Polycount Wiki: Texture Types


https://youtu.be/ZOHNRlrd1Ak


Original Source: Yanko Stefanov (2018)

Texturing is a rendering method that simulates light reflecting on an object. It is much faster than ray tracing. The texture stores the image of the distant environment surrounding the rendered object. There are many different types of textures, and each of their uses can be confusing for a beginner or mid-level artist. This page will seek to clarify what each type of texture does and how it is used. Note: In this page the words “texture” and “map” are used interchangeably.

           Diffuse                      Specular                 Displacement         Ambient Occlusion             Normal

       Diffuse                      Specular                 Displacement         Ambient Occlusion             Normal

Diffuse: This is the most common texture type. A diffuse texture is the same as if you were standing in front of the object and took a photo of it. The diffuse texture includes lighting and shadows information.

Diffuse: This is the most common texture type. A diffuse texture is the same as if you were standing in front of the object and took a photo of it. The diffuse texture includes lighting and shadows information.

Albedo: The albedo map is similar to the diffuse map. It is sometimes referred to as the "new diffuse map." However, an albedo map differs in a very significant way. It has had all lighting and shadows information removed from it, and is literally the color values of the image.

Albedo: The albedo map is similar to the diffuse map. It is sometimes referred to as the "new diffuse map." However, an albedo map differs in a very significant way. It has had all lighting and shadows information removed from it, and is literally the color values of the image.

ID: An ID map is a texture that defines areas of the model based upon the material ID color given to those areas in the UV unwrap map. It is most commonly used to easily mask (apply to specific areas based on a white/black texture input) different material types (e.g., steel, wool, rubber) in procedural texturing applications like Allegorithmic's Substance Painter and Substance Designer or Quixel's DDO.

ID: An ID map is a texture that defines areas of the model based upon the material ID color given to those areas in the UV unwrap map. It is most commonly used to easily mask (apply to specific areas based on a white/black texture input) different material types (e.g., steel, wool, rubber) in procedural texturing applications like Allegorithmic's Substance Painter and Substance Designer or Quixel's DDO.

Normal: A normal map is commonly used to fake high-resolution details on a low-resolution model. Each pixel of the map stores the surface information of the original high-res mesh. This creates the illusion of more surface detail or better curvature. However, the silhouette of the model doesn't change. The most commonly used is called a Tangent Space normal map, which is a blue-base map (128,128,128 in RGB space, or MikkT).

Normal: A normal map is commonly used to fake high-resolution details on a low-resolution model. Each pixel of the map stores the surface information of the original high-res mesh. This creates the illusion of more surface detail or better curvature. However, the silhouette of the model doesn't change. The most commonly used is called a Tangent Space normal map, which is a blue-base map (128,128,128 in RGB space, or MikkT).

Bump: Bump maps are greyscale images. They create the illusion of depth on the surface of a model using a very simple lighting trick. When values get brighter, details appear to pull out of the surface. When values get darker, they appear to be pushing into the surface. Bump maps are great for creating tiny details on a model. The problem with bump maps is that they break pretty easily if the camera views them from the wrong angle.

Bump: Bump maps are greyscale images. They create the illusion of depth on the surface of a model using a very simple lighting trick. When values get brighter, details appear to pull out of the surface. When values get darker, they appear to be pushing into the surface. Bump maps are great for creating tiny details on a model. The problem with bump maps is that they break pretty easily if the camera views them from the wrong angle.

Parallax: Creating parallax mapped materials always starts with creating a height map. Each pixel's texture coordinates are adjusted at render-time to create an illusion of depth as the viewer's eye moves across a scene. The virtue of parallax mapping is efficiency. It is the cheapest real-time technique for displacement mapping. As a rule, parallax mapping looks good on walls and floors.

Parallax: Creating parallax mapped materials always starts with creating a height map. Each pixel's texture coordinates are adjusted at render-time to create an illusion of depth as the viewer's eye moves across a scene. The virtue of parallax mapping is efficiency. It is the cheapest real-time technique for displacement mapping. As a rule, parallax mapping looks good on walls and floors.

Displacement: Technique that aims to render bumps as true geometry. Unlike bump mapping, normal, or parallax which tries to "fake" bumps, displacement mapping actually displaces the surface with real geometry based on black and white values of the texture map itself.

Displacement: Technique that aims to render bumps as true geometry. Unlike bump mapping, normal, or parallax which tries to "fake" bumps, displacement mapping actually displaces the surface with real geometry based on black and white values of the texture map itself.

Vector Displacement: Vector displacement displaces on 3 axes of 3 colors, while regular displacement or height along the same normal. Rarely used, but useful in some cases. An example would be a mushroom. Mushroom’s cap grows in a certain directions, it cannot be done properly with Height or Displacement map, because they pull the geometry only in one axis (Up) while Vector displacement works with multiple axes (Up and Sides).

Vector Displacement: Vector displacement displaces on 3 axes of 3 colors, while regular displacement or height along the same normal. Rarely used, but useful in some cases. An example would be a mushroom. Mushroom’s cap grows in a certain directions, it cannot be done properly with Height or Displacement map, because they pull the geometry only in one axis (Up) while Vector displacement works with multiple axes (Up and Sides).

Height: Height mapping is very similar to displacement map, but it's usually is applied on a terrain, mountains, and larger scale surfaces where the values are only used to modify the vertex height. Height mapping usually refers to large scale / tessellation based techniques, where displacement mapping usually refers to small scale / raytracing techniques.

Height: Height mapping is very similar to displacement map, but it's usually is applied on a terrain, mountains, and larger scale surfaces where the values are only used to modify the vertex height. Height mapping usually refers to large scale / tessellation based techniques, where displacement mapping usually refers to small scale / raytracing techniques.

Specular: A specularity map defines how strong the textured surface will 'shine' at a certain position. Most render engines use this information to define the appearance of specular highlights.

Specular: A specularity map defines how strong the textured surface will 'shine' at a certain position. Most render engines use this information to define the appearance of specular highlights.

Roughness (or Glossiness or "Gloss"): Represent how smooth or rough a surface is. Also called Microsurface. Roughness map controls the specular spread, what you need to consider is controlling where the spread happens and to what degree. Good to use for metals, rust, ice, skin, etc.

Roughness (or Glossiness or "Gloss"): Represent how smooth or rough a surface is. Also called Microsurface. Roughness map controls the specular spread, what you need to consider is controlling where the spread happens and to what degree. Good to use for metals, rust, ice, skin, etc.

Metallic: A metallic texture is a black/white texture that tells the rendering engine whether a particular area is metallic (reflective) or non-metallic (non-reflective). This map is used in the Physically Based Rendering workflow, and is a primary mask type for Unreal Engine 4. In this texture, white denotes metal and black denotes non-metal materials.

Metallic: A metallic texture is a black/white texture that tells the rendering engine whether a particular area is metallic (reflective) or non-metallic (non-reflective). This map is used in the Physically Based Rendering workflow, and is a primary mask type for Unreal Engine 4. In this texture, white denotes metal and black denotes non-metal materials.

Anisotropy: This is an uncommon map type. Anisotropy is used to simulate stretched out highlights. In the real world, they are caused by elongated micro-scratches or details that go in the same direction.

Anisotropy: This is an uncommon map type. Anisotropy is used to simulate stretched out highlights. In the real world, they are caused by elongated micro-scratches or details that go in the same direction.

Ambient Occlusion: Ambient occlusion is simply a simulation of the shadowing caused by objects blocking the ambient light. Because ambient light is environmental, unlike other types of lighting, ambient occlusion does not depend on light direction. As such, it can be pre-computed for static objects.

Ambient Occlusion: Ambient occlusion is simply a simulation of the shadowing caused by objects blocking the ambient light. Because ambient light is environmental, unlike other types of lighting, ambient occlusion does not depend on light direction. As such, it can be pre-computed for static objects.

Environment: A rendering method that simulates light reflecting on an object. It is much faster than ray tracing. The texture is used to store the image of the distant environment surrounding the rendered object.

Environment: A rendering method that simulates light reflecting on an object. It is much faster than ray tracing. The texture is used to store the image of the distant environment surrounding the rendered object.

Cavity: Cavity maps are narrower version of ambient occlusion, keeping dark shadows to crevices and sharp corners only. Useful for hard surfaces texturing.

Cavity: Cavity maps are narrower version of ambient occlusion, keeping dark shadows to crevices and sharp corners only. Useful for hard surfaces texturing.

Curvature: Curvature map stores the convexity/concavity of the mesh. It can be used to mask where the surface would get more wear or where sub-surface scattering might occur. You could achieve some interesting effects by just putting the curvature map on top of the diffuse and change the mode in Photoshop.

Curvature: Curvature map stores the convexity/concavity of the mesh. It can be used to mask where the surface would get more wear or where sub-surface scattering might occur. You could achieve some interesting effects by just putting the curvature map on top of the diffuse and change the mode in Photoshop.

Thickness (or Translucency): Thickness maps are a measurement of how thick/thin a surface is. Typically used for faking light passing through a surface when lit from behind, such as a cat's ear or a stretched hide. It can be used in a Sub Surface Scattering (SSS) shader or directly in the diffuse/albedo to fake a SSS effect.

Thickness (or Translucency): Thickness maps are a measurement of how thick/thin a surface is. Typically used for faking light passing through a surface when lit from behind, such as a cat's ear or a stretched hide. It can be used in a Sub Surface Scattering (SSS) shader or directly in the diffuse/albedo to fake a SSS effect.

Emissive (or Glow): Emissive maps mimic surfaces that emit light, like a computer monitor, a vehicle dashboard at night, or magical effects.

Emissive (or Glow): Emissive maps mimic surfaces that emit light, like a computer monitor, a vehicle dashboard at night, or magical effects.

Opacity (or Transparency): Transparency maps define how opaque a texture is at a certain position. Bright zones mean 100% opaque (solid) areas, dark zones are transparent areas.

Opacity (or Transparency): Transparency maps define how opaque a texture is at a certain position. Bright zones mean 100% opaque (solid) areas, dark zones are transparent areas.


Subsurface Scattering (SSS): SSS maps fake the behavior of light bounce inside of a semi-translucent surface, such as skin or wax. The brighter the area, the more light passes through.

Diffuse: Overall color

Diffuse: Overall color

Epidermal: Top-most layer of the skin, should look like the skin is drained of all blood (like a dead body), desaturated version of the overall color map.

Epidermal: Top-most layer of the skin, should look like the skin is drained of all blood (like a dead body), desaturated version of the overall color map.

Subdermal: The level underneath the skin, saturated version of the overall color map with a bit of blur on it. The lower dermal area contains more blood vessels and typically is the layer that contributes the most red because of this.

Subdermal: The level underneath the skin, saturated version of the overall color map with a bit of blur on it. The lower dermal area contains more blood vessels and typically is the layer that contributes the most red because of this.

Back Scatter: Back scatter is what controls the way the material handles light that comes from behind the object. This is due to the light absorption in specific parts of the face for example, ears nose, thin skin areas.

Back Scatter: Back scatter is what controls the way the material handles light that comes from behind the object. This is due to the light absorption in specific parts of the face for example, ears nose, thin skin areas.

Specular: A specularity map defines how strong the textured surface will 'shine' at a certain position.

Specular: A specularity map defines how strong the textured surface will 'shine' at a certain position.