[Radiance-general] genBSDF questions

[Rob Guglielmetti]

Just wanted to give a shout-out to Greg for making this conversation even possible. This BSDF capability is a tremendous addition to this thing we call Radiance! Happy Thanksgiving to those who celebrate, and Happy Saturday to all. 

- Rob


On Nov 24, 2011, at 11:48 AM, Greg Ward wrote:

> Hi Marija,
> 
> I'll put some answers inline...
> 
>> From: Marija Velickovic <maricanis at gmail.com>
>> Date: November 24, 2011 12:17:05 AM PST
>> 
>> Hi,
>> 
>> I've been experimenting with genBSDF for some blinds simulation and
>> while working on it I've found out that some things aren't clear
>> enough to me :)
>> 
>> I understand that BSDF generated for a blind can be used for all
>> blinds with the same size in the building model.
>> 
>> Should genBSDF be repeated for different dimensions of blinds or
>> already calculated BSDF file can be reused?
>> For example if in a building I have windows with few different sizes,
>> should I recalculate BSDF for each window/blinds size
>> 
>> If blinds are similar to venetian blinds, is it ok if I use existing
>> BSDF for window with new height/width?
> 
> If the blind behavior is close to uniform over its surface, i.e., the blind spacing and width is small relative to all of your windows, then you are probably OK with a single distribution on all your windows.  It's easy enough to run a comparison to be sure.
> 
> This brings up an important point regarding genBSDF, which is how you treat the edge conditions.  The best approach when you are trying to get the general behavior of something like venetian blinds is to create a system much larger than the area you actually sample with genBSDF, then use the "-dim" option to specify the region over which to sample.  This region should be an even multiple of the blind spacing, or whatever other repeating structures your system contains.
> 
> Another approach, which works for a system of a specific size that is fully contained within a rectangular window opening, is to model the edges of the opening along with the system you give to genBSDF.  If you make the mistake of leaving the edges open, then light that would NOT leak into your room would get by the geometry you give to genBSDF, and the generated BSDF would have too much transmission.
> 
>> #######
>> Another question is if I have blinds composed of 2 or more types of
>> lamelas (for example one type of lamelas goes on the bottom part of
>> window, and another on top part). Should these blinds be treated as
>> one unit and calculated common BSDF for them; or separated and then
>> separate BSDF surfaces included in Radiance model?
> 
> If your lamellas (slats) are different, then I expect your BSDFs will be different and should be calculated separately.
> 
>> #######
>> In BSDF material definition
>> 
>> mod BSDF id
>> 6+ thick BSDFfile ux uy uz funcfile transform
>> 0
>> 0|3|6|9 rfdif gfdif bfdif rbdif gbdif bbdif rtdif gtdif btdif
>> 
>> "An optional 'transform' is used to scale the thickness and reorient
>> the up vector. " What does this exactly mean?
>> 
>> For example if input to genBSDF was defined in cm and my scene is in
>> m, which units should be used for thickness and what should be put as
>> transformation.
> 
> All Radiance primitives need to be able to pass through the xform command and maintain correct behavior.  If you insert a BSDF into your model, later scaling and reorienting the model with xform, a transform will be added to the BSDF primitive to make sure the distribution is properly oriented and the thickness is correct in the new units.
> 
> You should specify the thickness in the units corresponding to the model or scene file your BSDF is used with.
> 
>> Hope my questions are clear enough :)
>> 
>> Thanks in advance,
>> Marija
> 
> No problem.
> -Greg
> _______________________________________________
> Radiance-general mailing list
> Radiance-general at radiance-online.org
> http://www.radiance-online.org/mailman/listinfo/radiance-general