[MUD-Dev] Wilderness

Ling Lo ling at slimy.com
Thu Aug 2 01:21:59 New Zealand Standard Time 2001

On Wed, 1 Aug 2001, Nathan F. Yospe wrote:
> John Buehler <johnbue at msn.com> said:

[snip snip]

>> and such.  To be honest, I haven't figured out an efficient way
>> to handle flowing water.  I think it'll boil down to a batch
>> simulator.

> Nope.  Least energy gradient simulation with random reseeding for
> flood overflow.  Easiest way to do it is with an n-propogated GA
> during basic terrain generation.  What gets tricky is handling
> *air* flow for cooler and warmer regions relative to mountains and
> lakes, wind eroded deserts and valleys, and, once the terrain is
> generated, maintaining it for the reduction-to-regeneratable of
> player effects, for runtime wind and rain patterns, and for scent
> propogation.

> Using an n-propogating genetic algorithm for fluid flow, with
> seeds for replication on speed and volume in a given node, gives a
> near-enough to work with, faster than vectorized, almost congruent
> with known behavior due to surface tension (you can adjust the

[snip again]

What is an n-propagating genetic algorithm?

>From what I can understand of the above, creating the effects of
water on the surface of the terrain is effectively a modified
shortest path algorithm using GA.

Visual demonstration of GA wrt TSP:


GA Introduction


  |   Ling Lo

MUD-Dev mailing list
MUD-Dev at kanga.nu

More information about the MUD-Dev mailing list