Someone care to explain the following from the manual:

"It's usually best to only use one interval with method 3. Too many 
intervals can lead to artefacts, and POV will create more intervals if 
it needs them."

It seems with one interval, I don't get good results. If I up the 
intervals to 5 or 10, it seems to work better, but is slower in general.

Anyone have any clue about the relationship and why the manual asks to 
use only 1 interval?


-- 
~Mike

Post a reply to this message

> It seems with one interval, I don't get good results. If I up the
> intervals to 5 or 10, it seems to work better, but is slower in general.
>
> Anyone have any clue about the relationship and why the manual asks to
> use only 1 interval?


Well, more intervals causes more sampling since there are the same number of
samples per interval.

What the documentation is saying is that, if you need better results, you
should increase the sample count rather than the interval count.

 - Slime
 [ http://www.slimeland.com/ ]

Post a reply to this message

Among other things, Slime wrote:

> What the documentation is saying is that, if you need better results, you
> should increase the sample count rather than the interval count.

What I don't quite understand is why the number of samples has so much
effect over the overall brightness of the media (I'm thinking of dense
scattering media to simulate SSS). Sure, the sample number should affect
the quality and presence of possible artifacts, but why is "samples 15" so
much brighter than "samples 75" (method 3, intervals 1)? Is this maybe due
to some bad coding on my part?

-- 
light_source{9+9*x,1}camera{orthographic look_at(1-y)/4angle 30location
9/4-z*4}light_source{-9*z,1}union{box{.9-z.1+x clipped_by{plane{2+y-4*x
0}}}box{z-y-.1.1+z}box{-.1.1+x}box{.1z-.1}pigment{rgb<.8.2,1>}}//Jellby

Post a reply to this message

Slime wrote:

> What the documentation is saying is that, if you need better results, you
> should increase the sample count rather than the interval count.

Interesting that the defaults for media would be set up as such:

aa_level     : 4
aa_threshold : 0.1
absorption   : <0,0,0>
confidence   : 0.9
emission     : <0,0,0>
 >> intervals    : 10
 >> method       : 3
ratio        : 0.9
 >> samples      : Min 1, Max 1
variance     : 1/128
SCATTERING
   COLOR        : <0,0,0>
   eccentricity : 0.0
   extinction   : 1.0

I'm beginning if they meant samples should be at 1 in the documentation 
and not intervals.

I'm hopelessly confused.

-- 
~Mike

Post a reply to this message

Mike Raiford wrote:

> I'm hopelessly confused.

The better question is: Is there a really good tutorial on using media?

-- 
~Mike

Post a reply to this message

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Jellby wrote:
| What I don't quite understand is why the number of samples has so much
| effect over the overall brightness of the media (I'm thinking of dense
| scattering media to simulate SSS). Sure, the sample number should
affect
| the quality and presence of possible artifacts, but why is "samples
15" so
| much brighter than "samples 75" (method 3, intervals 1)? Is this
maybe due
| to some bad coding on my part?
|
	That's because the darkening you see is an artefact :) It's due to
the fact that when you don't take many samples, a lot of them
proportionnaly get taken deep inside the media where it's dark.
Increasing the number of samples ensures that more and more get taken
in the thin layer near the surface where the SSS effect occurs which
leads POV to notice that the result should be bright. BTW, you
shouldn't increase the number of samples too much. Instead you should
increase the aa_level and possibly decrease the aa_threshold. This
will ensure that the additional samples get taken in places where
they will have a useful impact on the result.

		Jerome

- --
******************************
*      Jerome M. Berger      *
* mailto:jbe### [at] ifrancecom *
*  http://jeberger.free.fr/  *
******************************
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.4 (GNU/Linux)

iD8DBQFB2ZnwqIYJdJhyixIRAjCoAJ9Mh12iDLjNYBtmUPJZg5HthQ5zTwCeIjar
AuqaKLzmlYjCRPrfbUjynlA=
=2sxH
-----END PGP SIGNATURE-----

Post a reply to this message

> That's because the darkening you see is an artefact :) It's due to
> the fact that when you don't take many samples, a lot of them
> proportionnaly get taken deep inside the media where it's dark.
> Increasing the number of samples ensures that more and more get taken
> in the thin layer near the surface where the SSS effect occurs which
> leads POV to notice that the result should be bright.

According to this, more samples -> brighter media. But I see the opposite,
as if the influence of the samples was "averaged" (since most samples are
deep inside, the effect of the surface ones is weaker). I would have
thought that the sampling algorithm (and colour calculation) would be
smarter than that, but maybe that's not so easy...

> BTW, you 
> shouldn't increase the number of samples too much. Instead you should
> increase the aa_level and possibly decrease the aa_threshold. This
> will ensure that the additional samples get taken in places where
> they will have a useful impact on the result.

I couldn't get the same effect... Even if the "darkening" due to many
samples (or intervals) is an artifact, I didn't find a reliable way to get
what I needed other than increasing the number of samples.

-- 
light_source{9+9*x,1}camera{orthographic look_at(1-y)/4angle 30location
9/4-z*4}light_source{-9*z,1}union{box{.9-z.1+x clipped_by{plane{2+y-4*x
0}}}box{z-y-.1.1+z}box{-.1.1+x}box{.1z-.1}pigment{rgb<.8.2,1>}}//Jellby

Post a reply to this message

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Jellby wrote:
| According to this, more samples -> brighter media.
	That is correct. If you push this to the extreme, you'll get this
image (10 samples):
http://jeberger.free.fr/mpov/sphere_bl.png

	Increasing the number of samples (to 200) gives this one:
http://jeberger.free.fr/mpov/sphere_no.png

| But I see the opposite,
| as if the influence of the samples was "averaged" (since most
samples are
| deep inside, the effect of the surface ones is weaker). I would have
| thought that the sampling algorithm (and colour calculation) would be
| smarter than that, but maybe that's not so easy...
|
	Do you have a sample code that produces this effect?

|
|>BTW, you
|>shouldn't increase the number of samples too much. Instead you should
|>increase the aa_level and possibly decrease the aa_threshold. This
|>will ensure that the additional samples get taken in places where
|>they will have a useful impact on the result.
|
| I couldn't get the same effect... Even if the "darkening" due to many
| samples (or intervals) is an artifact, I didn't find a reliable way
to get
| what I needed other than increasing the number of samples.
|
	It takes some trial and error, but it should be possible to get the
same effect with a much faster render. Of course, it all depends on
whether you can afford to waste more time fiddling with the
parameters or simply letting the computer render in the background
while you're doing something else ;) To make accurate comparisons,
you also need to make sure that the maximum number of samples is the
same in both cases (I'm not talking of the second parameter of the
"samples" keyword here, but of the number of potential samples which
is something in the order of min_samples * 2^aa_level). Then if you
set aa_threshold to 0 the results should be the same (and take
*forever* to render) so all you need to do is find a value that's low
enough to get the effect you want while not being exactly 0.

	Note that this only applies to subsurface scattering simulations: eg
in cases when you have a very dense nearly uniform media. If you have
a density with areas of very dense media and other areas that are
empty, you will need the increased number of samples

		Jerome
- --
******************************
*      Jerome M. Berger      *
* mailto:jbe### [at] ifrancecom *
*  http://jeberger.free.fr/  *
******************************
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.4 (GNU/Linux)

iD8DBQFB2cctqIYJdJhyixIRAh5MAJwNx+lC7Jw+TJml4MXtT7bAedh8MgCdHgrM
GTqdVZYfJXewymGtk1yKX4Y=
=uGCQ
-----END PGP SIGNATURE-----

Post a reply to this message

> | But I see the opposite,
> | as if the influence of the samples was "averaged" (since most
> samples are
> | deep inside, the effect of the surface ones is weaker). I would have
> | thought that the sampling algorithm (and colour calculation) would be
> | smarter than that, but maybe that's not so easy...
> |
> Do you have a sample code that produces this effect?

Yep, try this (povray 3.6.1, custom compile):

---------------------------------------
camera { location <0, 0, -10> look_at 0 }

light_source { <-20, 20, -20>, rgb 1 }

sphere {
  0, 3
  hollow
  pigment { color transmit 1 }
  interior {
    media {
      scattering { 5, rgb 3 }
      method 3
      intervals 1
      samples 10 // try also with 200
    }
  }
}
---------------------------------------

I've attached two images: "test1" has samples 10, "test2" has samples 200.

> It takes some trial and error, but it should be possible to get the
> same effect with a much faster render. Of course, it all depends on
> whether you can afford to waste more time fiddling with the
> parameters or simply letting the computer render in the background
> while you're doing something else ;) To make accurate comparisons, 
> you also need to make sure that the maximum number of samples is the
> same in both cases (I'm not talking of the second parameter of the
> "samples" keyword here, but of the number of potential samples which
> is something in the order of min_samples * 2^aa_level). Then if you
> set aa_threshold to 0 the results should be the same (and take
> *forever* to render) so all you need to do is find a value that's low
> enough to get the effect you want while not being exactly 0.

In my experiments, I did not see any significant difference in the output by
fiddling with aa. According to your explanation, "samples 10 aa_level 8
aa_threshold 0.0001" should give similar results to "samples 160 aa_level 4
aa_threshold 0.0001", and I don't see that with the above example. Could
you give a more clear example?

> Note that this only applies to subsurface scattering simulations: eg
> in cases when you have a very dense nearly uniform media.

Yes, my experiments are confined to uniform media. I wonder if this adaptive
sampling and antialiasing is only valid for non-uniform media (opposite to
your sentence above), maybe uniform media makes adaptive sampling
unnecessary...

Or maybe I'm making some mistakes.

-- 
light_source{9+9*x,1}camera{orthographic look_at(1-y)/4angle 30location
9/4-z*4}light_source{-9*z,1}union{box{.9-z.1+x clipped_by{plane{2+y-4*x
0}}}box{z-y-.1.1+z}box{-.1.1+x}box{.1z-.1}pigment{rgb<.8.2,1>}}//Jellby

Post a reply to this message

Attachments:
Download 'test1.jpg' (3 KB) Download 'test2.jpg' (3 KB)

Preview of image 'test1.jpg'

Preview of image 'test2.jpg'

> Interesting that the defaults for media would be set up as such:


You should see the thread "default media setting inconsistancies (?)" on
12/27/2004 (about 15 threads down). Assuming I gave correct information in
that thread (and I may be wrong; no one's confirmed), it should clarify
things for you.

 - Slime
 [ http://www.slimeland.com/ ]

Post a reply to this message