3.3.15 Atmospheric Effects
The Graphics Libriry Simulates fog and haze effects required for visual simulation
applications bv blending the object color with a user-specified fog color.  The user also
controls the fog density through the Graphics Library interface.  In the VGXT versions,
this computation is calculated on a per-pixel basis through the special Image Engine
processors on the VGXT raster boards.

3.3.16 Subdivision

The realism of a scene increases as the size of geometric primitives decreases.  The
geometry subsystem automatically subdivides incoming geometric primitives based
on a Liser-specified tolerance.  This unique feature allows the user to control the quality
of the rendering without having to modify the database.  On VGXT systems, per-pixel
texture perspective and fog computations can help avoid the use of subdivision for
high performance applications that cannot afford the generation of additional, and
perhaps extraneous, vertices,

3.3.17 Ray-Tracing and Radiosity

The PowerVision architecture includes specialized hardware that can be used to
accelerate software-based ray-tracing models and radiosity global illumination
3.3.18 Fast Screen Clear
Rendering complex scenes with update rates of 30 Hz or more requires a very fast
screen clear capability.  At 30 Hz, the application must be able to clear the screen and
render the geometry in less than 33 milliseconds.  At these update rates, without a fast
screen clear insufficient time is left to render the scene.  The PowerVision architecture
uses parallel Image Engines to clear at a rate of 800 million pixels per second for a five
span system and 1.6 billion pixels per second for a ten span svstem.  Thus, at 30Hz,
clearing the entire 1280 x 1024 screen requires only 5 percent of the frame time (1.6
milliseconds) for a five span system or 2.5 percent of the frame time (.8 milliseconds)
for a ten span system.

3.3.19 Stencil Planes

The stencil bitplanes implemented in the Raster Subsystem depth buffer allow a new
mechanism for affecting the results of pixel algorithms, In many ways, the stencil can
be thought of as an independent, high-priority Z-buffer.  The stencil value can be
tested during each pixel write, and the result of the test affects both the resulting
stencil value, and whether the pixel algorithm produces any other result.

One application of the stencil is Z-buffered image copy.  With one pass, the stencil
planes record the result of depth comparisons between source and destination areas
of the frame buffer; with a second pass, the image is copied from source to
destination, with only the pixels that passed the depth comparison being updated.  As
an example, this method can be employed with a library of small 3-D images, such as
spheres and rods, to quickly construct molecular models in the frame buffer.