Brushed steel material

[tricky]

I've battled with this sort of material over the years. Usually, the camera is at a distance where I can 'fudge' the results. However, I now need to be very close to a metal leg of a table: a simple cylinder about 60mm diameter and 900mm high.

Here are a few snaps of the look I am going for:


You can see it has a vertical 'stretching' of the reflections along its length, but also a cylindrical brushed effect. The top piece has your typical concentric circular brushing.

I have been playing around with bitmaps in the diffuse slot, anisotropy, anisotropy rotation, hdr lighting etc and tied myself in a not.

Does anyone know of a good tutorial (or an 'off the peg' material to buy) that will help me out? I found a Corbell tutorial that deals with anisotropic rotation which might help with the top 'disc' of metal, but the results have never been particularly good.

[Pixelcon]

The top piece is easy using a gradient ramp set to spiral in the anis-rotation slot.

The other bit is easy too. The long stretches will happen regardless because of the cylindrical shape. That Anisotropic stretching also happens in the opposite direction from the grain. As you can see in your third image there appears to be a grain running in the opposite direction from the stretching.


I've attached my starting point metal which you might try. You can turn off the bitmap as it doesn't do much but it is useful to visualize how the anisotropic effect will appear.

metal.zip

[Syclone1]

This tutorial at evermotion seems to be quite popular

http://www.evermotion.org/tutorials/...1-5-final-sp-1

-dave

[tricky]

The top piece is easy using a gradient ramp set to spiral in the anis-rotation slot.

The other bit is easy too. The long stretches will happen regardless because of the cylindrical shape. That Anisotropic stretching also happens in the opposite direction from the grain. As you can see in your third image there appears to be a grain running in the opposite direction from the stretching.


I've attached my starting point metal which you might try. You can turn off the bitmap as it doesn't do much but it is useful to visualize how the anisotropic effect will appear.

[ATTACH]25489[/ATTACH]

Thanks James - I will take a look at that file.

[tricky]

This tutorial at evermotion seems to be quite popular

http://www.evermotion.org/tutorials/...1-5-final-sp-1

-dave

Looks promising. If I can achieve those sorts of results I'll be quite happy.

[Vizioen]

The top piece is easy using a gradient ramp set to spiral in the anis-rotation slot.

The other bit is easy too. The long stretches will happen regardless because of the cylindrical shape. That Anisotropic stretching also happens in the opposite direction from the grain. As you can see in your third image there appears to be a grain running in the opposite direction from the stretching.


I've attached my starting point metal which you might try. You can turn off the bitmap as it doesn't do much but it is useful to visualize how the anisotropic effect will appear.

[ATTACH]25489[/ATTACH]

I never really understood what either the rotation, anistotropy correlation is. Or even the UV vectors derivation.

I opened your scene and I was checking. Your anisotropy is at 0.4 and rotation at 90°. If I put the anisotropy at -0.4 and the rotation at 0°. I get the same result. Is that a normal thing?

[Pixelcon]

noooo idea :P

[mattclayton]

...If I put the anisotropy at -0.4 and the rotation at 0°. I get the same result. Is that a normal thing?

Yes that's normal...or rather expected i should say. I tend not to use the rotation too often, generally you want it reacting either horizontally or vertically so i just use negative or positive values accordingly.

[Vizioen]

Yes that's normal...or rather expected i should say. I tend not to use the rotation too often, generally you want it reacting either horizontally or vertically so i just use negative or positive values accordingly.

So in conclusion:

Anistropy stretches the highlights. Whereas the reflection map distorts it according to the bitmap.

Question is: the stretching of the highlight is this according to the camera/viewer's Z-axis or the object's?

[mattclayton]

...Question is: the stretching of the highlight is this according to the camera/viewer's Z-axis or the object's?

Couple of quick tests below...Same material on all objects - cylinders are instances - Material Anisotropy set to -0.5



As you can see from the variance in width of highlight on the cylinders it's not reliant on the object orientation.

Out of interest i did i second shot with the camera rotated round to see if that had any affect but clearly not...i assume this is because it's not technically a change of viewing angle, more just a rotation of the canvas...thought i'd put it up anyway just to be thorough

[Vizioen]

Couple of quick tests below...Same material on all objects - cylinders are instances - Material Anisotropy set to -0.5

[ATTACH=CONFIG]25636[/ATTACH][ATTACH=CONFIG]25637[/ATTACH]

As you can see from the variance in width of highlight on the cylinders it's not reliant on the object orientation.

Out of interest i did i second shot with the camera rotated round to see if that had any affect but clearly not...i assume this is because it's not technically a change of viewing angle, more just a rotation of the canvas...thought i'd put it up anyway just to be thorough

Great test, thank you. This might seem like a stupid question, but are there actually objects that stretch in another direction? Meaning why would you use positive anisotropy values?

[Macker]

Great test, thank you. This might seem like a stupid question, but are there actually objects that stretch in another direction? Meaning why would you use positive anisotropy values?

It depends on what direction the strokes of the brushed metal are going, not the underlying material. The effect is created by hundreds of tiny grooves. The highlight/lobe stretches at 90 degrees to the brush stroke/ grain.

[mattclayton]

It depends on what direction the strokes of the brushed metal are going, not the underlying material. The effect is created by hundreds of tiny grooves. The highlight/lobe stretches at 90 degrees to the brush stroke/ grain.

Yeah, that's how i understand it. Especially looking at the earlier reference images on this thread.

[Pixelcon]

It depends on what direction the strokes of the brushed metal are going, not the underlying material. The effect is created by hundreds of tiny grooves. The highlight/lobe stretches at 90 degrees to the brush stroke/ grain.

Which is how the material set up in the file I attached works...