[Maximus3D]

Thanks Tom but i must been falling down the Stupid Tree and hitting every branch on the way down last night because i'm not quite sure i get this part.. i make a velvet shader, and somehow magically that turns itself into a .r2 file, and then i can add that to my material so it does something special there and i get a cool effect. am i close ? eheh..

/ Max

[tom]

It's not specifically about velvet shader but there's a full information on r2 file which you can extract the information about "what's it" and "how can you use it"...

[jotero]

good anistropy test olivier thanks

[Olivier Cugniet]

thank's everyone

tom, would another setting of the r2 file give some good/interresting results ?

[tom]

tom, would another setting of the r2 file give some good/interresting results ?

Sure, why not... if you understand the benefit of 2 parameters explained in the tutorial, you can create many possibilities. Do not memorize, know how it works.

[beppeg]

Hi Olivier, can you share also a Open GL view in texture decal for the first pic?
Thanx for the test

[Sheik]

I did some testing to understand how to control anisotropy. Oliver, on the first image the cylinders at he back, is that the same material and mapping as on the top? I tested that way, with the JD´s angle map projected from the top also to the sides, but that doesn’t seem to work (cylinder in the middle).
The other one has the side mapped separately and doesn’t use the angle map.
On the others I just tested to find out size of the map doesn’t seem to matter on how rough the effect is, it is just about angle (right?).
But what about a straight brushed steel surface, planar and brushed in one direction?
The ground consists of stripes of the same metal with anisotropy, but they are mapped in different directions and scales. I would expect to see the stripes in clearly different colours, and also see the direction of the anisotropy, but it all looks the same .Only roughness (15) is visible but no anisotropy (99). Why?

….and the mapping in Studio:

Sheik

[pascaldechine]

Very nice test Olivier.
I'm trying to get the same result but not really easy.
First I'm using rhino and the plugin and it seems when I make a tube like yours, I have to explose my tube and set the anisotropy material only to the surfave of the tube (i have also to redo the surface to get a simple circle surface). The rest of the tube, I have to set the same material except I must cancel the anisotropy effect or I have some reflections very strange.
May I ask you how did you set your scene for the lights?

Pascal de chine

[tom]

Imagine the grooves on such a cylinder. On its sides the grooves are in one single direction, so angle mapping is not needed. Instead, you should assign a seperate material to sides which has a constant aniso angle that defines the direction of grooves. Why do we map the cap of the cylinder, because for simulating the concentric circular grooves, we need to change the angle of the aniso on the surface per pixel.

[pascaldechine]

ok I see. I just try and it works following what you said...
Thanks Tom

Pascal de chine

[joie]

Excuse me but I don´t get this anisotropic direction thing at all.

In many other 3D programs, the anisotropic direction is driven by the UV mapping of the object, so the UV direction IS the anisotropic direction, you can rotate the entire direction but you can´t map the anisotropic vector, UVs do it for you, and I find that way easier, because I usually model mechanical objects in NURBS with implicit UV mappings, so I get perfect anisotropic effects in Mental Ray with no problems.

[Mihai]

You can basically do the same in Maxwell, the angle map is there to give you more control. On the sides of a cylinder for example you don't need an angle map at all. But if you wanted to make patterns like on that cube on the right, how would you do it from nurbs objects?

[joie]

I´m talking more about the cilinder caps rather than the cube patterns...

For the cilinder caps, with NURBS, you´d only have to make all the CVs converge to one point and you have done it because the UVs would converge to that point also.

[Thomas An.]

This thread needs to be in the tutorial section. Please