[MUD-Dev] The Endeavour Map and MUD Applications

John Bertoglio jb at pulsepoll.com
Thu Feb 17 19:42:03 CET 2000

> Justin Rogers
> Sent: Wednesday, February 16, 2000 10:55 PM
> [John]

<cut description of use of bit mapped graphics to generate database>

>     I remember this thread and it was actually quite interesting.  The
> problem
> lied in the abundance of characters that could be used to specify
> various
> terrain types and the such.
>     This new approach you are talking about has some other neat
> ramifications
> that I'll talk about below.  But since you are making high resolution,
> beautified
> graphics of the world and landscape you also have these available to
> the user.

Actually, the entire map probably would not be directly available to users.
Pieces could be cut out and turned into local maps but discovery is part of
the fun of this design. It would be far more fun (at least for the designer)
to see maps posted by users and compare them to the actual world map whose
main function is to serve as a design template.

> [John]
> > You save the file as a PPM file. A program that parses each pixel of
> the
> > graphic creates the initial world, exits from one location to
> another and
> > defines the elevation.

>     How does it define the exits.  Obviously this is truly a *world*
> file in
> which every direction can be taken to reach every other direction.
> You
> would have to specify quite a bit more to get *real* type exits.

My world design uses a zoom metaphor. Each cell mapped using the graphic
layout is about 20km x 20km. When a player or group is moving at this level,
they are in "expedition mode" which allows large distances to be traversed
in a reasonable amount of time. The gateway between Regions know how
difficult it is to move from one to another. You can declare the use of road
or water movement (assuming other conditions are already met and there is a
road or major river connecting the two Regions).  Movement is abstract as is
the consumption of resources. How and where you are moving determines the
frequency and type of encounters which drop you down into Location level to
deal with. While you can move 2000km in a few seconds, you will starve to
death if the proper steps are not taken. There is an intermediate level
which further breaks down each of the Regions in to 400 1km Areas. This
creates 400 million unique places in the world. Each can be controlled by a
player bringing a strategic level of play for the ultimate in powergaming.

The program generates gateways between all cells in all 8 compass directions
as the cells are parsed. Obviously, some routes are easier than others.
"Real exits" are created at the Location level which uses a reasonably
standard "rooms" metaphor. They are created by a GUI online editor, by
character actions and by random generation at the point of entry.

> this
> height field type map there are definitely several programs out that
> can
> autogenerate unique worlds.  Which certainly makes more sense than
> plotting
> a couple million pixels...

I am very interested in the idea of autogeneration at more granular levels
but actually, a 1200 x 800 world can be built very quickly. Remember, in a
paint program you draw outlines and then flood fill. A million clicks would
be very tedious. Of course, you can spend days tweeking the world. My basic
point that it is much faster to draw a new island in a part of the globe
seems a bit barren than to hand code all the cells into the world database.

>     Your overplayed picture approach definitely has its merits...  What
> I fully
> propose would be the use of 16 color maps and actually define various
> levels
> that get transposed....  16 colors for elevation of certain heights.
> The same
> 16 colors for cities, same for minerals etc...

That is exactly how it works. Interestingly, there is no strict limit for 16
color palettes, it is just that I have never needed more than 16. It might
make more sense to define all the colors in a single palette since PSP
(paint shop pro) allows you to display the palettes in palettes order. It
also allows you to define the palettes in a text document (including
comments so you can remember what color does what).

> Then the user can ask
> for a
> certain type of map within the game.
>   >  map geographic
>   >  map mineral
>     That way you can use the basic set of colors, but you can have 20
> different
> types of maps.

This idea is interesting because it would allow you to use the same legend
for each map and just relabel for a different purpose. I have chosen to use
different palettes because it seems reasonable to use colors which relate to
the information. What we do is store the color value (in hex) with the
terrain (or whatever) type and display it when the player uses the map
making skill. Depending on the level of skill and/or the ownership of the
proper maps, the player can display an X by X display of the local data
inside of a window in their browser.

>     Also, have you looked into using a geographic database to create
> your world.
> I've looked into using GIS data to create a MUD exactly like the
> world.

I like this idea, but one of the reasons to build your own planet is to make
a "better" world. At the very least, one which is different. I suspect the
same methodology described in my original post would work great with GIS

> I
> defined about 100 regions and then defined programs that would
> populate those
> regions using biological systems.  Once I had a fully functional world
> model with
> GIS data and vegetation data built on top I put about 400 mini ai NPCs
> to work
> in groups of ten and let them build and populate.  It was an ants
> behind the glass
> type of MUD because there was no PC integration.  But it was rather
> neat to watch
> and the system was there in a way that I didn't need any world
> builders to create
> areas.  After about 3 months all my areas were created.

This is a really interesting idea.
I would be very interested in knowing how you did the world generation.

John A. Bertoglio

| PulsePoll.com <http://www.pulsepoll.com/>
| 503.781.3563
| jb at pulsepoll.com <mailto:jb at pulsepoll.com

MUD-Dev maillist  -  MUD-Dev at kanga.nu

More information about the mud-dev-archive mailing list