Nathan Kopp wrote:
> 
> It's not the media.  It's the "noise3d" fix.

I suspect there is something funny about media. I am working on a cloudscape and
switching to 0.5 immediately led to all sorts of weirdness. I'm still
researching the matter, but here is one observation: method 3 media tends to
render as black if minimum samples is set to 1. Raising it usually helps, but
I've had to go as high as 20 to eliminate some weird banding. Method 2 media
sometimes acts up too, but overall the behaviour seems kind of chaotic (i.e. I
haven't made the connection yet).
Could it perhaps be connected to the constant media speed enhancement? How does
this work?

-- 
Margus Ramst

Personal e-mail: mar### [at] peakeduee
TAG (Team Assistance Group) e-mail: mar### [at] tagpovrayorg

Post a reply to this message

Margus Ramst <mar### [at] peakeduee> wrote...
> Nathan Kopp wrote:
> >
> > It's not the media.  It's the "noise3d" fix.
>
> I suspect there is something funny about media. I am working on a
cloudscape and
> switching to 0.5 immediately led to all sorts of weirdness. I'm still
> researching the matter, but here is one observation: method 3 media tends
to
> render as black if minimum samples is set to 1. Raising it usually helps,
but
> I've had to go as high as 20 to eliminate some weird banding. Method 2
media
> sometimes acts up too, but overall the behaviour seems kind of chaotic
(i.e. I
> haven't made the connection yet).
> Could it perhaps be connected to the constant media speed enhancement? How
does
> this work?

It's possible.  The speedup works like this:

If the ray in question is a shadow ray (meaning that we are only determining
how much light is blocked as the light travels through the media) and all
participating media are constant, then POV will use one interval, and one
sample using media method 1 instead of the user-selected method.  The reason
is that no other samples are needed.  A single sample will give us the
density, and we really don't care how much light is hitting the media along
the path of the ray, because we only care how much light was blocked as the
beam of light traveled along the ray.  Does that make sense?

-Nathan

Post a reply to this message

I've hit two problems
    i) brightness, easily solved
    2) the interaction of two medias is completely different. the media
seems to almost dissapear.This is the one where I can't find the logic. or
get any control, Now that Nathan has explained I can see why, but not how to
compensate.

Mick


"Nathan Kopp" <Nat### [at] Koppcom> wrote in message
news:39236af5$1@news.povray.org...
>
> Margus Ramst <mar### [at] peakeduee> wrote...
> > Nathan Kopp wrote:
> > >
> > > It's not the media.  It's the "noise3d" fix.
> >
> > I suspect there is something funny about media. I am working on a
> cloudscape and
> > switching to 0.5 immediately led to all sorts of weirdness. I'm still
> > researching the matter, but here is one observation: method 3 media
tends
> to
> > render as black if minimum samples is set to 1. Raising it usually
helps,
> but
> > I've had to go as high as 20 to eliminate some weird banding. Method 2
> media
> > sometimes acts up too, but overall the behaviour seems kind of chaotic
> (i.e. I
> > haven't made the connection yet).
> > Could it perhaps be connected to the constant media speed enhancement?
How
> does
> > this work?
>
> It's possible.  The speedup works like this:
>
> If the ray in question is a shadow ray (meaning that we are only
determining
> how much light is blocked as the light travels through the media) and all
> participating media are constant, then POV will use one interval, and one
> sample using media method 1 instead of the user-selected method.  The
reason
> is that no other samples are needed.  A single sample will give us the
> density, and we really don't care how much light is hitting the media
along
> the path of the ray, because we only care how much light was blocked as
the
> beam of light traveled along the ray.  Does that make sense?
>
> -Nathan
>
>

Post a reply to this message

Nathan Kopp wrote:
> 
> It's possible.  The speedup works like this:
> 

No, I don't think this is the culprit. The same happens with both constant and
non-constant densities, emitting, absorbing and scattering media. The media is
opaque to both camera and shadow rays. I'm beginning to think that the adaptive
mechanism is broken somehow.

-- 
Margus Ramst

Personal e-mail: mar### [at] peakeduee
TAG (Team Assistance Group) e-mail: mar### [at] tagpovrayorg

Post a reply to this message

Margus Ramst <mar### [at] peakeduee> wrote...
> Nathan Kopp wrote:
> >
> > It's possible.  The speedup works like this:
> >
>
> No, I don't think this is the culprit. The same happens with both constant
and
> non-constant densities, emitting, absorbing and scattering media. The
media is
> opaque to both camera and shadow rays. I'm beginning to think that the
adaptive
> mechanism is broken somehow.

Does it really happen with emitting media?  There was a change made to
extinction/absorption of media method 3, but that shouldn't affect emitting
media.  (There may be some other issues that affect media method 3, though).

-Nathan

Post a reply to this message

Nathan Kopp wrote:
> 
> Does it really happen with emitting media?

It does, with the exception that the media won't cast a shadow (as expected).
I am not the only one affected, am I?
Here's a little scene that should demonstrate the problem:

#version unofficial megapov 0.5;

camera{location <2,3,-5> look_at 0}
light_source{<-50,50,-50> color 1 }

sphere{0,1
    pigment{rgbf 1}
    finish{ambient 0 diffuse 0}
    interior{media{method 3 samples 1 emission 1}}
    hollow
    translate y
}

plane{y,0 pigment{checker .3,.6} finish{diffuse .7 ambient 0}}

-- 
Margus Ramst

Personal e-mail: mar### [at] peakeduee
TAG (Team Assistance Group) e-mail: mar### [at] tagpovrayorg

Post a reply to this message

Margus Ramst <mar### [at] peakeduee> wrote...
>
> It does, with the exception that the media won't cast a shadow (as
expected).

Actually, emitting media is not supposed to cast a shadow.  Only absorbing
and scattering media will cast shadows.

-Nathan

Post a reply to this message

Margus Ramst <mar### [at] peakeduee> wrote...
> Nathan Kopp wrote:
> >
> > Does it really happen with emitting media?
>
> It does, with the exception that the media won't cast a shadow (as
expected).
> I am not the only one affected, am I?
> Here's a little scene that should demonstrate the problem:

There was a bug if you choose 3 or fewer samples, and I've found it and
fixed it.  There may be other bugs, though.

-Nathan

Post a reply to this message

Nathan Kopp wrote:
> 
> There was a bug if you choose 3 or fewer samples, and I've found it and
> fixed it.  There may be other bugs, though.
> 

Well, the demo scene works fine with samples >1, and I haven't noticed any
significant quality jump going from 3 to more samples in any of my tests.

-- 
Margus Ramst

Personal e-mail: mar### [at] peakeduee
TAG (Team Assistance Group) e-mail: mar### [at] tagpovrayorg

Post a reply to this message

Margus Ramst <mar### [at] peakeduee> wrote...
> Nathan Kopp wrote:
> >
> > There was a bug if you choose 3 or fewer samples, and I've found it and
> > fixed it.  There may be other bugs, though.
> >
>
> Well, the demo scene works fine with samples >1, and I haven't noticed any
> significant quality jump going from 3 to more samples in any of my tests.
>

Maybe the bug only happened for samples=1.  Anyway, it was a
division-by-zero problem because somewhere we were dividing by
(sampleCount - 1).

-Nathan

Post a reply to this message