Specularity Definition of Radiance Material

Kolomenski_Andrei_JS · October 20, 2015, 9:08pm

Hello Radiance Experts,

I have used a Hemispherical Spectral Reflectance meter to measure the Specular Included (SCI) and Specular Excluded (SCE) SPDs of a material. The obtained SPDs are in the range of 360 to 740 nm., commonly considered the photopic range. Then I compute the total photopic specularity using the equation below.

[cid:[email protected]]

For a black shiny metal sample, I compute a specularity of 21%. Is it correct to use a value of 0.21 for my specularity definition of a metal material primitive. I’ve read that specularity values greater than 0.1 are not realistic (http://radsite.lbl.gov/radiance/refer/ray.html#Materials) in Radiance. In the .hdr rendered image, the material appears to be highly specular to the degree that it acts like a mirror. So this why I’m concerned if I’m using the right procedure to define the specularity for the Radiance metal primitive.

Is it correct to use a hemispherical reflectance meter, and the equation above to define the Radiance specularity term.

Thank you for your input,
Andrei Kolomenski

···

From: Adriana Lira [mailto:[email protected]]
Sent: Tuesday, October 20, 2015 3:50 PM
To: Radiance general discussion
Subject: [Radiance-general] from a rad and a dat file to ies

Hi Everybody,

I was wondering if it is possible to create a *.ies file from a *.dat and a *.rad file. Originally, I used the ies2rad command to obtain a *.rad and a *.dat file from an IES file. Then I did some edits on the *.rad file and now I would like to create an IES file again. Is it possible?

Best,

Dra. Adriana Lira Oliver
Diseños Sustentables Integrales 3E, S.C.
Directora
w www.3esc.com.mx<http://www.3esc.com.mx/>
e [email protected]<mailto:[email protected]>
t 62 84 84 89
c 04455 21 07 61 41
d Oxtopulco 31-1, Oxtopulco Universidad,
Del. Coyoacán, C.P. 04318, D.F., MEXICO

[Image removed by sender.]

Greg_Ward · October 21, 2015, 4:48pm

Hi Andrei,

While specularities greater than 0.1 are not realistic for non-metals, specularity is limited only by physicality (i.e., 1.0) for metals.

Your method should work well enough for metallic surfaces with uncolored highlights. If you have colored specular reflections, then you may have to do something a little more sophisticated, like multiplying the CIE standard observer curves to get XYZ then convert to RGB for Radiance using xyz_rgb.cal.

Cheers,
-Greg

···

From: "Kolomenski, Andrei (JSC-SF311)[WYLE INTEG. SCI. & ENG.]" <[email protected]>
Subject: [Radiance-general] Specularity Definition of Radiance Material
Date: October 20, 2015 2:08:10 PM PDT

Hello Radiance Experts,

I have used a Hemispherical Spectral Reflectance meter to measure the Specular Included (SCI) and Specular Excluded (SCE) SPDs of a material. The obtained SPDs are in the range of 360 to 740 nm., commonly considered the photopic range. Then I compute the total photopic specularity using the equation below.

For a black shiny metal sample, I compute a specularity of 21%. Is it correct to use a value of 0.21 for my specularity definition of a metal material primitive. I’ve read that specularity values greater than 0.1 are not realistic (http://radsite.lbl.gov/radiance/refer/ray.html#Materials) in Radiance. In the .hdr rendered image, the material appears to be highly specular to the degree that it acts like a mirror. So this why I’m concerned if I’m using the right procedure to define the specularity for the Radiance metal primitive.

Is it correct to use a hemispherical reflectance meter, and the equation above to define the Radiance specularity term.

Thank you for your input,
Andrei Kolomenski

Kolomenski_Andrei_JS · October 22, 2015, 3:39pm

Hi Greg,

Thank you for your reply.

Unfortunately, as I pasted in the Specularity formula from MS Word into my original post, the division sign was lost in the process. The formula is intended to be a spectral power distribution (SPD) summation of the difference between a specular included (SCI) and a specular excluded (SCE) divided by the SCI.

This formula defines a relative specularity as a percentage of the total reflectance. As I understand, Radiance’s material specularity variable is the total specular reflectance normalized percentage, and not the relative specular percentage I was previously thinking.

So that: total reflectance = total specular reflectance + total diffuse reflectance

Using this methodology for a black material I compute a 5% Total Reflectance, and a 20% Relative Specularity, meaning that my Total Specular Reflectance is (5%*0.2) = 1%, and this value (0.01) is used to define Radiance’s specularity variable. Also, the Radiance RGB triplet is scaled by Total Reflectance (0.05).

Does this sound correct to you?

Thank you for the insight,
Andrei Kolomenski

···

From: Greg Ward [mailto:[email protected]]
Sent: Wednesday, October 21, 2015 11:49 AM
To: Radiance general discussion
Cc: Maida, James C. (JSC-SF311); Clark, Toni A. (JSC-SF)[WYLE INTEG. SCI. & ENG.]
Subject: Re: [Radiance-general] Specularity Definition of Radiance Material

Hi Andrei,

While specularities greater than 0.1 are not realistic for non-metals, specularity is limited only by physicality (i.e., 1.0) for metals.

Your method should work well enough for metallic surfaces with uncolored highlights. If you have colored specular reflections, then you may have to do something a little more sophisticated, like multiplying the CIE standard observer curves to get XYZ then convert to RGB for Radiance using xyz_rgb.cal.

Cheers,
-Greg

From: "Kolomenski, Andrei (JSC-SF311)[WYLE INTEG. SCI. & ENG.]" <[email protected]<mailto:[email protected]>>

Subject: [Radiance-general] Specularity Definition of Radiance Material

Date: October 20, 2015 2:08:10 PM PDT

Hello Radiance Experts,

I have used a Hemispherical Spectral Reflectance meter to measure the Specular Included (SCI) and Specular Excluded (SCE) SPDs of a material. The obtained SPDs are in the range of 360 to 740 nm., commonly considered the photopic range. Then I compute the total photopic specularity using the equation below.

Specularity ≝S=100Nλλ=360 nm.λ=740 nm SPD(λ)SCI- SPD(λ)SCESPD(λ)SCI

For a black shiny metal sample, I compute a specularity of 21%. Is it correct to use a value of 0.21 for my specularity definition of a metal material primitive. I’ve read that specularity values greater than 0.1 are not realistic (http://radsite.lbl.gov/radiance/refer/ray.html#Materials) in Radiance. In the .hdr rendered image, the material appears to be highly specular to the degree that it acts like a mirror. So this why I’m concerned if I’m using the right procedure to define the specularity for the Radiance metal primitive.

Is it correct to use a hemispherical reflectance meter, and the equation above to define the Radiance specularity term.

Thank you for your input,
Andrei Kolomenski

Greg_Ward · October 22, 2015, 5:07pm

Hi Andrei,

Actually, your original formula came through for me. Maybe it got mangled on the way back to your inbox.

The formula you had for the specularity, where you divided by the "SCI" component, was correct if you use the Radiance "metal" primitive and apply the overall reflectance as your RGB values. This is because the specular component is multiplied by the RGB color for metals. I don't recommend using the "plastic" type in your case, but if you did, you would want to leave off the denominator as you say.

Cheers,
-Greg

···

From: "Kolomenski, Andrei (JSC-SF311)[WYLE INTEG. SCI. & ENG.]" <[email protected]>
Subject: Re: [Radiance-general] Specularity Definition of Radiance Material
Date: October 22, 2015 8:39:14 AM PDT

Hi Greg,

Thank you for your reply.

Unfortunately, as I pasted in the Specularity formula from MS Word into my original post, the division sign was lost in the process. The formula is intended to be a spectral power distribution (SPD) summation of the difference between a specular included (SCI) and a specular excluded (SCE) divided by the SCI.

This formula defines a relative specularity as a percentage of the total reflectance. As I understand, Radiance’s material specularity variable is the total specular reflectance normalized percentage, and not the relative specular percentage I was previously thinking.

So that: total reflectance = total specular reflectance + total diffuse reflectance

Using this methodology for a black material I compute a 5% Total Reflectance, and a 20% Relative Specularity, meaning that my Total Specular Reflectance is (5%*0.2) = 1%, and this value (0.01) is used to define Radiance’s specularity variable. Also, the Radiance RGB triplet is scaled by Total Reflectance (0.05).

Does this sound correct to you?

Thank you for the insight,
Andrei Kolomenski

Clark_Toni_A_JSC-SF · October 22, 2015, 7:23pm

Greg, according to different ASTM measurement standards, a hemispherical reflectance meter is deemed as accurate for highly specular and non-specular materials. ASTM standards encourage the use of multi-angle measurements for materials that have medium gloss. So it is quite possible, per a given test standard that usage of the hemispherical meter may not yield accurate data.

A multi angle meter is analogous to a BRDF measurement, just less angles. With respect to Radiance terms for Total Reflectance, Total Specular Reflectance, and Total Diffuse reflectance, is there anything we could extract from test data from a multi-angle meter that would allow us to determine terms for Total Specular and Diffuse reflectance, aside from using a BRDF function?

For instance, per Radiance, what angle range around the specular angle constitutes the definition of a specular reflection? For Radiance, what angle range from the specular angle defines the diffuse reflectance? Please see attached image. Here, I am thinking that the definition of Total Specular Reflection should constitute some range of angles around the specular angle. All other angles should fall within the Total Diffuse Reflection category. For instance if I define the Total Specular reflection as the flux that reflects +/- 15 degrees from the specular angle would that be a good definition? Alternatively can we use an analogy from the lamp world for the definition of beam angle (% drop off by angle from maximum relative intensity)?

Thank you for your thoughts as we attempt to reconcile differences in measurement techniques and definitions for parameters for Radiance material classification.

Toni Clark, P.E.

···

From: Greg Ward [mailto:[email protected]]
Sent: Thursday, October 22, 2015 12:08 PM
To: Radiance general discussion
Cc: Maida, James C. (JSC-SF311); Clark, Toni A. (JSC-SF)[WYLE INTEG. SCI. & ENG.]
Subject: Re: [Radiance-general] Specularity Definition of Radiance Material

Hi Andrei,

Actually, your original formula came through for me. Maybe it got mangled on the way back to your inbox.

The formula you had for the specularity, where you divided by the "SCI" component, was correct if you use the Radiance "metal" primitive and apply the overall reflectance as your RGB values. This is because the specular component is multiplied by the RGB color for metals. I don't recommend using the "plastic" type in your case, but if you did, you would want to leave off the denominator as you say.

Cheers,
-Greg

From: "Kolomenski, Andrei (JSC-SF311)[WYLE INTEG. SCI. & ENG.]" <[email protected]<mailto:[email protected]>>

Subject: Re: [Radiance-general] Specularity Definition of Radiance Material

Date: October 22, 2015 8:39:14 AM PDT

Hi Greg,

Thank you for your reply.

Unfortunately, as I pasted in the Specularity formula from MS Word into my original post, the division sign was lost in the process. The formula is intended to be a spectral power distribution (SPD) summation of the difference between a specular included (SCI) and a specular excluded (SCE) divided by the SCI.

This formula defines a relative specularity as a percentage of the total reflectance. As I understand, Radiance's material specularity variable is the total specular reflectance normalized percentage, and not the relative specular percentage I was previously thinking.

So that: total reflectance = total specular reflectance + total diffuse reflectance

Using this methodology for a black material I compute a 5% Total Reflectance, and a 20% Relative Specularity, meaning that my Total Specular Reflectance is (5%*0.2) = 1%, and this value (0.01) is used to define Radiance's specularity variable. Also, the Radiance RGB triplet is scaled by Total Reflectance (0.05).

Does this sound correct to you?

Thank you for the insight,
Andrei Kolomenski

Greg_Ward · October 22, 2015, 10:48pm

Hi Toni,

Unfortunately, there's no reliable method for getting the specular component based on just a few measurements. What Andrei has done is similar to what I have done in the past for metallic surfaces, and works reasonably well when the specular highlight is (a) fairly well-defined as from a polished material and (b) large with respect to the diffuse portion. If assumption (a) is wrong, then the highlight will spill over into the "SCE" region of the spectrophotometer's measurement, and if (b) is wrong, then subtracting "SCE" from "SCI" may not be a very accurate measure of the specular portion.

Short of a full BRDF measurement, I don't know of any methods that work for all surface materials.

Best,
-Greg

···

From: "Clark, Toni A. (JSC-SF)[WYLE INTEG. SCI. & ENG.]" <[email protected]>
Subject: Re: [Radiance-general] Specularity Definition of Radiance Material
Date: October 22, 2015 12:23:33 PM PDT

Greg, according to different ASTM measurement standards, a hemispherical reflectance meter is deemed as accurate for highly specular and non-specular materials. ASTM standards encourage the use of multi-angle measurements for materials that have medium gloss. So it is quite possible, per a given test standard that usage of the hemispherical meter may not yield accurate data.

A multi angle meter is analogous to a BRDF measurement, just less angles. With respect to Radiance terms for Total Reflectance, Total Specular Reflectance, and Total Diffuse reflectance, is there anything we could extract from test data from a multi-angle meter that would allow us to determine terms for Total Specular and Diffuse reflectance, aside from using a BRDF function?

For instance, per Radiance, what angle range around the specular angle constitutes the definition of a specular reflection? For Radiance, what angle range from the specular angle defines the diffuse reflectance? Please see attached image. Here, I am thinking that the definition of Total Specular Reflection should constitute some range of angles around the specular angle. All other angles should fall within the Total Diffuse Reflection category. For instance if I define the Total Specular reflection as the flux that reflects +/- 15 degrees from the specular angle would that be a good definition? Alternatively can we use an analogy from the lamp world for the definition of beam angle (% drop off by angle from maximum relative intensity)?

Thank you for your thoughts as we attempt to reconcile differences in measurement techniques and definitions for parameters for Radiance material classification.

Toni Clark, P.E.

From: Greg Ward [mailto:[email protected]]
Sent: Thursday, October 22, 2015 12:08 PM
To: Radiance general discussion
Cc: Maida, James C. (JSC-SF311); Clark, Toni A. (JSC-SF)[WYLE INTEG. SCI. & ENG.]
Subject: Re: [Radiance-general] Specularity Definition of Radiance Material

Hi Andrei,

Actually, your original formula came through for me. Maybe it got mangled on the way back to your inbox.

The formula you had for the specularity, where you divided by the "SCI" component, was correct if you use the Radiance "metal" primitive and apply the overall reflectance as your RGB values. This is because the specular component is multiplied by the RGB color for metals. I don't recommend using the "plastic" type in your case, but if you did, you would want to leave off the denominator as you say.

Cheers,
-Greg

From: "Kolomenski, Andrei (JSC-SF311)[WYLE INTEG. SCI. & ENG.]" <[email protected]>
Subject: Re: [Radiance-general] Specularity Definition of Radiance Material
Date: October 22, 2015 8:39:14 AM PDT

Hi Greg,

Thank you for your reply.

Unfortunately, as I pasted in the Specularity formula from MS Word into my original post, the division sign was lost in the process. The formula is intended to be a spectral power distribution (SPD) summation of the difference between a specular included (SCI) and a specular excluded (SCE) divided by the SCI.

This formula defines a relative specularity as a percentage of the total reflectance. As I understand, Radiance’s material specularity variable is the total specular reflectance normalized percentage, and not the relative specular percentage I was previously thinking.

So that: total reflectance = total specular reflectance + total diffuse reflectance

Using this methodology for a black material I compute a 5% Total Reflectance, and a 20% Relative Specularity, meaning that my Total Specular Reflectance is (5%*0.2) = 1%, and this value (0.01) is used to define Radiance’s specularity variable. Also, the Radiance RGB triplet is scaled by Total Reflectance (0.05).

Does this sound correct to you?

Thank you for the insight,
Andrei Kolomenski