Has anyone figured out how to do metallic finishes in Poser?  I don't mean metals...I mean the complex anisotropic look of glitter finishes, flake finishes, and so forth.

Bagginsbill ;)
look for stencil shaders car shaders. here and RuntimeDNA.

Quote - Bagginsbill ;)
look for stencil shaders car shaders. here and RuntimeDNA.

Yep!  That man is a Guru when it comes to the Material Room! I don't think there is anything that has ever been asked that he hasn't been able to come up with a solution for!

Take a look at the sticky at the top of the forum for Material Room, Nodes & Shaders - Tutorials and Discussions (Bookmarks - Updated)

There's a section with tutorial links for creating car paints and similar procedurals that will do what you want.

Well, I've got my reading list for the week!

Downloaded Matmagic, read through a whole bunch of threads in the Node Wizard section at RDNA, followed the links here (those that weren't broken, that is) and read through several more threads.

At this point all I can see is that it looks plausible to fake something acceptable.   The car finishes are very interesting, but it isn't what I'm after...I'm after a more distinct look:  http://www.jamminsam.com/images/bluesparkleLeeCampenjs.jpg

Well, I have some leads on ideas.  Moving on to experiments now.

Quote - Well, I've got my reading list for the week!

Downloaded Matmagic, read through a whole bunch of threads in the Node Wizard section at RDNA, followed the links here (those that weren't broken, that is) and read through several more threads.

At this point all I can see is that it looks plausible to fake something acceptable.   The car finishes are very interesting, but it isn't what I'm after...I'm after a more distinct look:  http://www.jamminsam.com/images/bluesparkleLeeCampenjs.jpg

Well, I have some leads on ideas.  Moving on to experiments now.

I fixed the links later in that thread.

Matmagic????  Or did you mean MatMatic?

Awww.  I'd just figured it out myself and was going to test it.  But thanks so much for the help....and the numbers (which saves a lot of trial and error).

As emulated, the texture is isotrophic, showing more towards the viewer and also kicking back towards light sources.  The outer  layer of the drum wrap is a gloss coating (which gives rise to the more defined specular highlight and the reflection visible mostly on the edges).

The main issue I'm having now is several of the channels have absurd base values, which means the preview display in Poser looks very odd.

Poser's procedural pattern generators are a very limited subset of what is needed. I've asked them several times to let me at the C++ code so I can add more types of nodes.

You might want to think of using images, not as the finished work like so many do, but as data (patterns) to use in the shader.

Then you can get more accurate patterns using other tools, or perhaps by lifting bits and pieces from real drum wrappers.

Quote - Poser's procedural pattern generators are a very limited subset of what is needed. I've asked them several times to let me at the C++ code so I can add more types of nodes.

And they haven't taken you up on that offer?!!!!     I think you know more about that material room and the nodes and what is needed than anyone who currently works there, or even the creator of Poser for that matter.

Think us writing to them and petitioning them to let you would help? 

I don't think so.

It's a complex product management issue - not about the code. If, for example, I introduce 20 new nodes in 20 days, test plans have to be developed for them, they have to be tested on multiple platforms, both 32-bit and 64-bit, Mac and PC, they have to be documented in the manual, etc. This could easily consume 3 people for 2 months.

FilterForge is a pretty nifty Photoshop Plugin for creating all sorts of patterns and effects, which can be made into seamless tiles at the flick of a switch. It lets you save separate diffuse, specular, bump and normal maps if desired (handy for 3D apps). You produce the patterns and effects using a node-based system which anyone who's used Poser, Vue, Bryce etc. should get to grips with quite quickly.

Hello,
I learned to love the SDK of Carrara. That gives the ability to create new nodes without the need to compile the full application again and again.
I managed to make replacements for some of Poser's node functions which were not  available
in Carrara before. There are also a lot of 3rd party plugins available.
Having that in Poser could be one way to finally get some nodes implemented which are missing.
And no need for the SM-Team to do testing of those nodes.

Best regards,
   Michael

Heh.  The main reason I'm going with procedural is to allow the lighting and camera direction interaction of these sorts of materials.  To get that pattern that changes with the direction of light and the direction of view, I'd be using multiple texture maps and some crazy node stuff between them.

Actual drum wrappers are very much about materials that change color and luster radically with small shifts of viewing position or lighting angle.  So impossible to scan or photograph usefully, and (to my mind) difficult to paint.

I'm better off, really, using painted texture maps for my grime layers and logos, and using the procedurals to generate fairly simple patterns (like the glitter look) that are within their capabilities.

That said, I've kept playing with the one above and it's gotten to a rather nice fish-scale look.. Not anything I've seen on a real drum, but something I believe as a drum finish at least.

Quote - Hello,
I learned to love the SDK of Carrara. That gives the ability to create new nodes without the need to compile the full application again and again.
I managed to make replacements for some of Poser's node functions which were not  available
in Carrara before. There are also a lot of 3rd party plugins available.
Having that in Poser could be one way to finally get some nodes implemented which are missing.
And no need for the SM-Team to do testing of those nodes.

Best regards,
   Michael

Right - well now you know what I've been discussing with them for a year. :-)

Quote - Heh.  The main reason I'm going with procedural is to allow the lighting and camera direction interaction of these sorts of materials.  To get that pattern that changes with the direction of light and the direction of view, I'd be using multiple texture maps and some crazy node stuff between them.

Actual drum wrappers are very much about materials that change color and luster radically with small shifts of viewing position or lighting angle.  So impossible to scan or photograph usefully, and (to my mind) difficult to paint.

I think you're misinterpreting my point. When, for example, you use the Cellular node to generate a pattern, and then you run that into the specular node or multiply with the Reflect node, or mix it with any another, you are using the pattern to drive the light-and-direction sensitive effect. Similarly, you could use an image that has the same pattern and drive the same effect. The cellular node is not in and of itself sensitive to light and angles, but does generate an interesting pattern. And the specular node is not in and of itself capable of generating interesting patterns, but it is sensitive to light and angles. When you combine them, you get something with an interesting pattern that is sensitive to light and angles.

I understand that for the "pearl" effect, we need to have the material react to light direction and viewing direction. But it doesn't do that uniformly. It does so in a pattern. While the light sensitive effect is easy to make in nodes, the particular pattern is not. The reason it is not is because the node set in Poser is rather limited. Compare it to Filter Forge and you'll see that in fact Poser's node ops are incredibly limited. The pattern generators are few, and there aren't any image filter ops. With the addition of image filter ops and some more pattern generators, Poser's nodes could generate exactly the pattern found in a "pearl" drum wrapper.

Meanwhile, if you could draw or capture that pattern as any sort of image, then it could be used to drive the light-sensitive nodes in the right pattern.

Quote -
I think you're misinterpreting my point. When, for example, you use the Cellular node to generate a pattern, and then you run that into the specular node or multiply with the Reflect node, or mix it with any another, you are using the pattern to drive the light-and-direction sensitive effect.

No, I'm understanding it....that was the sense of my reply, I hope...I'm just having trouble wrapping my mind around it.  I understand I think how the materials behave in the real world, but the intuitive sense to how they translate into the Poser world isn't there for me.  It just "feels" like there is a difference between driving an angle-dependent color ramp, and having a pre-made color design that is dimmed by angle.  I can't intellectually understand the difference, but something in the back of my mind thinks there must be a difference.

Well, actually...there would be, for that!  But there's no reason you couldn't do some color math instead, to shift the colors as the angle changed.  But it still "feels" as if something isn't quite the same, even if I don't actually know why.

Which may not, itself, reflect the materials I was looking at.  As far as I understand, what they call "crushed glass" is tiny chips of plastic partially remelted and flattened.  So the micro-structure is quite complex...a bit like crushed velvet, except my guess is it reflects most where it is flat to the source (the reverse of velvet, which is darkest looking down into the roots).

But does this work out like the old fishscale materials on a micro-scale?  Down at those scales, I know technically some of the sorts of things happen but that is far from being able to handle them intuitively.  My intuition is trained on a larger scale; give me theatrical lighting fixtures and I know exactly how they are going to behave.  Microscopic corner reflectors, on the other hand...!

Quote -
Similarly, you could use an image that has the same pattern and drive the same effect. The cellular node is not in and of itself sensitive to light and angles, but does generate an interesting pattern. And the specular node is not in and of itself capable of generating interesting patterns, but it is sensitive to light and angles. When you combine them, you get something with an interesting pattern that is sensitive to light and angles.

Right.  That seems clear enough...that there isn't a difference between making a flat render of a cellular noise shader and applying that as a texture map, or using it in Poser...except it is possible the latter actually leads to longer render times!

And there's no reason you can't have different maps for the different effects.  Or, for that matter, have the texture map a greyscale that drives a color ramp; that way you get back total color flexibility.  (My feeling is actually that tonal shifts, not hue shifts, reflects some of the actual materials better).

But by the same token, unless you are using noise and filters that there is no equivalent for in Poser, there's not that much utility in replacing a function with a map, right?

Quote -
I understand that for the "pearl" effect, we need to have the material react to light direction and viewing direction. But it doesn't do that uniformly. It does so in a pattern. While the light sensitive effect is easy to make in nodes, the particular pattern is not. The reason it is not is because the node set in Poser is rather limited. Compare it to Filter Forge and you'll see that in fact Poser's node ops are incredibly limited. The pattern generators are few, and there aren't any image filter ops. With the addition of image filter ops and some more pattern generators, Poser's nodes could generate exactly the pattern found in a "pearl" drum wrapper.

Yah.  The poverty of anisotropic specularity in Poser.  It's there, but so nascent there's not much you can do with it (I've been reading up on almost every thread you've posted to in past year or two.)   It seems obvious to me that in many of these finishes, the specular reflection travels across the tiny ridges that are themselves extruded and folded.  So as you shift the light, each of a hundred little lenticular-shaped highlights would sweep along short paths before being blocked by a fold of the material.

One thing that occurred to me was to somehow detach a normal map from the shadow generation and plug that into the specularity.  I've had a fair amount of luck modeling certain textural effects and rendering them as combined height maps, normal maps, etc., etc.  There's a bit of that on this project (what, you think I'm going to spend the polys on actual snares?)

But that isn't a Poser thing yet.  And it most certainly isn't in Poser 6, which I'm making some small effort to stay back-compatible through.

Basically, though, this has been very informative.  I'm glad I took this sideline because I really wasn't making use of the Poser materials (such as they are!) properly.  I'm already using several of your tricks for chrome. 

But it isn't important to me to try (especially, to try and fail) to make an accurate simulation of a specific vintage drum wrap.  I don't see it worth the time to paint by hand.  I've got plenty of hand-painting to do already on more useful stuff, such as the beat-up "used" texture.  I'd rather use procedurals to as far as they work, and stop there.

After I get done cleaning up some of the UV maps I'll jump back to do the simpler "glitter" texture.  I think I'm gonna plug that into spec channel, tho....rather than have it dependent on the reflection items/map as with your example.  (Unless I'm so tired with all these renders I've gone completely blind, and I'm misreading your example shader!)

Regardless, this has been a lot of fun and you are wonderful for helping like this.

Sigh.  Filter Forge looks wonderful.  But I won't be able to afford it until well after this model is done.

I'm still on my first cup of coffee and may be misinterpreting what I see, but ...

The size of the glossy speculars on the sphere implies lights that are 30 to 40 feet across shining on the stage, but that magically do not cast any shadow.

The shadows indicate the strongest light source is behind the drum.

I can barely understand what I'm seeing with all the contradictions here.

When evaluating a shader, use one low IBL and one spot or infinite.

Oddly enough, it's just default Poser lights.  This is in no way a scene, or even perhaps a representative scene.  I'm not making a render...I'm making a prop.

Sorry, never mind.  It seems whatever I do it just makes you angry at me.  I am sorry that I have been a pain to you.  If it was within my power, I would this thread no longer existed.

My apologies.

I'm not in the least bit angry - I have no idea what you're talking about or reacting to.

I was just telling you what I see.

Yes you're working on a shaders for a prop. So the glossy is too big. That's not an indictment against you as a person - it simply means there is a number that you should make smaller.

I don't have a way to communicate tone here - I'm not spitting or anything. I'm just typing as I do several hundred times a day. I give advice on how to improve, not comfort. I make money as a consultant, so every minute that I'm typing to this forum, I am making less money. Yes I work every minute I can. That isn't bullshit. So I type fast, and what I say is curt - short - minimalist.

So - calm down and either react to my suggestions or don't - makes no difference to me. Don't react to my tone because there isn't any tone - you're making that up. But even if you're not getting a lot out of my posts, other people are  by lurking in this thread. You have no idea how many people learn from these exchanges.

Oh I forgot this:

"Oddly enough, it's just default Poser lights."

Not oddly at all. The default Poser lights are the worst light set ever made for CG work. Ever.

This is exactly my point. Using these lights during shader design is guaranteed to produce a strange shader.

In every new version of Poser I get, the very first thing I do is delete those default lights. I've seen so many users justify rendering with them, saying "these are the defaults", as if SM would not have chosen this default setup unless it was a good setup. That isn't the case.

The setup has been the same since Poser 4, with no justification whatsoever. Now that we have IDL, IBL and raytracing, those lights have no place in the universe.

And especially they should not be used for shader design.