The Persistence of Vision Raytracer (POV-Ray).
This is the legacy Bug Tracking System for the POV-Ray project. Bugs listed here are being migrated to our github issue tracker. Please refer to that for new reports or updates to existing ones on this system.
Opened by Jim Holsenback - 2013-03-21
Last edited by Christoph Lipka - 2013-03-25
FS#279 - area_illumination causes artifacts when used with radiosity
see my post titled: “area light and radiosity problem?” in povray.binary.images [Edit - copied that post’s text here - clipka]
wondering about what’s going on here with this series of images. the radiosity and area light settings are unchanged from image to image, and all I did was radiosity on/off and area_light on/off (btw: using rad_def “Normal” settings)
the 1st image is radiosity only, the 2nd is area light only, and the 3rd combines them:
what’s up with blotches? change #5819/5820 (octree) or maybe something still with area lights?
RadOnly.png
I've had a closer look at the code now, but the necessary tests to suppress highlights for radiosity secondary rays are all in place. Can you provide a minimal test scene to toy around with?
Attached is a scene file setup to reproduce the artifacts.
I've discovered what appears to really be the cause of the blotches! It's the sphere that I use to mark the light source while in (scene) Debug mode, it has emission 1 set.
Toggle area_illumination and Debug and see what I mean.
Maybe the marker sphere is another clue, because I'm still thinking something is is not right with area_illumination.
Almost, but not quite: The "emission 1" is benign. Turns out your real problem is the combination of the following:
So what happens is this: The looks_like object is illuminated by some "lightlets" of the extremely close (and therefore extremely bright) area light; as it has "diffuse 0.6", the looks_like object itself appears extremely bright; the radiosity algorithm picks up this object for a few samples, and the brightness is so high that even averaging with all the other secondary rays for that sample still yields very bright light.
The reason this does not happen when area_illumination is off is that classic area lights are computed a bit different: They are still treated as point lights for every purpose other than whether (and to what degree) a point on a surface is shadowed; the shading based on the object surface's orientation with respect to the light source is still done as if the area light was a point light. In your case this light source is at the center of a convex object, so for each point on the surface the light is considered to be coming from behind, and consequently ignored.
To sum it all up, this isn't a bug after all. Explicitly set "diffuse 0" on your looks_like object, and you'll be fine.
I ran into this issue on a scene and setting max_sample 1 in radiosity settings fixed my problem.