[hatts]

Hello, just want to run this material by the forum to see if I'm on the right track.

I'm trying to simulate an electroplated ("candy-coated") magenta metal. The nature of this finish is one of high reflectivity, but where the highlights never go true white, nor the lows true black.

So in my mind, it should be a two-layer material with a regular "mirror" on bottom and a coloration layer above that. But that's not the approach that's worked for me so far.

Behold:


My goal is to get result #3 without the need to colorize the highlights in post.
As stated, the single-BSDF-with-R2 approach does not match my mental model of the two-layer solution.

[Hervé]

adjust the colors as you wish..

http://www.divshare.com/download/21536752-838

[hatts]

Thanks Herve, that was a helpful material to learn from, as you have proper coloration of the highlights.
However I found it sort of inflexible to work with. Forcing fresnel is sort of limiting in my color control, but is necessary to avoid the color pooling documented in my attempt #1...

[bograt]

Try something like this: http://www.divshare.com/download/21547338-34e
Its probably not exactly what your after but it may be a starting point

[Half Life]

I came up against similar issues with the RAL "Anodized Aluminum" line I did a while back -- you can check those materials to see how I solved some of the issues: http://resources.maxwellrender.com/sear ... v2=0&tipo=

I've been saying for a long time that to get materials like these physically accurate we need to have dye coatings as part of the MXM creation toolset... any other solution is just "faking it" to various degrees of success.

Best,
Jason.

[bograt]

Presumably this is the material you are trying to create?

It looks quite contrasty in this light...
Its obviously quite difficult to make a low contast metallic material with roughness 0, but if you match the lighting of a photo reference perfectly you may find you can achieve the desired result just by tweaking the K value...

[hatts]

bograt, I commend your research. That example is actually painted porcelain, and is a more basic finish.

The material I'm going for is a little more difficult, and is for this piece, in electroplated stainless steel.


(EDIT: fixed image)

[hatts]

Also @bograt: your example material was pretty good, but not any different really from the existing approach

@Half Life: yeah, that's a good idea, maybe I'll +1 you if you have it as a wish list post

[bograt]

Your pic is not visible, did you link to an external location?
I know what material you mean, I played around with a similar material a while ago but found that it was pretty difficult to reduce the contrast at roughness 0 without making the whole thing unrealistically light.
here is the material I made with various K values:

Like as Half Life mentioned, my technique is still faking it

[bograt]

Also @bograt: your example material was pretty good, but not any different really from the existing approach

@Half Life: yeah, that's a good idea, maybe I'll +1 you if you have it as a wish list post

I know, I noticed that afterwards

[bograt]

Just for fun:

This is an aluminium ball with a fractionally larger ball with a material with 0 black 0deg/90deg, 1.4 Nd force fresnel, 1mm attenuation and coloured transmittance.
Obviously very impracticable but interesting an interesting test.
you get similar falloff to this:

[hatts]

bograt the result of your "just for fun" experiment is actually outstanding, I tried it as well.
Unfortunately you're correct, it is not geometrically practical.
It led me to try experimenting with coatings but that didn't pan out either.

[eric nixon]

This works quite well, its just tries to mimic the look of your reference. Its not 'correct' in any way, uses several layers with r2, smooth pinky R0 , rougher red R90's + also a pink metallic coating on the top layer, which also has R2 set around 20 degrees like the others.



Its quite hard to dial in, took at least 15 mins . If you want a copy get JEff to Pm me with some pics of his wife

[eric nixon]

Sorry scratch that, just remembered what year it is.


another scene, same mxm... still needs some work..

[rusteberg]

there is a degrading tendency when you approach something with too much thought.