How to make this glitter material?

[lsscpp]

@moony

Nice illustration, but what are the vectors like at the start (the textured cube is shown - but the vectors are only shown after they have had 0.5 subtracted).

they point in +1 x,y,z starting from origin 0,0,0

[lsscpp]

@alborz

Yes! that’s my problem as well! that is why i tried to add emission and bump to my original material! But maybe the best way for these materials is post processing in Photoshop (to add extra glare?)

As Secrop said, to keep it in the physical realm, no emission, but rather you should shoot more light, and let the Filmic tonemapping do the magic. You often discover how lower the albedo of materials have to be set!
Let’s suppose you use emission in the shadertree. Then you place the lipstick in a dark corner of the scene. What happens? You probably have to tweak the shader again!
In Secrop nodetree, to get brighter glitters, you might want to get rid of diffuse compnent (letting the glossy shine alone) and maybe crank up the ligt blue in the color-ramp.

[alborz]

@lsscpp

As Secrop said, to keep it in the physical realm, no emission, but rather you should shoot more light, and let the Filmic tonemapping do the magic. You often discover how lower the albedo of materials have to be set!
Let’s suppose you use emission in the shadertree. Then you place the lipstick in a dark corner of the scene. What happens? You probably have to tweak the shader again!
In Secrop nodetree, to get brighter glitters, you might want to get rid of diffuse compnent (letting the glossy shine alone) and maybe crank up the ligt blue in the color-ramp.

but if i want to keep it in the physical realm, shouldn’t i keep the diffuse component since all materials have a Diffuse reflection?

[moony]

@lsscpp

they point in +1 x,y,z starting from origin 0,0,0

All of them can’t though - since the voronoi texture will randomise their direction. I was wondering what the distribution of vectors looked like before subtraction of 0.5 which made the distribution spherical around the origin.

Would it have been a hemispherical distribution with it’s base on the surface?

[Secrop]

@lsscpp

they point in +1 x,y,z starting from origin 0,0,0

exactly!

@alborz

but if i want to keep it in the physical realm, shouldn’t i keep the diffuse component since all materials have a Diffuse reflection?

Not necessarily. It depends on the type of material. Materials like this one can be made of different substances… The flake particles are normally metallic, which the glossy alone should be enough. Some are also mixed with other substances, like wax, epoxy, etc. The diffuse is just for mimic that part, and can be removed if not needed.

The only thing important here, is to mimic how the flakes are oriented. The trick I posted is just pratical and fast. But also, as I stated, variations can be made.

[Secrop]

@moony

All of them can’t though - since the voronoi texture will randomise their direction. I was wondering what the distribution of vectors looked like before subtraction of 0.5 which made the distribution spherical around the origin.

Would it have been a hemispherical distribution with it’s base on the surface?

The vectors can be pointing to anywhere inside that cube, starting from the origin; since the color components of the voronoi-cell are uniformly distributed in the [0, 1] interval.

For a more exact ‘hemispherical distribution’, one need to normalize the result after the subtraction of [0.5,0.5,0.5], which will turn the cube into a sphere; and then flip the vectors that are pointing to the negative of the normal. This is fairly a bit more complex, but can be done.

[moony]

@Secrop

The vectors can be pointing to anywhere inside that cube, starting from the origin; since the color components of the voronoi-cell are uniformly distributed in the [0, 1] interval.

For a more exact ‘hemispherical distribution’, one need to normalize the result after the subtraction of [0.5,0.5,0.5], which will turn the cube into a sphere; and then flip the vectors that are pointing to the negative of the normal. This is fairly a bit more complex, but can be done.

Ah - I get it now - thanks.

[RickyBlender]

by normailze you mean make unit vectors => L = 1?

but this won’t change direction or angle !

thanks
happy cl

[Secrop]

@RickyBlender

by normailze you mean make unit vectors => L = 1?

but this won’t change direction or angle !

More or less… It won’t change the vectors direction, but as soon as you add the Normal vector, the result it’s quite different.

For example, if all vectors were very small, and we added the normal vector to them, the result would not be very different from the normal vector itself, no matter the small vectors direction; If they were bigger than the unit lenght and we add the normal, we will even get vectors pointing to bellow the surface.

[alborz]

@Secrop

More or less… It won’t change the vectors direction, but as soon as you add the Normal vector, the result it’s quite different.

For example, if all vectors were very small, and we added the normal vector to them, the result would not be very different from the normal vector itself, no matter the small vectors direction; If they were bigger than the unit lenght and we add the normal, we will even get vectors pointing to bellow the surface.

SO now i have two new problems!:no:

i have a weird shading on the bottom part of the cap (cap is just a simple cylinder) -> which is only visible in camera view <- :mad:

Also i don’t see the reflections on the label (in camera view) but i see them in normal view port!!

I don’t know why everything gets messed up when i try to render in camera view!

My label should look like the cap, shiny with lots of reflections…

[alborz]

having a decent product shot with mild reflections is something i need

[Secrop]

Blend file needed for better review. From the screenshot it’s not possible to determine the cause, since most of the settings are hidden (and the problem can be anywhere… like wrong light path config, wrong object visibility, and since the problem is with reflections, it can be anywhere else outside of your mesh!)

[moony]

The reflection problem on the cap could be due to smooth shading issues across a sharp boundary.

Do you have edge loops at the bottom of the cap to smooth the transition out?

[alborz]

@Secrop

Blend file needed for better review. From the screenshot it’s not possible to determine the cause, since most of the settings are hidden (and the problem can be anywhere… like wrong light path config, wrong object visibility, and since the problem is with reflections, it can be anywhere else outside of your mesh!)

Here is the blend file (without HDRI)
the object’s dimensions are based on real life size so that’s why it is so small.

Model.blend (665 KB)

[alborz]

@moony

The reflection problem on the cap could be due to smooth shading issues across a sharp boundary.

Do you have edge loops at the bottom of the cap to smooth the transition out?

[ATTACH=CONFIG]502123[/ATTACH]

Yes i do! but when i change my camera position that glitch also moves (become bigger or smaller)

[IkariShinji (Thies Schulz-Holland)]

The glossy areas are reflecting your shadow catcher… Make it invisible to glossy rays.

[alborz]

@IkariShinji

The glossy areas are reflecting your shadow catcher… Make it invisible to glossy rays.

[ATTACH=CONFIG]502137[/ATTACH]

Oh shoot! great catch! :eyebrowlift: also please feel free to point out to any node problems or other problems you see! i would really appreciate it

[alborz]

thank everyone for helping :eyebrowlift:

i changed the glass to glossy with low IoOR (clear plastic) and removed solidify and i think it looks closer to the real image!

[dudecon]

Followed the instructions, went a bit crazy, and woke up to this: