I’ve noticed that two materials with the same “dark” base color can look very different depending on their metallic value (1 or 0), and I’m trying to understand why, or if it’s a bug or mistake by my part.
Please, see this example:
The cube on the left (metallic set to 0) looks way brighter/greyish than the cube on the right (metallic set to 1). Both materials have the exact same values other than Metallic: base color is 1d1d1d and roughness is set to 0.4.
Why? Shouldn’t the rougher, non-metallic* material look darker?
Any help to understand this topic is appreciated. Thanks a lot!
I guess it’s the specular value that is being ignored when using the metalness workflow.
Both workflows, metalness and specular can produce similar results depending on the setup.
For more details have a look here
Thank you for your input, Lorenz, but still, how to prevent the colors from differing that much?
In both cases, the specular value is set to 0.5 (in Blender), and I’m using simple test maps generated for each of them in Subtance Painter.
Here’s the node setup for each material in Blender:
Material A (metallic: 0)
Material B (metallic: 1)
The setup is identical, except for the actual map files used. And again, the only difference is in the metallic map exported from Substance (one is flat black, the other flat white).
Thanks in advance for any advice on this topic.
Hi, I don’t know all the scientific details but what I can tell from experience and basic knowledge is that metallics and dielectrics are very different materials. Metals usually have a lighter albedo value than dielectrics, so indeed a dielectric with a .5 looks bright already but for a metallic, you go around .75 or higher (in most cases). Also common metals use just a bit saturation.
I mean, what you describe is not an issue in Blender, but it’s the difference between dielectrics and metals.
You’re mostly seeing a difference in specular.
As a rule of thumb:
With non-metallic surfaces, specular color is always the same, no matter the base color.
With metallic surfaces, specular color is multiplied with the base color.
As a result, the specular influence on your right cube is much lower, resulting in a darker cube.
This is correct behaviour.
Thanks for the replies, Thomas and Peetie. So this behavior is expected.
I understand dieletric and metallic materials react differently to light, but I wasn’t expecting the non-metallic one (Material A, in this case) to become that much brighter. It’s almost a light grey where the light hits more directly in the first image… I’ll give it a few more tries anyway, taking into account that the non-metallic material may need a darker base color, as you pointed.
An extra question that came up:
I’m using Principled BSDF, plugging in the basic PBR maps as image textures, and leaving the Specular value alone in the node setup (default 0.5). Should I tweak it or just leaving it alone is the correct course of action for a metalness workflow in Blender?
Quick update I’d like to share to anyone who may have a similar question in the future:
I tweaked the Specular value in Blender and in fact it made the color look darker as mentioned. Here’s the result:
This setup is slightly different, but it shows how the Specular value affects the materials. I always thought I should just leave it at default when setting up materials in Blender.
Thank you all very much for your help.
Four situations where I consider reducing the specular:
- If I think reflections are too bright at facing angles, this is purely an artistic choice.
- If I know the material structure is porous making specular reflections get absorbed. I.e. rubber.
- By viewing angle on bumpy materials to counter the “anti darkening fix”. Careful with this one.
- Procedurally set it to 0 in shadow gaps. I.e. gaps between wooden floorboards using a single plane.