I am testing an emission media for use as lava.  The problem is, I am
getting a solid white ball in my test sphere:

// Lava
#declare texLava = texture {
 pigment { rgbt 1 }
 //finish { ambient 0.7 }
}

#declare intLava = interior {
 ior 1.4
 media {
  emission 0.1
  samples 3, 8
  method 3
  intervals 2
  density {
   spherical
   triangle_wave
   frequency 2
   turbulence 0.6
   color_map {
    [ 0.02 rgb 1.2 ]
    [ 0.08 rgb <1, 1, 0> ]
    [ 0.16 rgb <1, 0, 0> ]
    [ 0.25 rgbt <0, 0, 0, 1> ]
   }
   scale 0.1
  }
 }
}

sphere {
 0, 10
 hollow
 texture { texLava }
 interior { intLava }
}

If I remove the 0.1 from emission, the media disappears entirely.  What have
I done wrong here?

Post a reply to this message

In article <3d528e71@news.povray.org>,
 "David Wallace" <dar### [at] earthlinknet> wrote:

> I am testing an emission media for use as lava.  The problem is, I am
> getting a solid white ball in my test sphere:

That's just because the media is too bright, all three colors go above 1 
so the color gets clipped to pure white in the final image. Reduce the 
value given to emission, try 0.01 or 0.05 and adjust until you get what 
you want.

-- 
Christopher James Huff <chr### [at] maccom>
POV-Ray TAG e-mail: chr### [at] tagpovrayorg
TAG web site: http://tag.povray.org/

Post a reply to this message

"Christopher James Huff" <chr### [at] maccom> wrote in message
news:chr### [at] netplexaussieorg...
> In article <3d528e71@news.povray.org>,
>  "David Wallace" <dar### [at] earthlinknet> wrote:
>
> > I am testing an emission media for use as lava.  The problem is, I am
> > getting a solid white ball in my test sphere:
>
> That's just because the media is too bright, all three colors go above 1
> so the color gets clipped to pure white in the final image. Reduce the
> value given to emission, try 0.01 or 0.05 and adjust until you get what
> you want.

Actually, having tried it out here, I see that the whole media statement
could be trouble for this lava idea. Not sure what the end result wanted is
exactly but that spherical pattern for density is all wrong, seems to me
anyway. The index at 0 is the outside and 1 is inside. Could think of it as
a Earth, having [0 rgb 1] being the core and [1 rgb 0] being the outer
atmosphere where it meets space. frequency -1 reverses the order so I used
that.

Also a couple things to note as well, method 3 intervals are done
automatically-- unless you really want to override that-- and absorption
could be needed too. No idea why ior 1.4 is there but David has his
reasons...

I could go on... (as i try stuff with it here)

How about that amount of empty space and the turbulence? Seems to me the
media would only be a sparkling mass and not solid. I think the amount of
solidness diminishes a lot with that turbulence. And I tried emission 0.3
absorption 0.2 which was okay-looking, as long as turbulence is 0.2 or less.
If more is used the frequency must be increased, or better yet, scale of the
density or a different object used than is now.

Interestingly enough, the media can change drastically as scale, turbulence,
frequency are used differently. The spherical pattern is pretty special as
it is anyway. I kind of knew all this already, of course  :-)  It isn't real
good with frequency, I guess, since by design it is only a unit size without
a repetition of itself to all other points in POV-space unlike most other
patterns.

Maybe the idea was to get a single textured appearance for other things than
that sphere, David will have to confirm this before I think it is pointless
to try and use the density_map as it currently stands. Oh... perhaps it was
intended to have transparent places with it overlaid into a part of a scene?
I should have thought of that, if so.

Post a reply to this message

"hughes b" <omn### [at] charternet> wrote in message
news:3d52c1e0$1@news.povray.org...
> "Christopher James Huff" <chr### [at] maccom> wrote in message
> news:chr### [at] netplexaussieorg...
> > In article <3d528e71@news.povray.org>,
> >  "David Wallace" <dar### [at] earthlinknet> wrote:
> >
> > > I am testing an emission media for use as lava.  The problem is, I am
> > > getting a solid white ball in my test sphere:
> >
> > That's just because the media is too bright, all three colors go above 1
> > so the color gets clipped to pure white in the final image. Reduce the
> > value given to emission, try 0.01 or 0.05 and adjust until you get what
> > you want.
>
> Actually, having tried it out here, I see that the whole media statement
> could be trouble for this lava idea. Not sure what the end result wanted
is
> exactly but that spherical pattern for density is all wrong, seems to me
> anyway. The index at 0 is the outside and 1 is inside. Could think of it
as
> a Earth, having [0 rgb 1] being the core and [1 rgb 0] being the outer
> atmosphere where it meets space. frequency -1 reverses the order so I used
> that.
>
> Also a couple things to note as well, method 3 intervals are done
> automatically-- unless you really want to override that-- and absorption
> could be needed too. No idea why ior 1.4 is there but David has his
> reasons...
>
> I could go on... (as i try stuff with it here)
>
> How about that amount of empty space and the turbulence? Seems to me the
> media would only be a sparkling mass and not solid. I think the amount of
> solidness diminishes a lot with that turbulence. And I tried emission 0.3
> absorption 0.2 which was okay-looking, as long as turbulence is 0.2 or
less.
> If more is used the frequency must be increased, or better yet, scale of
the
> density or a different object used than is now.
>
> Interestingly enough, the media can change drastically as scale,
turbulence,
> frequency are used differently. The spherical pattern is pretty special as
> it is anyway. I kind of knew all this already, of course  :-)  It isn't
real
> good with frequency, I guess, since by design it is only a unit size
without
> a repetition of itself to all other points in POV-space unlike most other
> patterns.
>
> Maybe the idea was to get a single textured appearance for other things
than
> that sphere, David will have to confirm this before I think it is
pointless
> to try and use the density_map as it currently stands. Oh... perhaps it
was
> intended to have transparent places with it overlaid into a part of a
scene?
> I should have thought of that, if so.
>
>
>
What I am trying to simulate is lava, a molten liquid, with an interior
consistent with a (dense and viscous) liquid.  The glow (emission) comes
from the extreme temperature.  I really need a bright, turbulent emission
media, similar to fire but more liquid in appearance (hence the ior).

The bright core is in the center (map entry 0) with a mostly clear periphery
(map entry 1), like fire.  Only the 0.00-0.02 part of the map has an entry
higher than 1; the vast majority, 0.25-1.00, is transparent.  This should
not result in a white sphere.  Actually, the lava jet I am simulating
consists of mostly stretched spheres but I need the media right before
continuing here.

Post a reply to this message

"David Wallace" <dar### [at] earthlinknet> wrote in message
news:3d540089@news.povray.org...
>
> The bright core is in the center (map entry 0) with a mostly clear
periphery
> (map entry 1), like fire.  Only the 0.00-0.02 part of the map has an entry
> higher than 1; the vast majority, 0.25-1.00, is transparent.  This should
> not result in a white sphere.  Actually, the lava jet I am simulating
> consists of mostly stretched spheres but I need the media right before
> continuing here.

Yes, what you described you want to do then isn't really what is being
created by that scene file script though.

spherical uses the map index opposite of what you are saying. 0 is the
outermost shell and 1 is the innermost core. frequency -1 would reverse that
or you need to move the entries around to change it.

Thing is, that transparency--- oops, first please allow me to mention what I
neglected before now. You used a transmit 1 in the density_map and
transparency has no use there. density is only in the form of the three rgb
values.
Anyway, to go on... Reason I brought that spacious transparent portion up
before is because the turbulence effectively swirls the media into oblivian.
I seriously doubt know the reason for it except that it appears to be a
quirk with that spherical pattern. I'm not sure at all. I just know that
pattern, being of only a unit size to begin with, doesn't normally go beyond
that unit size. Doesn't tile or repeat or continue outward from a 1 unit
radius, in other words. So that being the case (hopefully I'm telling this
right!) it seems to have problems at the surface when the boundry is
encountered.

But truthfully, I wish I knew why the white-out occurs. In checking on your
script I found blue was introduced somehow. Of course rgb <1,1,1> or White
does contain blue but I'd think it shouldn't come out of the mix when the
next map entries have none.

Well, that's my observations.

Post a reply to this message

"hughes b" <omn### [at] charternet> wrote in message
news:3d54281d@news.povray.org...
> "David Wallace" <dar### [at] earthlinknet> wrote in message
> news:3d540089@news.povray.org...
> >
> > The bright core is in the center (map entry 0) with a mostly clear
> periphery
> > (map entry 1), like fire.  Only the 0.00-0.02 part of the map has an
entry
> > higher than 1; the vast majority, 0.25-1.00, is transparent.  This
should
> > not result in a white sphere.  Actually, the lava jet I am simulating
> > consists of mostly stretched spheres but I need the media right before
> > continuing here.
>
> Yes, what you described you want to do then isn't really what is being
> created by that scene file script though.
>
> spherical uses the map index opposite of what you are saying. 0 is the
> outermost shell and 1 is the innermost core. frequency -1 would reverse
that
> or you need to move the entries around to change it.
>
> Thing is, that transparency--- oops, first please allow me to mention what
I
> neglected before now. You used a transmit 1 in the density_map and
> transparency has no use there. density is only in the form of the three
rgb
> values.
> Anyway, to go on... Reason I brought that spacious transparent portion up
> before is because the turbulence effectively swirls the media into
oblivian.
> I seriously doubt know the reason for it except that it appears to be a
> quirk with that spherical pattern. I'm not sure at all. I just know that
> pattern, being of only a unit size to begin with, doesn't normally go
beyond
> that unit size. Doesn't tile or repeat or continue outward from a 1 unit
> radius, in other words. So that being the case (hopefully I'm telling this
> right!) it seems to have problems at the surface when the boundry is
> encountered.
>
> But truthfully, I wish I knew why the white-out occurs. In checking on
your
> script I found blue was introduced somehow. Of course rgb <1,1,1> or White
> does contain blue but I'd think it shouldn't come out of the mix when the
> next map entries have none.
>
> Well, that's my observations.
>
Well, I ditched the spherical for bozo.  I still found that the media was so
bright that it appeared white until I got serious about dimming it:

#declare intLava = interior {
 ior 1.1
 media {
  emission 0.02
  samples 1, 20
  variance 1/1000
  confidence .995
  density {
   bozo
   turbulence 1.00
   omega 0.7
   lambda 1.7
   color_map {
    [ 0.0 rgb <0.0, 0.0, 0.0> ]
    [ 0.1 rgb <0.0, 0.0, 0.0> ]
    [ 0.3 rgb <0.2, 0.1, 0.1> ]
    [ 1.0 rgb <0.4, 0.1, 0.2> ]
   }
   scale 4
  }
 }
}

The size of the spheres has a big impact on the color (larger is brighter).
They still appear a glowing red, which is what I was after in the first
place.

Post a reply to this message