"Anthony D. Baye" <Sha### [at] spamnomorehotmailcom> wrote:
> Second note: Bryce 7 has the ability to bake Light Probes for HDRI, and I was
> wondering if this sort of thing could be done natively in povray using a fisheye
> camera.

I've created many HDR images in POV-Ray for use as image-based lighting (IBL)
and environment maps.

Use a spherical camera and render a scene at a 2:1 aspect ratio (e.g., +w2048
+h1024) and output the image as EXR using +fe

You can then wrap the image around a sphere using map_type 1 for the image_map
and a finish having some ambient and/or emissive value (e.g., ambient 1 emission
1)

-------------------------------------------------
www.McGregorFineArt.com

Post a reply to this message

clipka <ano### [at] anonymousorg> wrote:
> Materials from stock commercial 3D models (like, say, from Renderosity,
> DAZ3D or the like) suffer from this very same problem: Someone starts
> with their favourite light setup, tweaks a figure's material until it
> looks good, and then sells it for money. But it never fits with other
> people's models, because they are tweaked to look best with some other
> light setup.

This is precisely why I included the parameters RC3M_Ambient and RC3M_Diffuse in
my Object Collection module RC3Metal.

Post a reply to this message

"Robert McGregor" <rob### [at] mcgregorfineartcom> wrote:
> "Anthony D. Baye" <Sha### [at] spamnomorehotmailcom> wrote:
> > I have six light sources.  There is no difference in lighting between the two
> > samples. Granted the part I'm wondering about is in moderate shadow, but it
> > should be receiving some of the light.  The primary texture is at least the same
> > color when rendered with radiosity, which indicates to me that the color of the
> > secondary relies too much on it's ambient value.
>
> Adding to the finish(es) an emission value that matches the existing ambient
> value should fix that issue completely when using radiosity.
>
> -------------------------------------------------
> www.McGregorFineArt.com

So I guess the question is: Does the emission keyword supercede ambient? should
the textures in the libraries be updated to reflect that?

and thanks for the tips on HDRI.  Now I just need a step-by-step tutorial on how
to use it in scene.

Regards,
A.D.B.

Post a reply to this message

> "Robert McGregor" <rob### [at] mcgregorfineartcom> wrote:
>> "Anthony D. Baye" <Sha### [at] spamnomorehotmailcom> wrote:
>>> I have six light sources.  There is no difference in lighting between the two
>>> samples. Granted the part I'm wondering about is in moderate shadow, but it
>>> should be receiving some of the light.  The primary texture is at least the same
>>> color when rendered with radiosity, which indicates to me that the color of the
>>> secondary relies too much on it's ambient value.
>>
>> Adding to the finish(es) an emission value that matches the existing ambient
>> value should fix that issue completely when using radiosity.
>>
>> -------------------------------------------------
>> www.McGregorFineArt.com
>
> So I guess the question is: Does the emission keyword supercede ambient? should
> the textures in the libraries be updated to reflect that?

NO ! and NO !

>
> and thanks for the tips on HDRI.  Now I just need a step-by-step tutorial on how
> to use it in scene.
>
> Regards,
> A.D.B.
>

ambient is meant as a cheap alternative to diffuse inter-reflection that 
radiosity does.
emission is meant for textures that actualy *emit* light. Adding 
emission to most textures will wrongfully make them emit light in 
radiosity scenes.

emission never supercede ambient, they are totaly different beasts.
Most textures using high ambient are probably for simulating some amount 
of translucency. For example, PinkAlabaster is a fairly translucent 
stone and the texture for it have a high ambient. Beter done using 
subsurface that ambient, but much slower...

If you look at the metal textures, they have insane ambient that looks 
resonably good when alone with a black background. Add any environment, 
and it becomes unrealistic and often bad looking.

Post a reply to this message

"Anthony D. Baye" <Sha### [at] spamnomorehotmailcom> wrote:
> So I guess the question is: Does the emission keyword supercede ambient?

It doesn't supersede ambient per se, but supersedes one particular ad hoc usage
of ambient, which was to cause objects to glow.

The rule of thumb for POV-Ray 3.7 is that if your object glows in the dark, use
emission.  If it doesn't, use ambient.

> should
> the textures in the libraries be updated to reflect that?

I would say no, because then you'd have everything glowing when radiosity is
used.  This was already a problem with the stock textures in POV-Ray 3.6
(especially with the metals), though not everyone may have noticed it.

Post a reply to this message

Am 30.11.2014 15:25, schrieb Robert McGregor:
> "Anthony D. Baye" <Sha### [at] spamnomorehotmailcom> wrote:
>> I have six light sources.  There is no difference in lighting between the two
>> samples. Granted the part I'm wondering about is in moderate shadow, but it
>> should be receiving some of the light.  The primary texture is at least the same
>> color when rendered with radiosity, which indicates to me that the color of the
>> secondary relies too much on it's ambient value.
>
> Adding to the finish(es) an emission value that matches the existing ambient
> value should fix that issue completely when using radiosity.

Uh... don't! That would just be a nasty workaround, not a solution.

Post a reply to this message

Am 30.11.2014 18:40, schrieb Anthony D. Baye:

> So I guess the question is: Does the emission keyword supercede ambient? should
> the textures in the libraries be updated to reflect that?

Absolutely not - while technically ambient and emission essentially do 
the same thing, they reflect two entirely different use cases, which the 
previous ambient-only mechanism could not both cater for simultaneously 
without different material definitions for radiosity and non-radiosity 
scenes.

As a matter of fact, the two have been separated in POV-Ray 3.7 so that 
carefully designed textures can now be used for both radiosity and 
non-radiosity renders alike without change; the key is to design the 
texture to work in a radiosity scene first (for reflective materials 
make sure to use a light probe in your testing environment or otherwise 
provide some environment to reflect), then add some ambient to lighten 
up shadows in non-radiosity renders.

Post a reply to this message

clipka <ano### [at] anonymousorg> wrote:
> Am 30.11.2014 15:25, schrieb Robert McGregor:
> > Adding to the finish(es) an emission value that matches the existing ambient
> > value should fix that issue completely when using radiosity.
>
> Uh... don't! That would just be a nasty workaround, not a solution.

I completely agree it's not the "right" way to go about it, but it is a "quick
fix" for such a problem as he originally described. For example, here is "Basic
Scene 10 - Night moon partly cloudy sky:"

1) the top image is the original (non-radiosity)

2) the middle version uses a default radiosity {} block without changing
anything else

3) the bottom is the same as #2 but replacing ambient with emission. Quick fix.
Not identical, but to my eye renders using radiosity always looks better anyway.

-------------------------------------------------
www.McGregorFineArt.com

Post a reply to this message

Attachments:
Download 'basicscene10.png' (539 KB)

Preview of image 'basicscene10.png'

"Anthony D. Baye" <Sha### [at] spamnomorehotmailcom> wrote:

> and thanks for the tips on HDRI.  Now I just need a step-by-step tutorial on how
> to use it in scene.

There's not much to it really:

1) To create an HDR image for use as IBL just render your scene with a spherical
camera; here's the default from the Insert menu:

camera {
  spherical
  location <0,0,0>      // position
  look_at  <0,0,1>      // view
  angle 360             // horizontal degrees
        180             // vertical degrees
}

Render the scene using command-line (for example) +fe +w2048 +h1024 to produce
an EXR image.

2) Then you can use this resulting EXR image as the IBL for any scene by
wrapping it on a huge sphere to act as the environment lighting, something like
(assuming radiosity):

sphere { 0, 10000 hollow inverse
   pigment { image_map { exr "SomeEXR" interpolate 3 map_type 1 }}
   finish { emission 2 diffuse 0 }
}

If there are any unwanted speckle-artifacts from the EXR lighting in a scene
(likely due to sampling errors over very large differences in neighboring pixel
colors) I make a smaller, heavily blurred version of the EXR in GIMP or PS and
use that instead, say 512x256 pixels.

-------------------------------------------------
www.McGregorFineArt.com

Post a reply to this message

> clipka <ano### [at] anonymousorg> wrote:
>> Am 30.11.2014 15:25, schrieb Robert McGregor:
>>> Adding to the finish(es) an emission value that matches the existing ambient
>>> value should fix that issue completely when using radiosity.
>>
>> Uh... don't! That would just be a nasty workaround, not a solution.
>
> I completely agree it's not the "right" way to go about it, but it is a "quick
> fix" for such a problem as he originally described. For example, here is "Basic
> Scene 10 - Night moon partly cloudy sky:"
>
> 1) the top image is the original (non-radiosity)
>
> 2) the middle version uses a default radiosity {} block without changing
> anything else
>
> 3) the bottom is the same as #2 but replacing ambient with emission. Quick fix.
> Not identical, but to my eye renders using radiosity always looks better anyway.
>
> -------------------------------------------------
> www.McGregorFineArt.com
>

This is a special case where ambient was used to make the sky glow. In 
/THIS/ particuliar case, replacing ambient by emission is correct as 
this texture *is* meant to be a source of illumination.
There are a few similar textures, like the lightnings and various skys 
as well as the "Luminous" finish.



Alain

Post a reply to this message