Hey! I'm passing along a tip I picked up somewhere for the same website. For dielectrics, one single IOR value is the way to go because dielectrics get their color data from the albedo, as the specular should always be white in a dielectric material. If you're making a metal material AND you have a render engine that supports RGB IOR (if your IOR has 6 fields in total, 2 for R, G and B), you can instead do this: 1: In the top graph, click "line select". This is where you'll pick the wavelength of R(ed), G(reen) and B(lue) light. (For reference, red is "0.7065 µm (r)", green is "0.5461 µm (e)", and blue is "0.4358 µm (g)".) 2: after selecting your line, let's say red (0.7065 µm (r), you will have the refractive index value (n = 1.7303), and the extinction coefficient (k = 8.1065) 3: Go to your 3D software, and open the IOR properties. Your first two boxes will be for the red channel. Copy/paste the values for n and k into the boxes in that order. 4: Repeat the steps for the green and blue wavelengths. This will, as far as I've been told, give you a 100% accurate representation of a pure metallic material as far as the reflection goes. (Remember, a metallic object doesn't have albedo information, all the color data is produced in the reflection as it's a 100% reflective material by nature.)
Good to know eyeballing it is perfectly acceptable method. :) No need to get too nerdy on this complex topic, when in doubt, study a real world example. Great vid.
Awesome vid! Not sure if anyone remembers the old vray/grant warwick method from years back of using refractiveindex and literally creating custom reflectance curves with the RGB values but overtime ive learned simple is better :p Also excellent point i never realized when I would look up ior values for materials that is the purest/cleanest version of those materials which you rarely ever come across in the real world! If you ever need ideas for vids I would love to see your approach for tonemapping! Great vid cheers!
That’s the whole point, none of the materials are pure. Tap water, river water, all have different properties, therefore eyeballing is still a living breathing art :D About ideas, please share some more! I’ll see to cover tonemapping.
I love gold, everyone has their own interpretation of how it should look. I'm dealing with a client who wants the gold on his product to look 24ct even though it looks like plated metal ILR. just tried ior 30, it's pretty much the same as turning ior off in vray. Client was like it doesn't look glitzy enough and proceeded to show me badly comped renders of gold bars. I love my job.
Don't be a slave to numbers. I think Disney approached it best using a specular value instead of IOR, since the specular value can more easily be artistically controlled as decent ranges are in the 0-1 range; 0.25 = IOR 1.333 (water), 0.5 = IOR 1.5 (most dielectrics), 1.0 = IOR 2.3'ish (diamond) and so on. For metals I use Adobe's edgetint values driven by a facing factor. There are other sources that provide edgetint data, but the list I find is always lacking so I end up eyeballing it by just transitioning to a less saturated and brighter version of the face data. For dielectrics, it's important to know that the IOR values are "theoretical" representations of a perfect surface. While in the real world we use simplified geometry both on the macro and micro scale. We use normal/bump modifications without thinking about lack of shadowing that would occur; meaning there would be loss of energy. We plug in some looked up IOR 1.5 for rubber, even if microscopic pictures clearly show "many holes" in the surface where energy would get lost. If I observe rubber to be less reflective, then I'll adjust the specular artistically. Do *NOT* be a slave to numbers and disregard what you observe. Other places where I'd tweak specularity are within non modelled shadow gaps (i.e. floor boards represented on a flat plane - reflections *would* get lost in the dark cracks not only diffuse), and if a dielectric (metals are visually unaffected) have a coating; pouring epoxy over wood reduces the wood's reflectivity as the interface are much closer to each other than wood/air or epoxy/air. You can see this easily by submerging some glossy bubbly plastic into water - its reflections are severely dropped, but the same won't happen to a metal spoon.
Good study, using corona render's physical material (non-legacy) is also adding a bit of "translation" layer too cause it now really depends on you tweaking the base color and base values only (reflect, rough)
I dont really understand why Refractive Index is correct for some materials but you scroll down and convert the % of the Reflectance and the number dont match. If you want glass IOR 1.5 but why gold IOR is 30 and not the refractive index they put 5.15? I also swear it was 4.7 few months ago on the site :) IOR dont chnage lol
From what I understand no way, devs said to make things easier for users, they removed IOR from metals, but if you wish to play with IORs you can enable "Use Complex IOR for metals" in Advanced options in material which still doesn't give you a direct IOR input... So I guess only way if you want to be exact is to go with Legacy materials.
![ComfyUI Workflow Bauen & mit FLUX.1 [SCHNELL/DEV] + Loras Experimentieren](http://i.ytimg.com/vi/c3rSqj5Xzr8/mqdefault.jpg)
Hey! I'm passing along a tip I picked up somewhere for the same website.
For dielectrics, one single IOR value is the way to go because dielectrics get their color data from the albedo, as the specular should always be white in a dielectric material.
If you're making a metal material AND you have a render engine that supports RGB IOR (if your IOR has 6 fields in total, 2 for R, G and B), you can instead do this:
1: In the top graph, click "line select". This is where you'll pick the wavelength of R(ed), G(reen) and B(lue) light.
(For reference, red is "0.7065 µm (r)", green is "0.5461 µm (e)", and blue is "0.4358 µm (g)".)
2: after selecting your line, let's say red (0.7065 µm (r), you will have the refractive index value (n = 1.7303), and the extinction coefficient (k = 8.1065)
3: Go to your 3D software, and open the IOR properties. Your first two boxes will be for the red channel. Copy/paste the values for n and k into the boxes in that order.
4: Repeat the steps for the green and blue wavelengths.
This will, as far as I've been told, give you a 100% accurate representation of a pure metallic material as far as the reflection goes. (Remember, a metallic object doesn't have albedo information, all the color data is produced in the reflection as it's a 100% reflective material by nature.)
Excellent points there, thank you!
Great example. Its always best to try and see what works and what not.
Good to know eyeballing it is perfectly acceptable method. :) No need to get too nerdy on this complex topic, when in doubt, study a real world example. Great vid.
Very logical conclusion. And I want to say thank you because you are always straight go the point without much time waisting bla bla...
Awesome vid! Not sure if anyone remembers the old vray/grant warwick method from years back of using refractiveindex and literally creating custom reflectance curves with the RGB values but overtime ive learned simple is better :p Also excellent point i never realized when I would look up ior values for materials that is the purest/cleanest version of those materials which you rarely ever come across in the real world! If you ever need ideas for vids I would love to see your approach for tonemapping! Great vid cheers!
That’s the whole point, none of the materials are pure. Tap water, river water, all have different properties, therefore eyeballing is still a living breathing art :D
About ideas, please share some more! I’ll see to cover tonemapping.
I love gold, everyone has their own interpretation of how it should look. I'm dealing with a client who wants the gold on his product to look 24ct even though it looks like plated metal ILR. just tried ior 30, it's pretty much the same as turning ior off in vray. Client was like it doesn't look glitzy enough and proceeded to show me badly comped renders of gold bars. I love my job.
Don't be a slave to numbers. I think Disney approached it best using a specular value instead of IOR, since the specular value can more easily be artistically controlled as decent ranges are in the 0-1 range; 0.25 = IOR 1.333 (water), 0.5 = IOR 1.5 (most dielectrics), 1.0 = IOR 2.3'ish (diamond) and so on.
For metals I use Adobe's edgetint values driven by a facing factor. There are other sources that provide edgetint data, but the list I find is always lacking so I end up eyeballing it by just transitioning to a less saturated and brighter version of the face data.
For dielectrics, it's important to know that the IOR values are "theoretical" representations of a perfect surface. While in the real world we use simplified geometry both on the macro and micro scale. We use normal/bump modifications without thinking about lack of shadowing that would occur; meaning there would be loss of energy. We plug in some looked up IOR 1.5 for rubber, even if microscopic pictures clearly show "many holes" in the surface where energy would get lost. If I observe rubber to be less reflective, then I'll adjust the specular artistically. Do *NOT* be a slave to numbers and disregard what you observe. Other places where I'd tweak specularity are within non modelled shadow gaps (i.e. floor boards represented on a flat plane - reflections *would* get lost in the dark cracks not only diffuse), and if a dielectric (metals are visually unaffected) have a coating; pouring epoxy over wood reduces the wood's reflectivity as the interface are much closer to each other than wood/air or epoxy/air. You can see this easily by submerging some glossy bubbly plastic into water - its reflections are severely dropped, but the same won't happen to a metal spoon.
Thank you, I struggle a lot with IOR
Good study, using corona render's physical material (non-legacy) is also adding a bit of "translation" layer too cause it now really depends on you tweaking the base color and base values only (reflect, rough)
Thank you so much, man! You helped me !
I dont really understand why Refractive Index is correct for some materials but you scroll down and convert the % of the Reflectance and the number dont match. If you want glass IOR 1.5 but why gold IOR is 30 and not the refractive index they put 5.15? I also swear
it was 4.7 few months ago on the site :) IOR dont chnage lol
that's why i hate physics corona mat ) . i always used 8-12 ior to metal, but in physics mat all metal too reflectivity
So...what about the default settings in Corona Physical Mats when you switch on metal...IOR is grayed out.
From what I understand no way, devs said to make things easier for users, they removed IOR from metals, but if you wish to play with IORs you can enable "Use Complex IOR for metals" in Advanced options in material which still doesn't give you a direct IOR input... So I guess only way if you want to be exact is to go with Legacy materials.
@@RenderRam @mxmillo see my comment above for complex IOR values in metal materials! Metals SHOULDN'T use a direct IOR input!