Aluminum material - achieve dull without looking fake?

[sean72]

I have been trying to achieve a relatively flat aluminum but have not found a way to reduce the reflection without exaggerating the roughness to the point where the material doesn't look quite real. I have attached examples and my progress to date. Any suggestions or examples?

[Trixtr]

If you apply a different enviroment you should get some better results. Furthermore you could add an(or many) opacity label to add scratch marks and defects. I too find it very difficult to get a realistic result, but these tricks usually make it better.

[cbecks]

Excellent question. I find that a lot of KeyShot materials are just a little too reflective. Hope we can find a good answer here.

I hope we won't have to adjust the environment to make a material look right. If the part is already in the environment we like, which simulates the real one.

[RRIS]

You can try to liven up the material a little bit by switching from a metal material to a generic material. Then you can add a little bit of clearcoat (the first photo you posted has some nice shiny reflections, you could try to mimic that). Another option is to combine your roughness map with a second one (if you already use one). Something like a fractal noise map or a dirt texture.

[mattjgerard]

I've found that when I'm fighting with something like this, its always my lighting that is lacking. So, make sure you have a contrasting HDRI with some rather sharp black to white transitions. The way the HDRI transitions between light and dark really affects the apparent shininess of the object. The sharper the falloff, the shinier the surface.

[cbecks]

I've found that when I'm fighting with something like this, its always my lighting that is lacking. So, make sure you have a contrasting HDRI with some rather sharp black to white transitions. The way the HDRI transitions between light and dark really affects the apparent shininess of the object. The sharper the falloff, the shinier the surface.

The sharper the falloff, the shinier the surface.
Great tip Matt. I'll remember that. Thanks

[cbecks]

I have been trying to achieve a relatively flat aluminum but have not found a way to reduce the reflection without exaggerating the roughness to the point where the material doesn't look quite real. I have attached examples and my progress to date. Any suggestions or examples?

So sounds like maybe an HDRI with less contrast (smoother falloff) would make your parts look less shiny.
Did you find a good solution?

[Will Gibbons]

I'll add a comment here too. I will often use plastic for dull metal materials. Crank the IoR and then you have control over diffuse and specular values.

[HaroldL]

I find it odd that with KeyShot's "scientifically accurate" materials there is still a lot of work that needs to be done to get an accurate representation of a material. I'm not sure what scientifically accurate means in regards to materials especially metal. Maybe it's just a way of having a consistent starting point.  I think it creates a lot of frustration when a material doesn't look right for a given environment or scene and then you end up using a material that is not even what you want it to look like. 

[RRIS]

I find it odd that with KeyShot's "scientifically accurate" materials there is still a lot of work that needs to be done to get an accurate representation of a material. I'm not sure what scientifically accurate means in regards to materials especially metal. Maybe it's just a way of having a consistent starting point.  I think it creates a lot of frustration when a material doesn't look right for a given environment or scene and then you end up using a material that is not even what you want it to look like. 

Well, just like real life, you get most of your material expression from your lighting and environment. I find I can get really good results with careful attention to your studio, in that respect Keyshot is no different from any other render/3d software. Also, keep in mind that professional promo renders are often heavily photoshopped, with added/cleaned up reflections and shadows, boosted contrast and colors.
Ask any professional product photographer, and you will get an idea of the work and skill involved in creating spectacular imagery.

[mattjgerard]

Well, just like real life, you get most of your material expression from your lighting and environment. I find I can get really good results with careful attention to your studio, in that respect Keyshot is no different from any other render/3d software. Also, keep in mind that professional promo renders are often heavily photoshopped, with added/cleaned up reflections and shadows, boosted contrast and colors.
Ask any professional product photographer, and you will get an idea of the work and skill involved in creating spectacular imagery.

And this here is where some of the art and experience comes in, knowing what to fight for in the 3d app, and knowing what to let go of until you do some post processing. I have learned with our specific products what aspects of the image I will never be able to get to look right in KS, so I stop worrying about it. I have started thinking more about how to create assets for my photoshop comp than trying to get a 98% final render from keyshot. This all makes sense to me from my past days of shooting REDRAW linear on our old RED EPIC camera. Looks gross while shooting, flat and washed out, but all the info was there for manipulation in post. With KS, I'd rather render a flattish low contrast render and be able to tweak and tune in Photoshop. I'm really starting to see the benefit of the higher bit psd files, they allow so much more flexibility. I think that sort of thing would help especially in cases like this, as once KS renders it white, its 100% white, there's no data left and you can't make any adjustments to that highlight.

[Anindo Ghosh]

what scientifically accurate means in regards to materials especially metal

Aluminium is an interesting metal: That whitish surface effect seen in real-world aluminium samples consists of a layer of Aluminium III Oxide (alumina) 4 to 6 nanometers thick, which forms almost instantly (in picoseconds, according to Wikipedia) when a newly machined aluminium surface is exposed to air.

This form of aluminium oxide is an opaque white ceramic material with low specularity. To render it, one might try adding a low opacity label of a material with white diffuse and dark grey specular values over the aluminium.

I had expected the "anodized" parameters of the metal to yield some useful result, but at an anodization thickness of 4 to 5 nanometers, there seems to be no visible effect, and beyond that a yellowish tint appears. The white powdery effect doesn't show up.

The other such problematic metal to render is titanium: Again, real-world titanium instantly forms an abrasion-resistant, low specularity oxide layer of 1 to 2 nanometers, which 3D rendering engines don't seem to offer a way to simulate.

If someone does figure out a way to render that oxide coating realistically, it would be invaluable for some of my work, so please share!.

(n.b. I'm using the IUPAC spelling of Aluminium by choice, rather than the Americanized "Aluminum")

[RRIS]

what scientifically accurate means in regards to materials especially metal

Aluminium is an interesting metal: That whitish surface effect seen in real-world aluminium samples consists of a layer of Aluminium III Oxide (alumina) 4 to 6 nanometers thick, which forms almost instantly (in picoseconds, according to Wikipedia) when a newly machined aluminium surface is exposed to air.

This form of aluminium oxide is an opaque white ceramic material with low specularity. To render it, one might try adding a low opacity label of a material with white diffuse and dark grey specular values over the aluminium.

I had expected the "anodized" parameters of the metal to yield some useful result, but at an anodization thickness of 4 to 5 nanometers, there seems to be no visible effect, and beyond that a yellowish tint appears. The white powdery effect doesn't show up.

The other such problematic metal to render is titanium: Again, real-world titanium instantly forms an abrasion-resistant, low specularity oxide layer of 1 to 2 nanometers, which 3D rendering engines don't seem to offer a way to simulate.

If someone does figure out a way to render that oxide coating realistically, it would be invaluable for some of my work, so please share!.

(n.b. I'm using the IUPAC spelling of Aluminium by choice, rather than the Americanized "Aluminum")

Not sure how it would look, but you could always try adding the white material as a label with a texture to dictate opacity (so a uniform gray level). Benefit of this method is you have independent control over metal and oxide colors. Or, for newer versions of keyshot, use a generic shader with a metallic value inbetween 0 and 1 (instead of pure 0 or 1 like you'd do typically). But, you lack control over the oxide color, since both use the same base color.
And of course you can add details such as anisotropy and surface imperfections.

Attached image is rather quickly done with the generic shader, metallic value of 0.8, roughness 0.25, anisotropic of 0.65.

[mattjgerard]

RRIS that's pretty good for a quick study. i have not dived into the generic shader, I think if I spent the time to learn it, could be pretty powerful at creating a material from scratch just like you did here.

[RRIS]

RRIS that's pretty good for a quick study. i have not dived into the generic shader, I think if I spent the time to learn it, could be pretty powerful at creating a material from scratch just like you did here.

As long as you don't need SSS or transparency, it's pretty powerful (I think it's supposed to be the Keyshot version of the BSDF shader, well suited for PBR texture sets). I couldn't say if it's actually faster to render than the other Keyshot shaders though. But I now use it for 90% of my materials (I like the metallic slider like I used in the sample above, and the availability of a separate clearcoat layer).