Unit 04 Buses

Download as pdf or txt
Download as pdf or txt
You are on page 1of 24

Understanding PC and Troubleshooting Unit 4

Unit 4 Buses
Structure:
4.1 Introduction
Objectives
4.2 Bus
4.3 Industry standard architecture (ISA)
8-Bit-ISA
16-BIT-ISA
4.4 Peripheral component Interconnect (PCI)
PCI Bus configuration and signals
PCI layout
PCI signals
4.5 Accelerated graphics Port
AGP vs PCI
AGP layout
AGP signals
4.6 FSB
4.7 Dual Independent Bus(DIB)
4.8 Troubleshooting
4.9 Summary
4.10 Glossary
4.11 Terminal Questions
4.12 Answers

4.1 Introduction
In the previous units we discussed about many important components of the
computer system like memory, CPU, motherboard, memory slots, etc.,
which are required for a computer? But, when you have various important
components together the question arises as to how they could be linked and
communicated between each other so that the data in the form of
information or address of the memory location can be transferred. This
question is answered by the use of electric wires. These are called Buses.
Specifically Bus is an electric cable or collection of cables that are used to
transfer the data, information or the address of the memory location. In this
unit we will discuss the importance of this data travelling agent and list the
different types and parts of the Bus. We will explain the importance of AGP
Manipal University Jaipur B1416 Page No.: 66
Understanding PC and Troubleshooting Unit 4

and help you to use of expansion Buses to connect the expansion devices.
IBM came to a conclusion that connecting the expansion elements which
are required and leaving the other functions of the expansion devices which
can connect with the plugin through the Bus was a feasible choice.

Objectives:
After studying this unit you will be able to:
 Define a Bus.
 Identify a Bus in the computer system
 Classify and explain the different types of Bus in detail.
 Differentiate between the types of Bus
 Resolve the basic Bus problems through troubleshooting

4.2 BUS
As we discussed above Bus is a collection of cables that is used to
communicate between two or more components of the computer system.
Bus consists of three major types such as address Bus, data Bus and
control Bus. Address Bus is the one which carries the formation of the
location of the information. Data Bus is the one which carries data between
the different components of the computer. Control Bus is the one which
allows transferring the control instructions from CPU to memory or I/O
devices. A typical diagram of the connectivity of the Buses is as shown in
the figure 4.1. In this figure you can observe that the data Bus allow two way
transfers between CPU, I/O and the memory. Address Bus and Control Bus
allow only the control flow from CPU to I/O and memory.

Figure 4.1: Bus connection

Manipal University Jaipur B1416 Page No.: 67


Understanding PC and Troubleshooting Unit 4

The control Bus helps in performing the specific operations by the


respective devices. Address Bus is used to send and specify the address of
the specific location. The data bus is used to transfer instructions from
memory to the CPU for execution. It carries data to and from the CPU and
memory as required by instruction translation.
All Buses are measured in terms of clock speed and measured by MHz.
Width specifies the amount of data transferred in one time. For example 8
bit Bus can transfer 8 bit data. Buses are connected to number of devices
which are working at different speed, some are faster and some or slow.
When the devices operates which can perform slower than other devices
cannot send or receive the data. Therefore clock speed needs to set so that
all the devices interact with each other at same speed. A Bus is also called
as backbone because it connects the computers through one cable.
As discussed in unit 3 sections 3.2.2 most of the CPUs use the expansion
slots to connect the add-on rather than connecting the extended device.
Expansion Buses are used to connect the system with the peripheral
devices like printer, monitor, scanner, etc. This has made the PC a flexible
machine that can be configured to the requirement of the external devices or
add-ons. This has given rise to different types of Buses. In order to make
the application execution faster and the transfer of data bits faster, fast Bus
is used. Local Bus is used to directly connect the data transfer to the CPU.
Buses can be serial and parallel. Serial Buses are those which allow the
data to travel in single bit-serial form. Parallel Buses are those that allow
data transfer in multiple wires in a parallel form. Based on the type of Bus
used with the particular devices as mentioned above Bus can be
unidirectional and bidirectional. Unidirectional Buses allow data transfer
from one device to another in one way path. For example control Bus allows
transfer from CPU to I/O and Memory and not backward. Bidirectional
Buses allows data to travel through it from one device to another in both the
direction. For example data Bus allows the data to travel to and from the
CPU to I/O and memory. Internal and external Buses are used to connect
the internal and external devices. Internal Bus allows connection between
internal components like sound card and CPU. External Bus allows the
connectivity between any external devices like USB. Therefore different
architectures were designed to support the different kinds of Buses and the

Manipal University Jaipur B1416 Page No.: 68


Understanding PC and Troubleshooting Unit 4

computer system and its components. Generally with respect to this kind of
architecture, the Bus has been divided into three major types. They are ISA,
PCI and AGP.

Self Assessment Questions


1. ____________ is the electric wire that is used to answer the question
of linking and communication between different components.
2. State whether the following statements are true or false:
a. Control Bus is the one which carries data between the different
components of the computer.
b. Local Bus is used to directly connect the data transfer to CPU.
3. Buses are measured in terms of _____________ and measured by
_________.
4. ____________Buses are used to connect peripherals.
5. The three major types of Buses are ___________, _____________
and ___________.

4.3 Industry Standard Architecture (ISA)


As you have already studied meaning of ISA in unit 3 section 3.2.2 we will
directly discuss on the architecture of ISA. Industry Standard Architecture
Buses were also referred to as PC Buses. The use of standardized Buses
helped to support the operating system and many application software. ISA
made a path to the manufacturers to produce faster and supportable PCs
and expansion devices. Depending upon the architecture used by the
processor two versions of ISA was used.
They were:
 8-Bit ISA
 16-Bit ISA

4.3.1 8-Bit ISA


8-bit ISA was used by PC/XT Bus. The PC/XT is commonly referred to as
the XT, was IBM's successor to the original IBM PC. XT stands for extended
technology. The XT was designed to serve the enhanced machine for
commercial purpose.
8-Bit ISA was the original ISA found on IBM machines which was 8 bits wide
indicating the 8 bit data of the 8088 processor. 8-bit ISA consisted of 62 pins
Manipal University Jaipur B1416 Page No.: 69
Understanding PC and Troubleshooting Unit 4

out of which 20 address and 8 data lines were available. It ran with a speed
of 4.77 MHz. Diagram of Simple 8-bit Bus is given below in figure 4.2

Figure 4.2: Diagram of 8-Bit Bus

The 8-bit ISA Bus consists of single card edge connector with 62 pin
contacts. The Bus provides 8 data lines and 20 address lines which allow
the board to be placed in the conventional memory. The Bus supports
connection for 6 interrupts (IRQ2-IRQ7) and three DMA channels (DMA0-
DMA2).
All though each connector on the Bus is supposed to work in the similar
fashion, the PC which is designed with eight expansion slots requires any
card inserted in the eighth slot to provide a special signal called “card
selected” on pin labelled B8. Timing requirement for the eighth slot are also
tighter.
Interrupts and DMA (Direct memory access) channels are supported by this
Bus. The table below shows the DMA channel that has expansion with its
specific standard function. An interrupt is an event in hardware that triggers
the processor to jump from its current program counter to a specific point in
the code. DMA channels are system pathways used by many devices to
transfer information directly to and from memory.

Manipal University Jaipur B1416 Page No.: 70


Understanding PC and Troubleshooting Unit 4

Table 4.1: DMA channel expansion

DMA Channel Expansion Specific function


0 No It can refresh dynamic
RAM
1 Yes For Add-on cards
2 Yes Floppy disk controller
3 Yes Hard disk controller

The oscillator pin in both PC/XT and PC/AT ISA Bus configuration provides
the system oscillator signal to the expansion Bus. An oscillator is a circuit
which creates a waveform output from direct current input. When the PC
needs to be reset, the RESET DRV pin drives the whole system into the
reset state. The 20 address pins connects an expansion board to the
system address Bus. When address signals are valid, the Address Latch
Enable (ALE) signal indicates that the address may now be decoded.
When errors occur on the expansion board the -I/O channel Check (-I/O
CHCK) signal applies flags to the mother board. The minus sign before the
signal indicates that the active low logic is used in the signal. The 6
hardware interrupts are IRQ2, IRQ3, IRQ4, IRQ5, IRQ6, and IRQ7 which
are used by the expansion board to seek the attention of the CPU. Interrupt
0 and 1 handles the high priority of the timer chip and keyboard. Therefore
these interrupts will be busy handling timer chip and the keyboard and it is
not available to the Bus.
4.3.2 16-bit ISA
16-bit ISA is an expansion of 8-bit ISA with the connection of 36-pin
connector added along with the 62-pin connector as add-on. 8-bit Bus is
replaced by 16-bit Bus by doubling the number of data bits and adding pin
connectors. 16-bit was used in PC/AT Bus. The IBM Personal Computer/AT
are also called as PC AT or PC/AT, was IBM's second-generation PC,
designed around the 6 MHz Intel 80286 microprocessor and released in
1984 as machine type 5170. The name AT stands for "Advanced
Technology", and was chosen because the AT offered various technologies
that were then new in personal computers. . This diagram of 16-bit Bus is as
shown in the figure 4.3

Manipal University Jaipur B1416 Page No.: 71


Understanding PC and Troubleshooting Unit 4

Figure 4.3: diagram of 16-bit Bus

The diagram consists of 62 pin connector of 8 bit Bus and an additional 36


pin connector is used with label C and D. An extra 8 bits were used to
normalize the total 16 bit data bus. These 8 bits didn’t have any effect on the
working of the 16-bit Bus. The table shows the pin assignments of a 32 pin
connector.
In addition to 8-bit data, 16-bit Bus consists of extra 8 bits, five interrupts
and three DMA channels. One extra bit is used for +5dc volts which is not
included in number of bits. It is also added with four additional address lines.
The speed is increased from 4.77 MHz to 8.33 MHz.
You should observe that though in ideal situation PC/XT should run on
PC/AT expansion slots, it is far from reality with most of the older version of
XT slots.
Therefore there was an idea of mixing both 8-bit Bus and 16-bit Bus
architecture which consist of 32-pin connector and could be expanded
without starting from scratch for the sake of expansion. Now, ISA Bus
architecture has retired since most of the systems are adopting PCI and
AGP architecture.

Self Assessment Questions


6. ____________ Buses are also called as PC Buses.
7. The two versions of ISA are ___________ and ____________.
8. 8-bit ISA was used by __________ Bus.
9. 8-bit ISA consists of _________ pins out of which of ________ and
___________ are address and data liens respectively.
10. 16-bit Bus consists of extra _________ bits, ________interrupts and
_____________ DMA channels.
11. The speed is increased from ____________ to __________ MHz.
Manipal University Jaipur B1416 Page No.: 72
Understanding PC and Troubleshooting Unit 4

4.4 Peripheral component Interconnect (PCI)


As you have already studied meaning of PCI in unit 3 section 3.2.2 we will
directly discuss on the architecture of PCI. Graphics processing cards
became difficult to process with 32-bit Bus processors with 8.33 MHz speed.
During the development of versions there were two architectures which
were merged and named as VL and PCI. VL stands for VESA Local Bus.
(VESA) Video Electronics Standards Association local bus acted as high
speed path for DMA and memory mapped I/O. This is different from ISA Bus
as ISA Bus handles interrupts and port mapped I/O.
The disadvantage with VL was its Bus performance was dependent on CPU
speed. The manufacturer failed to give the standard specifications.
PCI was developed in mid-1992 by the integration of Intel and a group of
manufacturers. In section 4.4.1 and 4.4.2 you will study about the
configuration, layout and signals of PCI.
4.4.1 PCI configuration
PCI is a vast improvement over ISA Bus architecture with the increase in its
speed from 8.33 MHz. to 33 MHz frequency. This increased the data
transfer from 5MB/Sec to 132MB/sec. Another important characteristic was
that it gave automatic configuration property to the switches and peripherals.
This feature also automatically adjusts the interrupts requests, DMA
assignments, the addresses of the requests and data used by the PCI
peripheral. The following are the features supported by PCI Bus which are
used today.
 Linear Bursts: whenever memory sends data to the peripheral devices it
will be in linear address order. Linear Burst is a method of continually
filling the data line in the Bus by transferring the data. Linear address
order is the process of increasing the next byte in the order which read
or write the large amount of data into the single address. This is the
important feature of PCI. This helps in increasing the performance rate
of PCI when compared to Bus which does not have linear burst
capability by almost double the time.
 Low access latency: access latency is the time taken by the CPU to
permit the request access by the peripheral for its control. Reducing this
time increases performance of the PCI. Since this kind of expansion slot

Manipal University Jaipur B1416 Page No.: 73


Understanding PC and Troubleshooting Unit 4

support lower access latency the system can easily send the data to the
CPU.
 Bus mastering: Bus mastering is a feature of PCI which helps in any
intelligent peripheral to have a control of the Bus which is used in
boosting the throughput of the architecture. This supports in handling the
task according to the priority base. The high priority task can be handled
and throughput can be increased.
 Concurrency: when a processor chip works with BUS masters
simultaneously rather than waiting for them to complete, the process is
called concurrency. Wait state may be sometimes zero at the time of
concurrency
 Dual voltage architecture: dual voltage features expects two different
voltages to be supported. PCI Bus supports +5 volts as well as +3.3
volts which are called as low voltage mode.
4.4.2 PCI Bus Layout
For any architecture there must be proper borders and boundaries. This is
given to the PCI architecture by their layout. The layout of the figure to show
How it is arranged is given in figure 4.4.

Figure 4.4: PCI Bus layout

In the figure it is observed that +5 volts connector is divided into two


segments. A +3.3 volts connector is added with a key at the 12th position of
Manipal University Jaipur B1416 Page No.: 74
Understanding PC and Troubleshooting Unit 4

the Bus just to avoid any insertion of the +5 volts board to +3.3 volts Bus.
A +5 volts connector is added with a key at the 50th position to avoid the
wrong insertion of +3.3 volts board into the +5 volts slot.
In general the same system resources cannot be used by the other devices.
This is an exemption is PCI. PCI Bus has its own internal interrupts which
handle the requests on the BUS. To avoid the confusion PCI Buses are
named as A1, A2, A3,……., B1, B2, B3, …….. C1, C2, C3,…… and D1,
D2, D3,…….. PCI uses a technique which holds the signal dynamically and
whenever there is a requirement it can be reassigned. This technique is
called Interrupt ReQuest (IRQ). IRQ can be shared by two or more PCI
devices. These IRQs can be accessed during CMOS setup routine.
4.4.3 PCI signals
There are various signals which serve different functions in the PCI Bus.
The PCI configuration is differentiated as asynchronous and synchronous
based on the dependence of the FSB speed. FSB stands for Front sided
Bus which is responsible for data transfer between CPU and one of the two
chips of the Core logic chipset on the mother board. An asynchronous
configuration is found in the latest motherboards whose PCI speed can be
set independent of FSB.FSB derives the Bus clock of the PCI. You can see
some of the signals below in the table along with its functions.

Table 4.1: The function of PCI signals

Signals Functions
Clock signal Provides timing for the PCI Bus
It can be varied from 0Hz to 33 MHz
Reset Signal Reset all PCI devices
Command/-Byte Defines which of the bytes are transferred
enable signal
Parity/PAR64 Represents the Parity across address/data and byte enabled
signal lines
-Request Initiates the Bus mastering
(-REQ)
signal/line
-Grant (-GNT) Approves the Bus mastering request
signal/line

Manipal University Jaipur B1416 Page No.: 75


Understanding PC and Troubleshooting Unit 4

-Frame Signal Clock Frame is driven true when valid PCI Bus cycle is in
progress.
Target Ready Target Ready signal is driven true, if within the given Bus
(-TRDY) signal cycle the device is able to complete the transaction
Initiator Ready Indicates that the Bus is ready to accept the data or confirms
(-IRDY) signal that the data in the Bus is valid.
Stop (-STOP) This indicates the initiator to terminate or halt the current data
signal transfer.
Initialization This also works as Chip select during read and writes
Device Select transaction of PCI configuration.
(IDSEL) signal
Device Select As an Input, it has to indicate the current transfer status to
(DEVSEL) know whether it has assumed the control
signal As an output, it has to find out itself as the target for current
Bus transfer.
Lock (-LOCK) Prevents the other initiator to modify the locked addresses or
signal selected expansion device.
Primary and Reports data parity errors and address parity errors in the
secondary error PCI BUS.
reporting (PERR
and SERR)

Self Assessment Questions


12. ________ is a high speed intelligent Buses
13. State whether the following statements are true or false.
a. In VL architecture the performance of the Bus is independent of CPU
speed.
b. PCI gave automatic configuration property to switch and jumper-less
peripherals.
c. Linear bursts helps in increasing the performance rate of PCI when
compare to Bus which do not have linear burst capability by almost
double the time.
14. ___________ is the time taken by the CPU to permit the request
access by peripheral for its control.
15. ___________ is the technique which holds the signal dynamically and
wherever there is a requirement it can be reassigned.
16. ___________ derives the Bus clock of the PCI.
17. FSB stands for ____________.

Manipal University Jaipur B1416 Page No.: 76


Understanding PC and Troubleshooting Unit 4

4.5 Accelerated Graphic Port (AGP)


AGP stands for Accelerated Graphics Port which is designed to provide
throughput demands of 3 D graphics. The growth in the field of computer
science and engineering gave a significant improvement in the PC. The
visualization of the information is improved in its quality in colour depth
resolutions and wide range of special effects which creates the requirement
of more storage place. This is not only demands for more memory but also
requires huge bandwidth for data transfer. This was fulfilled by Accelerated
Graphics Port (AGP) which is well applicable for 3D applications.
You can visually experience the life like realism with the use of 3D images.
AGP helps you to accelerate between the text and 3D images and the PCs
graphic controller will be coordinating with the system memory to store and
transfer the graphics data. AGP has many additional characteristics which
make AGP a useful technology for graphics. The following are the
characteristics of AGP.
 It accelerates graphics performance by giving high speed Bus
 Graphic controller is enabled
 Helps in speeding up the data flow from CPU to graphics controller. In
this case the data is about the decoded data of video.
 It provides high bandwidth
Structure of AGP
Since the AGP’s performance is directly proportional to the bandwidth, the
device is fixed in such a position that increases the bandwidth. An
architecture is developed which act as an interface between chipset and
graphic controller. The architecture is as shown in the figure below:

Manipal University Jaipur B1416 Page No.: 77


Understanding PC and Troubleshooting Unit 4

Figure 4.5: Block diagram of AGP architecture

The AGPset is used to accelerate the graphic performance and is directly


connected to the system memory through a dedicated high speed Bus. 3D
texture data is provided to the AGPset through the system memory. AGP
enables the hardware called as graphic accelerated controller to execute
texture maps directly from system RAM or local memory. An AGP interface
is placed between PC’s chipset and graphic controller which helps in
increasing the bandwidth. The graphic accelerator controller and the
AGPset is connected through AGP Bus. This AGP helps to speed the flow
of decoded video from the AGPset to the graphic controller and vice-versa.
Another dedicated high speed Bus is connected to L2 cache through the
core data controller. In addition to the conversion of texture mapping,
offloading this tremendous data overhead from the PCI Bus leaves PCI free
to handle drive data transfers and other I/O controllers.
With the help of AGP that offers the bandwidth which is twice the bandwidth
of PCI, 2 bytes of data can be made to pass through the AGP CLOCK with a
bandwidth of 532 MB/s. More refinement or adjustment to the AGP handles

Manipal University Jaipur B1416 Page No.: 78


Understanding PC and Troubleshooting Unit 4

the data and allow 4 Bytes to be passed with the bandwidth of more than 1
GB/s on every AGP clock. This means that the bandwidth is doubled when
the AGP clock refines the data byte.
4.5.1 AGP vs PCI
Before studying the difference between AGP and PCI we should throw a
glance on its similarities. Both Buses use at least one PCI slot. AGP was not
created to replace PCI, but to create a new dedicated Bus which is used to
help the graphic cards functioning.
The important feature of AGP specification is that a 32-bit AGP will easily fit
into PCI configuration. But still there are many features which differentiate
AGP from PCI with advance advantages.
They are:
 The bandwidth capacity of AGP is 4 times greater than PCI with
pipelining or super-pipelining features.
 AGP supports optimized 3D graphics
 AGP uses clock speed up to 66 MHz where the PCI used a fixed clock
speed of 33 MHz.
 AGP makes multiple requests for the transaction and PCI will make only
one request at a time and wait until the earlier requests are executed
completely.
 There is no sharing of bandwidth with other devices in AGP whereas
PCI shares it with other devices.
The result of the significant improvement and additions in 32-bit AGP
intended in optimising for high performance 3D graphics. The most
important difference is the clock speed. The PCI users fixed 33 MHz Bus
whereas the clock speed of AGP varies up to 66 MHz.
The basic comparison between PCI and AGP has been shown in the below
table.
Table 4.2: Comparison between PCI and AGP
PCI AGP
Address/data is multiplexed Address/data is de-multiplexed
Non pipelined Request is pipelined
Data bytes will be high at 133 MB/s Data bytes will be peak at 533MB/s
with 32 bits with 32 bits

Manipal University Jaipur B1416 Page No.: 79


Understanding PC and Troubleshooting Unit 4

Multiple target and multiple master Single target and single master
Connected to the entire system Connected to only read/write memory.
this is not connected to I/O operations
There is High/low priority queues No priority queues

4.5.2 AGP layout


AGP is used on ATX and NLX machines. NLX (New Low Profile Extended)
was a form factor proposed for low profile, low cost, mass-marketed retail
PCs. NLX specification defines the motherboard size, hole mounting
locations, riser card location, and maximum component heights on
motherboard. It has three variations 3.3 V, universal and 1.5V. Though the
signal layout is similar the key locations are different. Therefore, these cards
arenot interchangeable. While checking compatibility between video card
and the motherboard In AGP you can ignore the speed multiplier as it is
negligible and interchangeable. The table below will help you in
understanding the important functions of the different graphics card.

Table 4.3: AGP cards and its functions

AGP card Type of Description


connector
It Supports 3.3V signalling and has the
3.3V Card 3.3V slot
available speeds1x, 2x.
It Supports 1.5V signalling and has the
1.5V Card 1.5V slot
available speeds1x, 2x, 4x.
It Supports 3.3V and 1.5V signalling and
Universal
Double slotted has the available speeds 1x, 2x at 3.3V and
Card
1x, 2x, 4x at 1.5V.
It Supports 0.8V signalling and has the
3.0 Card 1.5V slot
available speed4x, 8x.
Universal It Supports 1.5V and 0.8V signalling and
1.5V 3.0 1.5V slot Available speeds are 1x, 2x, 4x at 1.5V and
Card 4x, 8x at 0.8V.
Supports AGP 3.3v, 1.5V, and 0.8V
Universal 3.0 signalling. Speeds available are 1x, 2x at
Double slotted
Card 3.3V and 1x, 2x, 4x at 1.5V and 4x, 8x at
0.8V.

Note that different slots will have different key positions in the card.

Manipal University Jaipur B1416 Page No.: 80


Understanding PC and Troubleshooting Unit 4

4.5.3 AGP signals


In order to work in an advance performance mode the full bandwidth needs
to be utilized and with the Bus is queued by the target. The following are
some of the signals which helps in operating the AGP card by asserting
them.

Table 4.4: AGP Signal and its functions

Signal Function
PIPE # This request is asserted by the current
master
It indicates that the target has to queue
up the full width request.
Read Buffer Full signal (RBF#) Indicates whether the pre-requested low
priority read data can be readily accepted
by the Master.
Once RBF# is asserted, the arbiter
cannot initiate the request of taking back
the low priority read data to the master.
Write Buffer Full signal (WBF#) Indicates Whenever there is data keying
from the core program to ensure that
whether master is ready to accept the
data
Once WBF# is asserted the arbiter
cannot initiate the request to provide
data.
GNT# Indicates that the signals have some
meaning to the master
FRAME# Indicates the output form the core logic
AD Bus Strobe 0 signal (AD_STB0) 2x data transfer mode on address line is
provided
Data providing agent drives this signal
SideBand Strobe signal (SB_STB) Timing for SBA is provides and driven by
AGP master
CLOCK (CLK) Provides timing for both AGP signals

Clock divisors help in under clocking or overclocking the FSB in the


motherboard. Wider range of FSB gives greater variety of clock divisors.

Manipal University Jaipur B1416 Page No.: 81


Understanding PC and Troubleshooting Unit 4

Self Assessment Questions


18. _______________ Port is applicable for 3D applications.
19. State whether the following statements are true or false
a. Performance of AGP is directly proportional to the bandwidth
b. The bandwidth capability of AGP is 4 times greater than PCI
c. RBF# indicates the pre-requested low priority read data can be
readily accepted by the master.
d. GNT# indicates that the Whenever there is data keying from the
core program to ensure that whether master is ready to accept the
data
20. AGP is used on __________ and _______ machines.

4.6 Front Sided Bus (FSB)


FSB stands for Front Sided Bus and used to connect CPU to the computer
system and CPU with the various other components of the system. FSB is
two directional flows of data. It means various components send and
receivethe data from the processor. FSB is also called as processor Bus
because its acts as a main path for the data flow from processor to other
parts of the motherboard. The speed of FSB ranges from 66MHz, 100 MHz,
133 MHz, 266MHz, 400MHz and above. This is one of the most important
considerations while buying motherboard.
The important feature of the FSB is its bit size. It is observed that 32 bit FSB
performs 100% more than that of 16 bit FSB. Therefore the measure of data
transfer delivered in one tick (part of 1 CLOCK cycle) is an important
consideration of FSB.
The communication between various components is done by FSB through
the Chipset. The chipset consists of two processors namely Northbridge and
Southbridge. Both the bridges act as controllers and policemen between the
CPU and various components of the system which is responsible for data
transfer among them. Northbridge has its own heat sink since it works
continuously and easily gets heated up fast. Northbridge is also called as
Memory Controller Hub (MCH) and Integrated Memory Controller (IMC).The
Southbridge is the chip that controls all the functions of the computer
peripherals. The Southbridge can usually be distinguished from the north
bridge by not being directly connected to the CPU.

Manipal University Jaipur B1416 Page No.: 82


Understanding PC and Troubleshooting Unit 4

The main difference between Northbridge and Southbridge is that


Northbridge will take care of traffic of the faster components of the CPU and
the Southbridge will controls the slower traffic components. The faster
components of the CPU are video cards and random access memory.
Sometimes Northbridge also handles the control of CPU cache. The slower
components as I/O devices, DMA controller and system clock fall in to the
control of Southbridge and it is also called as I/O controller Hub (ICH). The
speed of these Buses depends on speed of FSB. When the CPU is fast and
it is connected with low performance FSB then there is a result of data
bottlenecks. In this case Processor performs faster than FSB for one activity
and waits for FSB to send the new instruction. This state of CPU is called
idle state. Therefore to utilize the faster CPU you should also have faster
FSB.
Self Assessment Questions
21. FSB is also called as ________________.
22. The chipset consists of two processors namely ___________ and
____________.
23. The state of CPU that it performs faster than FSB for the activity and
waits for FSB to send the new instructions is called ____________.

4.7 Dual Independent Bus (DIB)


Dual Independent Bus is architecture to connect the processor, memory,
and L2 cache. L2 cache is a set of memory circuits which is designed to
speed up access to frequently used data. One bus connects the processor
to L2 cache and a second connects the processor to main memory. The two
Bus features of DIB enables the data flow simultaneously or in parallel
processing method rather than queuing up in a single sequential method
and waiting for a long time to complete one instruction and CPU going to
idle state. In DIB one Bus connects the CPU to level-2 cache and other Bus
connects memory to level-2 cache. The main use of this connection is that
level-2 cache helps in increasing performance of the Bus externally.
The following are the features of DIB:
 Increase of speed due to double Bus
 Scale up of the performance externally independent of the speed of the
Bus.

Manipal University Jaipur B1416 Page No.: 83


Understanding PC and Troubleshooting Unit 4

 Allows access to faster cache.


 Allows multiple and parallel cache requests through the process of
pipelining cache to the processor.
Self Assessment Questions
24. DIB connects ____________ to _________ and ___________.
25. DIB allows single sequential method for processing. (True/ false)

4.8 Troubleshooting
Generally trouble shooting of the Busis not done since it is more of a
passive connector. It can be very easily replaced. But for some of the major
signals which are connected to the motherboard you might need to trouble
shoot. This happens at the POST board. Normally you could do this process
by using LED lights in the POST boards. A light-emitting diode (LED) is a
semiconductor light source. These are used as indicator lamps in many
devices, and are increasingly used for lighting. LED light is used in POST
board to indicate the status of the power supply, control signals and clock
time. When the LED is missing in the board then it means that there is
some error in connection in the motherboard.
You can note the following hints to rectify the Bus errors.
 You can use the multi-meter to check the voltage level in the Bus.
Irrespective of low voltage or standard Bus you have to find the -12 volt
dc and +5 volt dc. You should ensure that a low voltage Bus should
have +3.3 Volt dc. Otherwise, you need to conclude that there is fault in
the Bus and need to replace power supply.
 You can adjust the CLOCK signal between 0 Hz and 33 MHz. The
timing signals for the external device and slots is provided by CLOCK
signal. If the CLOCK signal is 0 then the expansion device will not work
and you need the check the motherboard circuitry and rectify it.
 After correction or rectifying the expansion device, it can be set to its
original settings by reinitializing with the help of Reset signal. You should
always note that Reset signal should be active only once or twice after
the power is supplied.

Manipal University Jaipur B1416 Page No.: 84


Understanding PC and Troubleshooting Unit 4

Some of the important considerations while connecting Bus to the


various components are as follows.
 You should take care of the Bus slot versions. For example a PCI
interface of 2.4 may not work with its full functionality when it is
connected to a PCI interface of version 2.3 or lower versions. They are
compatible with the only slots to which they are designed for.
 As the connectors are mechanical devices, they are more likely to worn
out. When you are connecting the device be careful while fixing the
motherboard because if the connections are wrong some of the
functionality of the motherboard may fail or it may also cause severe
motherboard damage. If you are more likely connecting and
disconnecting the expansion devices, it may result in unreliable
connections.
 It is always better to remove the faulty board or replace it. You should
always remove and reinsert the board since board is a metallic device. If
the board develops metallic oxides it will not work properly. By removing
metallic oxide you can remove the dust from the expansion slots.
 When you find a faulty board, after removing it from the system test it on
another board before replacing it to confirm the problem is related to
motherboard itself. Sometimes the signal in Bus may also cause for
different kind of component failure.
When the Bus slot gets failed, then very small methods you can apply.
 You can block the slot and damaged slot should not be used
 Replace the damaged Bus slot with the new Bus slot connector.
 Replace completely the motherboard.
Self Assessment Questions
26. Troubleshooting of Buses can be done at ____________.
27. _________ is used to find the faulty Buses.
28. CLOCK signal can be adjusted between _________and _______.

4.9 Summary
Bus has been the little passive interrupt which has the capability to be an
agent for communication between CPU, memory and various components
of the computer. Bus is divided into three types of Bus, address Bus, data

Manipal University Jaipur B1416 Page No.: 85


Understanding PC and Troubleshooting Unit 4

Bus and control Bus. It is a backbone of the network allowing the data to
pass through various computers and network. These are the electric wires
which help in transferring the data. There are various types of Buses based
on the type of its use such as expansion Bus, local Bus, internal and
external Bus, unidirectional and bidirectional Bus. The Bus architecture is
divided into 3 major types like ISA, PCI and AGP. The other kinds of Buses
are FSB and DIB which are responsible for connecting the various
components to motherboard. Industry Standard Architecture which are not
in use were used to connect the system with video cards, sound cards,
networking cards, etc. This is bigger in size than the PCI and is black in
colour. ISA Buses were also referred as PC Buses. ISA has its two versions
namely 8-bit ISA and 16-bit ISA.
Peripheral Component Interconnect (PCI) is a high speed, intelligent Bus.
PCI supports many features like, linear burst low access latency, Bus
mastering, etc.
AGP is best suited for 3D applications. AGP’s performance is directly
proportional to the bandwidth; its bandwidth is twice than that of PCI. The
important feature of AGP specification is that a 32-bit AGP will easily fit into
PCI configuration. It can process multiple request rather than waiting for a
long time for the new instruction and making the CPU sit idle.Bus is queued
by signals.
FSB stands for Front Sided Bus and used to connect CPU to the computer
system and CPU with the various other components of the system. FSB is
two directional flows of data, i.e. means various components send and
receives the data from the processor.
DIB which stands for Dual integrated Bus enables the data flow
simultaneously or in parallel processing method rather than queuing up in
a single sequential method and waiting for a long time to complete one
instruction and CPU going to idle state.
Troubleshooting with respect to Bus is done at POST boards and rectified
using LED lights.

Manipal University Jaipur B1416 Page No.: 86


Understanding PC and Troubleshooting Unit 4

4.10 Glossary
Term Description
I/O The term I/O is used to describe any program, operation or
device that transfers data to or from a computer and to or from
a peripheral device. Every transfer is an output from one device
and an input into another. Devices such as keyboard.
Back bone A central conduit designed to transfer network traffic at high
speeds.
Interrupt An interrupt is an event in hardware that triggers the processor
to jump from its current program counter to a specific point in
the code. It is a signal that says that an event has occurred.
LPT Port Line Print Terminal is the original, and still common, name of
the parallel port interface on IBM PC-compatible computers. It
was designed to operate a text printer that used IBM's 8-bit
extended ASCII character set.
PC/AT The IBM Personal Computer/AT are also called asPC AT or
PC/AT, was IBM's second-generation PC, designed around the
6 MHz Intel 80286 microprocessor and released in 1984 as
machine type 5170. The name AT stood for "Advanced
Technology", and was chosen because the AT offered various
technologies that were then new in personal computers
DMA Channel Direct memory access (DMA) channels are system pathways
used by many devices to transfer information directly to and
from memory.
Expansion IN the computer an opening is there where a circuit board can
be inserted to add new capabilities to the computer called
expansion slots.
COM3 A COM3 (or Communications 3) port is a communications
channel on the computer that is used to transmit data from
hardware devices to the processor.

IRQ IRQ stands for Interrupt ReQuest, and refers to special


numbered channels that are used by devices to get the
processors attention.
PC/XT The IBM PC/XT (also written PC-XT or PC XT), commonly
referred to as the XT, was IBM's successor to the original IBM
PC.

Manipal University Jaipur B1416 Page No.: 87


Understanding PC and Troubleshooting Unit 4

4.11 Terminal Questions


1. Explain the working of Bus connection.
2. Bring out the different DMA channels with its specific functions
3. After the development of two versions of architecture of Buses, why was
PCI Bus selected?
4. Explain the features supported by PCIBus?
5. List the different PCI signals along with their functions?
6. What are the characteristics of AGP?
7. Explain the architecture of AGP along with its figure?
8. Compare between PCI and AGP?
9. How do you troubleshoot Bus problem?

4.12 Answers
Self Assessment Questions
1. Buses
2. a. False , b. True
3. CLOCK speed and MHz
4. Expansion
5. ISA, PCI and AGP
6. ISA
7. 8-bit and 16-bit
8. PC/XT
9. 62, 20, 8
10. 8, 5, 4
11. 4.77 MHz to 8.33 MHz
12. PCI
13. a. False , b. True , True
14. Access time
15. Interrupt ReQuest (IRQ)
16. FSB
17. Front Sided Bus
18. AGP

Manipal University Jaipur B1416 Page No.: 88


Understanding PC and Troubleshooting Unit 4

19. a. True m b. True , c. True , d, False


20. ATX and NLX
21. Processor Bus
22. Northbridge and Southbridge
23. Idle state
24. Level-2 cache, CPU and Memory
25. False
26. POST board
27. LED lights
28. 0 Hz to 33 MHz

Terminal Questions
1. Refer Section 4.2. BUS
2. Refer section 4.3.1, 8-bit Bus, Table 4.2. DMA channel expansion
3. Refer Section 4.4, Peripheral component Interconnect
4. Refer Section 4.4.1, PCI configuration
5. Refer Section 4.4.3, PCI signal, table 4.1, the functions of PCI signals
6. Refer Section 4.5, Accelerated Graphic Port
7. Refer Section 4.5, structure of AGP
8. Refer Section 4.5.1, AGP vs PCI
9. Refer Section 4.8, Troubleshooting

References:
 Trouble shooting, maintaining and repairing PCs, the 1st author in PC
hardware, 5th edition by Stephen J. Bigelow, Tata McGraw Hill
Publications, 2001
 Upgrading and repairing PCs, 15th anniversary edition by Scott Mueller,
Que publishing, 2003

E-References:
 http://www.interfacebus.com/Design_Connector_PCMCIA.html
 http://parts.digikey.com/1/parts/601570-ic-8-bit-bus-sw-2port-20-tssop-
pi5c3244le.html
 http://www.javvin.com/hardware/DIB2.html
 http://www.developer.Intel.com/technology/agp/index.htm

Manipal University Jaipur B1416 Page No.: 89

You might also like