something similar. Woe to the adventurer who loses
> > Again, the question I posed was not about a small area.
> Then you have a lot of work ahead! Good luck ;)
Thanks for your response and well-wishing, Richard,
More information about the MUD-Dev