POV-Ray : Newsgroups : povray.general : problem with turbulence in 'fog' : Re: problem with turbulence in 'fog' Server Time
25 Apr 2024 19:42:08 EDT (-0400)
  Re: problem with turbulence in 'fog'  
From: Kenneth
Date: 4 Dec 2017 02:45:01
Message: <web.5a24fa5080800e7e89df8d30@news.povray.org>
"omniverse" <omn### [at] charternet> wrote:

> Hmmm. Not what I was thinking. The scene file is ..scenes\advanced\mist.pov
> and it was ground fog, but what it was the turbulence itself I had to
> increase drastically to see it good. Also increasing 'distance' to make
> it more apparent closest to camera, otherwise it tends to wash out in itself
> and scene background.
>
> There seems to be better parameter values to use for making the turbulent fog
> appear, and even so maybe it isn't doing what you would expect of it
> because of the inherent design of fog. Or to put it another way, that it
> isn't like media.

Yes, I'm finally seeing 'an effect' in the mist.pov file (by changing
some things)-- although I'm not exactly sure what it is that I'm seeing(!) I
agree with your assesment, that the fog itself may be 'hiding' the turbulence
effect-- unless the various parameters all work together in a particular way.
Try this instead:

fog{
    color Gray70
    fog_type 2
    fog_alt 0.001
    fog_offset 1.3
    distance 15
   // turbulence (1 + 50*clock)*<.15, .15, .15>
    turbulence 40*<.15, .15, .15>
    turb_depth 1.5 // try changing to 0.0 then to 4.0
    omega 0.001 // to get rid of any 'graniness'
   // lambda .001 // 1.25
   // octaves 10  // 5
   // scale 3  // has NO effect on turbulence (or on anything else, as far
   // as I can see)
}

The 'white blotches' (the turbulence) appear to be visible on object *surfaces*,
although I can't see anything within the misty fog itself. And turb_depth seems
to be a necessary addition-- and within a small range of values (I think!) BTW,
those splotches can't be scaled; if your scene uses larger objects, the
splotches appear smaller. That seems like a limitation in itself.

> Still can't see obvious turbulent fog above the water in that scene either,
> so it looks almost like a surface texture.

I agree. My own main expectation of turbulence was that it should cause the fog
to appear 'lumpy' at it's own higher 'horizontal surface' (or pseudo-surface,
that is-- somewhere between fog_alt and fog_offset, depending on the density),
climbing 'above and below' that imaginary surface like real rolling fog. But
maybe that's a wrong supposition(?) But it raises the question of what
turbulence *is* supposed to be doing.

> However... those omega, lambda and octaves parameters seem to cause trouble
> so leaving them out (or as default?) looked best to me.
>
I think some of the parameters in the mist.pov file (and the other files) are
actually fighting against the turbulence appearance...like the turbulence value
is set way too low, and the omega value is too high (even at .35 or whatever.)
It makes me think that something in the underlying fog or turbulence code may
have changed in the (distant) past-- because the example scenes (and the
documentation) apparently expect the turbulence to be readily visible, which it
isn't now.

Meanwhile, I'm running some animation experiments on mist.pov and one of the
other files, to see what happens when various parameters change.


Post a reply to this message

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