POV-Ray : Newsgroups : povray.general : radiosity clamping Server Time
17 Nov 2024 09:17:18 EST (-0500)
  radiosity clamping (Message 1 to 4 of 4)  
From: Peter Kryszkiewicz
Subject: radiosity clamping
Date: 15 Apr 2003 22:26:20
Message: <3e9cbf4c@news.povray.org>
I'm trying to simulate daylight coming through a window, like that classic
Vermeer scene on the cover of Computer Graphics by Foley et al.

I can't get my window pane to cast much light into the scene at all. I've
set ambient to 1 for the window and turned off all other ambient values in
the scene, then played with the variables. Is this a clamping problem in the
POV radiosity code, as I've seen mentioned before?

If I play with the POV source (volunteer to improve) is there a place where
I can easily add a capability to increase the effective brightness of an
arbitrary radiator to a value higher than ambient 1? I guess I'm talking
about simulating a very strong area light, which is what daylight coming
through a window is supposed to look like in a radiosity scene.


Post a reply to this message

From: Ive
Subject: Re: radiosity clamping
Date: 16 Apr 2003 03:20:49
Message: <3e9d0451$1@news.povray.org>
> I'm trying to simulate daylight coming through a window, like that classic
> Vermeer scene on the cover of Computer Graphics by Foley et al.
>
;) like this one?
http://news.povray.org/povray.binaries.images/27257/


> I can't get my window pane to cast much light into the scene at all. I've
> set ambient to 1 for the window and turned off all other ambient values in
> the scene, then played with the variables. Is this a clamping problem in the
> POV radiosity code, as I've seen mentioned before?
>
I do not think so.


> If I play with the POV source (volunteer to improve) is there a place where
> I can easily add a capability to increase the effective brightness of an
> arbitrary radiator to a value higher than ambient 1? I guess I'm talking
> about simulating a very strong area light, which is what daylight coming
> through a window is supposed to look like in a radiosity scene.
>
I did go with an ambient value of about 150 (if I remember right) for the
skysphere and it is lit only by radiosity to reproduce the way the Foley
cover was done.

good luck
-Ive


Post a reply to this message

From: Christoph Hormann
Subject: Re: radiosity clamping
Date: 16 Apr 2003 04:42:48
Message: <3E9D1788.FCC78AB2@gmx.de>
Peter Kryszkiewicz wrote:
> 
> I'm trying to simulate daylight coming through a window, like that classic
> Vermeer scene on the cover of Computer Graphics by Foley et al.
> 
> I can't get my window pane to cast much light into the scene at all. I've
> set ambient to 1 for the window and turned off all other ambient values in
> the scene, then played with the variables. Is this a clamping problem in the
> POV radiosity code, as I've seen mentioned before?

No.  The color values returned by sampling rays are not clipped in
POV-Ray.  Same applies for the samples stored in the cache (you can check
the cache file if you don't believe that).  The only place where values
are clipped is when reusing them.  Therefore you can use ambient finishes
with ambient > 1 without problems.  The only difficulty arises when the
stored illuminance values exceed 1.

> If I play with the POV source (volunteer to improve) is there a place where
> I can easily add a capability to increase the effective brightness of an
> arbitrary radiator to a value higher than ambient 1? I guess I'm talking
> about simulating a very strong area light, which is what daylight coming
> through a window is supposed to look like in a radiosity scene.

If you want an area light, use an area light.  There is no point in trying
generate light source effects with ambient objects.  Sunlight coming
through a window should be made with a light source placed outside,
skylight with an ambient sphere/sky_sphere.

Next MegaPOV will contain a patch for removing the clipping.  You will
however recognize that it won't make much difference in most situations.

Christoph

-- 
POV-Ray tutorials, include files, Sim-POV,
HCR-Edit and more: http://www.tu-bs.de/~y0013390/
Last updated 28 Feb. 2003 _____./\/^>_*_<^\/\.______


Post a reply to this message

From: Peter Kryszkiewicz
Subject: Re: radiosity clamping
Date: 16 Apr 2003 18:05:05
Message: <3e9dd391@news.povray.org>
Excellent result now!
I was suffering under the impression that I should normalize my ambients to
the range 0-1 and didn't realize I could use an unbounded value.

One of the scenes I'm trying to perfect can be seen at
http://www.worldwins.net/~peter/images/worldwins-sideview.jpg

My site at worldwins.net will  use a 'virtual office' based file system,
part of a new OS I'm developing so I want all my links to have a consistent
look and feel. POVRay is ideal for this, and also a lot of fun to work with.
This is only my second posting to any newsgroup (I've been an underground
cave dweller by necessity the last couple of years - the Real World has been
too distracting).

I'm doing a lot of work with XML, XSLT, SVG and  Schemas; all my POV source
files are in XML and are transformed back and forth from .xml to .pov to
SVG. If anyone else is doing work in this area I'd like to know about it.


Post a reply to this message

Copyright 2003-2023 Persistence of Vision Raytracer Pty. Ltd.