visit

http://www.lilysoft.org/CGI/SR/Spectral%20Render.htm

for additions and changes, have fun

-Ive

Post a reply to this message

Am 02.12.2013 09:04, schrieb Ive:
> visit
>
> http://www.lilysoft.org/CGI/SR/Spectral%20Render.htm
>
> for additions and changes, have fun

The images look great, I particularly like the one with the colored 
wine(?) glasses.

How are you addressing the problem of synthesizing spectra from RGB values?

One thought: Shouldn't it be possible to support image maps by wrapping 
them in a pigment function? It'll still be synthesized from RGB, but 
better than no support at all (or so I think).

Oh, and one more thing: I understand that you have the necessary 
equipment to measure materials' spectra. Would you mind providing a few 
selected plastic materials like, say... LEGO bricks? (both opaque and 
transparent, please) ;-)

Post a reply to this message

Am 02.12.2013 23:56, schrieb clipka:
> How are you addressing the problem of synthesizing spectra from RGB values?
>
I did use the blue, green, red patch of row 3 from the color checker 
chart - assuming that the color experts at Gretag have had reasons to 
choose these -  together with a "luminance" multiplier to make sure 
equal values do produce something like gray.
Someone more serious than me would have used one of the CIE color match 
functions, after all these represent what an average human calls pure 
red, green and blue.

> One thought: Shouldn't it be possible to support image maps by wrapping
> them in a pigment function? It'll still be synthesized from RGB, but
> better than no support at all (or so I think).
>
Hm, I did not think of that, but now that you mention it, this should 
certainly be possible. Maybe something for version 0.3.

> Oh, and one more thing: I understand that you have the necessary
> equipment to measure materials' spectra. Would you mind providing a few
> selected plastic materials like, say... LEGO bricks? (both opaque and
> transparent, please) ;-)
>
Funny that you think of LEGO bricks. I had exactly the same idea of 
using them as "reference colors" when I discussed lightsys with Jaime 
about 10 years ago. But meanwhile I do not have a single brick anymore.

But I did measure them and might still have the data on some old hard 
disk within some Pentium 3 machine, me thinks ;)
I will look for it, but I'm really not sure, so any other ideas...

-Ive

Post a reply to this message

Ive <ive### [at] lilysoftorg> wrote:
> Am 02.12.2013 23:56, schrieb clipka:
> > How are you addressing the problem of synthesizing spectra from RGB values?
> >
> I did use the blue, green, red patch of row 3 from the color checker
> chart - assuming that the color experts at Gretag have had reasons to
> choose these -  together with a "luminance" multiplier to make sure
> equal values do produce something like gray.
> Someone more serious than me would have used one of the CIE color match
> functions, after all these represent what an average human calls pure
> red, green and blue.
>
> > One thought: Shouldn't it be possible to support image maps by wrapping
> > them in a pigment function? It'll still be synthesized from RGB, but
> > better than no support at all (or so I think).
> >
> Hm, I did not think of that, but now that you mention it, this should
> certainly be possible. Maybe something for version 0.3.
>
> > Oh, and one more thing: I understand that you have the necessary
> > equipment to measure materials' spectra. Would you mind providing a few
> > selected plastic materials like, say... LEGO bricks? (both opaque and
> > transparent, please) ;-)
> >
> Funny that you think of LEGO bricks. I had exactly the same idea of
> using them as "reference colors" when I discussed lightsys with Jaime
> about 10 years ago. But meanwhile I do not have a single brick anymore.
>
> But I did measure them and might still have the data on some old hard
> disk within some Pentium 3 machine, me thinks ;)
> I will look for it, but I'm really not sure, so any other ideas...
>
> -Ive

Hey,my name is miaoyu, I am a student,I have read your article about "spectral
rendering with pov-ray".And I also have visited  your website
http://lilysoft.org. How beautiful your works are! But I want to beg a help from
you. I am studying pov to do my graduation project.I have confronted some
confusions,please give me a help,expecting your replying!Greate gratitudes!
My questions:
1: The value of solar irradiance in 817nm is 1.08279W/(m2*nm).How can I set
related statements in pov?
2:How can I make the value of each pixel of final image be radiance? not color.

My codes are as bolow:
#version 3.7;
global_settings{ assumed_gamma 1.0 }  #include "spectral.inc" #declare
SpectralWavelength = 817nm;
camera
{
angle  7.54371
location <camera_x,camera_y,camera_z >
look_at <0,0,0>
rotate <0,-31,0>
right x
up y
}
light_source {<sun_x,sun_y,sun_z>
            color rgb  1.08279 }

#declare geom_file_name = "pov-xyz.txt";
#declare spec_file_name = "pov-ref.txt";
#declare n=o;
#fopen   Input_geom_file geom_file_name read
#fopen   Input_spec_file spec_file_name read
#while(defined(Input_geom_file Input_spec_file))
    #read(Input_geom_file,x1,y1,z1)
    #read(Input_spec_file,ref)
        sphere
        {
        <x1-center_x,y1-center_y,z1-center_z>, //sphere location
         1.55     //sphere radius
        finish {specular ref}  //sphere BRDF
        }
    #declare n = n + 1;
#end

Can you help me to check it,Beg a help! Greate gratitudes!

Post a reply to this message

Ive <ive### [at] lilysoftorg> wrote:
> visit
>
> http://www.lilysoft.org/CGI/SR/Spectral%20Render.htm
>
> for additions and changes, have fun
>
> -Ive

Just stumbled across this site...I was wondering if you could use this to do
wavelength-dependent Rayleigh scattering (like we have in our real atmosphere,
for example), by specifying an array of colors or density multipliers for the
different wavelengths and then selecting based on WavelengthIndex, so you'd have
something like:

scattering {1, scattr[WavelengthIndex]}

I'm wondering if you had something like this if you could simulate the red
sunset/blue sky effect you get when the light from our sun interacts with our
atmosphere.

Post a reply to this message

"jceddy" <jce### [at] gmailcom> wrote:
> Ive <ive### [at] lilysoftorg> wrote:
> > visit
> >
> > http://www.lilysoft.org/CGI/SR/Spectral%20Render.htm
> >
> > for additions and changes, have fun
> >
> > -Ive
>
> Just stumbled across this site...I was wondering if you could use this to do
> wavelength-dependent Rayleigh scattering (like we have in our real atmosphere,
> for example), by specifying an array of colors or density multipliers for the
> different wavelengths and then selecting based on WavelengthIndex, so you'd have
> something like:
>
> scattering {1, scattr[WavelengthIndex]}
>
> I'm wondering if you had something like this if you could simulate the red
> sunset/blue sky effect you get when the light from our sun interacts with our
> atmosphere.

 I just got that a little while ago myself and just tested it out a little bit
so I'm not sure if you actually have to have an index of refraction in the sky
to do that, or simply a pattern.
 I remember you need to do several renders and combined them at least the way
parts of it are done, maybe that doesn't mean it has to be done for everything.
Hopefully someone else will know more   And reply here

Post a reply to this message