Hi all,
I know that questions regarding the trans model tend to be either repetitive or lead to never-ending threads, however here is one more in the series.
I wanted to model a translucent surface, where I know really nothing about reflection. My idea was to keep Rd and Rs set to zero and only define the transmission. A simple example would be a surface with perfect lambertian transmission (Td=1, Ts=0) and no reflection at all (ok, hard to produce such a material, but I had reasons to separate transmission from reflection information). According to my understanding that would be:
void trans puretransmission
0
0
7 1 1 1
0 0
1 0
However, being lazy, I thought this might be a good time to try the widget for cross-checking. Which made me realize that Cr, Cb, Cg are not asked for. Instead, I seam to need a value > 0 for Rd, which means I can never achieve the above definition.
Am I just doing it wrong? Did this problem never arise because noone else wanted such a strange trans model? Or - which would give me a headache - is my understanding of Cr, Cg, Cb wrong? I always understood A1-A3=1-absorption(r,g,b).
Thank you, cheers, Lars.
Hi all,
I know that questions regarding the trans model tend to be either repetitive or lead to never-ending threads, however here is one more in the series.
I wanted to model a translucent surface, where I know really nothing about reflection. My idea was to keep Rd and Rs set to zero and only define the transmission. A simple example would be a surface with perfect lambertian transmission (Td=1, Ts=0) and no reflection at all (ok, hard to produce such a material, but I had reasons to separate transmission from reflection information). According to my understanding that would be:
void trans puretransmission
0
0
7 1 1 1
0 0
1 0
However, being lazy, I thought this might be a good time to try the widget for cross-checking. Which made me realize that Cr, Cb, Cg are not asked for. Instead, I seam to need a value > 0 for Rd, which means I can never achieve the above definition.
Am I just doing it wrong? Did this problem never arise because noone else wanted such a strange trans model? Or - which would give me a headache - is my understanding of Cr, Cg, Cb wrong? I always understood A1-A3=1-absorption(r,g,b).
Thank you, cheers, Lars.
Hi Lars,
without testing it I would say your trans specification exactly does what
you want (100% diffuse tranmission). However, this is a border case where
the formulae for trans as in RwR on page 325/326 are not defined (dividing
0/0)...
Cheers,
David
···
2013/1/17 Lars O. Grobe <[email protected]>
Hi all,
I know that questions regarding the trans model tend to be either
repetitive or lead to never-ending threads, however here is one more in the
series.
I wanted to model a translucent surface, where I know really nothing about
reflection. My idea was to keep Rd and Rs set to zero and only define the
transmission. A simple example would be a surface with perfect lambertian
transmission (Td=1, Ts=0) and no reflection at all (ok, hard to produce
such a material, but I had reasons to separate transmission from reflection
information). According to my understanding that would be:
void trans puretransmission
0
0
7 1 1 1
0 0
1 0
However, being lazy, I thought this might be a good time to try the widget
for cross-checking. Which made me realize that Cr, Cb, Cg are not asked
for. Instead, I seam to need a value > 0 for Rd, which means I can never
achieve the above definition.
Am I just doing it wrong? Did this problem never arise because noone else
wanted such a strange trans model? Or - which would give me a headache - is
my understanding of Cr, Cg, Cb wrong? I always understood
A1-A3=1-absorption(r,g,b).
Thank you, cheers, Lars.
_______________________________________________
Radiance-general mailing list
[email protected]
http://www.radiance-online.org/mailman/listinfo/radiance-general