3D OBJECT VISUALISATION ON A PC DISPLAY

Methodology

A solid object is visualized on a display-screen by creating a 3D model using an app and rendering the 3D model to the 2D screen.
The modeling is done at a very low level to make the rendering process efficient.
A classic 3D modeling and rendering pipeline,
has a modeling component and a rendering component.

1. Modeling:
Any 3D object is described as a connected set of 3D surfaces.
A 3D surface is represented as a set of 3D triangle (location and orientation) units.
Camera viewing parameters are specified.
The oriented 3D triangle is associated with a corresponding 2D triangle on a 2D image called a texture.
Lighting conditions are specified with one or more light sources.

3D model-editing tools are available to create an oriented 3D triangle-mesh representation of any 3D object.
Examples are Maya and 3D Studio Max.

2. Rendering
All rendering is done by a dedicated GPU that is included in an SOC (in addition to a CPU)
or by an externally connected GPU.
Rendering is done, broadly speaking in one of two ways:

a) Texture-mapping
The oriented 3D triangle is mapped to a planar 2D triangle based on camera parameters.
A corresponding triangle in its associated texture is mapped onto this planar 2D triangle.
Lighting-conditions control the final display of each planar 2D triangle.
The display of these planar 2D triangles is ordered,
either using a (hardware) Z-plane buffer that orders the planar 2D triangles on a pixel py pixel basis based on their corresponding Z-coordinates,
or by software depth-sorting -- ordering the triangles first and then sequencing the rendering operation.

b) Ray-tracing
Ray-tracing, starts with the same model, including a 2D texture associated with an oriented 3D triangle.
Instead, of texture mapping and illumination control,
it traces the path of each light-ray from each-lighting source from the source to the camera
after multiple reflections from oriented 3D triangles in the scene and their associated surface textures.
This is a more realistic visualisation of a real-world object, but far more computationally expensive.
GPU advancements mainly target better ray-tracing efficiency.