The following is the first part in a
series of articles pertaining to the OpenGL Application Programming
Interface (API). The series will continue to expand and (hopefully)
provide useful information to the public in understanding an API that is
still evolving (though quite mature enough as it is).
Part One will deal with what has been a revolutionary step forward in
terms of providing "realism" in 3D without requiring massive
amounts of processing power - Texture Mapping.
Introduction
Texture mapping is simply a function
where you fit images onto polygons (a polygon is a 2D shape with at
least three sides) in any particular scene. It is very possibly the most
significant advancement in PC graphics for the last decade. Texture
mapping can enhance the realism of games, providing for very realistic
images for anything from the walls of a room to any monsters you may
encounter in a game. Prior to the introduction of texture mapping for
the PC (it was actually being used in high-end workstations for stuff
like simulations, prior to making it to the desktop), achieving high
levels of realism was a matter of "brute force" – using
nothing but thousands or millions of small lit and shaded polygons that
take an unusual amount of time to render and hence resulted in very poor
performance.
Texture mapping is a very intensive task however, requiring a CPU to
do lots of calculations and hence results in very slow performance.
Fortunately the introduction of affordable and powerful 3D hardware into
the consumer market (such as those offered by 3dfx, ATI, Matrox, NVIDIA,
etc.) has taken this burden off the CPU while doing the calculations
more efficiently and faster at the same time.