Is it possible to create a white non-emitting gas in radiosity 
environment without light sources (light_source)?

Post a reply to this message

> Is it possible to create a white non-emitting gas in radiosity
> environment without light sources (light_source)?

No, radiosity can't illuminate scattering media. You can still get the
absorption effect, of course.

However, you can fake the illumination with emitting media, if you carefully
construct a density pattern which has higher values in areas which would be
lit up. Unfortunately, this must be done manually.

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

Post a reply to this message

Slime wrote:
> However, you can fake the illumination with emitting media, if you carefully
> construct a density pattern which has higher values in areas which would be
> lit up. Unfortunately, this must be done manually.

In this case the faked scattering can brighten (or completely wipe) some 
shadowy areas, which can be inacceptable for some scenes. Also the gas 
(media) does not take into account the environmental lighting (for the 
scattering effect), because it does not scatter (diffusely reflect) light.

Will POV-Ray support at least type 1 scattering for radiosity environment?

Thank you,
William

Post a reply to this message

> In this case the faked scattering can brighten (or completely wipe) some
> shadowy areas, which can be inacceptable for some scenes. Also the gas
> (media) does not take into account the environmental lighting (for the
> scattering effect), because it does not scatter (diffusely reflect) light.


I'm not *exactly* sure what you're saying here, but I'm pretty sure that, if
scattering media interacted flawlessly with radiosity (and so was lit up
exactly correctly), it would be (theoretically) possible to sample it in
terms of density and illumination (as long as we're talking about scattering
type 1), and use the density data for an absorption media and the
illumination data for emission media, creating exactly the same result. So
what I was suggesting was that you guess at the illumination data and use it
in emission media to do just that.

Yes, this may brighten up some shadowy areas. But that would happen anyway
with scattering media that responded to radiosity: light bouncing off a
surface would brighten the media (that is, reflect off it) and then that
would hit the shadowed areas, brightening them.

> Will POV-Ray support at least type 1 scattering for radiosity environment?

I don't speak for the POV Team but AFAIK there aren't any plans for this. I
would also worry about its speed if it were implemented.

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

Post a reply to this message

William Peska wrote:
> 
> Will POV-Ray support at least type 1 scattering for radiosity environment?

You can use scattering media in radiosity scenes without problems and 
scattering media is taken into account by radiosity when you use 'media 
on'.  The only limitation is that for the media simulation the diffuse 
lighting is not taken into account.

Doing this would not be completely impossible but it would be quite 
different from how radiosity works on surfaces and increase in memory 
use and calculation time would be enormeous.  You can assume they will 
be similar to memory use of media photons compared to surface photons 
(just that for media photons only the memory use is much larger and not 
the calculation time).

Christoph

-- 
POV-Ray tutorials, include files, Sim-POV,
HCR-Edit and more: http://www.tu-bs.de/~y0013390/
Last updated 03 May. 2005 _____./\/^>_*_<^\/\.______

Post a reply to this message

Christoph Hormann wrote:
> You can use scattering media in radiosity scenes without problems and 
> scattering media is taken into account by radiosity when you use 'media 
> on'.  The only limitation is that for the media simulation the diffuse 
> lighting is not taken into account.

This limitation can be inacceptable for scenes which contain mainly/only 
diffuse lighting.

> Doing this would not be completely impossible but it would be quite 
> different from how radiosity works on surfaces and increase in memory 
> use and calculation time would be enormeous.

Maybe there is a resource effective way to do that.

William

Post a reply to this message

William Peska wrote:
> 
>> Doing this would not be completely impossible but it would be quite 
>> different from how radiosity works on surfaces and increase in memory 
>> use and calculation time would be enormeous.
> 
> Maybe there is a resource effective way to do that.

I thought i made it clear that the required time and memory would be 
enormeous no matter how efficient the implementation is.

Christoph

-- 
POV-Ray tutorials, include files, Sim-POV,
HCR-Edit and more: http://www.tu-bs.de/~y0013390/
Last updated 03 May. 2005 _____./\/^>_*_<^\/\.______

Post a reply to this message

Slime wrote:
>>In this case the faked scattering can brighten (or completely wipe) some
>>shadowy areas, which can be inacceptable for some scenes. Also the gas
>>(media) does not take into account the environmental lighting (for the
>>scattering effect), because it does not scatter (diffusely reflect) light.
> 
> I'm not *exactly* sure what you're saying here, but I'm pretty sure that, if
> scattering media interacted flawlessly with radiosity (and so was lit up
> exactly correctly), it would be (theoretically) possible to sample it in
> terms of density and illumination (as long as we're talking about scattering
> type 1), and use the density data for an absorption media and the
> illumination data for emission media, creating exactly the same result.

Faking scattering with emission will not create "exactly the same 
result" if the illumination of the nearby objects is different in the 
long run.

> Yes, this may brighten up some shadowy areas. But that would happen anyway
> with scattering media that responded to radiosity ...

The difference in the brightening can be significant - sufficiently to 
compromise the overall reality of the scene.

William

Post a reply to this message