In reflection mapping, a distortion in appearance resulting from two nearby vertices on a surface that have different normals, and therefore different texture images.

The portions of a surface where normals point away from the viewpoint.

Rendering specific portions of a scene in a distinctive color, indicating a portion of the scene graph that is ready to be picked and manipulated independently of other objects in the scene.

Level of detail. Usually refers to a csLOD scene graph node, a subclass of csSwitch, that allows you to select the accuracy with which you render an object. The csLOD node selects amongst its children based on the distance from the viewpoint to the node. The children are indexed by an integer. Typically, as the index increases, the rendering rate also increases, and the amount of detail in the child decreases.

For reflection mapping, the distance to the texture image environment map is finite; reflections do not depend solely on the direction of the reflection angle. Reflections from a large flat surface vary; they show the alternating lights in the room (see Figure 8-2).

For reflection mapping, the distance between the viewpoint and the surface is finite. The texture coordinates depend on the complete ray-path geometry: the location of the viewpoint and the location of the reflecting surface point and its normal. These quantities, and the distance to the texture image, define the point where a ray intersects the cylinder (see Figure 8-2).

Eliminating from the graphics pipeline objects that cannot be seen from the viewpoint because they are behind foreground objects.

Selecting objects from a scene and manipulating them independently from the rest of the objects in the scene. For example, removing a wheel from a rendered car and moving it about on the screen.

A traversal callback implemented after a traverser leaves a node.

A traversal callback implemented before a traverser enters a node.

A method of simulating a complex lighting environment in which you treat a surface as a reflector and follow one ray (from your eye and reflecting off the surface) to select a point on a texture image that defines the visual environment. As an object rotates in the environment, the image appears to move over the surface, in contrast to perhaps better-known texture-mapping techniques, which fix an image on a surface.

For reflection mapping, the reflection geometry is simplified so that only the direction of the reflection vector determines texture coordinates. Effectively, the texture map is very far away (see Figure 8-1).

For reflection mapping, the reflection geometry is simplified so that only the direction from the viewpoint to the center of the scene determines the ray direction for every point in the scene: all the rays from the viewpoint are parallel. Effectively, the viewer is very far away (see Figure 8-1).

Also known as representations. Higher-order geometric primitives, that objects not made simply from triangles. Typically a rep is more like a pure mathematical object and must be tessellated with triangles before rendering.

Organizing a scene graph to reflect the spatial relationships of the objects in the scene.

Defining a common boundary for two surfaces.

An object that approximates a higher-order geometric surface (a rep) with a set of triangles. Triangles are OpenGL primitives, but reps typically are not. Tessellation is a way to render a rep.

An image that is used in texture or reflection mapping. These operations map each point on the surface of an object to a point in the texture image. With a texture map, the association is done once; the texture image is fixed on the surface, even when the surface moves. With reflection mapping, the image appears as a reflection from a fixed environment, and slides over a surface as it rotates.

Also known as triangle fans. A trifan is made of a set of adjacent triangles with one common vertex. One vertex is required to add a triangle to a trifan. The other two vertices of the triangle are the one common to all triangles in the fan, and a vertex shared with only one other triangle. See Figure 3-1.

Also known as triangle strips. A tristrip is made of a series of adjacent triangles developed iteratively from one triangle by adding a vertex and sharing two vertices with a triangle already in the strip. See Figure 3-1.

Eliminating from the graphics pipeline objects that cannot be seen from the viewpoint because they are outside the viewing frustum, that is, outside the field of view.