Some really good glazing types, including highly diffuse museum glazings (one of the best being Opalika from Schott at http://www.us.schott.com/architecture/english/references.html) have no Radiance descriptions or are often not archived in the Window database. You can model those in different ways. The easiest way is to start out with the radiance material trans.
Example:
1) http://www.us.schott.com/architecture/english/products/colored-glass/opalika/index.html
2) Download the pdf file on the link below
3) See http://www.us.schott.com/architecture/english/download/schott_farbglas_us_kl.pdf and take a look at Opalika
4) Go to the product information which contains technical data:
http://www.us.schott.com/architecture/english/download/
5) Select the technical information at the bottom:
Technical Information
OPALIKA®-Physical and chemical properties
http://www.us.schott.com/architecture/english/download/opalikae.pdf
6) This product sheet tells you all optical properties, including a nice intensity distribution plot on p. 6.
7) The total transmittance is around 30%. Nevertheless, the material has a spike caused by the Radiance trans parameter number 7, which is the transmitted specular component.
Let's define this material. The general format of trans is:
void trans opalika
0 0 7
a1 a2 a3 a4 a5 a6 a7
We need additional parameters from the datasheet:
rd = diffuse reflectance of opalika, say 0.2
rs = specular reflectance of opalika, say 0.02
td = diffuse transmittance of opalika, say 0.28
ts = specular transmittance. Play with this value and see how close you can get to the intensity distribution displayed on the data sheet.
Let's start with ts = 0.01
a7=ts/(td+ts)
a6=(td+ts)/(rd+td+ts)
a5=0 (roughness - lets just assume 0 for convenience)
a4 = rs = 0.02 let's say (specular reflectance of Opalika)
a3 = a2 = a1 = rd/ ( (1-rs) *(1-a6) )
If you do the calculations, you will find the following:
void trans opalika
0
0
7 0.5 0.5 0.5 0.02 0.0 0.592 0.0345
Define this material in radiance as a polygon, and put a light source/horizontal sun rays behind it. Determine the luminance at nadir. Determine the luminance at 1, 2, 3 ... 20 degrees off nadir. See if you can get the plot listed in the data sheet. If not, change the ts parameter.
This is a good exercise since you should use data that are verifiable. Create your own values, verify them and don't rely too much on others 
I have found substantial transmittance mistakes in manufacturers published data. We discussed some concerns about the windows database before in this group.
Martin Moeck
···
________________________________
From: radiance-general-bounces@radiance-online.org on behalf of Ramana Koti
Sent: Fri 5/12/2006 4:07 PM
To: Radiance general discussion
Subject: [Radiance-general] diffuse skylight without well: continued
Hello,
Prof Moeck, thanks for the quick pointer. I couldn't ascertain if it can be done okalux products but I need glass types that I can model in Optics and Window for Radiance material format and DOE-2.1E glass library format. I looked up Optics and Window databases and could not locate Okalux in them. Please advise.
Regards,
Ramana.
On 5/12/06, Martin Moeck <MMoeck@engr.psu.edu> wrote:
Ramana,
Okalux has nice product specifications at
http://www.okalux.com/Okalux_2003/englisch/frames_e.html
Select "Products" at the bottom.
They have a good database since they specify the direct and diffuse light transmission, UV protection, U-value, sound reduction, etc. You can use those in pyramidal toplights. For a translation of the direct and diffuse light transmission into the Radiance trans material, see the radiance knowledge base and digests.I would also look into the glaze.cal script.
Dr. Martin Moeck, Pennsylvania State University
________________________________
From: radiance-general-bounces@radiance-online.org on behalf of Ramana Koti
Sent: Fri 5/12/2006 3:04 PM
To: Radiance general discussion
Subject: [Radiance-general] diffuse skylight without well
Hello all,
Can anyone point me to a commerical manufacturer's website where there's specifications of a diffuse skylight (preferably withouth well) that can be modeled in Optics and imported into Radiance material format? (Anything close or suggestions on an alternative way of doing it will work too!)
Regards,
Ramana.
--
Venkata Ramana Koti
1137 E Orange St Apt #35
Tempe AZ 85281 USA
Tel: 480.326.9275
www.daylit.org
MS in Building Design (Energy/Climate)
School of Architecture and Landscape Architecture
College of Design
Arizona State University, Tempe AZ 85287
_______________________________________________
Radiance-general mailing list
Radiance-general@radiance-online.org
http://www.radiance-online.org/mailman/listinfo/radiance-general
--
Venkata Ramana Koti
1137 E Orange St Apt #35
Tempe AZ 85281 USA
Tel: 480.326.9275
www.daylit.org
MS in Building Design (Energy/Climate)
School of Architecture and Landscape Architecture
College of Design
Arizona State University, Tempe AZ 85287
I have found substantial transmittance mistakes in manufacturers published data. We discussed some concerns about the windows database before in this group.