[Radiance-general] Modelling lightwell with translucent glazing

[Gregory J. Ward]

Hi Derek,

I didn't see any responses to this query, so I guess it fell through  
the cracks.  Sorry about that.  Some quick, feeble answers:

1.  Certain frits will transmit diffuse light, though imperfectly.   
Note that an ideal diffuser will transmit 50% of the light and  
reflect the other 50%.  To transmit more than 50% of the light, you  
will necessarily deviate from a Lambertian (diffuse) distribution.

2.  If you choose a dense enough frit for Lambertian diffusion, then  
a diffuse transmission/reflection model will suffice by definition.   
A reasonable diffuse transmission to expect is around 40% for a good  
quality frit, and about the same for refleciton.

3.  You will run into problems if you don't use the mkillum command  
and manually identify your sources.  See the basic tutorials in  
"Rendering with Radiance" on appropriate application.

-Greg

> From: "Derek Whitehead" <Derek.Whitehead at rwdi.com>
> Date: July 11, 2007 6:35:17 PM PDT
>
> Hello,
>
> I am a relative newcomer to Radiance, so please forgive my ignorance.
>
> I am interested in simulating a room with a deep lightwell as the only
> source. The top of the light well has a double glazed skylight and the
> bottom of the lightwell is an as-yet undefined product that produces
> diffuse light and has a very high visible light transmittance. My
> questions are:
>
> 1. Will a light frit or white paint coating on glass transmit highly
> diffused light? If not, any products to recommend that will?
> 2. Will Lamberts cosine law be a reasonable estimate for the BTDF and
> BRDF? And what are reasonable peak (zero angle) values to expect?
> 3. Am I going to run into problems with Radiance finding the light
> source? Which inputs will help Radiance find the light?
>
> Thanks for any guidance.