trans material - tspec comparison

Hello All,

For tspec comparison, I rendered the same scene for four different tspec
values of 10%, 50%, 90% and 100% (ideal diffuser).

I was expecting varying amount of diffusion, but I am getting puzzling
results. only 100% (ideal diffuse) value returns no direct sun-patch. All
the other values show a clear sun-patch in the space.

Any clue why this is happening? Would you expect such results? Everything
else is constant in the model. (images are attached)

Thanks,

Shri

Shri,

In the images you sent I suspect that the sunpatch is over exposed, so it
appears white in all three renderings. If you create a falsecolor rendering
you'll likely see that the sunpatch has different luminance values (make
sure to set the scale sufficently high).

One of the drawbacks of using Radiance's trans material to model
translucent fenestration materials is that the transmission is divided into
strictly specular and diffuse. The specular component is not scattered at
all, and the diffuse component is completely lambertian. You might have
expected the sunpatch to soften and broaden as the diffusion increases, but
with the trans material the shape of the sunpatch doesn't change as
diffusion increases, it just reduces in intensity.

To study diffusion like this, you'll really want to use a BSDF. Probably
also a tensor tree BSDF so you get sufficient accuracy in the sun patch.
And unfortunately LBNL hasn't measured BSDFs for diffusing glazing, so I
can't offer an immediate solution.

Best,
Andy

···

On Tue, Aug 12, 2014 at 3:51 PM, Shrikar Bhave <[email protected]> wrote:

Hello All,

For tspec comparison, I rendered the same scene for four different tspec
values of 10%, 50%, 90% and 100% (ideal diffuser).

I was expecting varying amount of diffusion, but I am getting puzzling
results. only 100% (ideal diffuse) value returns no direct sun-patch. All
the other values show a clear sun-patch in the space.

Any clue why this is happening? Would you expect such results? Everything
else is constant in the model. (images are attached)

Thanks,

Shri

_______________________________________________
Radiance-general mailing list
[email protected]
http://www.radiance-online.org/mailman/listinfo/radiance-general

Hi Shri and Andy,

I think you can "spread" the "direct" component when applying a roughness to the trans material. Then, the scattered light is not lambertian any more. You can control the scattering "width" by the roughness.
You can see the difference then if you look directly onto the material (smaller or larger spot towards sun direction). Getting different patches on the wall etc. is quite hard I guess. The reason for this is , that as soon as the "direct part" is scattered (when applying roughness), it is handled by the ambient calculation and it is not part of the direct calculation any more.

Using a BSDF with a fine resolution is in fact the best way to handle this.

good luck,

Jan

···

On 08/13/2014 01:26 AM, Andrew McNeil wrote:

Shri,

In the images you sent I suspect that the sunpatch is over exposed, so it appears white in all three renderings. If you create a falsecolor rendering you'll likely see that the sunpatch has different luminance values (make sure to set the scale sufficently high).

One of the drawbacks of using Radiance's trans material to model translucent fenestration materials is that the transmission is divided into strictly specular and diffuse. The specular component is not scattered at all, and the diffuse component is completely lambertian. You might have expected the sunpatch to soften and broaden as the diffusion increases, but with the trans material the shape of the sunpatch doesn't change as diffusion increases, it just reduces in intensity.

To study diffusion like this, you'll really want to use a BSDF. Probably also a tensor tree BSDF so you get sufficient accuracy in the sun patch. And unfortunately LBNL hasn't measured BSDFs for diffusing glazing, so I can't offer an immediate solution.

Best,
Andy

On Tue, Aug 12, 2014 at 3:51 PM, Shrikar Bhave <[email protected] > <mailto:[email protected]>> wrote:

    Hello All,

    For tspec comparison, I rendered the same scene for four
    different tspec values of 10%, 50%, 90% and 100% (ideal diffuser).

    I was expecting varying amount of diffusion, but I am getting
    puzzling results. only 100% (ideal diffuse) value returns no
    direct sun-patch. All the other values show a clear sun-patch in
    the space.

    Any clue why this is happening? Would you expect such results?
    Everything else is constant in the model. (images are attached)

    Thanks,

    Shri

    _______________________________________________
    Radiance-general mailing list
    [email protected]
    <mailto:[email protected]>
    http://www.radiance-online.org/mailman/listinfo/radiance-general

_______________________________________________
Radiance-general mailing list
[email protected]
http://www.radiance-online.org/mailman/listinfo/radiance-general

Ah, good point with the roughness. The only way I know how to get a good
value for the roughness parameter is trial and error it until it looks
about right in a rendering - which is probably why I avoid using it. If
anyone has an good method for setting roughness please share!

Andy

···

On Wed, Aug 13, 2014 at 4:38 AM, Jan Wienold <[email protected]> wrote:

Hi Shri and Andy,

I think you can "spread" the "direct" component when applying a roughness
to the trans material. Then, the scattered light is not lambertian any
more. You can control the scattering "width" by the roughness.
You can see the difference then if you look directly onto the material
(smaller or larger spot towards sun direction). Getting different patches
on the wall etc. is quite hard I guess. The reason for this is , that as
soon as the "direct part" is scattered (when applying roughness), it is
handled by the ambient calculation and it is not part of the direct
calculation any more.

Using a BSDF with a fine resolution is in fact the best way to handle this.

good luck,

Jan

On 08/13/2014 01:26 AM, Andrew McNeil wrote:

Shri,

In the images you sent I suspect that the sunpatch is over exposed, so
it appears white in all three renderings. If you create a falsecolor
rendering you'll likely see that the sunpatch has different luminance
values (make sure to set the scale sufficently high).

One of the drawbacks of using Radiance's trans material to model
translucent fenestration materials is that the transmission is divided into
strictly specular and diffuse. The specular component is not scattered at
all, and the diffuse component is completely lambertian. You might have
expected the sunpatch to soften and broaden as the diffusion increases, but
with the trans material the shape of the sunpatch doesn't change as
diffusion increases, it just reduces in intensity.

To study diffusion like this, you'll really want to use a BSDF. Probably
also a tensor tree BSDF so you get sufficient accuracy in the sun patch.
And unfortunately LBNL hasn't measured BSDFs for diffusing glazing, so I
can't offer an immediate solution.

Best,
Andy

On Tue, Aug 12, 2014 at 3:51 PM, Shrikar Bhave <[email protected]> > wrote:

  Hello All,

For tspec comparison, I rendered the same scene for four different
tspec values of 10%, 50%, 90% and 100% (ideal diffuser).

I was expecting varying amount of diffusion, but I am getting puzzling
results. only 100% (ideal diffuse) value returns no direct sun-patch. All
the other values show a clear sun-patch in the space.

Any clue why this is happening? Would you expect such results?
Everything else is constant in the model. (images are attached)

Thanks,

Shri

_______________________________________________
Radiance-general mailing list
[email protected]
http://www.radiance-online.org/mailman/listinfo/radiance-general

_______________________________________________
Radiance-general mailing [email protected]://www.radiance-online.org/mailman/listinfo/radiance-general

_______________________________________________
Radiance-general mailing list
[email protected]
http://www.radiance-online.org/mailman/listinfo/radiance-general

T&E's how I've always done it. =/

···

On 8/13/14, 5:37 PM, "Andrew McNeil" <[email protected]<mailto:[email protected]>> wrote:

Ah, good point with the roughness. The only way I know how to get a good value for the roughness parameter is trial and error it until it looks about right in a rendering - which is probably why I avoid using it. If anyone has an good method for setting roughness please share!

Andy

On Wed, Aug 13, 2014 at 4:38 AM, Jan Wienold <[email protected]<mailto:[email protected]>> wrote:
Hi Shri and Andy,

I think you can "spread" the "direct" component when applying a roughness to the trans material. Then, the scattered light is not lambertian any more. You can control the scattering "width" by the roughness.
You can see the difference then if you look directly onto the material (smaller or larger spot towards sun direction). Getting different patches on the wall etc. is quite hard I guess. The reason for this is , that as soon as the "direct part" is scattered (when applying roughness), it is handled by the ambient calculation and it is not part of the direct calculation any more.

Using a BSDF with a fine resolution is in fact the best way to handle this.

good luck,

Jan

On 08/13/2014 01:26 AM, Andrew McNeil wrote:
Shri,

In the images you sent I suspect that the sunpatch is over exposed, so it appears white in all three renderings. If you create a falsecolor rendering you'll likely see that the sunpatch has different luminance values (make sure to set the scale sufficently high).

One of the drawbacks of using Radiance's trans material to model translucent fenestration materials is that the transmission is divided into strictly specular and diffuse. The specular component is not scattered at all, and the diffuse component is completely lambertian. You might have expected the sunpatch to soften and broaden as the diffusion increases, but with the trans material the shape of the sunpatch doesn't change as diffusion increases, it just reduces in intensity.

To study diffusion like this, you'll really want to use a BSDF. Probably also a tensor tree BSDF so you get sufficient accuracy in the sun patch. And unfortunately LBNL hasn't measured BSDFs for diffusing glazing, so I can't offer an immediate solution.

Best,
Andy

On Tue, Aug 12, 2014 at 3:51 PM, Shrikar Bhave <[email protected]<mailto:[email protected]>> wrote:
Hello All,

For tspec comparison, I rendered the same scene for four different tspec values of 10%, 50%, 90% and 100% (ideal diffuser).

I was expecting varying amount of diffusion, but I am getting puzzling results. only 100% (ideal diffuse) value returns no direct sun-patch. All the other values show a clear sun-patch in the space.

Any clue why this is happening? Would you expect such results? Everything else is constant in the model. (images are attached)

Thanks,

Shri

_______________________________________________
Radiance-general mailing list
[email protected]<mailto:[email protected]>
http://www.radiance-online.org/mailman/listinfo/radiance-general

_______________________________________________
Radiance-general mailing list
[email protected]<mailto:[email protected]>http://www.radiance-online.org/mailman/listinfo/radiance-general

_______________________________________________
Radiance-general mailing list
[email protected]<mailto:[email protected]>
http://www.radiance-online.org/mailman/listinfo/radiance-general

Thanks a lot Andy, Jan et all, you guys make this list amazing!

It was indeed trial and error. Unfortunately, I don't think I can offer a
decent method for setting the roughness yet - Only good news is that it
does change proportionally as expected.
Just out of curiosity, how much of an effort would it be (re time and money
both) for translucent glazing manufacturers to get the testing done from
LBNL?

Thanks again,

Shri

···

On Thu, Aug 14, 2014 at 9:53 AM, Guglielmetti, Robert < [email protected]> wrote:

T&E's how I've always done it. =/

On 8/13/14, 5:37 PM, "Andrew McNeil" <[email protected]<mailto: > [email protected]>> wrote:

Ah, good point with the roughness. The only way I know how to get a good
value for the roughness parameter is trial and error it until it looks
about right in a rendering - which is probably why I avoid using it. If
anyone has an good method for setting roughness please share!

Andy

On Wed, Aug 13, 2014 at 4:38 AM, Jan Wienold <[email protected]<mailto: > [email protected]>> wrote:
Hi Shri and Andy,

I think you can "spread" the "direct" component when applying a roughness
to the trans material. Then, the scattered light is not lambertian any
more. You can control the scattering "width" by the roughness.
You can see the difference then if you look directly onto the material
(smaller or larger spot towards sun direction). Getting different patches
on the wall etc. is quite hard I guess. The reason for this is , that as
soon as the "direct part" is scattered (when applying roughness), it is
handled by the ambient calculation and it is not part of the direct
calculation any more.

Using a BSDF with a fine resolution is in fact the best way to handle this.

good luck,

Jan

On 08/13/2014 01:26 AM, Andrew McNeil wrote:
Shri,

In the images you sent I suspect that the sunpatch is over exposed, so it
appears white in all three renderings. If you create a falsecolor rendering
you'll likely see that the sunpatch has different luminance values (make
sure to set the scale sufficently high).

One of the drawbacks of using Radiance's trans material to model
translucent fenestration materials is that the transmission is divided into
strictly specular and diffuse. The specular component is not scattered at
all, and the diffuse component is completely lambertian. You might have
expected the sunpatch to soften and broaden as the diffusion increases, but
with the trans material the shape of the sunpatch doesn't change as
diffusion increases, it just reduces in intensity.

To study diffusion like this, you'll really want to use a BSDF. Probably
also a tensor tree BSDF so you get sufficient accuracy in the sun patch.
And unfortunately LBNL hasn't measured BSDFs for diffusing glazing, so I
can't offer an immediate solution.

Best,
Andy

On Tue, Aug 12, 2014 at 3:51 PM, Shrikar Bhave <[email protected] > <mailto:[email protected]>> wrote:
Hello All,

For tspec comparison, I rendered the same scene for four different tspec
values of 10%, 50%, 90% and 100% (ideal diffuser).

I was expecting varying amount of diffusion, but I am getting puzzling
results. only 100% (ideal diffuse) value returns no direct sun-patch. All
the other values show a clear sun-patch in the space.

Any clue why this is happening? Would you expect such results? Everything
else is constant in the model. (images are attached)

Thanks,

Shri

_______________________________________________
Radiance-general mailing list
[email protected]<mailto:
[email protected]>
http://www.radiance-online.org/mailman/listinfo/radiance-general

_______________________________________________
Radiance-general mailing list
[email protected]<mailto:
[email protected]>
http://www.radiance-online.org/mailman/listinfo/radiance-general

_______________________________________________
Radiance-general mailing list
[email protected]<mailto:
[email protected]>
http://www.radiance-online.org/mailman/listinfo/radiance-general

_______________________________________________
Radiance-general mailing list
[email protected]
http://www.radiance-online.org/mailman/listinfo/radiance-general