Author Topic: Accurate anodized aluminum  (Read 7160 times)

2019-06-06, 17:31:05

John.McWaters

  • Active Users
  • **
  • Posts: 268
    • View Profile
    • JohnMcWaters.com
After seeing Apple's website on the design of the new Mac Pro, I began to think about the means of achieving an accurate representation of anodized aluminum (such as that used on window mullions or apple products).

Are there any sources you know of where this has been pulled off? I believe it comes down to replicating the small bumps that the anodizing creates on the surface. Simply adjusting the level of reflectivity isn't enough. It also looks like the reflection color is slightly speckled, but this is probably just from the texture of the surface.

2019-06-06, 18:23:34
Reply #1

Juraj

  • Active Users
  • **
  • Posts: 4759
    • View Profile
    • studio website
You're correct, and that image is probably render as well :- ).

But one thing is tricky: The effect is only achievable currently when you fine tune it for particular resolution and angle. When those variables change, the effect will either sample away into smoothness or create specular artifacts (like darken the surface, break the terminator,etc..).

We still need something like this to be implemented:
It operates the normals on highly sub-pixel information.

I've been seing so many papers on rendering glint over the years but nothing to manifest yet in rendering packages. Right now you have to fine tune intensity and filter/blur every time you change resolution. Ideally you'll render much higher resolution and down-scale in post-production using bicubic.
Please follow my new Instagram for latest projects, tips&tricks, short video tutorials and free models
Behance  Probably best updated portfolio of my work
lysfaere.com Please check the new stuff!

2019-06-06, 20:53:46
Reply #2

John.McWaters

  • Active Users
  • **
  • Posts: 268
    • View Profile
    • JohnMcWaters.com
Thank you for the input. After watching the video you linked, I thought I would try and modify a metallic car paint material as it may behave similarly to the aluminum. I believe I was able to make some headway here by using a very fine noise texture with the bump level turned up higher than one would expect. The fine noise texture is causing some edging to occur on the texture, and I'm unsure why this is occurring.


2019-06-06, 21:04:28
Reply #3

Juraj

  • Active Users
  • **
  • Posts: 4759
    • View Profile
    • studio website
Could be anything but I would first try to do following:

load your noise texture if it's bitmap as CoronaBitmap and set filtering to bicubic. Place that into CoronaBumpMap and use those settings to control intensity and size(blur).
If your noise texture is procedural, also do the same.
Don't use crazy high values in bump. If you Bump map already has absolute White & Black, no reason to go above 1. If that doesn't give you already result you're after, lower the blur in CoronaBumpMap.

Like I said above, normal detail in Corona (or other rendering engines) will not provide 100perc. solution right now, so you have to spread the trick into multiple maps. Place the noise also in reflection and gloss map (doesn't need to be the same intensity or values just to help the normal(/bump) detail).
Please follow my new Instagram for latest projects, tips&tricks, short video tutorials and free models
Behance  Probably best updated portfolio of my work
lysfaere.com Please check the new stuff!

2019-06-06, 21:08:10
Reply #4

John.McWaters

  • Active Users
  • **
  • Posts: 268
    • View Profile
    • JohnMcWaters.com
Okay, I will try these. The Noise texture is being generated by the noise component. I haven't used the Corona Bump Map component much, so I'll give it a try.

2019-06-06, 21:18:48
Reply #5

lolec

  • Active Users
  • **
  • Posts: 179
    • View Profile
One thing that might help you is understanding how that material is achieved in real life.

The "anodized" part is not very important, as it is essentially changing the color of the surface (the resulting microscopic structure from anodizing is so small that we can just ignore it and assume the metal to be that color, just like gold or copper) 

The important part is the sandblasted texture. Accheived by shooting a stream of very fine "sand" into the surface, the sand is made from a material that is stronger than the metal itself, so every time it hits it, it creates a tiny crater, removing a tiny amount of material.

Here is a couple of images I took with a microscope.





2019-06-06, 21:21:56
Reply #6

lolec

  • Active Users
  • **
  • Posts: 179
    • View Profile
Cellular + Smoke is what has worked best for me in the past.

Noise map produces softer results.

2019-06-06, 21:23:42
Reply #7

John.McWaters

  • Active Users
  • **
  • Posts: 268
    • View Profile
    • JohnMcWaters.com
One thing that might help you is understanding how that material is achieved in real life.

The "anodized" part is not very important, as it is essentially changing the color of the surface (the resulting microscopic structure from anodizing is so small that we can just ignore it and assume the metal to be that color, just like gold or copper) 

The important part is the sandblasted texture. Accheived by shooting a stream of very fine "sand" into the surface, the sand is made from a material that is stronger than the metal itself, so every time it hits it, it creates a tiny crater, removing a tiny amount of material.

Here is a couple of images I took with a microscope.

Thanks for sharing these! This is why I think a fractal procedural noise bump would be best. I also came across the image below which further shows the micro texture.

edit: I'll try the cellular/smoke technique as well.


2019-06-06, 21:25:23
Reply #8

lolec

  • Active Users
  • **
  • Posts: 179
    • View Profile
I couldn't find a way to edit my comment.

If you pay close attention to the original image you presented, you can see how it is composed of very small but "sharp" reflections. 

Noise doesn't quite produce that effect, but smoke combined with cellular does.



2019-06-06, 21:34:44
Reply #9

John.McWaters

  • Active Users
  • **
  • Posts: 268
    • View Profile
    • JohnMcWaters.com
I couldn't find a way to edit my comment.

If you pay close attention to the original image you presented, you can see how it is composed of very small but "sharp" reflections. 

Noise doesn't quite produce that effect, but smoke combined with cellular does.

Do you have a recommendation on how to mix the two?

2019-07-09, 15:24:26
Reply #10

John.McWaters

  • Active Users
  • **
  • Posts: 268
    • View Profile
    • JohnMcWaters.com
If anyone is curious, this is the best I've done so far for anodized aluminum.


2019-07-09, 17:58:52
Reply #11

lolec

  • Active Users
  • **
  • Posts: 179
    • View Profile
That is actually pretty good!

What was the actual method you used to achieve that result?


2019-07-09, 18:43:40
Reply #12

John.McWaters

  • Active Users
  • **
  • Posts: 268
    • View Profile
    • JohnMcWaters.com
That is actually pretty good!

What was the actual method you used to achieve that result?

I used a small cellular map in the bump with the bump set pretty high...5.0 I think. I also modified the cellular map and plugged it into the reflection glossiness.

2019-07-09, 18:46:44
Reply #13

PROH

  • Active Users
  • **
  • Posts: 1219
    • View Profile
Pretty nice result :) I like it to. Thanks for sharing.

2019-07-10, 14:09:18
Reply #14

dj_buckley

  • Active Users
  • **
  • Posts: 874
    • View Profile
Looks great John

Are you happy to share how you've got the cellular map setup?