 |
|
|
|
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Continuing the RSOCP theme (see thread started at
<web.43eafcf196b04c3582fc96790@news.povray.org>), I rendered a radiosity
version, with 3 spheres. Appart from wrong settings for the media sampling
(and long render time ...), I was puzzled by the reflection of the top
sphere on the main one: the spherical pattern of the media appears black!!
And I also noticed that if I add a fourth sphere, no reflection occurs for
any sphere...
Here is the render as it was when I stopped it. If someone has suggestions
....
(I can post the source code if necessary).
Bruno.
Post a reply to this message
Attachments:
Download 'scr_rsocp_6.jpg' (69 KB)
Preview of image 'scr_rsocp_6.jpg'
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Sounds like max_trace_level to me. I like the image.
--
-s
5TF!
Post a reply to this message
|
|
| |
| |
|
|
From: Alain
Subject: Re: Another version of my RSOCP with media spies: problem with radiosity.
Date: 11 Feb 2006 10:25:30
Message: <43ee01ea$1@news.povray.org>
|
|
|
| |
| |
|
|
Bruno Cabasson nous apporta ses lumieres en ce 11/02/2006 09:21:
> Continuing the RSOCP theme (see thread started at
> <web.43eafcf196b04c3582fc96790@news.povray.org>), I rendered a radiosity
> version, with 3 spheres. Appart from wrong settings for the media sampling
> (and long render time ...), I was puzzled by the reflection of the top
> sphere on the main one: the spherical pattern of the media appears black!!
> And I also noticed that if I add a fourth sphere, no reflection occurs for
> any sphere...
>
> Here is the render as it was when I stopped it. If someone has suggestions
> ....
>
> (I can post the source code if necessary).
>
> Bruno.
>
>
> ------------------------------------------------------------------------
>
You probably need to increase max_trace_level.
Make the calculations:
Enter the first media container, reflection on the sphere, exiting the container = 3
levels
Add 3 more levels for the second sphere (2 if you only encounter the container) = 6 or
5 levels total.
Add 1 more for reaching the background.
So, you need a strict minimum max_trace_level of 7, 10 if you bounce on 2 spheres and
encounter a
third one.
Don't fear setting it to high.
--
Alain
-------------------------------------------------
WARNING: The consumption of alcohol may leave you wondering what the hell happened to
your bra and
panties.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Alain <ele### [at] netscape net> wrote:
> Bruno Cabasson nous apporta ses lumieres en ce 11/02/2006 09:21:
> > Continuing the RSOCP theme (see thread started at
> > <web.43eafcf196b04c3582fc96790@news.povray.org>), I rendered a radiosity
> > version, with 3 spheres. Appart from wrong settings for the media sampling
> > (and long render time ...), I was puzzled by the reflection of the top
> > sphere on the main one: the spherical pattern of the media appears black!!
> > And I also noticed that if I add a fourth sphere, no reflection occurs for
> > any sphere...
> >
> > Here is the render as it was when I stopped it. If someone has suggestions
> > ....
> >
> > (I can post the source code if necessary).
> >
> > Bruno.
> >
> >
> > ------------------------------------------------------------------------
> >
> You probably need to increase max_trace_level.
> Make the calculations:
> Enter the first media container, reflection on the sphere, exiting the container = 3
levels
> Add 3 more levels for the second sphere (2 if you only encounter the container) = 6
or 5 levels total.
> Add 1 more for reaching the background.
> So, you need a strict minimum max_trace_level of 7, 10 if you bounce on 2 spheres
and encounter a
> third one.
> Don't fear setting it to high.
>
> --
> Alain
> -------------------------------------------------
> WARNING: The consumption of alcohol may leave you wondering what the hell happened
to your bra and
> panties.
You're right. I increased max_trace_level, and it is ok for the reflection.
I should have found this myself, shame on me ;-) Now, my concern for this
scene is to set proper parameters for the media, in order to have no
artifacts and an aceptable render time.
Bruno
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
like the image to. :-)
Is there any way to increase max_trace_level above 255 ? or would that
require some code changes?
the make-cloud macro's use a ton of medial layers, and I almost at 255
witch a single cloud. I can only slightly overlap clouds if they are rotated
45 degrees.
I'm using this in a 3000 frame animation, and render speed has not been a
problem until now..
Post a reply to this message
Attachments:
Download 'wow2.png' (279 KB)
Preview of image 'wow2.png'
|
|
| |
| |
|
|
From: Alain
Subject: Re: Another version of my RSOCP with media spies: problem with radiosity.
Date: 12 Feb 2006 11:09:08
Message: <43ef5da4$1@news.povray.org>
|
|
|
| |
| |
|
|
Jaap nous apporta ses lumieres en ce 12/02/2006 04:58:
> like the image to. :-)
>
> Is there any way to increase max_trace_level above 255 ? or would that
> require some code changes?
>
> the make-cloud macro's use a ton of medial layers, and I almost at 255
> witch a single cloud. I can only slightly overlap clouds if they are rotated
> 45 degrees.
>
> I'm using this in a 3000 frame animation, and render speed has not been a
> problem until now..
>
>
>
> ------------------------------------------------------------------------
>
You can use adc_bailout to limit the deepth you reach. When you exeed max_trace_level,
the colour
returned is black. When you stop due to adc_bailout, the colour returned is whatever
you have up to
that point.
You need to change the code to exeed max_trace_level 255. Probably limited by the size
of some aray.
--
Alain
-------------------------------------------------
Light travels faster than sound. That is why some people appear bright until you hear
them speak.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Alain <ele### [at] netscapenet> wrote:
> Jaap nous apporta ses lumieres en ce 12/02/2006 04:58:
> > like the image to. :-)
> >
> > Is there any way to increase max_trace_level above 255 ? or would that
> > require some code changes?
> >
> > the make-cloud macro's use a ton of medial layers, and I almost at 255
> > witch a single cloud. I can only slightly overlap clouds if they are rotated
> > 45 degrees.
> >
> > I'm using this in a 3000 frame animation, and render speed has not been a
> > problem until now..
> >
> >
> >
> > ------------------------------------------------------------------------
> >
> You can use adc_bailout to limit the deepth you reach. When you exeed
max_trace_level, the colour
> returned is black. When you stop due to adc_bailout, the colour returned is whatever
you have up to
> that point.
> You need to change the code to exeed max_trace_level 255. Probably limited by the
size of some aray.
>
> --
> Alain
> -------------------------------------------------
> Light travels faster than sound. That is why some people appear bright until you
hear them speak.
OK. thanks for the tip! I'll try it this night. Today I rendered the
attached image. Still some artifacts, where the media is very thin I
think). Again, I use 1 interval and 200 samples. Render tme: 18:30 hours...
for 512*384 and radiosity.
Cheers.
Bruno.
Post a reply to this message
Attachments:
Download 'rsocp_7.jpg' (24 KB)
Preview of image 'rsocp_7.jpg'
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Here is the final 512*384 render. There remain a few little artifacts, but
I'll leave it so.
Regards.
Post a reply to this message
Attachments:
Download 'rsocp_8.jpg' (73 KB)
Preview of image 'rsocp_8.jpg'
|
|
| |
| |
|
|
|
|
| |
