[Radiance-general] Re: -ar, -ad, -as, ahh $#*@!

[Greg Ward]

Aha!  It's not a bug, it's a "feature!"

The reason this shows up in rpict but not in rview is because rpict 
resorts to a specialized set of routines (in srcdraw.c) to draw sources 
that are small in the rendered image.  This overrides the illum 
calculation for low-angle, small sources, who are deemed too small to 
sample correctly on the image plane.  Rather than substituting the 
alternate material, which would normally happen for view rays, 
drawsources() sends a "shadow" ray to the source that is being drawn.  
This invokes the illum rather than its substitute, giving very 
different results if the substitute is a poor one.  In other words, you 
saw it before in your IES sources because the source distribution 
either didn't match the simulated source.  It shows up well in this 
example because your alternate material is a dark diffuse one rather 
than something emitting.

I could eliminate illum's from the drawsources() calculations, but I 
think the net result would be worse than what we currently have, as 
illum sources then would be poorly sampled on the image plane when they 
got to be small and/or thin.

-Greg

> From: Georg Mischler <schorsch@schorsch.com>
> Date: Sun Sep 14, 2003  12:52:56  AM US/Pacific
>
> Ok, so after mentioning the discrepancy between rpict and rview I
> actually ran my new test scene through rpict, and bingo!
>
> Maybe it's possible, but within the last hour I couldn't manage
> to get the effect in rview. I also found that the appearance
> together with mapping data was a coincidence, the bug triggers
> with plane jane illums as well.
>
> I'm appending a very simple test case using the rpict defaults
> (plus -ab 1 for visuals). Something interesting I also just
> detected is that the size of the source appears to make a
> difference.
>
> Since this now turns into a development issue, I'm crossposting
> to the dev-list. I suggest that everybody involved also reply
> there.