High!


increase the thickness of the rings to 200 metres (rather than 20, which 
would be the real thing) to get the difference object rendered correctly 
at the scale I used for my Solar System (1 POV unit = 13477 kms).		

I wonder how to get the shadow of the planet also extending across the 
unlit side of the rings (whose ambient is 0.3, as in reality, some 
sunlight is scattered through to the "lower" side)... should I try a 
negative spotlight?

See you in Khyberspace!

Yadgar

Post a reply to this message

Attachments:
Download '2009-11-10 saturn, take 8.jpg' (27 KB)

Preview of image '2009-11-10 saturn, take 8.jpg'

> I wonder how to get the shadow of the planet also extending across the 
> unlit side of the rings (whose ambient is 0.3, as in reality, some 
> sunlight is scattered through to the "lower" side)... should I try a 
> negative spotlight?

If you're using POV-Ray 3.7, try the diffuse backside illumination 
feature: Just add another float to the "diffuse" statement. This will 
have part of the illumination filter through to the other side 
(presuming that the rings are implemented as infinitely thin disks, 
otherwise you'd have to have radiosity enabled to make it work).

Post a reply to this message

High!

clipka wrote:

> If you're using POV-Ray 3.7, try the diffuse backside illumination 
> feature: Just add another float to the "diffuse" statement. This will 
> have part of the illumination filter through to the other side 
> (presuming that the rings are implemented as infinitely thin disks, 
> otherwise you'd have to have radiosity enabled to make it work).

Sounds tempting... but does a version for Debian Linux already exist? 
Otherwise, I will have to try it on my slow old Pentium III-933 which 
runs Windows XP...

No, they are not disks (which would have no interior), but very thin 
(200 metres) cylinders!

See you in Khyberspace!

Yadgar

Post a reply to this message

>> If you're using POV-Ray 3.7, try the diffuse backside illumination 
>> feature: Just add another float to the "diffuse" statement. This will 
>> have part of the illumination filter through to the other side 
>> (presuming that the rings are implemented as infinitely thin disks, 
>> otherwise you'd have to have radiosity enabled to make it work).
> 
> Sounds tempting... but does a version for Debian Linux already exist? 
> Otherwise, I will have to try it on my slow old Pentium III-933 which 
> runs Windows XP...

No binaries, but Linux sources of beta.34 have long since been released 
by now.

> No, they are not disks (which would have no interior), but very thin 
> (200 metres) cylinders!

That might be a problem then without radiosity. While the rings are 
transparent, and therefore "leak" light through the whole volume and so 
you /will/ get an effect, it might end up less realistic.

As another option, with non-zero thickness media would be the most 
realistic way to go anyway.

Post a reply to this message

High!

clipka wrote:

> As another option, with non-zero thickness media would be the most 
> realistic way to go anyway.

Yes, I also thought about using media... but then, wouldn't it render 
dead slow?

See you in Khyberspace!

Yadgar

P.S. A funny situation - talking in the lingua franca of the Internet to 
a Colognian countryman living a mere 10 kilometres (6 miles) from here...

Post a reply to this message

> High!
> 
> clipka wrote:
> 
>> As another option, with non-zero thickness media would be the most 
>> realistic way to go anyway.
> 
> Yes, I also thought about using media... but then, wouldn't it render 
> dead slow?

I guess that'll depend a lot on the settings. Only light rays from a 
very shallow angle will experience any significant anisotropy while 
traversing the ring system, so presumably you can get away with very few 
samples.

> P.S. A funny situation - talking in the lingua franca of the Internet to 
> a Colognian countryman living a mere 10 kilometres (6 miles) from here...

Post a reply to this message

> High!
> 
> clipka wrote:
> 
>> If you're using POV-Ray 3.7, try the diffuse backside illumination 
>> feature: Just add another float to the "diffuse" statement. This will 
>> have part of the illumination filter through to the other side 
>> (presuming that the rings are implemented as infinitely thin disks, 
>> otherwise you'd have to have radiosity enabled to make it work).
> 
> Sounds tempting... but does a version for Debian Linux already exist? 
> Otherwise, I will have to try it on my slow old Pentium III-933 which 
> runs Windows XP...
> 
> No, they are not disks (which would have no interior), but very thin 
> (200 metres) cylinders!
> 
> See you in Khyberspace!
> 
> Yadgar

The disk object DO have an interior, just like a plane.
Altough they have an interior, I realy don't know how a media contained 
"in" a disk would behave.



Alain

Post a reply to this message

High!

clipka wrote:

> As another option, with non-zero thickness media would be the most 
> realistic way to go anyway.

Do you mean scattering media? I started experimenting with it, but 
wonder how to use a density_map (which should use the cylindrical 
pattern) with it... whatever overall scattering (while also having a 
density_map) color I tried, I never got any visible density pattern...

See you in Khyberspace!

Yadgar

Now playing: Television Man (Talking Heads)

Post a reply to this message