There are some more affordable skin displacement maps at surfacemimic dot com, patches as well as more specific scans. Also some cross-polarised colour photos.

I think that would either be separate or involve plugging more nodes in, which then kind of defeats the point of the PBSDF. I'm using it as a higher-quality-more-versatile alternative to the Physical Root, so am trying to avoid complicated setups.

Thanks ChromeStar - I'll try and remember to add examples from time to time!

So here's a load of balls. Plugging a Blackbody node into the emission chip on the PBSDF allows the colour temperature to be set in kelvin (K) - the render goes from 1800K (candlelight); to 3000K (lightbulb); to 4000K (moonlight); to 5500K (midday); and finally to 10,000K (blue skies).

@adp001 - yes, but that is kind of the point I was trying to get to - to approach photorealism you have to go from the general to the particular. It is much harder to go from the particular to the general.

@Okhi - no-one likes different size texels ;) Maybe tri-planar projection or something along those lines??

As a general observation, and purely my opinion, the problem with photorealism as a goal is that the closer you get to it the harder it becomes. Like RedPhantom says, if one thing is not quite right it will destroy the entire illusion. One reason that the scanned heads look so realistic is that they are quite dense topologically with a lot of detail in the geometry; they would be difficult and in some ways impossible to rig and animate. Every detail is particular to the way they are, like a statue.

A Poser figure is a very different case as it is designed to be generic and non-specialised - to be posed realistically or not, to be morphed and scaled, to be rendered as animation or still image, in high resolution or not, in close up or not. It is intended to be all things to all users, to be as un-particular as possible. It is impossible for creators to anticipate every particular, so to go from a versatile and generic figure to a particular photoreal render will inevitably take some work to achieve.

That's not to say that vendors can't improve, or figure creators, or Poser itself though.

@Okhi - good luck with Painter, it does have a learning curve and the online docs aren't great, but there are loads of video tutorials for it. I like the work you've been showing so I'm looking forward to seeing what you can do with it!

Skin shader - the Superfly bit anyway, I haven't used Firefly now in years ;)

Poser figure? OK, here is La Femme with a character called Shana by Deecey (not available yet but will be in her store here when finished). She has hand painted the skin in Substance Painter, starting with the bump first. This is rendered from the last work in progress files I had, using the same setup as the scan heads posted earlier - black background, two area lights - then run through Photoshop (Curves, colour balance, high-pass sharpen). I'll post the skin shader that I used for this render in the next post.

The various SSS controls all affect each other, and I have been adjusting them in the order here. The Subsurface value is the overall amount or scale of the SSS effect (I don’t believe that the values in this node are affected by what units are set in the preferences), and changing the SSS Method to random walk uses more accurate volumetric calculations (so should be used with a closed mesh). The Subsurface Radius values control the depth that Red, Green and Blue light will penetrate and scatter back. Subsurface Colour affects the Base colour which will be more or less noticeable depending on the scale as set by the Subsurface value.

Illustration of the relationship between Radius, Subsurface colour and Base colour, with Blue light set to penetrate to a greater depth and therefore scatter more and give a blue tint to the thinner parts of the mesh:

The shader is very simple. Using the Cycles root plus Principled BSDF means no preview, but the results are more than worth it in my opinion. I had a thread with a bunch of examples, but that seems to have disappeared, so I’ll add my notes here (note that I am perfectly capable of making mistakes!). First, there were only the two maps with this sample, which is why I’m using a normal map (bump is easier to make, set up and adjust).

Specular in Superfly is base reflectance, or direct reflection, or f0; with this node 0-2 are mapped to to 0-16%. Using 0.25 here means 2% base reflectance which seems to work well. I’ve never used a spec map in Superfly yet, they aren’t needed. (NB. with the PBR Metal/Rough workflow as used in apps like Substance Painter and many others, the default would be 0.5 which would be 4% for all non-metals.)

As bagginsbill says, with a good bump a roughness map doesn’t really seem to matter that much (not the case with most materials though, they are important as they define surface relief at a smaller scale than bump or normal maps).

Here's another free scan head (ten24 again, and there is a sample scan on their store website if you want to experiment), this time rendered in P12 Superfly. Below is a screenshot of the geometry and the final render with post processing. I'll put a screenshot of the shader in the next post.

Two area lights at 600% scale (scale up for softer shadows), one in front, one behind. Poser Ground has black material applied so no bounced light to speak of.

The fastest easiest way would be to work with scan data. That's what I think a lot of games studios will be doing - much more cost effective. From a Poser POV though, go look at the sort of prices for commercial usage ...

Anyway, this is what I could get from P11 back in 2015 with a free head from Ten24 (aka 3dscanstore dot com). The eyes are rough as it's a single mesh item, they aren't separate - image is post processed and eye highlight painted in. I did convert the 8 bit bump to a normal map, but the shader was ridiculously basic (and I wouldn't use it now!).

The closest would be to use Create Text Prop in the Object menu to, er, create some text which Poser will see as a prop item. You only get the choice of 3 fonts, but you can assign materials, make them visible for rendering or export these items. They are flat 2d objects though and there is no way to extrude them to 3d in Poser (that I know of).

Reading up on Lightwave it should just be a case of making the lettering into a full 3d mesh and exporting as OBJ or FBX to get it into Poser.

Checked & my units are set to meters. There is another new beta so I'm downloading now, will check when that's installed.

On 12.0.336 it's working ok for me, either spinning the dial or typing a number directly, on all the cameras I looked at. Max limit default is 262128.02.

Octaves are set to 1 on the Fractal Sum node, try increasing to 16 or decreasing to 0.1. I can see facets on the object so it doesn't look hi-res to me, maybe the data going in isn't fine enough? If that doesn't affect it, let's see a screenshot with wireframe?