Hardware Reference
In-Depth Information
Accelerated Graphics
Port (AGP)
Introduced in 1997, AGP was designed to provide a cost-effective means of
improving the video performance of a PC and of reducing the burden that
would otherwise be imposed on the PCI bus by having to cope with the fast
throughput of video data. AGP enhances the interface between the video chipset
and the processor, and also makes it possible for the video processor to have
access to the main system memory. The AGP slot is physically similar to the
PCI slots that may be fitted to a motherboard. However, the AGP slot is usually
offset further from the edge of the motherboard. The AGP specification is based
on the PCI 2.1 specification which includes a 66 MHz bus speed.
Like the PCI bus, the AGP bus is 32-bit wide however, instead of running
at half of the system (memory) bus speed, AGP runs at the full speed of the
bus (66 MHz). AGP also benefits from the fact that, as only one slot is present,
there is no need to share the available bandwidth with any other devices!
In addition to doubling the speed of the bus, AGP has defined a double speed
(2
) mode that allows twice as much data to be sent over the port at the same
clock speed. In this mode, the hardware places data on the bus on both the rising
and falling edges of the clock signal. In contrast, the PCI bus places data on
only one of these transitions. The theoretical bandwidth is thus increased to a
little over 500 MB/s.
In the context of data acquisition, control and instrumentation, and as a poten-
tial means for interfacing to external hardware the AGP has obvious limitations,
not least of which is that there is only one slot available and this may already
be occupied by a graphics card. Happily, for most applications the PCI bus is
capable of providing sufficiently fast throughput with the added bonus that it is
well supported by a huge range of I/O cards.
The Universal
Serial bus
Offering true plug-and-play capability, the Universal Serial Bus (USB) has
become the
de-facto
future standard for the interconnection of a host computer
to a wide range of simultaneously accessible peripheral devices that share the
available USB bandwidth through a host-scheduled, token-based protocol. Fur-
thermore, unlike most other forms of computer bus, USB allows peripherals to
be attached, configured, used, and detached while the host and other peripherals
are in operation.
The Universal Serial Bus was originally specified as an industry-standard
extension to the PC architecture with a focus on Computer Telephony Integra-
tion (CTI), consumer, and productivity applications. In framing the original
specification, the following criteria were applied in defining the original USB
specification:
•
Ease-of-use for PC peripheral expansion
•
Low-cost solution that supports transfer rates up to 12 MB/s
•
Full support for real-time data for voice, audio, and compressed video
•
Support for various types of data transfer
•
Ability to cope with diverse system configurations, form factors, and host
computers.
One of the principal advantages of USB is the speed at which it operates.
USB supports two data transfer rates; 12 MB/s (described as
high-speed
) and
an alternative (but still quite respectable) 1.5 MB/s (described as
low-speed
).
Figure 2.7 shows how USB's two data transfer rates compare with those offered
by other interface types and standards.
