Cineversity

I’ve been trying for some time to improve my procedural foam material, closer to photo-realistic if possible.  I’ve been using the physical renderer because it seems more realistic.

The attached scene (R16) has my material so far applied to an example mattress, and two camera presets for close and far renders.  My far renders are OK, but up close the effect isn’t there yet.  And I’ve included a foam image that is pretty typical as a target.

I’m using the bump channel to create the foam texture, though I would be willing to use e.g. the normal channel instead, with some guidance.  I could also obtain some good high resolution flat foam photos for texturing too, but only if procedurals are completely out for this.

Any further hints, tips and links to material/texturing tutorials are most welcome as well.

Thanks,

Sean

Hi sean72,

As a starting point, create a paper based list of all channels of the C4D Material, keep it in you pocket, and every time you have a few minutes, observe the objects around you and what channels they might need to use to look in that way. Go along the list. :o)

===========

The Physical Render will not help here a lot, it is more for Camera specific aesthetics: but yes, that will help on the end to sell the shot.

I avoid when ever I can the term Photorealistic or Physical correct material. It is a longer discussion, but since a friend of mine was pulled to the judge, as his client had a different idea about Photorealistic, he didn’t wanted to pay. My friend is first class in with his awarded work. As a consultant in that case I mentioned only that a lens cap left on the camera produces a “photorealistic image”—it is just black. Easy to recreate.

Materials are never ever physical correct, it is nearly impossible to produce. We simulate such stuff. Think alone of the light in C4D, it is RGB based, not spectral based. So, I will not use those terms, nor answer to such ;o) jsut so you understand where I’m coming from.

BUT—of course I get your question, you want to deliver something that comes closer to the, e.g., image above.

To create great materials is a question of observation, and I really means exploration in DETAIL, not just a look and “I got it”. It might take hours to explore a material in depth. You might be surprised. Get reference images, shoot them by your self, with a MacBeth chart or at least a gray card next to it. Profile your camera, and color manage your pipeline. Calibrate your monitor.
Avoid Internet stuff as reference! Avoid 8bit/c images without color profile. JPGs are not acceptable for high end work. I could go on and on with this, but I stop here, I will publish soon a fifty part texture series. I hope that will set some standards.

This list will be not complete or aim to be in that way!

First step, modeling. The edges in your model are razor sharp. You will not find Foam that is in that way, use a “Fillit” or Bevels. Just very tiny, if you take a cube in the size of your model, the starting point for the fillet would be 0.03cm. It is with all objects, a sharp edge without the options to catch light, it just doesn’t work. (1)—>image

Next: Light, be clear what you really need in terms of light. Avoid Soft shadows, and get clear about the parameters inside of the light, e.g., Contrast—and why it is in that way. Lighting is based on a few simple rules, but the combinations are endless. There are some nice Cinematography books about light available.

The material in your case, study in detail “Subsurface Scattering”. You ahve the shader in C4D already. I think that is certainly one of the KEY elements here. As well (1)!

Check where the material is dull and where it is highly reflective (2) can you see the little sparks in the circle? These might bloom to a nice Bokeh if used outside the focus area, if set up in the Physical Render.

Again the edge (3), can you see how diffuse the edge is here, not only based on the light set up, but also based on Subsurface Scattering.

How many blu tones do you see (excluding the hand shadows) between point (1) and (2)? Just set your Photoshop eyedropper to a 5x5 or 11x11 matrix and measure, it is quite a variance in it. Based on many influences. One could be that the different light sources are uneven, different in color temperature (slightly, perhaps different in age). To make it easier I have placed a white bar into it. Quite amazing how that little area goes from pale blue to a different tone in that short distance!? This is what I have in mind when I say—explore.

The list could go on and on. (Digital Lighting and Rendering (3rd Edition) by Jeremy Birn, would be my suggestion to start with. I have to admit, I have only read edition one and two when they came out, starting around 13 [?] or 14 years ago. It’s a classic by now, but fresh in the 3rd edition, Nov 2013)

I hope I could point you in some direction, I hope I was not to straight with my opinions, but after all, waht you use to create your work will always “shine” though. Quality work is based on good material from the start.

My best wishes.

Sassi

I appreciate your detailed reply.  I have taken a close look at the SSS and it indeed goes a great step closer to realism.  I keep forgetting SSS because if often seems to make little difference, but I am probably not using it right.  I need to find some more technical tutorials on using SSS though, it is difficult to control.  For example, I rarely get any noticeable results unless my path is very low, like 0.1 or 0.2 (cm), which I don’t think quite makes sense.  And of course, I want to better understand the difference between single/simple and multiple SSS modes.  Also, the help suggests that color + luminance = 100 is vital, but what blend of the two is best is not suggested?

I’m also going to try to get Jeremy Birn’s book, sounds like a good one.  And I would be very, very keen to see your new series on textures, I like to understand how/why to make something work.

Hi sean72,

The best way to explore such set ups is first, to get clear what you like to see. Example, a SSS effect with a light close to the camera will not work as well as a light set up more from the opposite side. You might have a different idea later on, but for the exploration, that would be my suggestion. It is all a simulation, to come closer to the “reality”. Reality can bounce light around endlessly, not in million or trillions time, beyond that, and all of course in realtime. Any affordable computer is not able to do that. So we try always to get the best simulation, with the least amount of render-time. The more we get to know “the inner mechanics” of such algorithm, the better.
Scale is certainly another factor in the equation and good modeling.
The next main idea that I like to share is to set up parameter animations. If you like to explore something, change only one parameter at the time. When ever your cpu has some idle time, render a 0-100 frame animation with the parameter 0-100% . With the right naming of such, you get a little library that makes it easier to understand the parameters. While you scrub through, you know frame 46 is 46%. Sometime parameters are not in %, hence the naming of the clip to know what is what.
Such little clips will be more memorable than the old “change, render and repeat” method. Sometimes you might explore first what the min and the max parameter for an animation might be.
If possible use always the same set up (light, camera, model) so it becomes more comparable. But of course—sometimes the only change should be the light position.

What makes a material “more realistic” is the little stuff, what is nearly not visible, but we are sued to it. Most people can instantly tell when a shadow is wrong, or that something is wrong, but they can’t tell why. Similar to that is the typically judgement that it looks “CGish”, without the ability to say what to do next. Why? Because we have seen the “real” thing over a long time, and everything has more than what is directly obvious. Example, everything is reflective! There is not so much on this planet that isn’t. Check out if you can find 100% absorptive materials, you will find that the heat they collect from the light alone makes them dangerous. ;o)
The idea here is that this reflectivity is related to our view (perpendicular/tangential), the surface roughness vs scale, and the distance of the observer. But who would put a dull reflection on foam in 3D. It is nearly not visible.

What ever you find to be true, the more analytic and observatory work you invest—the more your materials becomes “realistic”, the more you can set up the fine nuances. If you just buy stuff or take other peoples set up, the more you make yourself dependent. If you have insight in one material, you get better with the next and so on. Here is the success—to be unique in your skills.

To support that—perhaps you request some tutorials about SSS here.

My best wishes.

Sassi