Forum: Poser - OFFICIAL


Subject: VSS Skin Test - Opinions

bagginsbill opened this issue on Apr 23, 2008 · 2832 posts


RobynsVeil posted Sat, 17 January 2009 at 9:29 PM

Thank you so much for your reassuring post, CobaltDream! I read your thread and the discussion you had with Bill way back in 2006 about Clay, Skin and Velvet and whilst you and he seemed to be on the same plane, it might not have been so, perhaps? If not, your experiments appear to be far more structured - taking into account the core element Light - than mine by far, so I gathered you knew what you were doing and just enlarging on it.

I really think there needs to be a wiki on nodes and nodal behaviour: everyone doing their own research and not knowing whether the conclusions they are arriving at are erroneous or not. I'm at that point in my experiments on the main channels of the Poser Surface node. Enlightening, true, but am I arriving at the right conclusions? Since no one else is publishing anything anywhere (except what's here on the Poser forum, and it's fairly higgledy-piggledy) I haven't been able to draw any comparisons to corroborate my conclusions.

Speaking of perversions: I've taken that initial node of his and whilst I left all of Bill's nodes and connections intact, tweaked the values massively... on some. Some things I am getting my head around, I'm elated to say. For instance, the simple_color node receives the texture map, which then gets slightly coloured - the original value of 189,219,219 of which I modified to 232,242,242, which made the skin tint just slightly less blue... this outputs to a spots node, which behaviour I understand better now too, because there's a fair bit of documentation on it. To go from textureMap to a tinting node to the spots node makes sense to me. Tweak the values and increase or decrease the size of the spots and it's all good. You are absolutely right, CobaltDream: it does get clearer... all of this colour stuff makes complete sense, well to me, or at least I think I understand it. It's the maths that I find completely baffling.

The spots node then outputs to a color_math Pow node, which also accepts input from a math_functions Add node. That is where I get lost. In my experiments I tried outputting into a math_functions Pow node Value 1 from that spots node, with a math_function Add (Value 1 is 1) plugged into Value 2. Plugged this into the Alternate_Diffuse channel. Did a render. Took the math_functions Pow out of the loops and did a render. No discernible difference. Now what? Here's where I would want to be reading up on math_functions Pow. My experiments have failed... probably for more than one reason:
1> I don't understand the maths behind the node
2> I don't understand what the node's purpose is
3> Because of these issues, I have no idea where to plug it or why I need it

Yet, quite obviously, Bill used it and to good effect. This is where I need to be doing most of my experimenting. Now, judging from what I've seen Bill do, you can actually use these maths nodes in a variety of ways. When they are used to input a value (sheesh, I wish I spoke the language correctly... you're all probably wondering what is she ON about??) ... okay, say I have a math_functions Add node. Value_1 is set to 2.2, Value_2 to 0.0. I output to that mysterious color_math Pow Value_2 (which is white). So, color_math Pow is accepting the value 2.2 in channel Value_2. From what I understand, Pow is supposed to kinda act like "to the power of"... so it would be whatever the value was in Value_1 to the power of whatever the value is in Value_2, in this case 2.2.
No idea if I'm on the right track, here, but like in my experiments, I have to make some assumptions and then either prove them right or wrong.

Does that sound right?

And then, I just have to sort out why one would do this sort of maths to colours at this stage of the game, before it gets plugged into the Blender node, which then gets plugged into this Diffuse node and then: whoa, Nellie. We have a bit of Ambient_Occlusion maths added at this point. This product then goes both into Comp - gotta find out what Comp does - and into a color_math Add which I'm going to assume by the colour in Value_1 addresses SSS.

For now, I think I have enough to explore.
Here's that shader... not that Bill will recognise it - with nodes rearranged (grouped)  to where I could make some sense of them:

Since it really can't be properly seen reduced, I've made it a clickable link to the full-sized image.

The more I study this set of nodes, the more I am impressed by its cleverness.

Monterey/Mint21.x/Win10 - Blender3.x - PP11.3(cm) - Musescore3.6.2

Wir sind gewohnt, daß die Menschen verhöhnen was sie nicht verstehen
[it is clear that humans have contempt for that which they do not understand] 

Metaphor of Chooks