i know for what is this node used. for brushed metal right? or something similar.
but please explain to me  why is there xdir_x,xdir_y and xdir_z ?
for what is it used?
dont tell me that i should use only u and v hightlight size. i want to know what it means.
this node is in poser for years now. 

 Nothing in nature has a perfectly circular specular highlight like you'd find on a Phong shader, so Anisotropic is probably the most underrated specular type of all.

Traditionally it is used to create brushed metal, stainless steel but can also be used for (procedural) hair or any other object that has "irregular" specular highlights.

Normally xDir, yDir and zDir (don't know the exact names in PoserSpeak) distort the direction of your highlight while highlight size will affect intensity or width. Really the best way to understand this is to load a sphere, apply an Anisotropic and tweak the settings.

 replicand has it right, as far as I know.

I don't think I've ever used Anisotropic for metal, but I use it all the time in hair and eyes.  To me, it has a more natural effect than other specularity settings.

Yes, that's right, anisotropic looks great when used for wet skin and hair!!!

Quote -  Nothing in nature has a perfectly circular specular highlight like you'd find on a Phong shader, so Anisotropic is probably the most underrated specular type of all.

Traditionally it is used to create brushed metal, stainless steel but can also be used for (procedural) hair or any other object that has "irregular" specular highlights.

Normally xDir, yDir and zDir (don't know the exact names in PoserSpeak) distort the direction of your highlight while highlight size will affect intensity or width. Really the best way to understand this is to load a sphere, apply an Anisotropic and tweak the settings.

you think i didnt do tests? :)
thats why i opened this thread. there are no changes. is my poser software broken or wht?there is no change in the redner when i change xDir, yDir and zDir

 What is your lighting setup, ice-boy?  That may have something to do with why you're not seeing changes.

 I'm seeing the same thing with infinite lights, Anisotropic plugged into alternate specular and normal specular turned off. I can sort of control the direction using U or V highlight size but x-zDir doesn't appear to work. Not very intuitive, this one.

its tehre for years and i dont remember if it worked. and since i dotn know how it works why isnt it explained? 
you wont tell me that its just there  and not working,

p.s. who uses the skin node? why its still there? you have bagginsbill's skin shader for free and its like .........200 times better.

 skin is useful for things other than skin ;)  makes a nice cloth.

Quote -  skin is useful for things other than skin ;)  makes a nice cloth.

i am using gamma corrected shaders which are mroe realistic then the standard shader settings. and i dont know why i would use skin for cloth. it has build ambient in. i can not use this. the specular behaves very strange. 

 Well just because YOU don't have any use for it doens't make it useless in general ;)

I've never used the skin shader for cloth but I can see how it might work for certain kinds of fabric.

I've once had a very nice (pre BB) Skin Shader that relied mostly on the clay shader. Looked very weird unrendered but had a wonderfully velveteen sheen to it. Very nice skin.

But... replicand.. are you saying that the Anisotropic is broken in P7? I've always only used it for the specular highlights in the eyes, and I haven't noticed that it behaved differently... 

i think its broken.

 @ TrekkieGrrl, I'm not saying that it's broken but in P7 (OSX 10.5) I get some pretty wonky results, unlike my example above. If you don't mind my saying so, if you're using it for eye highlights - that's a pretty small surface area to determine whether or not it's working properly.

Why has not BagginsBill commented yet?

Quote - Why has not BagginsBill commented yet?

It might be his one day of rest in seven.

Skin node works fine in Poser 5, in Poser 6 and 7 is really ugly, but I don't think much people really know how to use it, and yes, makes a better velvet than the velvet node :-)

* *I haven't commented because, indeed, I was resting on the seventh day. Heheheh.

But seriously, I don't have an answer here. Out of my 10,000 hours of practice to become an expert and know everything, I've only completed about 4,000 hours. Determining the exact functionality or lack of functionality of those 3 parameters on the Anisotropic node is somewhere in the next 6,000 hours of work.

However, I'll tell you that in the first 4,000 hours of work, I HAVE NEVER SEEN THOSE PARAMETERS DO ANYTHING AT ALL. I CANNOT FIND ANY CHANGE IN OUTCOME.

can we use texture images to control when the node is using U and V ? 

for example controling where it is?

Of course you can. You can modulate just about any parameter by plugging in some nodes.

lets say i have an object that is in reality brushed material. so its blury.
thats because of the microscopic lines.

so if i paint in photoshop a black and whtie image i can control this direction?
for example sometimes the metal is brushed more round then straigh lines.

BB is this similar to the shader that you made for the CD at runtimedna. runtimedna doesnt work so i can not look .

this would be nice in poser
www.neilblevins.com/cg_education/aniso_highlights/aniso_highlights.htm

We have that. What do you mean?

it would be nice if it would look like that in poser. yes the word anisotropic is on the node. but is it acting like it should? 
i am  trying to make it work  with hair. i was trying something with the edge blend. or maybe combining or substracting with another specular node  . but no luck

Here are two of my perfect UV mapped sphere (U horizontal, V vertical).

Both have an Anisotropic node for specular.

Left has u_highlight_size = 1, v_highlight_size=.05.
Right has the opposite, u size = .05, v size = 1.

but the specular goes in the center. is that normal?

Attached Link: http://www.neilblevins.com/cg_education/aniso_highlights/example_sphere_vertical_grooves.jpg

Poser Pro render.

Ice-boy you showed me a bunch of pictures that I can make in Poser. Are you saying you can't make those? Because I make those.

Quote - Still not understanding what you mean?

Poser Pro render.

those are your balls? or poser balls? 

My balls. Poser primitives have bad UV maps. You have to use props like those other guys were using. The anisotropic effect follows the UV contours.

I have tried to explain why the Poser primitives have bad UV maps to the SM people. The content guys there don't understand CG. They remapped the new ones completely different than what I was saying. Thank goodness they never released them. The old ones were bad enough. The new ones were worse.

So now I make my own primitives.

then the primitives are the reason? 

could we also make good hair specular with good UV's ?

thanks

OMG. i never looked at the UV from the sphere.

is this some sick joke or what?

Yes I told them. That is why I had to build the Environment Sphere myself. And the Orb.

The Poser torus is worse. The cylinder isn't too great either.

Quote - could we also make good hair specular with good UV's ?

Yes. The example hair render I posted has good UV.

OMG i now looked at the UV from the cube. there is space between.
in the name of god who did this?  he he he

so when you said that you made a shader where you light individual hairs. is this done with a better UV map or just with nodes? is this some new trick ? 

just trying to start a conversation he he : -)

You make me laugh out loud.

OK here's a hint. I use a Fractal_Sum to make long tiny grooves, too small for bump but visible anyway. I drive the Anisotropic with this. Only works if all the hairs are in the same direction, i.e. vertical or horizontal. This is my definition of a good hair UV - all hairs in the same direction ON THE UV MAP, so procedurals can do their job.

Kozaburo did not do this so I don't like his UV maps. Some are horizontal, some vertical, in the same map. He did this to use space.

try the ALL BACK HAIR. all hair are in the same direction.

i actually did this already. i changed teh UV's from some hair that i have on my computer. like oyu said. but i used a textur emap that i made.  always happy to see that sometimes i am thinking right.

thanks.

Quote - The anisotropic effect follows the UV contours.

.

i dont understand this. what are contours? sorry but thats by bad english. how does it follow it ?

thanks.

A contour line is when you have a 2-dimensional function and you draw a line or curve through the 2-D space following a path where the function has a constant value.

http://en.wikipedia.org/wiki/Contour_line

In my case, I was talking about lines of constant U or lines of constant V (UV coordinates) mapped onto the geometry of our models.

For the Anisotropic node to work properly, the fine microscopic lines we are hoping to simulate must line up with a U contour or a V contour. If not, it will not work.

Quote - > Quote - could we also make good hair specular with good UV's ?

Yes. The example hair render I posted has good UV.

does every hair need to be in line or just in the same direction? 
for example is this good? they are in the same direction.

specular

so is it normal that the anisotropic goes in the middle of the model? 
this is the UV.

Attached Link: http://www.renderosity.com/mod/forumpro/showthread.php?message_id=3153609&ebot_calc_page#message_3153609

thanks will try out.

Thanks for that node clip, there, ice-boy. Had me a bit of fun with it. First of all, I have NO blessed idea what I'm doing. I'm attempting to leverage what I think I know about shaders in building this one. Probably my first mistake, since my knowledge is still on pretty shaky pilings.

Anyway, here's the code:

  addNaught = Add(0, 0)
  hairTxtr = FractalSum(
    .000004 * u_Texture_Coordinate(),
    4 * addNaught, 
    4 * addNaught, 
    8, 
    .3, 
    .5, 
    .5).labelled("FractalSum")

  setSpecular = Anisotropic(
    hairTxtr, 
    4, 
    .5, 
    .22, 
    0, 
    1, 
    0).labelled("Anisotropic")

Not exactly your original numbers. Here's the shader in a form you're more familiar with:

...and the hair file I cobbled together quickly in the GIMP to use with Tequila Hair:

..and a piccie:

Not sure if the weird artifacts in the hair is due to the poor colourMap or if I need to muck around with the shader. Still have a long way to go with this one... but thanks for the clues, BB. Probably totally mis-using them and everything, but I'm having a great time doing it!

I think the color map needs to be plugged into the bump or displacement, as well, to really bring out the anisotropic use

i think hte fractal sum should be used for the bump

Right.
Trying to sort out what the different nodes are doing. Anisotropic I think I've kinda been playing enough with to know at least how Specular_Value and u and v highlight interact. Which is so cool... done some interesting things with shiny clothing using that and Edge_Blend.
And u_texture_coordinate is just a directional thing. Plugging it into the xScale is a bit weird to me, because the x-axis says side-to-side, not vertical. In the colourMap, the hair strands run vertically. Even the preview image is vertical. But hey, the stripes u_texture_coordinate creates don't do anything appropriate in the y scale or z scale channels - I tried - so x scale it is. I guess we must be talking about:
**Generally on any figure or prop, the U coordinate is chosen to go horizontally around the geometry and the V node goes vertically on the geometry.
**Meaning, the vertical lines are arranged horizontally across the geometry?
Think of it as latitude and longitude, like on a map. The value usually goes from 0 to 1 in both dimensions, but that is not an intrinsic limitation.
The same rules apply as to how to control the characteristics of the stripe. Play with this one for a while. Also try using the V node instead of the U node.
So I did.

Started to run into trouble. It was time to put up a test square, group and assign a material to the group, and try the texture against the group:

s = Surface()
s.Alternate_Diffuse = Blender(
  Spots(WHITE, IColor(0,8,190),.9,0, .25) * IColor(78,78,78),
  IColor(214,0,61),
  .8)
s.Alternate_Specular = Anisotropic(
  Specular_Colour = Fractal_Sum(.000004 * u_texture_coordinate(), 4, 4, 8, .3, .5 .5)
  Specular_Value = 4,
  u_Highlight_Size = .2,
  v_Highlight_Size = .6,
  x-dir_x = 0,
  x-dir_y = 0,
  x-dir_z = 0)

Gets me this:

No idea what is causing that disruption in the texture 3/4 of the way down. Any ideas?