Hellow,

I've recently begun using povray and although I'm finding solutions for a lot of
problems at the documentation and the newsgroups, there is something I'm finding
difficult to understand: If I use radiosity and as light source a sphere of 1 mm
of diameter, at a distance of 10 mm I obtain the brightness predicted by physics
(or that I think): 1% of the original , but if I use a pointlight with
fade_distance=1 and fade_power=2 deactivating radiosity, the brightness doubles,
while it should be the same.

I'm using Imagej to measure the brightness in a 16 bits gray scaled image
without gamma correction.

The code:
#version 3.7;

global_settings {ambient_light 0
   //radiosity {count 100000  max_sample -1 gray_threshold 0.0 brightness 1
normal
on media on}
    assumed_gamma 1.0
}

camera {orthographic location <5,0,0>  look_at <10,0,0> angle 90}

//Source (2 options)
//sphere {<0,0,0> 1 texture {pigment {color rgb<1,1,1>} finish {emission 1
ambient 0 reflection 0 diffuse 0} } }

light_source { <0,0,0> color rgb <1,1,1> fade_distance 1 fade_power 2}

//Surface
plane {<1,0,0> 10
    texture {pigment {color rgb<1,1,1>} finish {emission 0 ambient 0 reflection
0 diffuse 1 } }
}

Thanks

Post a reply to this message

I'll just say that your sphere has a radius, and the point source doesn't.
So it's 1 radius closer than the point light.
Move the sphere 1 radius further away from your surface and see what happens.

Work from there.

Post a reply to this message

"Bald Eagle" <cre### [at] netscapenet> wrote:
> I'll just say that your sphere has a radius, and the point source doesn't.
> So it's 1 radius closer than the point light.
> Move the sphere 1 radius further away from your surface and see what happens.
>
> Work from there.

No, that's not the problem. The sphere gives 1/2 of the brightness of the point
light.

Now I've reviewed the attenuation law of light at
http://www.povray.org/documentation/view/3.6.2/317/ and the trick is that I
should divide the intensity of the point light by 2 according to the formula at
that page if I want to reproduce the behaviour I obtain with the sphere.

Many thanks for your help anyway.

Post a reply to this message

On 7/11/2016 12:52 PM, fsv2712 wrote:
> "Bald Eagle" <cre### [at] netscapenet> wrote:
>> I'll just say that your sphere has a radius, and the point source doesn't.
>> So it's 1 radius closer than the point light.
>> Move the sphere 1 radius further away from your surface and see what happens.
>>
>> Work from there.
>
> No, that's not the problem. The sphere gives 1/2 of the brightness of the point
> light.
>
> Now I've reviewed the attenuation law of light at
> http://www.povray.org/documentation/view/3.6.2/317/ and the trick is that I
> should divide the intensity of the point light by 2 according to the formula at
> that page if I want to reproduce the behaviour I obtain with the sphere.
>
> Many thanks for your help anyway.
>
>

your code example indicated that you're using v3.7, but the link you 
referenced is for v3.6.2

check out: 
http://wiki.povray.org/content/Reference:Light_Source#Light_Fading

there's some additional information ... /maybe/ it's relevant

Post a reply to this message

Jim Holsenback <spa### [at] nothanksnet> wrote:
> there's some additional information ... /maybe/ it's relevant

Yes, there seems to be more useful information. I'll read it.

Thanks.

Post a reply to this message

Am 11.07.2016 um 17:03 schrieb fsv2712:

> I've recently begun using povray and although I'm finding solutions for a lot of
> problems at the documentation and the newsgroups, there is something I'm finding
> difficult to understand: If I use radiosity and as light source a sphere of 1 mm
> of diameter, at a distance of 10 mm I obtain the brightness predicted by physics
> (or that I think): 1% of the original , but if I use a pointlight with
> fade_distance=1 and fade_power=2 deactivating radiosity, the brightness doubles,
> while it should be the same.

You're probably seeing the effect of a decades-long legacy quirk of
POV-Ray's light intensity handling.

The original effective calibration of any brightness-ish properties was
such that unity values would give unity results in the output images:

- A light source with unity brightness, shining vertically on a white
surface with unity pigment and unity `diffuse` setting (at a distance of
`fade_distance` if applicable), would give a 100% white pixel in the
resulting image (presuming all other finish properties would be set to 0).

- Similarly, a surface with unity pigment and unity `ambient` setting
would also give a 100% white pixel in the resulting image (again
presuming all other finish properties would be set to 0).

This was all that could originally be said about the calibration of
light intensity.

- With the introduction of radiosity, the calibration of the `diffuse`
property was effectively formalized such that a material with unity
pigment and unity `diffuse` setting corresponds to unity _hemispherical
reflectance_.

- The faily new `emission` setting uses the same calibration as the
`ambient` setting (presuming the global `ambient_light` setting is left
at its default of 1.0), for ease of transition between the two mechanisms.

The result is that the calibrations for light source brightness and the
`emission` setting are such that the radiant flux of a light source with
unity brightness and fade_distance R does /not/ match the radiant flux
of a sphere with radius R and both unity pigment and unity `emission`
setting.

Post a reply to this message

Many thanks for the explanation, clipka.

I'm not sure I have fully understand what you are trying to tell, but even if
that was the case, this has helped me understand why this behaviour appears.

Post a reply to this message