[MUD-Dev] RP thesis...
Caliban Tiresias Darklock
caliban at darklock.com
Wed May 21 13:18:49 New Zealand Standard Time 1997
Matt Chatterley wrote:
> On Tue, 20 May 1997 clawrenc at cup.hp.com wrote:
> > In <Pine.LNX.3.96.970518090712.185C-100000 at mpc.dyn.ml.org>, on
> > 05/18/97
> > at 09:07 AM, Matt Chatterley <root at mpc.dyn.ml.org> said:
> > >and set a scale for each room - this map would be converted into
> > >normal rooms with randomly set sizes, descriptions, weather suited to
> > >the area, and so forth.
Replace 'randomly' with 'algorithmically'. A perfect seed for this would
be the X and Y coordinates of the player's position on the map; this
way, the description of the room at 10,20 would *always* be the same,
but would not necessarily need to be saved.
> > The problem with virtual rooms, which can be handled is when players
> > make a change to a virtual room which requires them to become
> > permanent, or the notice something in a virtual room and refer to it
> > later expecting it to be permanent.
It's still easier to save a list of changes to only those rooms which
have been changed, including a time stamp, and removing them after some
period of time. Sort of like an ASCII database. If the database grows
too large, the
oldest entries are removed to bring the database into line. If the entry
is too old, it is deleted. Rationales for this could be something like
'if the PCs are very active in the area, then so are the NPCs'. There
are almost endless unmentioned people; rarely does the main town include
an appropriate population, because most of the people in the town are
> > > l
> > You are in a forest...yada yada something about an oak tree.
> > ...much later...
> > > say "Just go east in the forest until you see the Oak tree,
> > and then head west..."
Algorithmically generated descriptions would prevent this problem quite
> <g> I think this would either be disallowed, or chances of seeing
> landmarks on a small scale would vary with the size of the room (a room is
> being given to represent an area of land, most likely, on a set scale,
> which could be quite large). It depends really. :)
An advantage to algorithmically generated descriptions is that they
could also describe surroundings with some degree of accuracy.
More information about the MUD-Dev