 |
|
|
|
|
|
| |
| |
|
|
|
|
| |
| |
|
|
I tried to use media but cant get it to work properly. Still when my xscan
paramter = 4 the reflection is too bright, it should be much weaker than when
xscan = 3. Here is how i modified the absorber object. The only other change I
made was to the photons global setting where i added "media 100". Am i doing
something wrong?
object{
box
{
<-extent,-extent,t_st> <extent,extent,t_st+t_abs>
hollow
pigment{rgbt 1}
finish
{
reflection{0,1 fresnel}
conserve_energy
}
interior{ior cmt_ior fade_distance cmt_abs fade_power 1000
media{
absorption <1,1,1>
intervals 3
samples 1,1 }
}
PhotonTarget(no, yes, yes, yes)
}
}
Alain <kua### [at] videotron ca> wrote:
> > Thanks for looking in to this clipka,
> >
> > can you explain a little more about how i could use media as a workaround?
> >
> > also, is the bug regarding TIR easy to explain? I will be relying on TIR in my
> > scenes which will have different angles than those in the code I included in my
> > earlier message. If it might go wrong its useful to know before so I can keep an
> > eye out for problems
> >
> > stu
> >
> >
> >
> > clipka <ano### [at] anonymousorg> wrote:
> >
> >>
> >> Examining your problem, I did indeed find a bug in the handling of the
> >> interior fade feature (wrong interior being applied under certain
> >> circumstances), which I suspect to be relevant in your case. (I also
> >> found a flaw regarding total internal reflections, but so far I did not
> >> manage to come up with a situation where it would manifest as a bug, so
> >> that one can probably be ruled out.)
> >>
> >> The interior fade bug is also present in POV-Ray 3.6.2, so I don't think
> >> it will be fixed in POV-Ray 3.7.0 release proper.
> >>
> >> One possible workaround might be to use media instead.
> >
> >
> >
> >
> You add "hollow" to the object. This enable the object to contain a media.
> Next, in the interior block, you add
> media{absorbtion Colour}
>
> If you want to let the red pass but absorb the green and blue, you need
> {absorbtion <0,1,1>}
>
> If you also use fade_dolor Colour, use <1,1,1>-Colour for the media.
>
>
>
> Alain
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Am 13.07.2012 18:24, schrieb Crippso:
>
> I tried to use media but cant get it to work properly. Still when my xscan
> paramter = 4 the reflection is too bright, it should be much weaker than when
> xscan = 3. Here is how i modified the absorber object. The only other change I
> made was to the photons global setting where i added "media 100". Am i doing
> something wrong?
The "media" keyword in the global photons block is irrelevant for your
case; it is intended for scattering media.
> object{
(BTW, while wrapping your "box" in an "object" statement doesn't hurt,
it's totally unnecessary; you need "object" only when you have
#declare'd an object and want to add it to the scene.)
> box
> {
> <-extent,-extent,t_st> <extent,extent,t_st+t_abs>
> hollow
> pigment{rgbt 1}
> finish
> {
> reflection{0,1 fresnel}
> conserve_energy
> }
>
> interior{ior cmt_ior fade_distance cmt_abs fade_power 1000
Remove the "fade_distance" and "fade_power" statements; after all,
you're trying to replace them with media.
> media{
> absorption <1,1,1>
You may need to increase absorption; note that its parameter is not
limited to the range [0..1].
Theoretically, you should set:
absorption = <1,1,1> / cmt_abs
to get the same results as you'd get with properly computed fade.
> intervals 3
> samples 1,1 }
Remove the "intervals" and "samples" statements; the default values
should be perfectly fine for uniform media as in your case.
>
> }
>
> PhotonTarget(no, yes, yes, yes)
> }
> }
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
I tried to include these changes but didn't get the right results.
if i put xscan=0, i have the simple case of photons travelling through the
absorber and reflected back through it, always in the z direction. Using your
media approach i dont seem to be able to change the brightness of the reflected
scene by changing cmt_abs paramter. I also noticed if i set cmt_abs = 0 there is
no error flagged up because of the divide by zero!
clipka <ano### [at] anonymousorg> wrote:
> Am 13.07.2012 18:24, schrieb Crippso:
> >
> > I tried to use media but cant get it to work properly. Still when my xscan
> > paramter = 4 the reflection is too bright, it should be much weaker than when
> > xscan = 3. Here is how i modified the absorber object. The only other change I
> > made was to the photons global setting where i added "media 100". Am i doing
> > something wrong?
>
> The "media" keyword in the global photons block is irrelevant for your
> case; it is intended for scattering media.
>
>
> > object{
>
> (BTW, while wrapping your "box" in an "object" statement doesn't hurt,
> it's totally unnecessary; you need "object" only when you have
> #declare'd an object and want to add it to the scene.)
>
> > box
> > {
> > <-extent,-extent,t_st> <extent,extent,t_st+t_abs>
> > hollow
> > pigment{rgbt 1}
> > finish
> > {
> > reflection{0,1 fresnel}
> > conserve_energy
> > }
> >
> > interior{ior cmt_ior fade_distance cmt_abs fade_power 1000
>
> Remove the "fade_distance" and "fade_power" statements; after all,
> you're trying to replace them with media.
>
> > media{
> > absorption <1,1,1>
>
> You may need to increase absorption; note that its parameter is not
> limited to the range [0..1].
>
> Theoretically, you should set:
>
> absorption = <1,1,1> / cmt_abs
>
> to get the same results as you'd get with properly computed fade.
>
> > intervals 3
> > samples 1,1 }
>
> Remove the "intervals" and "samples" statements; the default values
> should be perfectly fine for uniform media as in your case.
>
> >
> > }
> >
> > PhotonTarget(no, yes, yes, yes)
> > }
> > }
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
ah, it looks like i need to make other objects hollow too
"Crippso" <nomail@nomail> wrote:
> I tried to include these changes but didn't get the right results.
>
> if i put xscan=0, i have the simple case of photons travelling through the
> absorber and reflected back through it, always in the z direction. Using your
> media approach i dont seem to be able to change the brightness of the reflected
> scene by changing cmt_abs paramter. I also noticed if i set cmt_abs = 0 there is
> no error flagged up because of the divide by zero!
>
> clipka <ano### [at] anonymousorg> wrote:
> > Am 13.07.2012 18:24, schrieb Crippso:
> > >
> > > I tried to use media but cant get it to work properly. Still when my xscan
> > > paramter = 4 the reflection is too bright, it should be much weaker than when
> > > xscan = 3. Here is how i modified the absorber object. The only other change I
> > > made was to the photons global setting where i added "media 100". Am i doing
> > > something wrong?
> >
> > The "media" keyword in the global photons block is irrelevant for your
> > case; it is intended for scattering media.
> >
> >
> > > object{
> >
> > (BTW, while wrapping your "box" in an "object" statement doesn't hurt,
> > it's totally unnecessary; you need "object" only when you have
> > #declare'd an object and want to add it to the scene.)
> >
> > > box
> > > {
> > > <-extent,-extent,t_st> <extent,extent,t_st+t_abs>
> > > hollow
> > > pigment{rgbt 1}
> > > finish
> > > {
> > > reflection{0,1 fresnel}
> > > conserve_energy
> > > }
> > >
> > > interior{ior cmt_ior fade_distance cmt_abs fade_power 1000
> >
> > Remove the "fade_distance" and "fade_power" statements; after all,
> > you're trying to replace them with media.
> >
> > > media{
> > > absorption <1,1,1>
> >
> > You may need to increase absorption; note that its parameter is not
> > limited to the range [0..1].
> >
> > Theoretically, you should set:
> >
> > absorption = <1,1,1> / cmt_abs
> >
> > to get the same results as you'd get with properly computed fade.
> >
> > > intervals 3
> > > samples 1,1 }
> >
> > Remove the "intervals" and "samples" statements; the default values
> > should be perfectly fine for uniform media as in your case.
> >
> > >
> > > }
> > >
> > > PhotonTarget(no, yes, yes, yes)
> > > }
> > > }
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
>
> I tried to use media but cant get it to work properly. Still when my xscan
> paramter = 4 the reflection is too bright, it should be much weaker than when
> xscan = 3. Here is how i modified the absorber object. The only other change I
> made was to the photons global setting where i added "media 100". Am i doing
> something wrong?
>
> object{
>
> box
> {
> <-extent,-extent,t_st> <extent,extent,t_st+t_abs>
> hollow
> pigment{rgbt 1}
> finish
> {
> reflection{0,1 fresnel}
> conserve_energy
> }
>
> interior{ior cmt_ior fade_distance cmt_abs fade_power 1000
> media{
> absorption <1,1,1>
> intervals 3
With the default sampling method, intervals MUST stay at it's default of
1. Changing it only make the render MUCH slower.
If you need more samples, increase the samples count.
> samples 1,1 }
Also, with the default sampling method, samples should be at least 3
defaulting to 10.
There is only a single samples value used, the second is always silently
ignored.
>
> }
>
> PhotonTarget(no, yes, yes, yes)
This is a NOP statement. It set NOT a photons target, the default state.
The refraction and reflection switches are only relevent IF the object
is a target. Collect on is the default.
ANY reflective or refractive object that receive secondary photons WILL
reflect or refract them.
> }
> }
>
Alain
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Am 14.07.2012 03:05, schrieb Alain:
>> media{
>> absorption <1,1,1>
>> intervals 3
> With the default sampling method, intervals MUST stay at it's default of
You make it sound like any non-default value would cause the computer to
explode :-)
Then again, yes - it /does/ cause /rendering times/ to explode :-P
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
|
|
