Im rendering with SSS and IDL Poser 2012pro. I get these darkened areas shown by the arrows. The areas where this occurs have black artifacts appear during the SSS pass. Anyone know what might be causing this?

You'll have to solve this by a process of elimination.

Does this occur without your custom maps? If not then there might be an issue with the maps themselves, although it doesn't look like a bump of specular problem. Do the artifacts stay in place when the model is moved to a different position, or when you move the light emmitters? I take it that the face texture map is clear. Best of luck Chris.

Make sure you are using SR2.1 for poser and the latest version of EZSkin. In SR2.1 a new function (Scatter group) was added and EZSkin makes use if it. If you already have those installed, start the process of elimination

One other thing to check... Shadow Min Bias. I render with this set to .1 and it fixes a number of weird skin artifacts.

This turned out to be irradiance cache- 40 or below it doesnt make the dark marks, the higher you go above 40 the worse it gets. I also installed PP2012 SR2.1 made little to no difference. You can really see the marks during the SSS pass as "broken" or irregular black artifacts

Interesting. IC is a lot of grief. Thanks for the info.

I've done another render and had to increase IC to get rid of the dark smears.

For this render I had rotated the figure approx 90 Deg Y axis to get a profile shot.

So based on very limited data at this point it seams as though there is a correlation

possibly between camera angle and the degree of dark marks. The dark marks appeared on the forehead in the previous render, with the forehead roughly perpendicular to the camera image plane. Second render (profile shot) side of nose was roughly perpendicular to camera image plane and thats where the marks occurred. However there were plenty of other surfaces that were perpendicular to the camera that did not exibit this anomaly.

So there seems to be a correlation, but not consistently so, so I'm not really sure what that tells us at this point, except that playing with IC settings may get rid of it.

if both raytrace IC and IDL IC are set using d3d's script, and IDL is used, then raytrace IC (the IC in poser render settings in current versions) is ignored.  however, increasing raytrace IC in poser render settings also increases IDL IC.  it might ameliorate those blotches by setting IDL samples higher, e.g. 1024 or 2048.  IDL samples variable is affected by IDL quality in poser render settings, but is set more easily via d3d's renderFF script.

Issue not resolved, I thought I had found the answer. IC certainly effects it, but in continued testing I found that I was getting this pattern where the dark areas occur

The pattern was a black point surrounded by a lighter circle and they appeared in a grid like fashion. So I'm thinkin if it's in a grid or a somewhat regular pattern it must have something to do with the mesh. I sculpted this in Zbrush so I'm using displacement maps to retain the greater details. When I turn off displacement maps, it completely go's away.

And the black marks don't appear untill the SSS pass. So it seems to be an issue when SSS, IDL and displacement maps are used together. I did not mention displacement maps earlier because it just hadnt occured to me that they played any role in this. I wondered if displacement bounds settings might correct this. I started out at .3 went to 1.0 and aborted because it was taking forever. Tried a render at .60 DB's stilled slowed render and marks were still there.

Must be "ghost in the machine" or sunspots, I was going to do a render and stop it during the SSS pass and do a screen grab to show you all, and now I cant get it to do it.

Beats the hell out me!

My issue with IDL emitters and SSS were entirely due to some interaction with displacement maps. The problem didn't show up very clearly on small resolutions, but on big ones, it was very apparent.

I submitted a bug report to SM and there was no discussion at all of my adjusting settings, or hardware, or anything on my end to alleviate it. I assume it's a real bug.

I really wish there would be an emiter that you see (like a window or photoraphers lamp) and then the real emiter (invisible to scene  , and scalable in reflection node. Big emiters make huge inaccurate reflections) that actually produced the light . It would be nice if these two things could be the same... But we are not there yet.

By the time the emiter is big enough to stop making artifacts, it is washed out and lacking contrast.

You are right. There are anomolies from Zbrush as well. I see some of mine where polys are raised out in spline like fashion in precalc and later produce missreads and artifacts.

SR3 might fix some of ths.

I also only see it on real close up shots as well ( I think, I'm reaching the point where I'm not sure of anything)

given that the earlier scattering issue (scatter errors due to colour) and the displacement/scattering issue have been recognised as bugs, I can understand why users wouldn't want to try increasing IDL samples, as it increases render time by approximately the square of the sample size.  in the past, users were opposed to setting IDL IC = 100 for a similar reason, but the latter variable may only shrink blotches and not eliminate them. 

however, increasing IDL samples for a small area render in a spot known for displacement blotches might not be prohibitive, assuming the displacement blotches always occur in the same spots in duplicate renders that use IDL IC < 100.

Miss Nancy, in my primitive cube test, I rendered half the scene with IC=100 (it took several hours), and it did not eliminate the problem, though my blotches were, admittedly, not very subtle.

I dont know if that could be caused by voltage fluctuations or memory fluctuations. Ive never seen 2 renders vary with no changes in settings or camera. The dark shadows around the eyes are also messed up.

IDL with IC turned on always produces two differeent renderes in a row. You obsrevation skills are insufficient if you never noticed this before.

Note I am on ipad and have not keys. Please forgive the terrible typing and short resnse.

ive seen them pop up now and then. most often for me they appear on the legs, and look like little bruises - so i leave them alone.

out of hundreds of renders ive only ever had a couple in which i had to go back and screw with lighting and render settings to make them go away.

i use point lights a lot, and there are still some workspace preview issues with point light shadows -- perhaps this also has something to do with it?

For the record:

Does IC do any good at all when using IDL? I've always been told to turn it down, but howe about turning it off?

I just had this appear again on a render with displacement, using IDL (but no emitters).

Adjusting IC made no difference. Adjusting IDL quality made no difference.

What made a difference was changing the resolution of the image. Small resolution (1200x900px) showed very minimal splotches. Large resolution (5000x3750) showed very strong splotches.

My solution: render at both sizes. Layer small over large, resize to fit, adjust layer styles to "lighter", mask out everything except the parts covering the worst splotches.

The other option would be to render a second at full size but without displacement and merge as above.

Hopefully this was addressed in SR3.

Quote - For the record:

Does IC do any good at all when using IDL?

speed

does the SSS precalc algorithm also use this monte carlo method?  if so, that would add to the uncertainty of the blotch location and appearance.  I don't know which variables could be set by the user for SSS that would be analogous to setting IDL IC = 100 (giMaxError=0, skips IDL precalc step) and setting IDL samples to 2048 or 4096, but apparently the latter are not helping with the SSS displacement blotch bug.

p.s. ipad users may benefit from the bluetooth keyboard, however a note of caution: take out the batteries from the keyboard before putting it in one's briefcase and going for a ride somewhere.

SR 3 seems to have fixed this

Quote - SR 3 seems to have fixed this

Yes, it was intended to be fixed (see "Improved quality when combining subsurface scattering, indirect diffuse light and displacement mapping" readme entry). Glad to hear you're able to confirm.