In your global_settings block, try these photon values instead:

photons {count 200000 media 600, 2}

That seems to eliminate the 'banding' and fuzziness in the reflected laser
light.

Also, just as a safety measure, I added 'collect off' to the photon blocks in
your media box and the mirror box. According to the docs, it is 'on' by default.
But those object don't actually need photon effects to show up *on* them-- if I
correctly understand what 'collect' means. Although, I didn't notice any
render-time difference by turning it off.

In your two light sources, I added 'refraction off' to the photon blocks there.
Since your mirror is only 'reflecting' light, refraction is not needed. As far
as I can tell, the docs do not mention what the default behavior is-- on or off.

The most important change I made was in your media box:

media {
    scattering { 1, 1 extinction 0 }
    samples 6 // originally 100
      }

With this lower samples value, I don't see much (if any) difference in the
appearance of the media/light interaction...and the scene renders at least 10X
faster!

> Also, it is possible that the mirror reflect only the light directly from the
> source and not the scattered one (so we do not see the scattered beam on the
> mirror)?

If I understand your question, the only way to get a reflected light beam to
show up in POV-ray is to use photons (as well as media, to interact with the
'photons'.) Or do you mean the reflected appearance of the two light beams *in*
the mirror? That's just typical behavior of the mirror itself.

Post a reply to this message

Le 2025-12-06 à 00:57, Kenneth a écrit :
> In your global_settings block, try these photon values instead:
> 
> photons {count 200000 media 600, 2}
> 
> That seems to eliminate the 'banding' and fuzziness in the reflected laser
> light.
> 
> Also, just as a safety measure, I added 'collect off' to the photon blocks in
> your media box and the mirror box. According to the docs, it is 'on' by default.
> But those object don't actually need photon effects to show up *on* them-- if I
> correctly understand what 'collect' means. Although, I didn't notice any
> render-time difference by turning it off.
> 
> In your two light sources, I added 'refraction off' to the photon blocks there.
> Since your mirror is only 'reflecting' light, refraction is not needed. As far
> as I can tell, the docs do not mention what the default behavior is-- on or off.


Since version 3.6. By default, all light_source, except shadowless 
lights, have the following photons block :
{
	reflection on
	refraction on
}

Up to version 3.5, the states where the opposite.

Post a reply to this message

Alain Martel <kua### [at] videotronca> wrote:
>
> > In your two light sources, I added 'refraction off' to the photon blocks.
> > Since your mirror is only 'reflecting' light, refraction is not needed. As far
> > as I can tell, the docs do not mention what the default behavior is--
> > on or off.
>
>
> Since version 3.6. By default, all light_source, except shadowless
> lights, have the following photons block :
> {
>  reflection on
>  refraction on
> }
>

In the v3.8 documentation, I did finally find the default behavior at "3.4.3.4.3
Photon Global Settings"-- but it actually states the opposite(!) :

All photons default values:

Light_source:
area_light    : off
refraction    : off
reflection    : off

So there is some confusion here. I haven't yet looked at the online wiki docs to
double-check...

Post a reply to this message

On 12/5/25 09:51, radmac wrote:
> How does one create perfect mirror to reflect beam of light in scattering media?
> In attached scene, the reflected beam is clearly different from the incident
> one.

In your posted scene, mostly, what is causing the difference in 
intensity is that the beam for the light source to the mirror path is 
being calculated by the scattering media mechanism, while the reflected 
beam is being calculated by the photon mechanism.

Attached a modified version of your posted scene and four images (No 
AA). The scene has been updated to calculate both paths light intensity 
with the photon mechanism. For all changes look for '***'.

In the attached image the left column is POV-Ray version 3.8 beta 2; The 
right column is my yuqk fork - which has a fix for the photon banding.

The top row has 'media_interaction on' for the beam light - with the 
photons now used for both paths, the difference in intensity between 
paths due the scattering component is even more apparent. The bottom row 
turns 'media_interaction off' for the beam light - both beam path 
segments are now calculated solely by the photon mechanism.

Bill P.

Aside: The general rule with light sources and media is that the light 
source should be outside any media container by some small value.

Post a reply to this message

Attachments:
Download 'radmacphotonbeam_story.png' (110 KB) Download 'radmacphotonbeam.pov.txt' (2 KB)

Preview of image 'radmacphotonbeam_story.png'

hi,

"Kenneth" <kdw### [at] gmailcom> wrote:
> Alain Martel <kua### [at] videotronca> wrote:
> > ...
> > Since version 3.6. By default, all light_source, except shadowless
> > lights, have the following photons block :
> > {
> >  reflection on
> >  refraction on
> > }
> >
> In the v3.8 documentation, I did finally find the default behavior at "3.4.3.4.3
> Photon Global Settings"-- but it actually states the opposite(!) :
>
> All photons default values:
>
> Light_source:
> area_light    : off
> refraction    : off
> reflection    : off
>
> So there is some confusion here. I haven't yet looked at the online wiki docs to
> double-check...

thank you.  (before changing/updating the docs) can someone looking at the
sources confirm the defaults, please ?


regards, jr.

Post a reply to this message

"jr" <cre### [at] gmailcom> wrote:
>
> "Kenneth" <kdw### [at] gmailcom> wrote:
> >
> > So there is some confusion here. I haven't yet looked at the online wiki
> > docs to double-check...
>
> thank you.  (before changing/updating the docs) can someone looking at the
> sources confirm the defaults, please ?
>

In v3.8 beta 1, I've been running a series of tests (on just the photon
'refraction' defaults, not 'reflection' so far), using a modified version of the
OP's code here-- and there seems to be a mixed-up interaction of on/off/default
in the photons blocks of the light_source and the refracting object. In other
words, the *default* behavior changes(!) when other changes are made in one or
the other of the blocks. Sorry that I cannot be more specific yet, as I am still
testing and trying to make sense of the results. But as it's off-topic here,
I'll try to post a follow-up in a new thread.

I don't yet know if these odd results are specific to the 3.8 betas...and/or to
the OP's specific cylinder light_source...

Post a reply to this message

Le 2025-12-06 à 15:41, Kenneth a écrit :
> Alain Martel <kua### [at] videotronca> wrote:
>>
>>> In your two light sources, I added 'refraction off' to the photon blocks.
>>> Since your mirror is only 'reflecting' light, refraction is not needed. As far
>>> as I can tell, the docs do not mention what the default behavior is--
>>> on or off.
>>
>>
>> Since version 3.6. By default, all light_source, except shadowless
>> lights, have the following photons block :
>> {
>>   reflection on
>>   refraction on
>> }
>>
> 
> In the v3.8 documentation, I did finally find the default behavior at "3.4.3.4.3
> Photon Global Settings"-- but it actually states the opposite(!) :
> 
> All photons default values:
> 
> Light_source:
> area_light    : off
> refraction    : off
> reflection    : off
> 
> So there is some confusion here. I haven't yet looked at the online wiki docs to
> double-check...
> 
> 

And, in my experience, photons work just right WITHOUT any photons block 
in any light_source.

I only add one when I want a light to NOT shoot photons or to activate 
the area_light support.

Post a reply to this message

hi,

"Kenneth" <kdw### [at] gmailcom> wrote:
> ... In other words, the *default* behavior changes(!) when other changes
> are made in one or the other of the blocks. Sorry that I cannot be more
> specific yet, ...

thank you for digging up info / "investigating", appreciated.


regards, jr.

Post a reply to this message

On 12/7/25 10:34, Kenneth wrote:
> "jr" <cre### [at] gmailcom> wrote:
>>
>> "Kenneth" <kdw### [at] gmailcom> wrote:
>>>
>>> So there is some confusion here. I haven't yet looked at the online wiki
>>> docs to double-check...
>>
>> thank you.  (before changing/updating the docs) can someone looking at the
>> sources confirm the defaults, please ?
>>
> 
> In v3.8 beta 1, I've been running a series of tests (on just the photon
> 'refraction' defaults, not 'reflection' so far), using a modified version of the
> OP's code here-- and there seems to be a mixed-up interaction of on/off/default
> in the photons blocks of the light_source and the refracting object. In other
> words, the *default* behavior changes(!) when other changes are made in one or
> the other of the blocks. Sorry that I cannot be more specific yet, as I am still
> testing and trying to make sense of the results. But as it's off-topic here,
> I'll try to post a follow-up in a new thread.
> 
> I don't yet know if these odd results are specific to the 3.8 betas...and/or to
> the OP's specific cylinder light_source...
> 

I believe for the v3.7 / v3.8 versions, including yuqk, the 
documentation which is 'likely' wrong isn't with the light_source{} 
photon{} block, but the object's photon{} block - where the reflection 
and refraction defaults are 'on'.

---

What Kenneth is seeing is true too - there are some complicated 
dependencies and detail when dealing with photons.

For example, and using the updated scene I posted moving the beam light 
source outside the media box and using the photon mechanism for all 
light intensity calculations, if we set the beam light_source photon 
block's refraction to 'off' we get no photons deposited anywhere. The 
photon tracing stops at the first media box's transparent surface.

The same is true if we set the media box's photon block refraction 'off' 
while the light source's refraction is 'on' - no photons. The object's 
setting overrides the light's in that case.

However, if we turn the light source's photon refraction 'off' and the 
media box's refraction 'on', the light source's setting wins at the 
first media container's surface rather than the object's setting.

OK behavior I think, but not simple.

---

Secondary questions pop into my head: Does having an ior of other than 
one changes the specific behavior above? When we have transparent 
surfaces and ray reflection/fraction ray splitting, do the light's 
photon block's reflection/refraction settings continue to propagate in 
the same ways? What happens when we use, 'magic light transport', pass 
thru objects with light sources?

As Kenneth mentioned, there are too some differences in behavior 
depending upon the light source type. The cylindrical light working (at 
least sometimes) within objects containing media - bending our media 
light sources must be outside media containers rule.

Bill P.

Post a reply to this message

hi,

just to say.

William F Pokorny <ano### [at] anonymousorg> wrote:
> ...

thank you, and Kenneth (and CR just now) for the follow up.

@Kenneth.  got the code, *thank you* for investigating.

will get back to NG in a couple of days or so.


regards, jr.

Post a reply to this message