|
Graphics
Accelerators
A graphics accelerator
is a type of video card that has its own processor to boost performance.
With the demands that graphics applications make in a multimedia
environment, graphics accelerators have become not just an enhancement,
but also a common necessity. The processor on a graphics accelerator card
is specifically designed to manage video and graphics. Some features
included on a graphics accelerator are:
MPEG decoding
3D graphics
Dual porting
Color space conversion
Interpolated scaling
EPA Green PC support
Digital output to flat
panel display monitors
Applications support
for popular high- intensity graphics software such as AutoCAD, Quark,
Windows 9x, Windows NT, and Windows 2000. All these features are designed,
in some way, to reduce the burden on the systemboard CPU and perform the
function much faster than the systemboard CPU. For more information about
graphics accelerator cards, see these manufacturers at their Web sites:
ATI Technologies at http://www.ati.com, Matrox Graphics, Inc, at
http://www.matrox.com,
and 3-Dfx Interactive, Inc. at http://www.3-Dfx.com (makers of the
popular Voodoo graphics card).
Video Memory
Older video cards do
not have memory, but video memory is necessary today to handle the large
volume of data generated by increased resolution and color. Video memory
is stored on video cards as memory chips. The first video cards to have
memory all used DRAM chips, but now there are several technologies for
video memory chips.
The amount of data
received by a video card from the CPU for each frame (or screen) is
determined by three factors:
1. The screen
resolution, measured in pixels
2. The number of
colors, called the color depth, measured in bits
3. Enhancements to
color information, called alpha blending. The more data needed to generate
a single screen, the more memory is required to hold the data. This memory
is called the frame buffer. Besides the frame buffer, some video cards
need memory to store font or other graphical information.
Color Depth
Color depth is directly
related to the number of bits used to compose each pixel. Four, 8, 16, or
24 bits per pixel may be used. The larger the number of bits allocated to
storing each piece of data, the more accurate the value can be; in like
manner, the greater the number of bits allocated to store the value of
pixel color, the greater the number of color shades you can use and color
depth you can have. To determine the number of colors that can be
represented by each number of bits, use the number of bits as the exponent
of the number 2. For example, to calculate the number of colors
represented by 4 bits per pixel, you would raise 2 to the 4th power, which
equals 16 colors. (Note that the largest 4-bit number is 1111, which
equals 15 in decimal. If you include 0, then the number of values that can
be stored in a 4-bit number is 16.) A color depth of 24-bits per pixel
equals 2 to the 24th power or 16.7 million colors.
<Previous>
<Home>
<Next>
|
