1
$\begingroup$

The tutorials I used use layer weight with fresnel (I'm currently using facing because fresnel doesn't seem to work in my case) as the factor to mix diffuse emission. This looks fine for them but it does not seem to work for me.

Here is my attempt at it:enter image description here

On the left is my reference image. The pieces are modeled using solidified planes with beveled edges then remeshed, subdivided and displaced to add bumps.


Nodes:
enter image description here

Layer weight with facing is used as the factor to mix the Diffuse BSDF and Emission, then I added Ambient Occlusion to give shadow to the edges.



I don't know what's wrong but for me it doesn't look right,Maybe its because the models are flat compared to other SEM models but something looks to be wrong.

.blend file:

$\endgroup$
3
  • $\begingroup$ Could you share your blendfile over blendexchange? That is a more direct and easyer solution and the link is less likely to break $\endgroup$
    – Phönix 64
    Commented May 3, 2023 at 6:58
  • $\begingroup$ Sorry, added it. $\endgroup$
    – plin
    Commented May 3, 2023 at 8:09
  • $\begingroup$ Yes, your main issue is that your peaces are flat... @MartynasŽiemys answer is efficient base, but Ambient Occlusion will be a part of tree too ... also It really looks like one directional light play some role in final result. $\endgroup$
    – vklidu
    Commented May 4, 2023 at 18:39

2 Answers 2

0
$\begingroup$

You might want to change your color management settings from Filmic to "Standard". Filmic color transform is not designed for this and might be the reason it looks unusual since it compresses highlights.

I also don't think you need any sort of complex shader for it. I think a simple Layer Weight or Fresnel node should do it. I don't think the real electron microscope images use complex shaders.

enter image description here

$\endgroup$
5
  • $\begingroup$ Thanks for the reply. But there is no notable difference between the filmic and standard view. But there is a slight improvement Filmic:imgur.com/QYNRVeK Standard: imgur.com/soGYQvj $\endgroup$
    – plin
    Commented May 3, 2023 at 9:38
  • $\begingroup$ @Martynas I don't think your efficient setup will result in the (reference)[imgur.com/ldKN1JQ] result (and I don't mean AO) ... edges looks like they fits the Fresnel shader, but if you check two brightest convex peaces they should be dark at top, but they aren't ... I tried also manipulate normal, but didn't get even close to the reference (even I tried to rotate parts or sculpt them to be closer reference shape). $\endgroup$
    – vklidu
    Commented May 4, 2023 at 19:21
  • $\begingroup$ Well, maybe you are right. :D Maybe my answer is a bit useless this time. I don't really like the idea of looking at another render as reference, though. Maybe it would be better to look at the real thing(they are probably still renders, but at least coming out of real electron microscope). Unfortunately, I don't really know how they are rendered. It would probably be a good idea to do some research. It's interesting. I would like to look into it, I just couldn't find the time for it yet. $\endgroup$ Commented May 4, 2023 at 21:02
  • $\begingroup$ Sorry for the link above (wrong format) I was referring to OP's reference i.imgur.com/ldKN1JQ.png that seems to be real shot (or do you think it is a render?) Never mind in general it is helpful answer. $\endgroup$
    – vklidu
    Commented May 5, 2023 at 7:17
  • $\begingroup$ I assumed it's a render from the tutorial. It's interesting, how the scanning works. I thought some sort of 3d model is reconstructed from the data and then just rendered, but it seems the image is made out of electrons detected. So there should be a "light" source. That's the beam of electrons and there are also shadows. I wonder how that beam is different from light in it's interactions with surfaces. It might make sense that electrons don't bounce as much as light, but looking at images the "light" seems to be soft so that means the electrons come from different directions maybe. $\endgroup$ Commented May 5, 2023 at 8:27
0
$\begingroup$

Yes, your main issue is that your peaces are flat... @MartynasŽiemys answer is efficient base, but Ambient Occlusion will be a part of tree too ... also It really looks like one directional light play some role in final result.

enter image description here

For some reason light seems to be still too much even ... hm

enter image description here

I got the node tree quite messy, will share if I clean it up :)

BTW this is probably tut people used.

$\endgroup$

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .