[Radiance-general] RE: Prismatic glazing simulation in PMAP using Siteco data

[Roland Schregle]

Anthony J. Farrell wrote:
> Thanks Roland, Martin for input,
> 
> I meant to make to reference to the Prism1 and Prism2 commands in the 3.5
> release of Radiance. I am uncertain of the function file command line
> 
> mod prism1 id
> 5+ coef dx dy dz funcfile transform
> 0
> n A1 A2...An where the new direction variables
> 
> So if I wanted to redirect light from a vertical glazed window down at an
> angle of 45 degrees onto the working plane what values would I substitute in
> for A1 etc?

According to the manual, dx, dy, dz are the redirection coefficients, 
presumably applied to the incident ray's local coords. A1.. An are 
optional variables passed to the funcfile. However, having never used 
the PRISM material, I dunno for sure -- better check with Raphael Compagnon.

> Roland, Is it possible to use Prism1 or Prism2 commands in the PMAP version
> of Radiance to forward raytrace from the photon port (the target light
> (*luminance* source?)to the target illuminance source (the working plane)?

Beats me. :^)
Seriously, I never tried. I only included support for the more common 
materials. Looking at the code, it would appear that the PRISM material 
simply acts as a secondary light source -- a concept that is invalidated 
by the photon map anyway. Furthermore, the source vector [DxA, DyA, DzA] 
would be undefined, since it's only available specifically for this 
material. My guess is it simply won't work and immediately bail out 
since there's no scattering routine for this specific material.

You'll probably have to go the hard way and model the prism explicitly 
using instances, like the Y-glass example on the pmap webpage. This is 
extremely ugly, but it's the only solution I can envisage in your case.

--Roland


-- 
Roland Schregle
PhD candidate, Fraunhofer Institute for Solar Energy Systems
RADIANCE Photon Map page: www.ise.fhg.de/radiance/photon-map

END OF LINE. (MCP)