Notice the black halo around the green areas. This is where the transparency is failing.

These black halos are caused by anti-aliasing and attempting to render at transparent image out of 3dsMax.  When the image is smoothed during the render it blends colors together.  If our color map is anti-aliased separately from the alpha map, we get color bleeding.  In the image above, I’ve replaced the back background

Summary

I detail the following steps below, but for people looking for the quick answer, here it is.

1. Under Customize > Preferences… > Rendering tab, select “Use Environment Alpha”.
2. Render against a flat, black background – this isn’t optional.
3. Save to 32-bit .tga with “Pre-Multiplied Alpha” turned off.
4. Load the image in Photoshop and select the alpha channel via Select > “Load Selection…”
5. Create a layer mask for the background layer.

Render with a Black Background & Transparency

Now I won’t tell you how to set a black background since it’s the default and, if you’ve changed the background from the default, you know how to change it back.

Go to the Customize menu and select "Preferences…"

Select the "Use environment alpha" option.

Save the Render to .tga with Pre-Multiplied Alpha Disengaged

Disengage the pre-multiplied alpha setting

Now you’re ready to open the file in Photoshop.  Just open your new .tga file, and re-apply the alpha map that has been saved separately embedded within the image file.

Once you've opened the new .tga file, go to Select and pick "Load Selection…"

Load the image's alpha channel. It should be the default.

With your alpha channel as your selection, create a layer mask.

And here’s our final result without any halos.  I’ve added an image of clouds to the areas where the background was black.  The areas in purple were a purple background that was anti-aliased into the image.

Halos are gone- the object can be anti-aliased against any background.

That’s a wrap.  Enjoy!

Let’s begin with my warehouse sample scene.  There are several aspects of the scene that make it perfect for testing depth of field.

1. There are numbers around the scene that represent a distance from the camera in feet.  ”2″ is 2 feet from the camera, “4″ is 4 feet from the camera, etc.
2. There are lots of detail around like girders and windows that will accentuate any blurring effects.
3. The floor and wall have a checker pattern that is 1-foot by 1-foot squares.  They also help us understand the blurring effect.

This is a basic render of the warehouse scene. Not great.

We can do better than this, but let’s set up our depth of field effect first.  Once we get an effect we like, we can improve the quality.

Select the camera, and engage the depth of field (mental ray) effect.

Engage the effect.

When we take another render, we can see that some elements have gotten blurry.  With a target distance of 32 feet, we can see the number “32″ clearly, but the number “2″ is blurrier.

Warehouse render with default depth of field.

Notice that the image is still pretty grainy, and we’re even getting sampling artifacts on the tile flooring and garage doors.  Let’s improve our render sampling and get a comparison to see how we can improve the quality.  The most direct way to improve the quality is to increase our sampling from {1/4 and 4} to 1 and 16.  We’ll also want to use a better filter type.

Increased render sampling and using a better filter.

Below you can see the new render plus a comparison of key areas before and after the Samples per Pixel change.

A finer render of the warehouse.

A clear improvement in details.

The depth of field effect improved, too.

Now there’s a catch to all of this.  The original render only took 2:41 while the finer render took 6:53.   Remember not to turn up your depth of field quality settings until you’re ready to take a final render!

This scene was used in my mental ray depth of field quality tutorial.

Enjoy!

Final Gather Performance

The final gather algorithm in mental ray 3.5 is vastly improved from earlier versions, especially in its adaptiveness. This means you can often use much lower ray counts and much lower densities than in previous versions of mental ray.

In many cases, you can render still images with such extreme settings as 50 rays and a density of 0.1. If this causes “oversmoothing” artifacts, you can use the built-in ambient occlusion to solve those problems.

When using final gather together with GI (photons), make sure the photon solution is fairly smooth by first rendering with Final Gather disabled first. If the photon solution is noisy, increase the photon search radius until it “calms down,” and then enable Final Gather.

Quick Guide to Some Common Materials

Following are some quick rules of thumb for creating various materials. Each assumes the basic default settings as a starting point.

General Rules of Thumb for Glossy Wood, Flooring, and So On

These are the kind of “hybrid” materials you might require for architectural renderings; lacquered wood, linoleum, etc.

For these materials, set BRDF to Custom Reflectivity Function; that is, you’ll define a custom BRDF curve. Start out with 0 degree reflectivity of 0.2, 90 degree reflectivity of 1.0, and apply a suitable texture map to the Diffuse Color. Set Reflectivity between 0.6 and 1.0.

How glossy is the material? Are reflections clear or blurry? Are they strong or weak?

• For clear, fairly strong reflections, keep Reflection Glossiness at 1.0.
• For slightly blurry but strong reflections, set a lower Reflection Glossiness value. If performance becomes an issue try turning on Fast (interpolate).
• For slightly blurry but also very weak reflections, you can “cheat” by applying a lower Reflection Glossiness value for broader highlights while setting Reflection Glossy Samples to 0. This shoots only one mirror ray for reflections, but if they are very weak, the viewer can often not really tell.
• For moderately blurry surfaces, set an even lower Reflection Glossiness value and maybe increase the Reflection Glossy Samples value. Again, for improved performance turn on Fast (interpolate).
• For extremely blurry surfaces or surfaces with very weak reflections, try turning on Highlights+FG Only.

A typical wooden floor could use Reflection Glossiness of 0.5, Reflection Glossy Samples of 16, Reflectivity of 0.75, a nice wood texture for Diffuse Color, perhaps a slight bump map. If bumpiness should appear only in the lacquer layer, turn on Special Purpose Maps rollout > Do Not Apply Bumps To The Diffuse Shading.

Linoleum flooring could use the same settings but with a different texture and bump map, and probably with slightly lower Reflectivity and Reflection Glossiness values.

Ceramics

Ceramic materials are glazed; that is, they’re covered by a thin layer of transparent material. They follow rules similar to the general materials mentioned above, but set the BRDF method to By IOR (fresnel reflections) and set IOR to about 1.4 and Reflectivity to 1.0.

Set the Diffuse Color to a suitable texture or color, such as white for white bathroom tiles.

Stone Materials

A stone object usually has a fairly matte finish, or has reflections that are so blurry they are nearly diffuse. You can simulate the “powdery” character of stone with the Diffuse Roughness parameter; try 0.5 as a starting point. Porous stone such as brick would have a higher value.

Stone would have a very low Reflection Glossiness (lower than 0.25) and one can most likely use Highlights+FG Only to good effect for very good performance. Use a nice stone texture for Diffuse Color, some kind of bump map, and perhaps a map that varies the Reflection Glossiness value.

The Reflectivity would be around 0.5-0.6 with By IOR (fresnel reflections) off and 0 degree reflectivity at 0.2 and 90 degree reflectivity at 1.0

Glass

Glass is a dielectric, so By IOR (fresnel reflections) should definitely be on. The IOR of standard glass is 1.5. Set Diffuse Level to 0.0, Reflectivity to 1.0 and Transparency to 1.0. This is enough to create basic, completely clear refractive glass.

If this glass is for a window pane, turn Thin-walled on. If this is a solid glass block, turn Thin-walled off and consider if caustics are necessary or not, and set Refractive Caustics accordingly.

If the glass is frosted, set Refraction Glossiness to a suitable value. Tune the Refraction Samples for good quality or turn on Fast (interpolate) for performance.

Colored Glass

For clear glass, use the tips in the preceding section. Colored glass, however, is a different story.

Many shaders set the transparency at the surface of the glass. And indeed this is what happens if one simply sets a Refractive Color to some value, such as blue. For glass with Thin-walled turned on this works perfectly. But for solid glass objects this is not an accurate representation of reality.

The scene in following illustration contains two glass blocks of different sizes, a sphere with a spherical hole inside it, and a glass horse.

With a blue refraction color: Glass with color changes at the surface

The problems are evident:

• The two glass blocks are of different thicknesses, yet they are exactly the same level of blue.
• The inner sphere is darker than the outer one.

Why does this happen?

Consider a light ray that enters a glass object. If the color is located at the surface, the ray is colored somewhat as it enters the object, retains this color through the object, and receives a second coloration (attenuation) when it exits the object:

Diagram for glass with color changes at the surface

In the above illustration the ray enters from the left, and at the entry surface it drops in level and gets slightly darker (the graph illustrates the level schematically). It retains this color throughout its travel through the medium and then drops in level again at the exit surface.

For simple glass objects this is quite sufficient. For any glass using Thin-walledit is by definition the correct thing to do, but for any complex solid it is not. It is especially wrong for negative spaces inside the glass (like the sphere in our example) because the light rays have to travel through four surfaces instead of two, getting two extra steps of attenuation at the surface.

In real colored glass, light travels through the medium and is attenuated as it goes. In the Arch & Design material this is accomplished by turning on Advanced Rendering Options > Refraction > Max Distance, setting the Color At Max Distance, and setting the Refraction Color to white. This is the result:

Glass with color changes within the medium

The result is clearly much more satisfactory: The thick glass block is a deeper blue than the thin one, and the hollow sphere now looks correct. In diagram form it looks as follows:

d=Max Distance where attenuation is Color at Max Distance

The ray enters the medium and is attenuated throughout its travel. The strength of the attenuation is such that precisely the Max Distance (d in the figure) the attenuation matches that of Color At Max Distance. In other words, at this depth the attenuation is the same as was received immediately at the surface with the previous scene. The falloff is exponential, so at double the Max Distance value the effect is that of Color At Max Distance squared, and so on.

There is one minor tradeoff:

To render the shadows of a material correctly using this method, you must either use caustics or make sure mental ray is rendering shadows in Segments mode (see Shadows & Displacement Rollout (mental ray Renderer)).

Using caustics naturally gives the most correct-looking shadows (the above image was rendered without caustics), but requires that the scene has caustic photons enabled and contains a physical light source that shoots caustic photons.

On the other hand, the mental ray Segments shadows have a slightly lower performance than the more widely used Simple shadow mode. But if it is not used, the shadow intensity will not take the attenuation through the media into account properly. However, the image might still look pleasing.

Water and Liquids

Water, like glass, is a dielectric with an IOR of 1.33. Hence, the same principles as for glass (above) apply to bodies of water, which truly need to refract their environment. An example is water running from a tap. Colored liquids use the same principles as colored glass.

Water into wine

To create a liquid in a container, as in the preceding image, it is important to understand how the Arch & Design material handles refraction through multiple surfaces vs. the real-world behavior of light in such circumstances.

What is important for refraction is the transition from one medium to another with a different IOR. Such a transition is known as an interface.

For lemonade in a glass, imagine a ray of light travelling through the air (IOR=1.0). When it enters the glass, it is refracted by the IOR of the glass (1.5). The ray then leaves the glass and enters the liquid; that is, it passes through an interface from a medium of IOR 1.5 to another medium of IOR 1.33.

One way to model this in computer graphics is to make the glass one separate closed surface, with the normals pointing outward from the surface of the glass and an IOR of 1.5, and a second, closed surface for the liquid, with the normals pointing outward and an IOR of 1.33, leaving a small air gap between the container and the liquid.

This approach works, but can cause a problem: When light goes from a higher IOR to a lower there is a chance of an effect known as total internal reflection (TIR). This is the effect you see when diving into a swimming pool and then looking up: You can see the objects above the surface only in a small circle straight above. Anything below a certain angle shows only a reflection of the pool and things below the surface. The larger the difference in the IOR of the two media, the greater the chance of TIR.

So in our example, as the ray goes from glass (IOR=1.5) to air, there is a large chance of TIR. But in reality the ray would move from a medium of IOR=1.5 to one of IOR=1.33, which is a much smaller step with a much smaller chance of TIR. This looks different:

Left: Correct refraction; Right: the “air gap” method

The result on the left is the correct one, but how it is obtained?

The solution is to rethink the modeling, and not to think in terms of media, but in terms of interfaces. In our example, we have three different interfaces, where we can consider the IOR as the ratio between the IORs of the outside and inside media:

• air-glass interface (IOR=1.5/1.0=1.5)
• air-liquid interface (IOR=1.33/1.0=1.33)
• glass-liquid interface (IOR=1.33/1.5=0.8)

In the most common case of an interface with air, the IOR to use is the IOR of the media (because the IOR of air is 1.0), whereas in an interface between two different media, the situation is different.

To correctly model this scenario, then, we need three surfaces, each with a different Arch & Design material applied:

• the air-glass surface (blue), with normals pointing out of the glass, covering the area where air directly touches the glass, having an IOR of 1.5
• the air-liquid surface (green), with normals pointing out of the liquid, covering the area where air directly touches the liquid, having an IOR of 1.33
• the glass-liquid surface (red), with normals pointing out of the liquid, covering the area where the glass touches the liquid, having an IOR of 0.8

The three interfaces for a liquid in a glass

By setting suitable Max Distance and Color At Max Distance values for the two liquid materials (to get a colored liquid), we obtain the glass on the left in the preceding rendered image.

Ocean and Water Surfaces

A water surface is a slightly different matter than a visibly transparent liquid.

The ocean isn’t blue; it is reflective. Not much of the light that penetrates the surface of the ocean gets anywhere of interest. A small amount of light is scattered back up again, doing a bit of literal subsurface scattering.

To make an ocean surface with the Arch & Design material, follow these steps:

1. Set Diffuse Level to 0.0, Reflectivity to 1.0, and Transparency to 0.0. That’s right: No refraction is necessary.
2. Set IOR to 1.33 and turn on By IOR (fresnel reflections). Apply an interesting wobbly shader to Bump (Ocean (lume) works well here) and your ocean is basically done.

This ocean has reflections guided only by the IOR. But this might work fine; try it. Just make sure there is something there for it to reflect. Add a sky map, objects, or a just a blue gradient background. There must be something or it will be completely black.

The ocean isn’t blue; the sky is.

For a more tropical look, try setting Diffuse Color to a slightly blue-green color, set the Diffuse Level to a fairly low number such as 0.1, and turn on Do Not Apply Bumps To The Diffuse Shading.

Now you have a base color in the water that emulates the small amount of scattering that occurs in the top level of the ocean.

Enjoy the tropics.

Metal

Metals are reflective, which means they need something to reflect. The best looking metals come from having a true HDRI environment, either from a spherically mapped HDRI photo, or something like the mental ray physical sky.

To create classic chrome, turn off By IOR (fresnel reflections), set Reflectivity to 1.0, 0 degree reflectivity to 0.9 and 90 degree reflectivity to 1.0. Set Diffuse Color to white, and turn on Metal Reflections.

This creates an almost completely reflective material. Tweak the Reflection Glossiness parameter for various levels of blurry reflections. Also consider using the Round Corners effect, which tends to work very well with metallic objects.

Metals also influence the color of their reflections. Because you turned on Metal Reflections, this is already happening; try setting the Diffuse Color to a golden color to create gold.

Try various levels of Reflection Glossiness (with the help of Fast (interpolate) for performance, when necessary).

You can also change the Reflectivity value. This has a slightly different meaning when Metal Material is on; it blends between the reflections (colored by the Diffuse Color) and normal diffuse shading. This allows a blend between the glossy reflections and the diffuse shading, both driven by the same color. For example, an aluminum material would need a bit of diffuse blended in, whereas chrome would not.

Gold, silver, and copper

Brushed Metal

Brushed metal is an interesting special case. In some cases, creating a brushed metal requires only turning down the Reflection Glossiness to a level where you obtain a very blurred reflection. This is sufficient when the brushing direction is random or the brushes are too small to be visible even as an aggregate effect.

For materials that have a clear brushing direction and/or where the actual brush strokes are visible, creating a convincing look is slightly more involved.

The tiny grooves in a brushed metal surface all work together to cause anisotropic reflections. This can be illustrated by the following schematic, which simulates the brush grooves by modeling many tiny adjacent cylinders, shaded with a simple Phong shader:

Many small adjacent cylinders

As you can see, the specular highlights in the cylinders work together to create an aggregate effect which is the anisotropic highlight.

Also note that the highlight isn’t continuous; it is actually broken up into small, adjacent segments. So the primary visual cues that a material is brushed metal are:

• Anisotropic highlights that stretch out in a direction perpendicular to the brushing direction
• A discontinuous highlight with breaks in the brushing direction

Many attempts to simulate brushed metals have looked only at the first effect: the anisotropy. Another common mistake is to think that the highlight stretches in the brushing direction. Neither is true.

Hence, to portray brushed metals, it is necessary to simulate these two visual cues. The first is simple: Use Anisotropy and Anisotropy Rotation to make anisotropic highlights. The second can be done in several ways:

• with a bump map
• with a map that varies the Anisotropy or Reflection Glossiness values
• with a map that varies the Reflection Color

Each has advantages and disadvantages, but the one we suggest here is the last one. The reason for choosing this method is that it works well with interpolation.

1. Create a map for the brush streaks. The possible ways to do this include painting a map in a paint program, or using a Noise map that has been stretched heavily in one direction. The map should vary between middle-gray and white.
2. Apply this map to the Reflection Color in a scale suitable for the brushing.
3. Set Diffuse Color to white (or the color of the metal) but set Diffuse Level to 0.0 (or a small value).
4. Make sure Metal Material is on.
5. Set Reflection Glossiness to 0.75.
6. Set Anisotropy to 0.1 or a similar value. Use Anisotropy Rotation to align the highlight properly with the map. If necessary use Anisotropy Channel to base it on the same texture space as the map.

Brushed metal

The Arch & Design material presets.

In October of 2006, Autodesk implemented a material type called the “Arch & Design” material for the 3d Studio Max implementation of mental ray.  This material type is an elaborate construct of mental ray shaders and includes a useful set of preset material configurations that- with proper lighting- can make any render look special.

Today we’re going to take a deep dive into each one of these mental ray material presets.  By the end of this tutorial you’ll have a firm grasp over what each material does and when to use it.  One thing I’d like to note is that the 3dsMax documentation in this area is spectacular.

The material has a number of significant features:

• Easy to use, yet flexible – more complicated than VRay, but much simpler than the usual mental ray standard.
• Templates – allow fast access to settings combinations for common materials.
• Physically accurate – the material is energy conserving, making it impossible to create shaders that break the laws of physics.
• Glossy performance – advanced performance boosts including interpolation, emulated glossiness, and importance sampling.
• Tweakable BRDF – the user can define how reflectivity depends on angle.
• Transparency – “Solid” or “thin” materials: transparent objects such as glass can be treated as either solid or thin.
• Round corners – simulate fillets to allow sharp edges to still catch the light in a realistic fashion.
• Indirect Illumination control – set the final gather accuracy or indirect illumination level on a per-material basis.
• Built-in Ambient Occlusion – for contact shadows and enhancing small details.

The Arch & Design Material Window

The Arch & Design Default

Let’s take a render of the mental ray Arch & Design material’s default configuration.  For this tutorial, we’ll be using the Stanford Dragon Model scene.  The grey material has some glossiness and specularity, but no special effects like blurring or transparency.

Arch & Design defaults to a glossy grey material.

The Arch & Design Presets

There are 28 individual presets in the 3d Studio Max Arch & Design material, of which only 23 are full materials.  Let’s take a quick look at each one and get a feel for what it’s used for.  Below is a high resolution render showing you what each material looks like relative to one another.  While some of the more elaborate ones like masonry aren’t too impressive, the basic materials like copper and solid glass are excellent.

The Arch & Design material presets.

Let’s look more closely at each one of these presets and how they differ from each other.  Remember that the render above is using one of the free high-resolution HDRI maps and smart lighting, while the images below are using a plain white background and some big area lights.  That’s a recipe for some pretty ugly renders, so don’t look at these materials and gag right away.  Remember that they look good under the right lighting.

Satin Varnished Wood

Satin Varnished Wood

The satin varnished wood material is a reasonable wood material that is fine for most purposes.  It includes a diffuse map that gives it a grain, but you’ll need UVW mapping in order for it to look right.  The satin finish means that there’s little glossiness/specularity so this might be appropriate for construction sites or old toys.

Glossy Varnished Wood

The glossy varnished wood material is very similar to the satin wood above.  The main difference is in the reflection, glossiness, and specularity parameters.  Whereas you’d select satin varnished wood for more rural wood selections, you’d pick glossy varnished wood as a base material for flooring or for polished wood furniture.

Glossy Varnished Wood

Rough Concrete

Rough concrete is a strong selection for more distant 3d objects like roads, sidewalks, and buildings.  However, you probably would select your own bitmap for close up objects, or use a different material entirely.

Rough Concrete

Polished Concrete

Polished concrete removes the bump map from the regular rough concrete material, and has higher specularity and glossiness settings.  This is a great choice for floors in showr0oms or clean warehouses.

Polished Concrete

Glazed Ceramic

Glazed ceramic is your basic DGS material in mental ray.  It’s a typical medium glossiness/specularity material with some light reflections.

Glazed Ceramic

Glazed Ceramic Tiles

Glazed ceramic tiles is a twist on the ordinary glazed ceramic material where a tile bump/color map has been added.  While you have to dig down into the material in order to access the tile parameters like size and spacing, it does give you a quick way to get to tiled materials that look pretty convincing.

Glazed Ceramic Tile

Glossy Plastic

Glossy plastic is similar to the ceramic material but with different reflection falloff parameters.

Glossy Plastic

Matte Plastic

Matte plastic removes most of the reflection from the plastic material above, and tones down the specularity and reflection significantly.  This is approximately the same as the “Wall Paint” pro material type.

Matte Plastic

Masonry

The masonry material can be difficult to work with since getting realistic results often requires manipulating diffuse maps.  Masonry is another good choice for distant objects.

Masonry

Rubber

The rubber material is a little too specular for my tastes, but ultimately a convincing material under realistic lighting conditions.  This material is more like industrial rubber rather than tire rubber, so for automotive renders consider reducing the specularity/glossiness further to get a dirtier look.  The Oren-Blinn shader should help also.

Rubber

Leather

The leather material is a dark, rough, but glossy material suitable for a wide variety of skins and clothing.  You’ll need a convincing diffuse map in order to achieve superior results.

Leather

Glass (Thin Geometry)

The thin geometry glass material is made for objects that are made of a single sheet of polygons or very thin closed geometry like soap bubbles.  In the case of thin geometry, the material is instructing mental ray to render the surface as two-sided such that the render is extremely fast for translucent objects and light is still colored as it exits the object.

Glass, Thin Geometry

Glass (Solid Geometry)

The solid geometry glass material is the usual one-sided glass material we’re all used to.  As light enters one side of the object it undergoes refraction and this process occurs again as light exists the other side of the object.  In this case, the material is still an approximation, so light traveling through the object isn’t colored based on distance, but instead based on the number of times it enters/exits the 3d object.  You’ll notice that thick and thin areas of the object below undergo the same coloration.

Glass, Solid Geometry

Glass (Physical)

The physical glass material takes the longest to render, but yields excellent results.  The difference with this material is that light is colored based on how long it travels inside of the object.  Thus thick parts of the object below are colored with a deeper blue while thinner areas get less coloration.

Glass, Physical

Translucent Plastic Film (Thin)

The translucent plastic film material is similar to the thin geometry glass material in that it is meant for extremely thin objects and for “one-way trips” through geometry.  The thin translucent material allows a face illuminated on it’s normal side to illuminate it’s backside, too.

Translucent Plastic Film, Thin

Translucent Plastic Film (Opaque)

The opaque translucent plastic film material allows solid geometry to have a translucent effect.  This is useful for objects with closed geometry, but is computationally more expensive than the equivalent thin material.

Translucent Plastic Film, Opaque

Water, Reflective Surface

The reflective water surface material is actually not transparent and a great proxy for water at shallow angles and/or at a distance from the camera.  Consider using it whenever possible to save render time.

Water, Reflective Surface

Chrome

The chrome material is simply a 100% reflective material.  This is useful for extremely clean, polished metal surfaces.

Chrome

Brushed Metal

The brushed metal material is a chrome material with blurred reflections.  This is suitable for most metal objects.

Brushed Metal

Satin Metal

The satin metal material is like the brushed metal material but with even more blurred reflections.  This is suitable for stainless steel.

Satined Metal

Copper

The copper material is a very shiny colored metal.  It’s closer to brass in my opinion, but it still responds well to realistic lighting and HDRI.  Use it in darker renders if possible since with this white background it looks like gold.

Copper

Patterned Copper

The patterned copper material is a darker, duller version of the copper material with an added glossiness, specularity, and reflection shader.

Patterned Copper

This topic serves as an introduction to the Arch & Design material for mental ray.

A range of material effects available with the Arch & Design material.

What Is the Arch & Design Material?

The mental ray Arch & Design material is a monolithic material shader designed to support most materials used in architectural and product-design renderings. It supports most hard-surface materials such as metal, wood and glass. It is especially tuned for fast glossy reflections and refractions (replacing the DGS material) and high-quality glass (replacing the dielectric material).

The major features are:

• Easy to use, yet flexible – controls are arranged logically in a most-used-first fashion.
• Templates - allow fast access to settings combinations for common materials.
• Physically accurate – the material is energy conserving, making it impossible to create shaders that break the laws of physics.
• Glossy performance – advanced performance boosts including interpolation, emulated glossiness, and importance sampling.
• Tweakable BRDF (bidirectional reflectance distribution function) – the user can define how reflectivity depends on angle.
• Transparency – “Solid” or “thin” materials: transparent objects such as glass can be treated as either solid (refracting, built out of multiple faces) or thin (nonrefracting, can use single faces).
• Round corners – simulate fillets to allow sharp edges to still catch the light in a realistic fashion.
• Indirect Illumination control – set the final gather accuracy or indirect illumination level on a per-material basis.
• Oren-Nayar diffuse – allows “powdery” surfaces such as clay.
• Built-in Ambient Occlusion – for contact shadows and enhancing small details.
• All-in-one shader – photon and shadow shader built in.
• Waxed floors, frosted glass and brushed metals – all fast and easy to set up.

Physics and the Display

The Arch & Design material attempts to be physically accurate, hence its output has a high dynamic range. How visually pleasing the material looks depends on how colors inside the renderer are mapped to colors displayed on the screen.

When rendering with the Arch & Design material it is highly recommended that you operate through a tone mapper/exposure control such as the mr Photographic Exposure Control in conjunction with gamma correction, or at the very least use gamma correction.

A Note on Gamma

Describing all the details of gamma correction is beyond the scope of this topic; this is just a brief overview.

The color space of a normal, off-the-shelf computer screen is not linear. The color with RGB value 200 200 200 is not twice as bright as a color with RGB value 100 100 100, as one might expect.

This is not a bug because, due to the fact that our eyes see light in a nonlinear way, the former color is actually perceived to be about twice as bright as the latter. This makes the color space of a normal computer screen roughly perceptually uniform. This is a good thing, and is actually the main reason 24-bit color (with only 8 bits or 256 discrete levels for each of the red, green and blue components) looks as good as it does to our eyes.

The problem is that physically correct computer graphics operates in a true linear color space where a value represents actual light energy. If one simply maps the range of colors output to the renderer naively to the 0–255 range of each RGB color component it is incorrect.

The solution is to introduce a mapping of some sort. One of these methods is called gamma correction.

Most computer screens have a gamma of about 2.2 (known as the sRGB color space), but 3ds Max defaults to a gamma of 1.8, which makes everything look too dark (especially midtones), and light does not “add up” correctly.

Using a gamma of 2.2 is the theoretically correct value, making the physically linear light inside the renderer appear in a correct linear manner on screen.

However, because the response of photographic film isn’t linear either, users find that this theoretically correct value looks too bright and washed out. A common compromise is to render to the default gamma of 1.8, making things look more photographic; that is, as if the image had been shot on photographic film and then developed. However, when exporting and importing images (for example, as texture maps) with external image-editing programs, for best results set all gamma values on Preferences > Gamma and LUT Preferences to 2.2.

Tone Mapping

Another method for mapping the physical energies inside the renderer to visually pleasing pixel values is known as tone mapping. You can accomplish this either by rendering to a floating-point file format and using external software, or with a plug-in that allows the renderer to do it on the fly. In 3ds Max such plug-ins are known as exposure controls and are accessed from the Environment dialog.

Use Final Gathering and Global Illumination

The Arch & Design material is designed to be used in a realistic lighting environment; one that incorporates full direct and indirect illumination.

mental ray provides two basic methods for generating indirect light: Final Gathering and Global Illumination. For best results, be sure to use at least one of these methods.

At the very least, enable Final Gathering, or use Final Gathering combined with Global Illumination (photons) for quality results. Performance tips for using Final Gather and Global Illumination can be found here.

If you use an environment for your reflections, make sure the same environment (or a blurred copy of it) is used to light the scene through Final Gathering. In 3ds Max this means you should include a Skylight in your scene set to Use Scene Environment, or use Daylight system with Skylight set to mr Sky.

Use Physically Correct Lights

Traditional computer-graphics light sources live in a cartoon universe where the intensity of the light doesn’t change with the distance. The real world doesn’t agree with that simplification. Light decays when leaving a light source due to the fact that light rays diverge from their source and the intensity of the light changes over distance. This decay of a point light source is 1/d2; in other words, light intensity is proportional to the inverse of the square of the distance to the source.

One of the reasons for this traditional oversimplification is the fact that in the early days of computer graphics, tone mapping was not used and problems of colors blowing out to white in the most undesirable ways was rampant. (Raw clipping in sRGB color space is displeasing to the eye, especially if one color channel clips earlier than the others. Tone mapping generally solves this by “soft clipping” in a more suitable color space than sRGB.)

However, as long as only Final Gathering (FG) is used as indirect illumination method, such traditional simplifications still work. Even light sources with no decay still create reasonable renderings. This is because FG is concerned only with the transport of light from one surface to the next, not with the transport of light from the light source to the surface.

It’s when working with Global Illumination (GI) (that is, with photons) the troubles arise.

When GI is enabled, light sources shoot photons. For the Arch & Design material (or any other mental ray material) to be able to work properly, it is imperative that the energy of these photons to match the direct light cast by that same light. And since photons model light in a physical manner, decay is built in.

Hence, when using GI:

Light sources must emit photons at the correct energy.
The direct light must decay in a physically correct way to match the decay of the photons.
Therefore it is important to make sure the light shader and the photon emission shader of the lights work well together.

In 3ds Max this is most easily solved by using the photometric lights. All of these lights are guaranteed to have their photon energy in sync with their direct light. It is built in and automatic and one does not need to worry about it.

Features

The Shading Model

From a usage perspective, the shading model consists of three components:

• Diffuse - diffuse channel /including Oren Nayar “roughness”).
• Reflections - glossy anisotropic reflections (and highlights).
• Refraction - glossy anisotropic transparency (and translucency).

The Arch & Design material shading model

Direct and indirect light from the scene cause diffuse reflections as well as translucency effects. Direct light sources also create specular highlights.

Ray tracing is used to create reflective and refractive effects, and advanced importance-driven multi-sampling is used to create glossy reflections and refraction.

The rendering speed of the glossy reflections/refraction can further be enhanced by interpolation as well as “emulated” reflections with the help of Final Gathering.

Conservation of Energy

One of the most important features of the material is that it is automatically energy conserving. This means that it makes sure that diffuse + reflection + refraction <= 1. In other words, no energy is magically created and the incoming light energy is properly distributed to the diffuse, reflection and refraction components in a way that maintains the first law of thermodynamics.

In practice, this means, for example, that when adding reflectivity, the energy must be taken from somewhere, and hence the diffuse level and the transparency will be automatically reduced accordingly. Similarly, adding transparency happens at the cost of the diffuse level.

The rules are as follows:

• Transparency takes energy from diffuse; that is, at 100% transparency, there is no diffuse at all.
• Reflectivity takes energy from both diffuse and transparency; that is, at 100% reflectivity there is neither diffuse nor transparency.
• Translucency is a type of transparency, and the Translucency Weight parameter defines the percentage of transparency vs. translucency.

From left to right: Reflectivity=0.0, 0.4, 0.8, and 1.0

From left to right: Transparency=0.0, 0.4, 0.8, and 1.0

Conservation of energy also means that the level of highlights is linked to the glossiness of a surface. A high Reflection Glossiness value causes a narrow, intense highlight, while a lower value causes a wider, less intense highlight. This is because the energy is now spread out and dissipated over a larger area.

BRDF: How Reflectivity Depends on Angle

In the real world, the reflectivity of a surface is often view-angle dependent. A fancy term for this is bidirectional reflectance distribution function (BRDF); that is, a way to define how much a material reflects when seen from various angles.

The reflectivity of the wood floor depends on the view angle.

Many materials exhibit this behavior. The most obvious examples are glass, water, and other dielectric materials with Fresnel effects (where the angular dependency is guided strictly by the index of refraction), but other layered materials such as lacquered wood and plastic display similar characteristics.

The Arch & Design material allows this effect to be defined by the index of refraction, and also allows an explicit setting for the two reflectivity values for:

• 0 degree faces (surfaces directly facing the camera)
• 90 degree faces (surfaces 90 degrees to the camera)

Reflectivity Features

The final surface reflectivity is in reality caused by the sum of three components:

• The diffuse effect
• The actual reflections
• Specular highlights that simulate the reflection of light sources

Diffuse, reflections, and highlights

In the real world, highlights are just glossy reflections of the light sources. In computer graphics it’s more efficient to treat these separately. However, to maintain physical accuracy the material automatically keeps highlight intensity, glossiness, anisotropy, etc. in sync with the intensity, glossiness and anisotropy of reflections. Thus, there are no separate controls for these as both are driven by the reflectivity settings.

Transparency Features

The material supports full glossy anisotropic transparency and includes a translucent component, described in detail here.

Translucency

Solid versus Thin-Walled

The transparency/translucency property can treat objects as either solid or thin-walled.

If all objects were treated as solids at all times, every window pane in an architectural model would have to be modeled as two faces: an entry surface that refracts the light slightly in one direction, and immediately following it an exit surface, where light is refracted back into the original direction.

Not only does this entail additional modeling work, it is a waste of rendering power to simulate refraction that has very little net effect on the image. Hence the material allows modeling the entire window pane as a single flat plane, foregoing any actual refraction of light.

Solid vs. thin-walled transparency and translucency

In the preceding illustration the helicopter canopy, the window pane, the translucent curtain, and the right-hand sphere all use thin-walled transparency or translucency, whereas the glass goblet, the plastic horse, and the left-hand sphere all use solid transparency or translucency.

Cutout Opacity

Beyond the “physical” transparency, which models an actual property of the material, the material provides a completely separate, non-physical “cutout opacity” channel to allow “billboard” objects such as trees, or to cut out objects such as a chainlink fence with an opacity mask.

Special Effects

Built-in Ambient Occlusion

Ambient Occlusion (AO) is a method spearheaded by the film industry for emulating the look of true global illumination by using shaders that calculate the extent to which an area is occluded, or prevented from receiving incoming light.

Used alone, an AO shader, such as the separate mental ray Ambient/Reflective Occlusion shader, creates a grayscale output that is dark in areas light cannot reach and bright in areas where it can:

The following image illustrates the main results of AO: dark crevices and areas where light is blocked by other surfaces, and bright areas that are exposed to the environment.

An example of AO applied to a scene

One important aspect of AO is that the user can how far it looks for occluding geometry.

AO looked up within a shorter radius

Using a radius creates a localized AO effect: Only surfaces within the given radius are considered as occluders. This also speeds up rendering. The practical result is that the AO provides nice “contact shadow” effects and makes small crevices visible.

The Arch & Design material gives you two ways to utilize its built-in AO:

• Traditional AO for adding an omnipresent ambient light that is then attenuated by the AO to create details.
• Use AO for detail enhancement together with existing indirect lighting methods such as Final Gathering or photons.

The latter method is especially interesting when using a highly smoothed indirect illumination solution, such as a high photon radius or an extremely low final gather density, which could otherwise lose small details. By applying the AO with short rays these details can be brought back.

Round Corners

Computer-generated imagery tends to look unrealistic, partly because edges of objects are geometrically sharp, whereas most edges in the real world are slightly rounded, chamfered, worn, or filleted in some manner. This rounded edge tends to “catch the light” and create highlights that make edges more visually appealing.

The Arch & Design material can create the illusion of rounded edges at render time. This feature is intended primarily to speed up modeling, so that you need not explicitly fillet or chamfer edges of objects such as a tabletop.

Left: No round corners; Right: Round corners

The function is not a displacement; it is merely a shading effect, like bump mapping, and is best suited for straight edges and simple geometry, not advanced, highly curved geometry.

Performance Features

Finally, the Arch & Design material contains a large set of built-in functions for optimal performance, including but not limited to:

• Advanced importance sampling with ray rejection thresholds
• Adaptive glossy sample count
• Interpolated glossy reflection/refraction with detail enhancements
• Ultra-fast emulated glossy reflections (Highlights+FG Only mode)
• The option to ignore internal reflections for glass objects
• The choice between traditional transparent shadows, suitable for objects such as a window pane, and refractive caustics, suitable for solid glass objects, on a per-material basis.

Just like before, I’ll be explaining things from the perspective of a beginner/intermediate user.  You should know what mental ray is, how to enable it, and how to create new materials.  I’ll provide you with downloadable source files along the way so you can have a starting point for any complex effects.  I strongly recommend that you at least glance over the first part of this  SSS guide in order to make sure you have a strong understanding of the basic concepts before tackling this second section.

Remember that this guide is geared toward discussing what options are available in the 3dsMax implementation of mental ray, followed by detailed guidance on each material and shader type.  Toward the end of this section we’ll put these concepts into practice by manipulating renders step-by-step.

In part 1 of this tutorial we discussed

• What is Sub-Surface Scattering,
• The Fast SSS Material Type, and
• The Fast SSS Skin Material.

Then, in Part 2 of this tutorial we’ll discuss

• The SSS Physical Material & Shader, and
• The Parti-Volume Shader.

You can download my starter scene here, though I’m not including the texture files because they’re copyrighted so you may get an error message.  Any complex materials I create will be provided through a scene file, too.

The SSS Physical Material & Shader

This material type is significantly different from the Fast SSS Material and Fast SSS Skin Material in that it’s a true scattering solution based on photons.  That makes it significantly slower and more difficult to work with and should only be used where absolutely necessary.  Consider the following from the mental ray manual:

The fast shader is recommended for:

• leaves, grass, plastic, wax, butter,
• human skin, such as backlit ears,
• when rendering speed is important,
• when memory is important,
• materials into which light does not penetrate deeply.

The physical shader is recommended for:

• jade, emerald, and other highly translucent minerals,
• milk, blood, ketchup, ivory soap,
• thick slabs of translucent materials,
• anything where light scatters deeply,
• when rendering accuracy is important,
• experienced users to set up scenes for global illumination.

You’ll want to use the physical shader for highly refined renders like close-up products or more scientific renders and visualizations.  Otherwise the fast shader family has everything that you need.

Also I wanted to include a disclaimer for the images in this section.  The SSS Physical Material requires a lot of calibration in order to look good.  However this runs contrary to what I’m trying to do in this tutorial; to provide you with images that demonstrate only one change in a material with otherwise default settings.  That’s why some of these renders might be kind of ugly.

Now that the warnings are out of the way, let’s dig into the SSS Physical Material and get comfortable with the controls. In order for the SSS Physical Material to work properly, you must enable caustics generation.  Don’t forget to enable “All objects generate & receive caustics” if you don’t want to set up a more complex solution.

SSS Physical Material Parameter Definitions

We’ll begin by going over the SSS Physical material parameters and how they impact the material during renders. One thing to notice is that the SSS Physical material and the SSS Physical shader have the same parameter set.  This is because they’re basically the same thing.  The SSS Physical material is actually just a mental ray material that has the SSS Physical shader assigned to the “surface” and “photon” slots.  Since this is the case, we’ll only look at the material and I’ll leave exploring the shader component to you as homework.  See the image below.

The SSS Physical Material is identical to a mental ray material with instanced SSS Physical shaders in the Surface and Photon slots.

As with before, parameters that take color maps- like unscattered color- will be tested using the “rainbow map” you see along the bottom of the image below.  Parameters that take numerical maps- like front-surface scatter weight- will be tested using the “black-to-white” gradient ramp above the rainbow map.  Wherever I state values below an image, it’s referring to the gradient ramp and is read from left to right.  I.E. “Values are 0.0 to 1.0″ means that the left side of the dragon is 0.0 and the right side of the dragon is 1.0, with escalating values in between.

This is the original render with the gradient ramps along the bottom.

Material

The material parameter controls what the surface properties of the SSS Physical material should be.  For example, if you put a DGS Material (3dsMax) shader in this slot, you’ll get a diffuse-glossy-specular control that you can use to make a ceramic-looking effect.  The image below uses our rainbow map.

This image has the rainbow map in the material map slot.

Transmission

Transmission controls the color that light takes as it passes through the material.  For example a character’s skin might have a reddish color for the transmission swatch, while a green wax candle would have a green color.  Remember that a white or grey color in the transmission slot will take on photons of all colors, rather than photons of one specific color.  This is why you’ll tend to see multi-colored splotches in the default SSS Physical material.

This image has the rainbow map in the transmission color map slot.

Index of Refraction

You should already be familiar with what index of refraction controls.  This will change how light is “bent” or refracted as it passes through the material.  A value of 1 is the equivalent to air and 1.33 is equivalent to water.

This image has a gradient ramp in the index of refraction map slot. Values are 0.0 to 1.5.

Absorption Coefficient

The absorption coefficient is a 3 part parameter that controls the absorption of light as it passes through the material.  Each component represents an RGB color value, not an XYZ direction value.

This image has a gradient ramp in the absorption coefficient map slot. Values are 0.0 to 1.0.

Scatter Coefficient

Similar to the absorption coefficient, the scatter coefficient is also a 3 part parameter except this one controls the scattering of light as it passes through the material.  Again, each component represents an RGB color value, not an XYZ direction value.

This image has a gradient ramp in the scatter coefficient map slot. Values are 0.0 to 1.0.

Scale Conversion Factor

The scale conversion factor operates the same way as it does for the SSS Fast Material.  Check out it’s effects below.  Notice that values too close to 0.0 cause errors- black areas where mental ray refuses to render the material.

This image has a gradient ramp in the scale conversion factor slot. Values are 0.0 to 3.0.

Scattering Anisotropy

The scattering anisotropy parameter controls how light travels once it enters the object.  The range of this parameter is between -1 and 1.  A value of zero means it travels the same in all directions.  A value above zero favors “forward scattering” where light prefers to continue going the way it’s is already going, while a value below zero favors “back scattering” where light prefers to bounce back the way it came.

This image has a gradient ramp in the scatter anistropy slot. Values are 0.0 to 1.0.

Depth

The depth parameter controls how deeply light should penetrate the object.  As you can see in the image below, low values will force light to stay very shallow in the object making it glow and sparkle.  Values that are too high will cause light to get pretty buried in the object and dull the effect.  Finding a happy medium is key!

This image has a gradient ramp in the scatter depth slot. Values are 0.0 to 1.0.

Max Samples

The max samples parameter controls how many samples to take during the rendering process.  You can increase this value in order to get smoother results but you’ll rarely have to go above 512.  I like to set this at 256 until I’m ready for a final render.

This image has a the maximum samples value set to 256.

Max Photons

The max photons parameter controls how many photons should be sampled within the max radius during rendering.  I.E. if you have max photons set to 1,000 and max radius set to 5cm, then during render each sample will look around up to 5cm for photons and collect up to 1,000 photons.  Values up to about 1,000 should be fine.  I like to set this to 512 until I’m ready for a final render.

This image has a the maximum photons value set to 512.

Max Radius

The maximum radius controls how far each lookup should go in searching for photons during render.  A low value means very little smoothing will happen to the solution so you’ll need high photons and samples.  A higher value will smooth out the effect allowing for faster renders, but if set too high it could look unrealistic or dull.

This image has a gradient ramp in the max radius slot. Values are 0.0 to 1.0.

Lights

This is pretty straightforward.  It’s a list of include/exclude lights to help you refine the effect.  You can add lights to this list and control how they’re considered using the “mode” parameter.

Mode

Mode is an enumeration with only 3 settings so I’m not sure why it’s a spinner.  See the table below for what each mode means.

• 0: The material considers all lights.  This is the default.
• 1: The material uses the lights you specified inclusively.  I.E. mental ray will only use the lights you provided and nothing else.
• 2: The material uses the lights you specified exclusively.  I.E. mental ray will use all lights except the ones you provided.

Using the Parti-Volume Shader for SSS

Now that we’ve gotten familiar with the SSS Physical material, we can look at one of the most over-looked SSS techniques- the Parti-Volume shader.  What’s tricky about the Parti-Volume shader is that it follows a different paradigm to the SSS Fast Material and SSS Physical Material.  Instead of considering light traveling through a medium it calculates light as though it were traveling through some murky volume like dusty air or muddy water.  This technique was originally presented by Jeff Patton in the CGTalk forums and it’s a pleasure to formalize his method in this tutorial.

The Parti-Volume Shader tends to be used for things like smoke or dust in the air, or cloudy water like in the ocean.  In our case, we’re going to use it for cloudy glass materials like muddy water, jade, or smoked glass.

The caveat to this technique is that it is painfully slow to render and is more of a theoretical exercise than anything else. You will be forcing your computer to render a really difficult material.  This is the price that you pay for the impressive realism of the technique.

Parti-Volume Shader Parameter Definitions

The diagram below shows you a basic mental ray material that’s been set up with “glassy” shaders in order to create a basic Parti-Volume material.  All of the changes made in this section of the tutorial are with the Volume Shader (Parti Volume (physics)).

Part-volume as SSS solution setup.

By setting the Extinction parameter in the Parti-Volume and Parti-Volume Photon shader to 0.35, (and all other parameters set to defaults) we get the image below.

This is the original parti-volume render.

You’re probably wondering what’s going on here.  That’s not sub-surface scattering in the way we’ve been discussing it so far.  With caustics disabled, this isn’t a photon-based solution and it’s not Fast SSS either.  Instead it is a shader that allows light to pass through it in the form of ray-tracing and falloff, making it a different kind of sub-surface illumination.  You’ll notice that it has that same “jelly-like” look that a strong SSS Physical material might provide.  Thus, in a way, this tool should be kept in the same toolbox as your sub-surface scattering abilities.  Take a look at the image below where you can really see the light scattering and diffusion in effect.  With the lights brought lower to a simpler model, you can really see the diffusion.  If we wanted to get really fancy, we could even apply blurry transparency to this material through the surface shader.

This image was taken with a gradient ramp in the extinction slot. Values are 0.0 to 1.0.

Mode

This mode parameter determines whether or not the height parameter is used.  If Mode is set to 0 then the material is the same throughout the entire volume.  If the mode is set to 1, then there is only cloudiness at the specified height or lower.

• 0: Do not use the height parameter.
• 1: Use the height parameter.

Scatter Color

The scatter color proved difficult to pin down since using the rainbow map in this slot caused the entire material to result in errors.  It turns out that procedural maps do alright, so I used a marble map below.

The documentation says that this map determines the color of the direct and indirect light that gets scattered. Apparently this is a multiplier of the photon energy for the photons in the photon volume map.

This image has a marble map in the scatter color slot. It wouldn't take a bitmap for some reason.

Extinction

This controls how much light is absorbed or scattered in the material.  A value of 0.0 means means totally clear air or material.  In this case, it would be clear glass. The higher you set this value, the denser the material will become (and the more scattering you’ll see). Note that a high extinction value won’t allow photons to enter deep into the object since they will have already been scattered.

This image uses a gradient ramp to control the Extinction value. Values are 0.0 to 1.0.

r

g1

g2

These three parameters control the scattering process.  If g1 and g2 are set to “0″ then you’ll get what’s known as isotropic scattering or diffuse scattering.  That’s where light is diffused in every direction as it passes through the material.  The reason there are three parameters is because mental ray uses a two-part model.

Think of g1 and g2 as scatter anisotropy from before.  They control whether light bounces mostly forward (high values of g) or mostly backward (values of g between 0.0 and -1.0).  r is a weighting factor that lets mental ray decide the importance of g1 vs g2.  g1 is multiplied by r while g2 is multiplied by (1 – r).

For example,

r = 0.5

g1 = 0.5

g2 = -0.25

will yield the anisotropic values of 0.25 and -0.125.

This image uses a static r value (1) and g1 values are -1.0 to 1.0.

Non Uniform

This parameter spans between 0 and 1 .  Lower values mean a uniformly-distributed material (like haze or mist) while higher values introduce noise into the material like cloudy quartz.

This image has a gradient ramp controling the non-uniform slot. Values are 0.0 to 1.0.

Height

The height parameter is used in conjunction with the mode parameter.  When mode is set to 1, the height parameter determines at what point the “fogginess” of the material should stop.  I.E. a value of 1cm means that above 1cm the material should be clear.

This image has the height controlled by a gradient ramp. Values are 3.0 to 15.0.

Minimum Step Length

Maximum Step Length

Minimum and maximum step length control the quality of the effect.  This is very similar to how ordinary 3dsMax volumetric effects work.  A longer step length will decrease the quality of the effect and speed up your render.  A shorter step length will increase the quality of the effect while increasing render times, too.  Think of these in the same way as min/max samples in mental ray.

Below is a side-by-side comparison of a low quality render, the original render, and one with substantially higher quality (and a very long render time).

This image has three significant differences in quality, the far right is the highest.

This is the full high-quality render.

Light Distance

Light distance has to do with area lights and what level of sampling is used.  While there’s not much documentation on this subject, it has to do with how far lights are from the material in order to increase the area light sampling during render.

Min Level

Is ignored and does nothing.

No GI When Direct

When enabled, this will turn off rendering of GI information in the material.  Only direct illumination would be used in that case and it could speed up your renders if you don’t need GI in the material.

Lights

The light list you’re already familiar with by now.  It acts as an include-only list and, when deactivated, all lights are considered.

Conclusion

Well that’s the conclusion of this tutorial. We’ve gone over the major ways that you can get sub-surface scattering and volumetric effects in your materials. Be sure to leave me comments and I’ll update this guide if needed!

SSS via parti-volume shader (Part 2)

I’ll be explaining things from the perspective of a beginner/intermediate user.  You should know what mental ray is, how to enable it, and how to create new materials.  I’ll provide you with downloadable source files along the way so you can have a starting point.

We’ll begin by getting a strong understanding of what options are available in the 3dsMax implementation of mental ray, followed by detailed guidance on each material and shader type.  Finally we’ll put these concepts into practice by manipulating renders step-by-step.

In part 1 of this tutorial we’ll discuss

• What is Sub-Surface Scattering,
• The Fast SSS Material Type, and
• The Fast SSS Skin Material.

Then, in Part 2 of this tutorial we’ll discuss

• The SSS Physical Material & Shader, and
• The Parti-Volume Shader.

You can download my starter scene here, though I’m not including the texture files because they’re copyrighted so you may get an error message.  Any complex materials I create will be provided through a scene file, too.

What is Sub-Surface Scattering

If you’re already familiar with what SSS is you can safely skip this section.

Wikipedia really does the best job of explaining what subsurface scattering is:

Subsurface scattering (or SSS) is a mechanism of light transport in which light penetrates the surface of a translucent object, is scattered by interacting with the material, and exits the surface at a different point. The light will generally penetrate the surface and be reflected a number of times at irregular angles inside the material, before passing back out of the material at an angle other than the angle it would have if it had been reflected directly off the surface. Subsurface scattering is important in 3D computer graphics, being necessary for the realistic rendering of materials such as marble, skin, and milk.

Sub-surface scattering is most visible at thin geometry like between fingers.

You’ve probably seen this effect before- like when you hold a flashlight over your fingers or the sun shines through your eyelids.  The light travels through your skin and gets scattered, taking on a reddish color.

When enabled in a render it can have a subtle but dramatic impact on how your renders look.  Direct light sources will shine light on your object and, when configured correctly, mental ray will be able to scatter and color that light the same way that light would behave in real life.  Consider the images below.

Stanford dragon with sub-surface scattering enabled.

Stanford dragon without Sub-Surface Scattering but similar glossiness and specularity.

What Sub-Surface Scattering Options Are Available

mental ray in 3d Studio Max 2009 has four material-based sub-surface scattering modes and one “faked” mode.  These are:

• SSS Fast Material
• SSS Physical Material
• SSS Fast Skin Material
• SSS Fast Skin Material + Displace
• Parti-Volume Material

For the purposes of this tutorial, we’ll be considering the SSS Fast Skin Material and the SSS Fast Skin Material + Displace as virtually identical.  They’re only differentiated by the ability to add a displacement map.  Examples of each are shown below at default settings appropriate for the scene size.  We’ll go over material settings later in this tutorial.

Sub-Surface Scattering Examples

SSS Fast Material

SSS Physical Material

SSS Fast Skin Example

SSS Fast Skin + Displace Example

SSS Through Parti-Volume Example

A Deep Dive into the SSS Fast Material

Let’s begin with the simplest SSS material- the SSS Fast Material.  I’ll use a gradient ramp, a rainbow map, and a single render to demonstrate the effect of various parameters.  Note that I won’t render all of them because many of these parameters have to be pretty extreme in order to change the render.  Besides, 99.9% of the time you won’t have to alter parameters like Lightmap size or Falloff strength.

Parameters that take color maps- like unscattered color- will be tested using the “rainbow map” you see along the bottom of the image below.  Parameters that take numerical maps- like front-surface scatter weight- will be tested using the “black-to-white” gradient ramp above the rainbow map.  Wherever I state values below an image, it’s referring to the gradient ramp and is read from left to right.  I.E. “Values are 0.0 to 1.0″ means that the left side of the image is 0.0 and the right side of the image is 1.0.

This is the original render with the gradient ramps along the bottom.

SSS Fast Material Parameter Definitions

Below are definitions for each parameter in the SSS Fast Material rollout.

The Full SSS Fast Material

SSS Fast Material (mi) Parameters

The SSS Fast Material  (mi) Parameters give you high-level control over the quality of the SSS effect.

Scatter Group

All objects that should scatter light into each other should be in the same scatter group. To conserve memory, use as few scatter groups as possible. A person’s hands, face, etc. can use the same scatter group; even hands and faces of different people. In general, different scatter groups should be used only when using the same group would cause visible problems due to objects incorrectly scattering into each other. Two characters shaking hands, for instance, would need to have their hands in different scatter groups.

Lightmap Size

Lightmap size is the size of the lightmap as a percent of the total render size. The lightmap is based on screen space and 50% or less is usually enough (and it’s the default), but if edge artifacts start to appear, you can  increase this value.  You’ll rarely need to alter this value and it’s tough to get a change in your render while altering this parameter.

Number of Samples

Number of samples sets how many samples from the lightmap are used (maximum) per rendered ray.  Ideally this should be a power of two. 32 is probably the lowest useful value and 128 is plenty of resolution. However, if the scattering radii are really large then you may need to increase the samples in order to avoid any noisy artifacts.

10 samples on the left and 128 samples on the right.

Bump Shader

This should already be familiar to you.  You can apply a bump map to your material in order to add detail to your material.  One important thing to note is that bump mapping does not affect how your sub-surface scattering is calculated.  It only affects the color and specularity of the material. You should already know what bump mapping would do to this model so I think we’re good without a render.

Diffuse Sub-Surface Scattering

This parameter group is responsible for controlling the color and spread of light through the material.

Ambient / Extra Light

The ambient light controls is the ambient component as though it were an ordinary Standard material. This light will also be included in the lightmap and get scattered like any other light source which makes it the perfect place to add any HDRI light.  Here’re some good HDRI maps you could use.

An ambient light example using the rainbow map. Notice the purple color on the belly.

Overall Diffuse Coloration

All diffuse colors below are multiplied with this global color to yield the final result. This parameter is useful for coloring the final color of the material or to apply maps that change all light that hits the material. Moles on a character’s skin, for example, will block almost all underlying light (even scattered) and would make sense applied here.

This image has the rainbow map in the overall diffuse coloration map slot.

Unscattered Diffuse Color

This is pretty straightforward.  Adding color to this map will only affect unscattered light and will color the model but less-so than overall diffuse coloration.  Compare the image above with the one below.

The unscattered diffuse coloration using the rainbow map.

Unscattered Diffuse Weight

The weight control handles how significant the unscattered diffuse color should be in the final material.  In the image below, the low weights on the left reveal a material almost exclusively made up of sub-surface scattering. On the right the material has a high weight and is mostly the “white” color we used in the diffuse color swatch.

The unscattered diffuse weight parameter controlled by a gradient ramp. Values are 0.0 to 1.0.

Front-Surface Scatter Color

Front-Surface Scatter Weight

These control the color and weight for the front surface scattering.  Front-surface scattering affects surfaces that are facing the light source and are directly illuminated.  Other than that, the controls are very similar to the ones we used on the unscattered component above.  Notice that when the weight parameter is very low (left side of the lower image), the material becomes almost an ordinary Standard material except for the back-scattering component.

Front-surface scatter color controlled by the rainbow map.

Front-surface scatter weight controlled by a gradient ramp. Values are 0 to 1.0.

Front-Surface Scatter Radius

The scatter radius controls how deeply the light should be scattered along the surface for light-facing areas of your material.  This value is in scene units so if you’re working in feet like the downloadable example then .5′ means 6 inches of spread.  Remember that this will get multiplied by the scale conversion factor discussed in the advanced parameters section.

Front-surface scatter radius controlled by a gradient ramp. Values are 0.0 to 2.0'.

Back-Surface Scatter Color

Back-Surface Scatter Weight

These are the color and weight for the back-surface scattering.  Back-surface scattering affects surfaces that aren’t facing the light source but are being indirectly illuminated through the surface of the object.

Back-surface scatter color controlled by the rainbow map.

Back-surface scatter weight controlled by a gradient ramp. Values are 0 to 2.0'.

Back-Surface Scatter Radius

Back-Surface Scatter Depth

These are the scatter radius and depth for light that reaches the back surface from an illuminated front surface. Normally, the radius and depth are set to the same value (and if the depth is not specified, it defaults to the radius).  I’ll include an extra render below.  Note that it’s difficult to get this parameter to invoke a change in the render- the default is 0.03′ and these renders go to 1 foot.  If you’re setting radius/depth to really high values you should consider changing the scale conversion factor instead!

Back-surface scatter radius controlled by a gradient ramp. Values are 0 to 1.0'.

Back-surface scatter depth controlled by a gradient ramp. Values are 0 to 1.0'.

Back-surface scatter radius and depth controlled by a single gradient ramp. Values are 0 to 1.0'.

Specular Reflection

Specular Color

Shininess

The color and shininess control a regular Phong shader for the material specularity.  This should be pretty familiar to you from working with Standard materials.  The higher the shininess parameter, the higher the glossiness of your final material.  The specular color generally doesn’t need to be edited but there are some cases when it can be useful.

Specular color controlled by the rainbow map.

Specular level turned down to 10 from the default of 33.

Advanced Options

Lightmap Gamma Curve

The lightmap gamma curve is the gamma of the lightmap. If this is 1.0, normal diffuse light is stored. If it is less than 1.0, the curve is ‘flattened’ causing the light to spread out towards areas not directly facing the incoming light. If it is higher than 1.0, the curve is ‘narrowed’ causing the light to concentrate in areas directly facing the light source. In most cases, a value between 0.4 and 0.8 is fine.

Scatter Indirect Illumination

If this checkbox is checked then the material includes any indirect lighting (photons, final gathering, etc) in the lightmap to be scattered. This takes additional render time and one should judge on a case-by-case basis if the indirect light is significant enough that seeing it scattered or not makes a big difference to justify the extra rendering time.  You also must turn this on in order to have your SSS material react to HDRI lighting via Final Gather or to caustics being cast by other objects.

Scale Conversion Factor

The scale conversion factor is the first place you should go when creating a new material.  It’s a simple multiplier for all distances in the material.  Scattering is distance dependent so loading a material designed for a model made in inches will not work on a model where the unit is meters, and vice versa.

Scale conversion factor controlled by a gradient ramp. Values are 1.0 to 0.

Scatter Bias (+/- 1.0)

If this is set to 0.0 the scattering is completely uniform.  Positive values favor forward scattering, and negative values favor back scattering.  Backscattering is where light begins traveling through a surface but then turns around and bounces back the way it came.The allowed range is -1.0 to 1.0, but the useful range is much smaller. A value of 0.1 (a slight bias toward forward scattering) is a good start.

Scatter bias on the left image is unchanged, and on the right is -0.50. This biases light backwards.

Falloff Strength

This parameter sets the shape of the distance falloff along the scatter radius. Higher values will give you a more sudden falloff and lower values give a less sudden falloff.  Lower values will also make the scatter distance look shorter, so you might want to compensate by increasing the actual scatter distance for a ‘softer’ look. For high values (1.0 to 10.0), almost all of the samples in the scatter radius are equally weighted. For low values (0.1 to 1.0), the samples near the edge of the scatter radius are weighted less.

Falloff strength controlled by a gradient ramp. Values are 0 to 1.0.

‘Screen’ (soft) Compositing of Layers

When this is turned on the system will choose a softer compositing of the sub-surface scattering layers.  This must be turned off for photometric lights & photographic exposure control to look correct!

Looking into the SSS Fast Skin Material

That was intense. Now that we have a good hold on the basics of sub-surface scattering, let’s take a deeper look at the other material types and what makes them unique and useful. The SSS Fast Skin Material is very similar to the SSS Fast Material that we just learned about. This is because the SSS Fast Skin Material is like two materials layered on top of each other.

The displacement map in the SSS Fast Skin Material + Displace is a simple 3d Displacement shader. There isn’t much information about what this shader does so I’ve written a quick post about the 3d displacement shader here.

For the sake of keeping this tutorial as short as possible we’re going to look at the SSS Fast Skin Material + Displace because it’s a more functional version of the original Fast Skin Material.

SSS Fast Skin Material unaltered render (20% conversion factor).

SSS Fast Skin Material Diffuse Layers

The first major difference between the SSS Fast Skin Material and the SSS Fast Material is the extra front-surface scatter layer.  If you remember from earlier, the front-surface scattering has to do with light-facing surfaces and how light spreads across that surface rather than through it.  The reason for this extra layer is because skin and flesh diffuse light in a unique way compared to more homogeneous materials like stone or milk.

Let’s have a look at what each of these layers looks like independently (without changing the defaults).

Epidermal (Top) Layer Scatter

Good values for the epidermal layer are a slightly yellowish color, a weight of 0.5, and a radius of around 0.2 to 0.4 inches.  Notice what a huge impact the epidermal layer has on the overall coloration of this material!  It’s pervasive but the default scatter depth is so low that much of the detail in the dragon’s scales is still there.

This is the SSS Fast Skin Material epidermal layer isolated from the other layers.

Subdermal Layer Scatter

The subdermal layer operates in pretty much the same way as the epidermal layer (and the original SSS Fast Material).  This tends to be a deeper, reddish scattering effect.  A reddish/orangeish color with a weight of around 0.5 and a radius of 0.4 to 1.0 inches would be a good place to start experimenting.  Remember that if you’re setting the scatter radius unusually high or low, you may need to adjust your scale conversion factor!

Notice in this render how much detail we’ve lost due to this component’s high scatter radius.  This is probably one of the biggest color components in the material.

This is the SSS Fast Skin Material sub-dermal layer isolated from the other layers.

Back-Surface (Through) Scatter

As with the SSS Fast Material, this scattering represents light going entirely through the material.  You can set this scattering as deep as you need, but generally light doesn’t go through flesh very far.  In the image below you can see that the back-surface scattering isn’t a very heavy part of this example.  We could increase the weight manually in order to get more scattering in the dragon’s tail and claws.

This is the SSS Fast Skin Material epidermal layer isolated from the other layers.

SSS Fast Skin Material Specularity

Primary Specularity

The skin specularity functions are two-layered, allowing simulation of both the broad soft specularity of skin and any near-reflective specularities of top layer oiliness and wetness.  Notice that this layer is inherently blue-ish and is less glossy than the secondary specular layer.

This is the SSS Fast Skin Material primary specular layer isolated from the other layers.

Secondary Specularity

For skin, a good start is to have the primary specular layer with a very low shinyness (3.0 to 8.0) and a low level (0.1 to 0.3) and a slightly blueish color.  Then the secondary layer should have a high shinyness (20 to 100) and a medium level (0.3 to 0.6). Introducing mapping into the specularity channels can significantly enhance realism, too!

This is the SSS Fast Skin Material secondary specular layer isolated from the other layers.

Whenever possible you’ll want to use the SSS Fast Material, yielding to SSS Fast Skin Material whenever you’re working on actual skin.  These materials are highly optimized for working with day-to-day needs like candle wax, soap, or stone.  In part 2 of this sub-surface scattering tutorial we’ll take a look at the SSS Physical material and shader, followed by how you can use Parti-volume to create more specialized effects.  You’ll want to advance to those materials for highly refined renders like close-up products or more scientific renders or visualizations.

• Turn on “Advanced options” > “Scatter indirect illumination”.
• Turn off “Advanced options” > “‘screen’ (soft) compositing of layers”.
• “Advanced options” > “scale conversion factor” is 0.3

Consider rendering this with the Free High-Res HDRI Pack I’ve posted.

The 3d Displacement Shader Controls

Object Independent

When this is turned on the displacement effect is independent of the size of the object’s bounding box. When it’s off the displacement effect is scaled according to the size of the object.  Scaling the displacement based on object size is the standard behavior for regular 3ds Max displacement mapping so if you leave this checkbox checked then the displacement length will behave like a normal material would.

Left has Object Independent enabled. Right does not.

Displacement Length

This is the length of displacement when Object Independent is checked, the extrusion map is at 100 percent (white), and the Extrusion Strength equals 1.0. Lower gray levels in the extrusion map, or other values of Extrusion Strength, scale the amount of displacement.  Think of this as a way to alter the displacement map’s contrast.

0.083 (default), .25, and .35 displacement lengths

Extrusion Strength

Controls the height of the displacement. This value is a multiplier: at the default value of 1.0, the map’s effect is unchanged. Greater values will make the displacement more severe while lower values will make it more subtle.

1, .5, and 1.5 extrusion strengths

Extrusion Map

Click to display the Material/Map Browser and choose a map to use for the displacement. Displacement maps apply the gray scale of the map to generate the displacement. Lighter colors in the 2D image push outward more strongly than darker colors, resulting in a 3D displacement of the geometry.

Checker map, dent map, marble map

Direction Strength

Controls the strength of the direction shader.  Adding a direction shader has no real effect unless you have the Direction Strength set above zero.

0, 1, and 2 direction strength values

Direction Map

The direction map is a lot like a normal map.  The direction of the displacement is changed according to the RGB values of the map you use here. Red values offset in the U axis, Green values offset in V, and Blue values offset in W (using the object-local UVW coordinates).

Checker patterns, Red/Green, Red/Blue, Green/Blue

This week I’ll be showing you how to use the submerge (lume) shader in 3dsMax and mental ray. It’s an easy way to make your scenes look like their underwater while harnessing the power of caustics and global illumination. Try these techniques in conjunction with the waterbox we made last week for a really great one-two punch!

Also, you may have noticed that I included a transcript of last week’s video.  I’m going to try adding that in to the mix for the next few weeks to see if people find that useful.  Remember that you can always pass this page through Google Translate to get the transcription in your desired language!

I’m on vacation right now, but I’ve queued up this Monday Movie so that you can start your new year right!  This week I’m showing you how to create a water box in 3dsMax using the mental ray renderer.  For those of you who don’t know it, a water box is a great way to practice your rendering technique and for learning how the different settings work in mental ray.

Sorry for the delay.  This week’s video tutorial is part 2 from last week where we talked about the matte/shadow material type in the scanline renderer.  This week, we’re looking at how you can use matte/shadow materials in the mental ray renderer and we’ll use a quick-and-dirty camera matching technique along with it.

This week we’re looking at how you can create a reel-to-reel film material in 3dsMax. I’ll walk you through how to blend opacity mapped materials with translucent mental ray materials, and I’ll also give you a few pointers about how to animate the technique!

I’ll post the scene files for download in a little bit. Enjoy!