POV-Ray : Newsgroups : povray.newusers : Fireballs Server Time
13 Nov 2024 20:34:17 EST (-0500)
  Fireballs (Message 1 to 2 of 2)  
From: David
Subject: Fireballs
Date: 17 Jun 2000 21:20:13
Message: <B57171FB.12B5%davidmccabe@mac.com>
I've been trying to create a fire ball using media.
It looks OK so far, however, it looks to much like, well, a ball.
How can I make it's shape more random looking? I remember in the old 3.0
halo, the POV manual had a tutorial in which we made a fireball, and they
did something or other that involved scaling the sphere larger, but I don't
remember quite how it worked. Can anyone give me some help?

    Thanks
                   
        David McCabe
      dav### [at] maccom


Post a reply to this message

From: Bob Hughes
Subject: Re: Fireballs
Date: 18 Jun 2000 02:14:38
Message: <394c68ce@news.povray.org>
"David" <dav### [at] maccom> wrote in message
news:B57171FB.12B5%dav### [at] maccom...
|     I've been trying to create a fire ball using media.
| It looks OK so far, however, it looks to much like, well, a ball.
| How can I make it's shape more random looking? I remember in the old 3.0
| halo, the POV manual had a tutorial in which we made a fireball, and they
| did something or other that involved scaling the sphere larger, but I don't
| remember quite how it worked. Can anyone give me some help?

Yep, you are probably thinking of how the container object needs to be larger
than the unit-sized media (or halo) when turbulence is used because of the way
it goes beyond the volume of a non-turbulent media (or halo).  This of course
depends on the pattern type used though.  spherical and cylindrical for
example, but not planar, will be of unit dimension because of their similarity
to the objects sphere and cylinder.
With media having a 'density_map' you could include a variety of patterns into
just a spherical shape also, but you still will need to set the first index
inward from the edge.  0 to 1 in a index means outer surface to center of the
container object.

#declare None=density {rgb 0}
#declare SomeFringe=
 density {granite density_map {[0 rgb 0][1 rgb 1]} scale 1.25}
#declare SomeFill=
 density {spotted turbulence 0.25 density_map {[0 rgb 0][1 rgb 1]} scale .75}

sphere {0,1 hollow // always remember "hollow" :-)
 pigment {rgbf 1}
  interior {media{intervals 10 samples 1,1
    emission <1,.5,.1>*2 absorption <.6,.3,.1>
  scattering {4,<1,.5,.1>}
   density {spherical turbulence 0.5
    density_map {
      [.25 None] // helps to use larger than zero for this first one
      [.33 SomeFringe] // keep this away from zero
      [.75 SomeFill]
      }}
     }}
}

Hope this helps and I haven't confused the question.  turbulence and scale
(and density index ranges) are key to it all.  In fact you could use a
non-unit-sized sphere instead but the media considers the larger (or smaller)
size when sampling it and so it distributes the media in a sort of
thickness/thinness way.  Seeing is believing so just give it a try, make the
sphere 2 and test render then try a 0.5 radius too.  Something to keep in
mind.

Bob


Post a reply to this message

Copyright 2003-2023 Persistence of Vision Raytracer Pty. Ltd.