Can this technique be used to do things like putting shiny spots on exposed surfaces, add speckling to toon shaders, control slime locations when combined with my slime shader, etc? Yes it can. There are some drawbacks however. While the AO algorithm is as good for finding exposed surfaces as it is for finding protected surfaces, the implementation of it in Poser 6 appears to be geared toward protected surfaces. I don't think it can tell the difference between a really exposed bit, like the end of a nose and a moderately exposed bit, like the middle of a forehead. "Can this be used to make a brick texture look dirty? (Each individual brick, as opposed to only the corners of the mesh)" Not really. This approach is really for putting dirt in protected corners. If you just want dirt evenly spattered over each brick, there are ways to do it and they're probably easier than this approach. Texture baking? I wish. I don't expect it to every happen in Poser though. Still, the fact that I can make stuff like this for P6 users makes me very happy. Will there be a dirt pack? Not sure yet, but certainly I'm always storing away shaders and it's probably getting to be time I did some more shader packs. I want to take a look at some of the other new nodes first though. They quietly introduced a bunch of new rate of change variable nodes that look really interesting. I'll have to see what I can do with those. If they work how I think they do, I should be able to use them to colour sharp bits differently from blunt bits. Thanks for the interest, everybody. Enthusiastic people keep me going :-)

Any ideas for using the new Variable nodes ?

Here's some dirty chrome :-) Gee, I really better put this down and go and do some work. "Any ideas for using the new Variable nodes?" Yeah, but only vague ideas at the moment. As I was saying, they should allow you to tell areas of rapid surface change (ie sharp bits and creases) from areas of slow surface change. They all seem to be UV map dependent, which limits things a bit. My idea is to use the U and V rates of change to get a sort of average local rate of change and then use that to drive a blender so sharp bits get one colour and blunt bits get another. Average magnitude of the two should do it.

OOO I like that dirty chrome as well.. I think you need to make a pack of the materials you've created already.

Ajax, your big free pack of P5 shaders really helped me solve some problems and taught me a lot. Now this article has me fired up to go make some pristine props really filthy! Many, many thanks! :D

Could you post the material seting for the dirty chrome? I LOVE that!

Gareee, Settings for dirty chrome are above, as requested. The reflection map is one of the IBL probe images that come with P6, but pretty much any reflection map you like will do the job. The key thing going on here is that AO drives the reflection value and it's inverse drives the diffuse value, so you always have diffuse + reflection =1. madmaxh, Thanks :-) I really like to hear that people get right into those shaders and learn stuff by looking at them. That's what I really wanted that set to do - make the whole process of figuring out the material room a bit faster for people.

I had a feeling there would be more to it then just adding reflection like P5. Thanks much for the explanation, and shot! BTw something I thought of this morning.. is there a math setting set that could be used to just remove color saturation from a material setting?

Ajax..I tried this today & it worked quite well..how would we go about adding this setting to an already textured object? Stefan

I'm just guessing, but maybe insert a second blender node at the begnning, and have one half of it oint to the texture, and the other half connect to the existing blender2?

Gareee, For desaturation, just run your coloured input through a maths node and add 0 to it. That'll convert it to black and white. Stef, Just plug your texture map into input 2 of blender 2 (ie the blender that drives the diffuse colour). You can also use blenders to drive the spec, bump etc if you want. You drive the blending on them with the math functions clamp node, exactly the same way as the blender for diffuse. On each blender input 1 is dirt and input 2 is clean. The principal in the shaders I've shown here is simple when you break it down. There are three groups of nodes in the shader. One group provides colour for the clean areas, one group provides colour for the dirty areas and the final group is the AO that drives the blending between dirty and clean. If you want to add bump (for example) to the mix, it's just a matter of coming up with a group of nodes for clean bump and a group for dirty bump and then blending them with the same AO you already have, then plugging the blend into the bump channel of the root node. You can do that for as many different characteristics as you want (spec, ambience etc), just driving the blend each time with your original AO group.

That simple, huh? Man, you need a major in math to know what's going on with the nodes.... ;)

Scary thing is Garee, you are almost right. Writing good shaders requires a decent comprehension of mathematical principles. I have coded Renderman shaders and let me tell you - they can become really complex mathematically and code-wise. Poser (or really Firefly) gives us end-users a very good sub-set of Renderman's power with a MUCH easier interface. Trust me, Ajax's skills can come at a decent price in the graphics industry (look up "technical director" roles sometime).

AKmaterialroombookmark

bookmark ;-))

Really cool dirty shaders ! Very usefull !