> Then, it may be that the high reflection value (0.9) make them almost
> invisible. The more reflective a surface is, the less it will show
> photons effects. That's the main reason why it's recomended to set
> collect off for highly reflective surfaces.

Additionally, check if max_trace_level for the photons is high enough to
actually reach the plane. If the glass object has lots of holes in it, a
photon won't pass through, but rather get stuck somewhere along the way.
Note that you can define different max_trace_level for photons and the
actual raytraced rays, but I'm unsure if they use the same keyword... You'd
have to look at the docs for that.

Regards,
Tim

-- 
"Tim Nikias v2.0"
Homepage: <http://www.nolights.de>

Post a reply to this message

Does your light_source emit photons?

-S
5TF!

Post a reply to this message

stm31415 wrote:
> Does your light_source emit photons?
> 
> -S
> 5TF!
> 
There is no lightsource as such. This is a scene running on radiosity. 
The light is caused by a HDR image.

-- 
Maurice

Post a reply to this message

That explains everything. Only lightsources cause photons, if you'd want to
have caustics created by radiosity, you'd have to use ridiculously high
settings for the tracing depth.

A workaround would be to do a two-pass render. On the first pass, you only
render with a lightsource and photons, which you save to disk using
save_file in the photons-block. On the second pass, you render using
radiosity, but instead of placing an actual lightsource, you just load the
photons using load_file. You'd get the photons without the lightsource that
way.

-- 
"Tim Nikias v2.0"
Homepage: <http://www.nolights.de>

Post a reply to this message

Alain wrote:
> Maurice nous apporta ses lumieres ainsi en ce 2005-01-02 10:30... :
> 
>> Alain wrote:
>>
>>> Maurice nous apporta ses lumieres ainsi en ce 2005-01-02 04:37... :
>>>
>>>> Hi,
>>>>
>>>> I have this blob made up of 5000 glass spheres inside a spherical 
>>>> HDR. [blabla...]
>>>> }
>>>>
>>>> I can't find out what is wrong. Any suggestions are welcome.
>>>>
>>>> ------------------------------------------------------------------------ 
>>>>
>>>>
>>> What are the photons settings for the floor? Does it collect them?
>>
>>
>>
>> The floor:
>>
>> plane{ <0,1,0>, 0
>>   texture{
>>     pigment{checker color rgb<1,1,1>*1.2 color rgb<1,1,1>}
>>     finish{ambient 0.1 diffuse 0.8 metallic reflection{0.9}}
>>   }
>>   no_radiosity
>> }
>>
>> It has no photons section...
>>
> No photons section = collect on  target off  OK
> Then, it may be that the high reflection value (0.9) make them almost 
> invisible. The more reflective a surface is, the less it will show 
> photons effects. That's the main reason why it's recomended to set 
> collect off for highly reflective surfaces.

I think you might be right. When I switch off the reflective finish of 
the floor I see a hint of caustics on the floor. Is there any way to 
increase the strenght of the caustics? Can't seem to find anything in 
the manual that promises to do this. I found the caustics keyword for 
interior but whatever value I use it doesn't seem to give the right effect.

-- 
Maurice

Post a reply to this message

Do what Tim said, using saved data. Radiosity does give very pale caustics,
but if you use the two pass method they will be clear and bright.

-S
5TF!

Post a reply to this message

stm31415 wrote:
> Do what Tim said, using saved data. Radiosity does give very pale caustics,
> but if you use the two pass method they will be clear and bright.
> 
> -S
> 5TF!
> 

Thanks Tim and stm31415 for suggesting the two pass method. In the last 
few trial renders I've gotten some interesting results. Now, at least 
photons are being shot. Obviously more experimenting is needed with the 
light placement and stuff like that, but at least I'm back on the road 
again.

Thanks to all
-- 
Maurice

Post a reply to this message

Hi Maurice,

    Just curious how things went and if you have an image to show us :)

Regards,
George


Maurice <cel### [at] nospamhotmailcom> wrote:
> stm31415 wrote:
> > Do what Tim said, using saved data. Radiosity does give very pale caustics,
> > but if you use the two pass method they will be clear and bright.
> >
> > -S
> > 5TF!
> >
>
> Thanks Tim and stm31415 for suggesting the two pass method. In the last
> few trial renders I've gotten some interesting results. Now, at least
> photons are being shot. Obviously more experimenting is needed with the
> light placement and stuff like that, but at least I'm back on the road
> again.
>
> Thanks to all
> --
> Maurice

Post a reply to this message

George Pantazopoulos wrote:
> Hi Maurice,
> 
>     Just curious how things went and if you have an image to show us :)
> 
> Regards,
> George
> 
> 
> Maurice <cel### [at] nospamhotmailcom> wrote:
> 
>>stm31415 wrote:
>>
>>>Do what Tim said, using saved data. Radiosity does give very pale caustics,
>>>but if you use the two pass method they will be clear and bright.
>>>
>>>-S
>>>5TF!
>>>
>>
>>Thanks Tim and stm31415 for suggesting the two pass method. In the last
>>few trial renders I've gotten some interesting results. Now, at least
>>photons are being shot. Obviously more experimenting is needed with the
>>light placement and stuff like that, but at least I'm back on the road
>>again.
>>
>>Thanks to all
>>--
>>Maurice
> 
> 
> 
> 

Well, not very proud of the current result really.
I rendered the scene with the HDR image off and a cylinderical 
lightsource over the blob. Saving the photon and radiosity data.

The caustics seemed to look okay.

Then I switched off the lightsource put the HDR image back on and upped 
the reflection of the floor to 0.5.

The result is attached.

:-(

The caustics are now transfomred to shadows... I'm now renderingwith a 
less reflective floor (0.1). But I'm afraid I'll loose the effect I was 
going for.

-- 
Maurice

Post a reply to this message

Attachments:
Download 'newt.jpg' (72 KB)

Preview of image 'newt.jpg'

>
> Well, not very proud of the current result really.
> I rendered the scene with the HDR image off and a cylinderical
> lightsource over the blob. Saving the photon and radiosity data.
>

Pardon my ignorance, but I don't see a need for radiosity in your image. Why
is it there? I would try disabling radiosity for now and tweak your photons
settings. This would simplify your life by eliminate a big variable from
the equation :)


> The caustics are now transfomred to shadows... I'm now renderingwith a
> less reflective floor (0.1). But I'm afraid I'll loose the effect I was
> going for.
>

What exact effect are you looking for, and what is the diffuse component of
your floor?

-George

Post a reply to this message