An area light with size vectors 24*x and 24*z, when placed fairly close to a
surface should brightly illuminate a roughly square area on the surface,
should it not?  The diffuse illumination on nearby surfaces from an area
light should be related to the size and shape of the area light, but this
does not appear to be the case, and I believe that this has been listed in
the POV-Ray FAQ section.

Why is this important?  I frequently create interior architectural scenes
where I want to show a fluorescent cove light.  I'd like to model the cove
and put a long linear area light inside of it.  Because the diffuse
illumination is the key here, the result, unfortunately, looks more like I
put a point light inside of the cove.

So, should I resort to an actual array of point light sources?

Post a reply to this message

richardellison wrote:
> An area light with size vectors 24*x and 24*z, when placed fairly close to a
> surface should brightly illuminate a roughly square area on the surface,
> should it not?  The diffuse illumination on nearby surfaces from an area
> light should be related to the size and shape of the area light, but this
> does not appear to be the case, and I believe that this has been listed in
> the POV-Ray FAQ section.
> 
> Why is this important?  I frequently create interior architectural scenes
> where I want to show a fluorescent cove light.  I'd like to model the cove
> and put a long linear area light inside of it.  Because the diffuse
> illumination is the key here, the result, unfortunately, looks more like I
> put a point light inside of the cove.
> 
> So, should I resort to an actual array of point light sources?
> 
> 
> 
 From 3.4.7.5 discussing area lights in the manual
"The array-effect applies to shadows only. The object's illumination is 
still that of a point source"

So yes you need to produce your own array of point_lights or maybe work 
something with a high ambient shape

Post a reply to this message

Jim Charter wrote:

For instance you could try this:

#include "rad_def.inc"
global_settings {
         assumed_gamma 1.0
         max_trace_level 10
         radiosity {
                 Rad_Settings(Radiosity_Final, off, off)
         }
}

#default{finish{ambient 0}}


plane { y, 0
	pigment { rgb 1 }
	finish {diffuse 1 }
}

box { <-6, 4.5, -3> <6, 4.51 ,3>
	no_image
	pigment { rgb <1,1,0> }
	finish { ambient 1 }
}

sphere { .1, .1
         pigment { rgb 1 }
         finish { diffuse .8 }
}

camera {
   	location  vrotate ( <0.0, 0, -15.0>, <12, 1,0> )
   	look_at   < 0, 0,   0>
   	right     x*image_width/image_height
   	angle 34
}

Post a reply to this message

richardellison nous apporta ses lumieres en ce 2005-07-29 13:05:
> An area light with size vectors 24*x and 24*z, when placed fairly close to a
> surface should brightly illuminate a roughly square area on the surface,
> should it not?  The diffuse illumination on nearby surfaces from an area
> light should be related to the size and shape of the area light, but this
> does not appear to be the case, and I believe that this has been listed in
> the POV-Ray FAQ section.
> 
> Why is this important?  I frequently create interior architectural scenes
> where I want to show a fluorescent cove light.  I'd like to model the cove
> and put a long linear area light inside of it.  Because the diffuse
> illumination is the key here, the result, unfortunately, looks more like I
> put a point light inside of the cove.
> 
> So, should I resort to an actual array of point light sources?
> 
> 
> 
I had the same problem lately. My workaround was to use a few arealights lined up. It
gives you 
smoother shadows and the tiffuse illumination blendings did a nice job of hiding the
point-like 
diffuse areas. I used 3 area_light with intensities of 1/3 the original one.

Alain

Post a reply to this message

Thanks, Alain,
Glad to know it's not just me.  I was going to try a macro which would place
a string of area lights in the cove, but I've read in the FAQ that POV-Ray
area lights are more efficient than that.  Perhaps I have a wish-list
item...
Rchard

Post a reply to this message

Your suggestion is to use radiosity and an object with high ambient value as
a light source.  Not a bad idea, I'll be interested to see if that's a more
efficient technique than a row of area lights.
Richard

Post a reply to this message

rellison <ric### [at] yahoocom> wrote:
> Glad to know it's not just me.  I was going to try a macro which would place
> a string of area lights in the cove, but I've read in the FAQ that POV-Ray
> area lights are more efficient than that.

  POV-Ray's "area_light" gives you faster renders (when you use "adaptive")
but it only affects shadows. A true grid of point lights will also affect
diffuse and specular lighting, but you won't get the advantages of
adaptive sampling, which will result in slower renders.

-- 

                                                          - Warp

Post a reply to this message