[MUD-Dev] Re: Dynamic muds

Katrina McClelan kitkat at marcus.pants.nu
Fri Sep 24 05:28:54 New Zealand Standard Time 1999

On Thu, 23 Sep 1999, [iso-8859-1] Emil Eifr=E9m wrote:

> If I understand it correctly, you're dynamically generating every room
> description from all the objects within viewing distance from that room.

Well for starters, he probably doesn't use a traditional room.  See my
recent post for one way to do this.  JC may well have a different, perhaps
even better way to implement it in code.  If you start getting away from
the idea (and limitations) of a traditional room, generating the list of
nearby information is not as difficult as you seem to imply it would be.

> Sounds like CPU-cycles to me. Even if you optimize it and just generate
> descs whenever objects move and not every time someone enters a room or
> requests a room description ("look"), even then it sounds very resource
> intensive. Is your world big? Do you sometimes end up with "strange" room
> descriptions?

Perhaps moreso than a static MUD, but my static description MUD runs well
enough on a 486, so if you buy yourself a dual-processor pentium III 450
or so, with 128M of ram, and I think it'll be more than enough.  Plus,
provided th objects themselves have reasonable descriptions, the algorithm
isn't that bad.  A typical diku mud merges like items as it is, which is
more CPU intensive, either in pre-sorting or in run time sorting than it
would be to splice text.  It's an ugly algorithm to implement in code, but
that has nothing to do with its run time performance.  A good counter
example is bubble sort, which is simple in logic when compared to a quick
sort, but is much more expensive.


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

More information about the MUD-Dev mailing list