Motherboard
Content
PROJECT REPORT ON
“MOTHERBOARD”
SUBMITTED BY MD. ASHAB AHMAD ROLL NO. : L1/CHIP.3.1/JAN12/1001 COURSE : CHIP. 3.1 Submitted to : ANKUR AWASTHI (Teacher)
COMPUTER CONSULTANCY LTD. LUCKNOW
1
ACKNOWLEDGEMENT
I would like to express my gratitude towards Mr. Ankur Awasthi who gave me the opportunity to work. I would like to thanks to my UPTECH, COMPUTER CONSULTANCY LTD. for their guidance and all other members for their support. I thank to my God, my parents and all my friends who help me lot in project.
MD. ASHAB AHMAD ROLL NO. : L1/CHIP.3.1/JAN12/1001
2
TABLE OF CONTENT
MOTHERBOARD COMPONENTS OF A MOTHERBOARD ARCHITECTURE OF CHIPSET WORKING OF CHIPSET PROCESSOR WORKING OF CPU PROCESSOR GENERATION AND FAMILIES COMPUTER MEMORY BIBLIOGRAPHY
3
MOTHERBOARD
A motherboard is the main printed circuit board in a computerthat contains the central processing unit, appropriatecoprocessor and support chips, device controllers, memory, andalso expansion slots to give access to the computer‘s internalbus. The motherboard is the PCs center of activity. All devicesin a computer are in some way connected to the motherboard.
4
After CPU, motherboard is the second most important component in the system and therefore, it definitely needs special attention. • The design of a motherboard is dependent on the typeof CPU and mainly oriented around the chipset presentonboard. • The motherboard features decide the systemperformance to a great extent, upgrade potential, etc. • As there are numerous types of processors, there aredifferent types of motherboards as well. • The classification is usually done on the basis of typesof CPU sockets carried by these boards.
Functions of Motherboard
• Motherboard provides a substrate upon which othercomponents of a system such as CPU, RAM, ROM, Chipset,and Expansion Slots can reside. • Motherboard provides the electrical connection betweenvarious components in the system. • Motherboard provides interface for various add-on cardssuch as 3D graphics cards, NIC, Sound cards, etc. throughvarious expansion slots such as PCI, ISA, AGP, etc. • Motherboard provides the necessary interface with a hostof I/O devices. E.g., on-board IDE or SCSI interface forhard-disks, CD-ROM drives etc. • They also provide other traditional I/O connectors such
asPS/2interface, RS232 serial COM ports, Bi-directionalparallel port
5
(LPT), Joystick connection through game port,floppy disk interface, etc. • The battery driven RTC chip on the motherboard storesCMOS setup information.
• They provide IR port, CNR slot, IEEE 1394, USB forattaching emerging high-speed serial devices.
6
COMPONENTS OF A MOTHERBOARD
There are various components of a motherboard which fixedtogether into a single unit leads to the proper functioning of amotherboard.(1) Chips: The active devices on the motherboard aregathered together in chips . These are tiny electronic circuitswhich are crammed with transistors. • They provide IR port, CNR slot, IEEE 1394, USB forattaching emerging high-speed serial devices. There are various components of a motherboard which fixedtogether into a single unit leads to the proper functioning of amotherboard.(1) Chips: The active devices on the motherboard are gathered together in chips . These are tiny electronic circuits which are crammed with transistors.(2)
Socket:
These are holders, which have been soldered to the motherboard. The sockets are built to exactly match a card or a chip. Plugs, Connectors, and Ports: The motherboard alsocontains a number of inputs and outputs, to which variousequipment can be connected. Most ports (also called I/O ports)can be seen where they end in a connector at the back of thePC. These are: • Ports for the keyboard and mouse.
7
• Serial ports, the parallel port, and USB ports.Sockets for speakers/microphone etc.
Motherboard Form Factors
A motherboard form factor just describes the dimensions orsize of the motherboard and what the layout of the motherboard components is. It
8
is important to understand thedifferent motherboard form factors, because you cannot takeany motherboard and place it in a computer case. You must putan ATX board in an ATX case.Since the beginning of PCs, the following types of motherboardForm Factors appeared in the market: • Active and Passive Backplane. • Full size AT • Mini or Baby AT • LPX • ATX • Mini-ATX • NLX Active and Passive Backplane: • In Passive Backplane, PC‗s major Motherboardcomponents such as processor, memory chips, supportcircuitry, etc. are not placed on a single board; ratherthey are placed on a expansion boards plugged intoslots of another board. • Allow easy upgrade of entire system. • Active Backplane contain all typical components foundon a typical motherboard except the Processor and thecomponents which are directly attached to theprocessor such as cache memory, system clock, etc. • Backplane systems didn‘t gain much popularity.
9
Active and Passive Backplane Full Size AT: • Released in 1984. • 12 Inches Wide and 13.8 Inches in length. • Can fit into full size tower or desktop. • Provision for 8 expansion slots. • No uniform standard followed in component layout. • Some older AT motherboards do not contain anyonboard I/O ports and contained I/O ports in the form of
10
Full Size AT Baby AT: • 8.66 inches wide and 13 inches long. • Compatible with almost every type of case. • Mounting in newer Baby-AT boards. • Problem full length expansion cards.
11
Baby AT LPX: • 8.66 inches wide and 13 inches long. • Developed for Slimline or Low Profile cases used by somebranded manufactures • Western Digital first introduced this form factor in some of their systems. • Single Slot motherboards and Riser cards. • Built in video and I/O connectors. • Limited Upgradability and poor cooling.
12
LPX ATX: • Released by Intel in 1995. • The latest Pentium 4 motherboards are based on ATXversion 2.3. • 9.6 inches wide and 12 inches long.
13
ATX Mini-ATX: • 11.2 inches in length and 8.2 inches wide. • More cost-effective. • Can fit into both ATX and mini-ATX cabinets
14
Mini-ATX NLX: • Quickly replaceable motherboards. • Larger size processors and memory modules. • Backplane flexibility. • Additional features such as built-in multimedia solutions,video playback, extended audio, etc
15
Chipset The Chipset is the glue that connects the microprocessor to therest of the motherboard and therefore to the rest of thecomputer. On a PC, it consists of two basic parts -- the Northbridge and the Southbridge. All
16
of the various components of the computer communicate with the CPU through the chipset.
Chipset History
• At one time, multiple, smaller controller chips performeddifferent types of functions. • There was a separate chip (often more than one) for eachfunction: controlling the cache, performing direct memoryaccess (DMA), handling interrupts, transferring data overthe I/O bus, etc.
17
• Over time these chips were integrated to form a single setof chips, or chipset that implements the various controlfeatures on the motherboard.
What is Chipset?
• A chipset is just a set of chips • Logic circuits that are the intelligence of themotherboard. • Controlling data transfers between the processor, cache,system buses, and peripherals—basically everythinginside the computer. • A highly integrated circuit used to perform a set of functions The term ―chipset‖ is also used to refer to the mainprocessing circuitry on many video cards.
Chipset Features and Functions
Chipset processor support: • A chipset is designed to work with certain set of processors. • Most chipsets support one ―class‖ or generation of processor. • Processor speed support is also controlled by chipsets asfaster processors require chipset control circuitry capableof handling them. • Some chipsets are capable of supporting more than oneprocessor. Chipset Cache support: • Chipset determines how much secondary cache (L2) issupported. • Determines secondary cache type support likeasynchronous, synchronous burst & pipeline burst Secondary cache write policy – Write back or Writethrough. • Controls the maximum amount of memory the systemcan cache. Chipset Memory Support :
18
• The chipset dictates the maximum amount of RAM thatyou can have on the system. • The chipset controls the type of RAM that can be used. Itdetermines whether our motherboard can have EDO,SDRAM, DDR SDRAM or RDRAM etc. • Error correction logic is provided as part of the memorycontrol circuits of the chipset. A chipset supports eitherparity, ECC or both. Some of the desktop does notsupport any of the above logic.
Chipset Timing and Flow control:
• One of the chipset‘s most important functions iscontrolling memory reads and writes, and transfers to thelocal bus (usually PCI and/or AGP) by the processor. • The chipset performs address decoding. • Cache and memory data transfer to and from theprocessor. • The chipset controls the flow information from the localbus (PCI) to memory as well as from PCI directly toprocessor. • Responsible for managing memory system timing –reducing the processor wait state and inserting wait stateswherever necessary to ensure that the processor is notgoing ahead of memory. • Auto detection of memory. Peripheral and I/O Bus control: • The chipset controls the various types of buses (PCI, AGP,ISA etc.) and transfers information to and from them andthe processor and memory. • The chipset dictates what types of buses the system cansupport.
19
• The chipset has bus bridges to connect together thedifferent bus types that it controls. • The chipset integrates the IDE/ATA Hard disk controller. The data transfer rate of the IDE devices depends on thechipset. • Direct Memory Access and Bus Mastering of PCI devices isprovided by the chipset. • The interrupt controller provides the means by which I/Odevices request attention from the processor to deal withdata transfers. This work is performed by Interruptcontrollers which are integrated in the chipset. • USB support (USB controller) is implemented as part of thechipset. • Plug and Play – is a specification which enables device tohave their system resource usage (IRQ, DMA) setautomatically. Power Management support: • Most recent chipsets support a group of features that worktogether to reduce the amount of power used by the PCduring idle periods. • There are a number of different protocols that workstogether to make power management work like EnergyStar, APM, DPMS, SMM, ACPI etc.
20
ARCHITECTURE OF CHIPSET
21
WORKING OF CHIPSET
22
Expansion Slots The most visible parts of any motherboard are the expansionslots . These look like small plastic slots, usually from 3 to 11inches long and approximately 1⁄2 inch wide. As their namesuggests, these slots are used to install various devices in thecomputer to expand its capabilities. Some expansion devicesthat might be installed in these slots include video, network,sound, and disk interface cards.If you look at the motherboard in your computer, you will morethan likely see one of the main types of expansion slots used incomputers today: • ISA • PCI • PCIe • AGP • AMR • CNR Each type differs in appearance and function. In this section,we will cover how to visually identify the different expansionslots on the motherboard. ISA Expansion Slots: If you have a computer made before 1997, chances are themotherboard has a few Industry Standard Architecture (ISA)expansion slots . They‘re easily recognizable because they areusually black and have two parts: one shorter and one longer.Computers made after 1997
23
generally include a few ISA slotsfor backward compatibility with old expansion cards
(althoughmost computers are phasing them out in favor of PCI).
ISA expansion slotsPCI Expansion Slots: Most computers made today contain primarily PeripheralComponent Interconnect (PCI) slots. They are easilyrecognizable because they are short (around 3 inches long) andusually white. PCI slots can usually be found in any computerthat has a Pentium-class processor or higher.
24
PCI Expansion SlotsPCIe Expansion Slots: The newest expansion slot architecture that is being used
bymotherboards is PCI Express (PCIe)
25
. It was designed to be areplacement for AGP and PCI. It has the capability of beingfaster than AGP while maintaining the flexibility of PCI. Andmotherboards with PCIe will have regular PCI slots for backwardcompatibility with PCI. PCIe Expansion Slots AGP Expansion Slots: Accelerated Graphics Port (AGP) slots are very popular for videocard use. In the past, if you wanted to use a high-speed,accelerated 3D graphics video card, you had to install the cardinto an existing PCI or ISA slot. AGP slots were designed to be adirect connection between the video circuitry and the PC‘smemory. They are also easily recognizable because they areusually brown, are located right next to the PCI slots on themotherboard, and are shorter than the PCI slots
AGP Expansion Slots There are seven different speed levels for PCIe, and theyare designated 1X, 2X, 4X, 8X, 12X, 16X, and 32X. Thesedesignations roughly correspond to similarly designated AGPspeeds. The slots for PCIe are a
26
bit harder to identify than other expansion slot types because the slot size corresponds to its speed. For example, the 1X slot is extremely short (less than an inch). The slots get longer in proportion to the speed; the longer the slot, the higher the speed. The reason for this stems from the PCIe concept of lanes, which are the multiplied units of communication between any two PCIe components and are directly related to physical wiring on the bus. Because all PCIe communications are made up of unidirectional coupling between devices, each PCIe card negotiates for the best mutually supported number of lanes with each communications partner. AMR Expansion Slots: As is always the case, Intel and other manufacturers are constantly looking for ways to improve the production process. One lengthy process that would often slow down the production of motherboards with integrated analog I/O functions was FCC certification. The manufacturers developed a way of separating the analog circuitry, for example, modem and analog audio, onto its own card. This allowed the analog circuitry to be separately certified (it was its own expansion card), thus reducing time for FCC certification. This slot and riser card technology was known as the Audio Modem Riser , or AMR . AMR‘s 46-pin slots were once fairlycommon on many Intel motherboards, but technologiesincluding CNR and Advanced
Communications Riser (ACR) areedging out AMR. In addition and
27
despite FCC concerns,integrated components still appear to be enjoying the mostsuccess comparatively
An AMR slotCNR Expansion Slots: The Communications and Networking Riser (CNR) slots that canbe found on some Intel motherboards are a replacement forIntel‘s AMR slots. Essentially, these 60-pin slots allow amotherboard manufacturer to implement a motherboardchipset with certain integrated features. Then, if the built-infeatures of that chipset need to be enhanced (by adding DolbyDigital Surround to a standard sound chipset, for example), a CNR riser card could be added to enhance the onboardcapabilities. Additional advantages of CNR over AMR includenetworking support, Plug and Play compatibility, support forhardware acceleration (as opposed to CPU control only), and noneed to lose a competing PCI slot unless the CNR slot is in use.
28
29
PROCESSOR
A processor is the logic circuitry that responds to andprocesses the basic instructions that drive a computer. Theterm processor has generally replaced the term centralprocessing unit (CPU). The processor in a personal computer orembedded in small devices is often called a microprocessor. The ―brain‖ of any computer is the central processing unit (CPU) . This component does all the calculations and performs90 percent of all the functions of a computer. • From the 2MHz Intel 4004 launched in 1971 to the mindboggling 2 GHz Pentium 4 in 2002 from the samecompany-the Microprocessor technology has comeacross a long way over these thirty years. • A microprocessor is an integrated circuit (IC) thatcontains a complete CPU on a single chip. • All the processors are backward compatible with 8086,and therefore appropriately called as x86 processors. • The original IBM PC design was based on CPUsincorporating CISC architecture while its nearest rival,the Apple Macs were designed on Motorola 680x0 andIBM PowerPC processors features featuring RISC architecture • Though initially all x86 processors came with CISCarchitecture, present day x86 processor architecture isactually a perfect blending of RISC and CISC.
30
WORKING OF CPU
The CPU is centrally located on the motherboard. Since the CPUcarries out a large share of the work in the computer, data pass continually through it. The data come from the RAM and the units (keyboard, drives etc.). After processing, the data is sendback to RAM and the units. The CPU continually receives instructions to be executed. Each instruction is a data processing order. The work itself consists mostly of calculations and data transport:
Data have a path to the CPU. It is kind of a data expresswaycalled the system bus.
CISC and RISC instructions CISC (Complex Instruction Set Computer):
31
The first CPUs had a so called Complex Instruction Set Computer (CISC) .This means that the computer canunderstand many and complex instructions. The X86 instruction set, with its varying length from 8 to 120 bit, was originallydeveloped for the 8086 with its mere 29000 transistors. RISC (Reduced Instruction Set Computer) : The RISC instructions are brief and the same length (forexample 32 bit long, as in Pentium Pro), and they process muchfaster than CISC instructions. Therefore, RISC is used in allnewer CPUs. However, the problem is that the instructionsarrive at the CPU in 8086 format. Thus, they must be decodedfor every new CPU generation; the instruction set has beenexpanded. The 386 came with 26 new instructions, the 486with 6 new instructions, and Pentium with 8 new instructions. These changes mean that some programs require at least a386 or a Pentium processor to work.
32
PROCESSOR GENERATION AND FAMILIES
Inside a Modern Processor • BIU • Internal Registers • L1 Cache • Control Unit • ALU • Integer Execution Unit • FPU • Branch Prediction Unit • Virtual Memory Support Processor Data Bus • There are data busses both inside and outside theprocessor. • Usually by the term Data Bus we mean External DataBus.
33
• Internal data bus width is largely determined by thewidths of the CPU Internal Registers. • Each wire in the data bus carries one bit information. Processor Address Bus • The Front Side Bus: FSB is also known as the Processor Bus, Memory Bus , or System Bus and
connects the CPU with the main memory and is used toconnect to other components within the computer. • The FSB can range from speeds of 66 MHz, 133 MHz, 100MHz, 266 MHz, 400 MHz, and up. • The FSB speed can generally be set either using the system BIOS or with jumpers located on the computer motherboard. • The address bus part of a host processor bus (FSB) or memory bus physically consists of a set of wires that carry the addressing
34
information used to select a memory or information used to select a memory or I/O port location to which the data is being retrieved by the CPU. • As with the data bus, each wire in the Address bus carries a single bit of address information. Backside Bus • The back side bus is a special bus that allowscommunication between the CPU and the l2 cache, whichis a device that offloads some specialized computing tasksto make the CPU operate more quickly. Nothing besidesthe cache and CPU are connected to the back side bus. Processor Control Bus • The control bus represents the different control signals such as Memory and I/O Device Read, Memory Write, DMA Hold, Interrupt, etc. • Some Lines are output only, some lines are input only. Processor numbers are based on a variety of features that may include the processor's underlying architecture, cache front side bus, and clock speed • Architecture: Basic design of a microprocessor. May include process technology and/or other architectural enhancements. • Cache (MB/KB): A temporary storage area for frequently accessed or recently accessed data. Having certain data stored in a cache speeds up the operation of
35
the computer. Cache size is measured in megabytes (MB) or kilobytes (KB). • Front Side Bus: The connecting path between the processor and other key components such as the memory controller hub. FSB speed is measured in GHz or MHz • Clock Speed: Speed of the processor's internal clock, which dictates how fast the processor can process data. Clock speed is usually measured in GHz (gigahertz, or billions of pulses per second). Hyper Threading (HT) Hyper Threading is a new technology in Intel P4 processor to increase the CPU performance to immense level. Hyper Threading means1 physical processor and 2 logical processor. Normally More than one processor systems will give a high performance than single processor. But introducing multiple processors in a system involves high cost. HT is a solution which is cost-effective and High-Performance. See the video of HYPER-THREADING to understand about this topic.
PROCESSOR SOCKETS
36
37
Processor Packaging • DIP • PGA • CPGA • SPGA • FC-PGA DIP • MCM • LCC • PLCC • PQFP • BGA • SEC
38
Standardized Sockets and Slots • Originally the purpose of providing a CPU socket on the motherboard was just to provide a place to insert a processor onto the motherboard.
39
• However over the last few years Intel and AMD, the two major processor makers in the PC world, have defined several socket interface standards for PC motherboards. • These are standardized Socket and Slot specificationsto be used
with matching processors that arespecifically designed for them. Inserting a CPU There are several types of CPU sockets available. Todayvirtually all desktop PCs come with some variation of the SECpackaging. Other CPUs are generally not worth upgrading andmay be one of two common types of package: • Low-insertion-force (LIF) • Zero-insertion-force (ZIF)
40
COMPUTER MEMORY
The term computer memory refers to any form of electrical storage device inside a computer. However, most often the term refers to fast, temporary forms of storage. If the processor needs to retrieve each and every piece of data from the hard disk drive, the speed of the processor will become considerable slow. But when the same piece of data is stored in the computer memory, the processor can access it more quickly. Most forms of memory are intended to store data temporarily.
Types of Computer Memory • Cache and registers • RAM • ROM • FLASH Memory
41
• EMBEDED Memory • OTPICAL Memory Cache: • To cache is to set something aside, or to store foranticipated use. • Caching, in PC terms, is the holding of a recently orfrequently used code or data in a special memorylocation for rapid retrieval. • Speed is everything when it comes to computers. • While working, the CPU is constantly requesting andusing information and • Executing code. The closer the necessary data is to theCPU, the faster the system can locate it and executethe operation. • High-speed memory chip generally used for caching is called static RAM (SRAM) L1 Cache: • Starting with the 486 chips, a cache has been includedon every CPU. This original on-board cache is known as the L1 (level 1) or internal cache. • All commands for the processor go through the cache. The cache stores a backlog of commands so that if await state is encountered, the CPU can continue to process using commands from the cache. • Caching will store any code that has been read and keep it available for • The CPU to use. This eliminates the need to wait for fetching of the data from DRAM.
42
• Registers: These are memory cells built right into the CPU thatcontain specific data needed by the CPU, particularly thearithmetic and logic unit (ALU).
L2 Cache: • Additional cache can be added to most computers, depending on the motherboard. This cache is mounted directly on the motherboard, outside the CPU. • The external cache is also called L2 (level 2) and is the same as L1, but larger. L2 can also (on some motherboards) be added or expanded. L3 Cache : • Tertiary (third) caches originated out of server technology, where high-end systems use more than a single processor. • Initially an L-3 cache memory chips were built directly into the North Bridge but now it comes built-in with the processor.
43
44
RAM • The capacitor can retain charge for a fraction of a second, so DRAM requires an entire circuitry to keep the capacitors charged. The process of recharging the capacitors is known as refreshing. • Each transaction between the CPU and memory is called a bus cycle • Memory controller handles the movement of data to and from the CPU and the system memory banks. The memory controller is also responsible for the integrity of the data as it is swapped in and out. o Parity o ECC Types of RAM • Static RAM (SRAM): • Dynamic RAM (DRAM): • FPM DRAM: • EDO DRAM: • SDRAM: • DDR SDRAM: • RDRAM: • DDR2: • VRAM: SRAM: SRAM memory requires no refresh at all, it will maintain its information so long as it has sufficient power to keep it. This is due to the fact that internally, the SRAM component is made up of flip-flop circuitry, which does not depend on refreshing.
45
DRAM: The Dynamic RAM like SDRAM too consists of transistors, but it has only a pair of transistor unlike four to six transistors in SRAM. Another difference is that, DRAM stores each bit of data in a separate capacitor within an integrated circuit. Now a capacitor leaks charge, because of which data is stored in a DRAM for on it a tiny fractions of seconds before getting lost. To overcome this problem, the DRAM needs to be refreshed periodically. Because of periodic refreshing it gets its name as ―dynamic‖ and hence called Dynamic RAM (DRAM). The DRAM is slower than SRAM, but the advantage is that it requires less power is also inexpensive. FPM DRAM: The Fast Page Mode DRAM (FPM DRAM) is slightly faster than conventional DRAM. This is because of the fact that FPM DRAM works by eliminating the need for a row address if data is located in the row previously accessed. It is sometimes called page mode memory.
46
EDO DRAM: Extended data-out DRAM (EDO RAM) is much faster version of DRAM. Unlike conventional DRAM which can only access one block of data at a time, EDO RAM usually start fetching the next block of memory as soon as it sends the previous block to the processor. It is about five percent faster than FPM. Maximum transfer rate to L2 cache is approximately 264Mbps.
SDRAM: This type of memory synchronizes its input and output signals with the incoming clock that is used in the system board. By doing so, data transactions can continually take place with each successive rising edge of the clock. SDRAM is about five percent faster than EDO RAM and its transfer rate to L2 cache is approximately 528 Mbps.
47
DDR RAM: Double Data Rate SDRAM also known as DDR RAM is just like SDRAM except that is higher bandwidth, meaning greater speed. This is achieved by transferring data on the up and down tick of clock cycle. Maximum transfer rate to L2 cache is approximately 1064 Mbps for 133 MHz
RDRAM: Rambus DRAM is a radical departure from the previous DRAM architecture. It was designed by Rambus and uses a special high-speed data bus called Rambus channel to transfer data between memory and processor. RDRAM memory chips works in parallel to achieve a data rate of 800 MHz or 1600 Mbps.
48
DDR2: DDR2 is the recent version of DDR SDRAM. It offers newfeatures and functions that enable higher clock and data rate operations. DDR2 transfer 64 bits of data twice every clock cycle. However, a drawback of DDR2 is that it is not compatible with the DDR SDRAM memory slots.
VRAM: Video RAM is a type of RAM which can be read from and written to at the same time. This is called dual-ported memory. On the other hand DRAM is single ported, which means that the memory can be written to and read from, but one at a time and not simultaneously. VRAM is most commonly used on video accelerator because it outperforms the other memory type by being dual ported. Flash Memory: Flash Memory is used to quickly store data from electronic devices such as digital cameras and MP3 players. Unlike DRAM or SRAM, data written to flash memory doesn't require power to maintain the stored contents. Embedded Memory: Embedded Memory is a small, dense format frequently used on electronic devices with small form factors, such as PDAs and cell phones. While limited in storage capacity when compared with DRAM modules used on personal computers, embedded memory plays a crucial role in many electronic devices due to its small size.
49
Optical Memory: In Optical Memory, data is stored on an optical medium (i.e., CD-ROM or DVD), and read with a laser beam. While not currently practical for use in computer processing, optical memory is an ideal solution for storing large quantities of data very inexpensively, and more importantly, transporting that data between computer devices. ROM • ROM is a nonvolatile memory that is installed by the vendor of the computer during the process of manufacturing the motherboard or secondary components that need to retain the code when the machine is turned off. • With the use of ROM, information that is required to start and run the computer cannot be lost or changed. • ROM is used extensively to program operation of computers, as well as devices like cameras and controls for fuel injectors in modern cars. • In Computers ROM generally holds instructions for performing owner on Self Test (POST) routine, and the BIOS information used to describe system configuration Types of ROMs • PROM • EPROM • EEPROM
50
PROM: A programmable read-only memory ( PROM) or field programmable read-only memory ( FPROM) is a form of digital memory. Such PROMs are used to store programs permanently. They are frequently seen in computer games or such products as electronic dictionaries, where PROMs for different languages can be substituted
EPROM: An EPROM , or erasable programmable read-only memory , is a type of computer memory chip that retains its data when its power supply is switched off. In other words, it is non-volatile normally used in electronic circuits. Once programmed, an EPROM can be erased only by exposing it to strong ultraviolet light.
51
EEPROM: EEPROM such as Flash memory (Electrically ErasableProgrammable Read-Only Memory) allow the entire ROM (or selected banks of the ROM) to be electrically erased (flashed back to zero) then written to without taking them out of the computer (camera, MP3 player, etc.).Flashing is much slower (milliseconds) than writing to RAM(nanoseconds) (or reading from any ROM).EPROMs are easily recognizable by the transparent window in the top of the package, through which the silicon chip can be seen, and which admits UV light during erasing.
Memory Packaging Memory is available in various physical packaging. Roughly in order of their appearance, the major types of packaging include: • DIP • SIPP • SIMM • DIMM • SODIMM • RIMM
52
Dual Inline Pin Package (DIP): • Early versions of RAM were installed as single chips,usually 1-bitwide DIP (dual inline package) • To upgrade or add memory, new chips had to beindividually installed on the motherboard (eight or ninechips per row—nine chips if using parity). This could bechallenging, because each chip had 16 wires that neededto be perfectly aligned before insertion into the base.
Single in-line Pin Package (SIPP): • One of the first module forms of DRAM, the SIPP is a printed circuit board with individual DRAM chips mounted on it. • A few 80286-based computers used (often non-standard) memory modules like SIPP memory (single in-line pin package). • SIPP's 30 pins often bent or broke during installation,which is why they were quickly replaced by SIMMswhich used contact plates rather than pins.
53
Single Inline Memory Module (SIMM): • Used in personal computers. • A 30-pin. The first variant of SIMMs has 30 pins and provides 8 bits of data (9 bits in parity versions). • SIMMs have 30 contacts in a single row along the lower edge • A 30-pin SIMM can have as few as two or as many asnine individual DRAM chips. • The second variant of SIMMs—also called PS/2 has 72pins and provides 32 bits of data (36 bits in parity versions).
54
Dual Inline Memory Module (DIMM): • A DIMM or dual in-line memory module comprises a series of random access memory integrated circuits. • These modules are mounted on a printed circuit board and designed for use in personal computers. •DIMMs began to replace SIMMs (single in-line memory modules) as the predominant type of memory module as Intel's Pentium processors began to control the market. • The main difference between SIMMs and DIMMs is that SIMMs has a 32-bit data path, while DIMMs have a 64-bitdata path.
55
Small Outline DIMM (SODIMM): • Small outline DIMM (SO-DIMM). Smaller version of theDIMM, used in laptops. Comes in versions with 72 (32 bit),144 (64 bit), 200 (72 bit) pins. • SO-DIMMs are a smaller alternative to a DIMM, beingroughly half the size of regular DIMMs. As a result SO-DIMMs are mainly utilized in laptops.
Rambus Inline Memory Module (RIMM): • Direct Rambus DRAM or DRDRAM (sometimes justcalled
Rambus DRAM
or RDRAM ) is a type of synchronous dynamic
RAM, created by the RambusCorporation. • The first PC motherboards with support for RDRAMdebuted in 1999. They supported PC800 RDRAM, whichoperated at 800 MHz over a 16 bit bus using a 184 pinRIMM form factor. • High speed 1066, 800, 711 and 600 MHz RDRAM storage • Overheating causes the problem of hanging.
56
Memory Allocation • How the memory is allocated for use by the CPU is called memory mapping • It uses hexadecimal addresses to define ranges of memory. • The original processors developed by Intel were unable to use more than 1 MB of RAM, and the original IBM PC allowed only the first 640 KB of memory for direct use. • MS-DOS applications were written to conform to this limitation. • The first 640 KB was reserved for the operating system and applications (designated as conventional memory). The remaining 384 KB of RAM (designated as upper memory) was earmarked for running the computer'sown housekeeping needs (BIOS, video RAM, ROM, and so on) Memory Mapping: • Conventional (base) memory • Upper memory area (UMA) • Expanded memory( become obsolete nowadays) • High memory area (HMA)
57
• Extended memory (XMS)
Conventional Memory: • Conventional memory is the amount of RAM, typically640 KB, addressable by an IBM PC or compatiblemachine operating in real mode.
58
• Conventional memory is located in the area between 0and 640 KB. Without the use of special techniques,conventional memory • is the only kind of RAM accessible in DOS mode andDOS mode programs. Upper Memory Area: • The upper memory area (UMA), the memory block from640 KB to 1024 KB, is designated for hardware use, such as video RAM, BIOS, and memory-mapped hardware drivers that are loaded into high memory. Expanded Memory: • Expanded memory can provide up to 32 MB of additional memory, and because it is loaded from a 64-KB section, it is below the 1-MB limit and therefore recognizable by MS-DOS. • MS-DOS applications must be specifically written to take advantage of expanded memory. • 80386 and newer processors can emulate expanded memory by using memory managers such asEMM386.EXE and HIMEM.SYS. High Memory Area: • An irregularity was found in the Intel chip architecture that allowed MS-DOS to address the first 64 KB of extended memory on machines with 80286 or faster processors. This special area is called the high memory area (HMA). • A software driver called an A20 handler must be run toallow the processor to access the HMA.
59
Extended Memory Specification (XMS): • RAM above the 1-MB address is called extended memory • With the introduction of the 80286 processor, memorywas addressable up to 16 MB. Starting with the memory was addressable up to 4GB. • Extended memory is Accessed through an extended memory manager(HIMEM.SYS). Shadow RAM: • Many high-speed expansion boards use shadow RAM toimprove the performance of a computer. Shadow RAM rewrites (or shadows) the contents of the ROM BIOSand/or video BIOS into extended RAM (between the640-KB boundary and 1 MB). • This allows systems to operate faster when applicationsoftware calls any BIOS routines. In some cases, systemspeed can be increased up to 400 percent. 80386DX processor,
60
Memory Errors: Detection and Correction Almost all computers check every memory bit at systemstartup to determine whether there are any memory errors. There are two methods available to detect and correct thememory errors. These are: • Parity checking • Error-correction code (ECC) Parity Checking: In parity checking, the computer manufactures add an extra bitto each byte of memory. This additional bit is known as a paritybit and is used to check the integrity of data contained in theRAM. A parity bit can be either odd or even.An even parity bit is set to 1 if the total count of 1s in the givenset of bits is odd (making total count of 1s including parity bitto even). If the total count of 1s is even, the even parity bit isset to 0. On the other hand, the odd parity bit is set to 1 if thecount of 1s in the
61
given set of bit is even(making total count of 1s including parity bit to odd). If the total count of 1s is odd, theodd parity bit is set to 0.Whether it is an odd or even parity bit, the algorithm todetermine the memory error is same. The parity bit checks thatif a given byte of memory has the right number of binary zerosand ones in it. If the count changes, your computer knows anerror occurred.However, remember that parity checking is only an errordetection mechanism and not an error correction mechanism.And because of not having any error correction mechanism, theparity checking is rarely used in modern day memory modules. Error-Correction Code (ECC): A more sophisticated mechanism of error detection and also to remove such memory errors is the error-correction code (ECC)mechanism. The ECC mechanism works in conjunction with the memory controller. The ECC mechanism functions in the following way: • When a data is stored in memory, the controller using ECC mechanism calculates a code that describes the bit sequence of the data to be stored in the memory. This code is stored along with the unit of data. • When the unit of data is requested for reading, the memory controller again calculates the code for about-to-be-read data using the original algorithm. • The memory controller then compares the newly generated code with the code generated when the data was stored. • If the codes match, the data is free of errors and is sent.
62
• If the codes don‘t match, there is an error in the data that is being read. The memory controller then determines the missing or erroneous bits through the code comparison. • The memory controller then corrects the erroneous data using some kind of error correction algorithm built into ECC. The ECC scheme requires extra bits per byte of storage to store the code that contains the bit sequence of data to be stored or to be read. ECC memory requires five extra bits to protect an 8-bit byte, six to protect a 16-bit word, seven to protect a 32-bitword, and eight to protect a 64-bit word.
63
BIBLIOGRAPHY WEBSITES : www.google.com www.en.wikipedia.org www.motherboards.com www.rams.com
---- THANKYOU
64
“MOTHERBOARD”
SUBMITTED BY MD. ASHAB AHMAD ROLL NO. : L1/CHIP.3.1/JAN12/1001 COURSE : CHIP. 3.1 Submitted to : ANKUR AWASTHI (Teacher)
COMPUTER CONSULTANCY LTD. LUCKNOW
1
ACKNOWLEDGEMENT
I would like to express my gratitude towards Mr. Ankur Awasthi who gave me the opportunity to work. I would like to thanks to my UPTECH, COMPUTER CONSULTANCY LTD. for their guidance and all other members for their support. I thank to my God, my parents and all my friends who help me lot in project.
MD. ASHAB AHMAD ROLL NO. : L1/CHIP.3.1/JAN12/1001
2
TABLE OF CONTENT
MOTHERBOARD COMPONENTS OF A MOTHERBOARD ARCHITECTURE OF CHIPSET WORKING OF CHIPSET PROCESSOR WORKING OF CPU PROCESSOR GENERATION AND FAMILIES COMPUTER MEMORY BIBLIOGRAPHY
3
MOTHERBOARD
A motherboard is the main printed circuit board in a computerthat contains the central processing unit, appropriatecoprocessor and support chips, device controllers, memory, andalso expansion slots to give access to the computer‘s internalbus. The motherboard is the PCs center of activity. All devicesin a computer are in some way connected to the motherboard.
4
After CPU, motherboard is the second most important component in the system and therefore, it definitely needs special attention. • The design of a motherboard is dependent on the typeof CPU and mainly oriented around the chipset presentonboard. • The motherboard features decide the systemperformance to a great extent, upgrade potential, etc. • As there are numerous types of processors, there aredifferent types of motherboards as well. • The classification is usually done on the basis of typesof CPU sockets carried by these boards.
Functions of Motherboard
• Motherboard provides a substrate upon which othercomponents of a system such as CPU, RAM, ROM, Chipset,and Expansion Slots can reside. • Motherboard provides the electrical connection betweenvarious components in the system. • Motherboard provides interface for various add-on cardssuch as 3D graphics cards, NIC, Sound cards, etc. throughvarious expansion slots such as PCI, ISA, AGP, etc. • Motherboard provides the necessary interface with a hostof I/O devices. E.g., on-board IDE or SCSI interface forhard-disks, CD-ROM drives etc. • They also provide other traditional I/O connectors such
asPS/2interface, RS232 serial COM ports, Bi-directionalparallel port
5
(LPT), Joystick connection through game port,floppy disk interface, etc. • The battery driven RTC chip on the motherboard storesCMOS setup information.
• They provide IR port, CNR slot, IEEE 1394, USB forattaching emerging high-speed serial devices.
6
COMPONENTS OF A MOTHERBOARD
There are various components of a motherboard which fixedtogether into a single unit leads to the proper functioning of amotherboard.(1) Chips: The active devices on the motherboard aregathered together in chips . These are tiny electronic circuitswhich are crammed with transistors. • They provide IR port, CNR slot, IEEE 1394, USB forattaching emerging high-speed serial devices. There are various components of a motherboard which fixedtogether into a single unit leads to the proper functioning of amotherboard.(1) Chips: The active devices on the motherboard are gathered together in chips . These are tiny electronic circuits which are crammed with transistors.(2)
Socket:
These are holders, which have been soldered to the motherboard. The sockets are built to exactly match a card or a chip. Plugs, Connectors, and Ports: The motherboard alsocontains a number of inputs and outputs, to which variousequipment can be connected. Most ports (also called I/O ports)can be seen where they end in a connector at the back of thePC. These are: • Ports for the keyboard and mouse.
7
• Serial ports, the parallel port, and USB ports.Sockets for speakers/microphone etc.
Motherboard Form Factors
A motherboard form factor just describes the dimensions orsize of the motherboard and what the layout of the motherboard components is. It
8
is important to understand thedifferent motherboard form factors, because you cannot takeany motherboard and place it in a computer case. You must putan ATX board in an ATX case.Since the beginning of PCs, the following types of motherboardForm Factors appeared in the market: • Active and Passive Backplane. • Full size AT • Mini or Baby AT • LPX • ATX • Mini-ATX • NLX Active and Passive Backplane: • In Passive Backplane, PC‗s major Motherboardcomponents such as processor, memory chips, supportcircuitry, etc. are not placed on a single board; ratherthey are placed on a expansion boards plugged intoslots of another board. • Allow easy upgrade of entire system. • Active Backplane contain all typical components foundon a typical motherboard except the Processor and thecomponents which are directly attached to theprocessor such as cache memory, system clock, etc. • Backplane systems didn‘t gain much popularity.
9
Active and Passive Backplane Full Size AT: • Released in 1984. • 12 Inches Wide and 13.8 Inches in length. • Can fit into full size tower or desktop. • Provision for 8 expansion slots. • No uniform standard followed in component layout. • Some older AT motherboards do not contain anyonboard I/O ports and contained I/O ports in the form of
10
Full Size AT Baby AT: • 8.66 inches wide and 13 inches long. • Compatible with almost every type of case. • Mounting in newer Baby-AT boards. • Problem full length expansion cards.
11
Baby AT LPX: • 8.66 inches wide and 13 inches long. • Developed for Slimline or Low Profile cases used by somebranded manufactures • Western Digital first introduced this form factor in some of their systems. • Single Slot motherboards and Riser cards. • Built in video and I/O connectors. • Limited Upgradability and poor cooling.
12
LPX ATX: • Released by Intel in 1995. • The latest Pentium 4 motherboards are based on ATXversion 2.3. • 9.6 inches wide and 12 inches long.
13
ATX Mini-ATX: • 11.2 inches in length and 8.2 inches wide. • More cost-effective. • Can fit into both ATX and mini-ATX cabinets
14
Mini-ATX NLX: • Quickly replaceable motherboards. • Larger size processors and memory modules. • Backplane flexibility. • Additional features such as built-in multimedia solutions,video playback, extended audio, etc
15
Chipset The Chipset is the glue that connects the microprocessor to therest of the motherboard and therefore to the rest of thecomputer. On a PC, it consists of two basic parts -- the Northbridge and the Southbridge. All
16
of the various components of the computer communicate with the CPU through the chipset.
Chipset History
• At one time, multiple, smaller controller chips performeddifferent types of functions. • There was a separate chip (often more than one) for eachfunction: controlling the cache, performing direct memoryaccess (DMA), handling interrupts, transferring data overthe I/O bus, etc.
17
• Over time these chips were integrated to form a single setof chips, or chipset that implements the various controlfeatures on the motherboard.
What is Chipset?
• A chipset is just a set of chips • Logic circuits that are the intelligence of themotherboard. • Controlling data transfers between the processor, cache,system buses, and peripherals—basically everythinginside the computer. • A highly integrated circuit used to perform a set of functions The term ―chipset‖ is also used to refer to the mainprocessing circuitry on many video cards.
Chipset Features and Functions
Chipset processor support: • A chipset is designed to work with certain set of processors. • Most chipsets support one ―class‖ or generation of processor. • Processor speed support is also controlled by chipsets asfaster processors require chipset control circuitry capableof handling them. • Some chipsets are capable of supporting more than oneprocessor. Chipset Cache support: • Chipset determines how much secondary cache (L2) issupported. • Determines secondary cache type support likeasynchronous, synchronous burst & pipeline burst Secondary cache write policy – Write back or Writethrough. • Controls the maximum amount of memory the systemcan cache. Chipset Memory Support :
18
• The chipset dictates the maximum amount of RAM thatyou can have on the system. • The chipset controls the type of RAM that can be used. Itdetermines whether our motherboard can have EDO,SDRAM, DDR SDRAM or RDRAM etc. • Error correction logic is provided as part of the memorycontrol circuits of the chipset. A chipset supports eitherparity, ECC or both. Some of the desktop does notsupport any of the above logic.
Chipset Timing and Flow control:
• One of the chipset‘s most important functions iscontrolling memory reads and writes, and transfers to thelocal bus (usually PCI and/or AGP) by the processor. • The chipset performs address decoding. • Cache and memory data transfer to and from theprocessor. • The chipset controls the flow information from the localbus (PCI) to memory as well as from PCI directly toprocessor. • Responsible for managing memory system timing –reducing the processor wait state and inserting wait stateswherever necessary to ensure that the processor is notgoing ahead of memory. • Auto detection of memory. Peripheral and I/O Bus control: • The chipset controls the various types of buses (PCI, AGP,ISA etc.) and transfers information to and from them andthe processor and memory. • The chipset dictates what types of buses the system cansupport.
19
• The chipset has bus bridges to connect together thedifferent bus types that it controls. • The chipset integrates the IDE/ATA Hard disk controller. The data transfer rate of the IDE devices depends on thechipset. • Direct Memory Access and Bus Mastering of PCI devices isprovided by the chipset. • The interrupt controller provides the means by which I/Odevices request attention from the processor to deal withdata transfers. This work is performed by Interruptcontrollers which are integrated in the chipset. • USB support (USB controller) is implemented as part of thechipset. • Plug and Play – is a specification which enables device tohave their system resource usage (IRQ, DMA) setautomatically. Power Management support: • Most recent chipsets support a group of features that worktogether to reduce the amount of power used by the PCduring idle periods. • There are a number of different protocols that workstogether to make power management work like EnergyStar, APM, DPMS, SMM, ACPI etc.
20
ARCHITECTURE OF CHIPSET
21
WORKING OF CHIPSET
22
Expansion Slots The most visible parts of any motherboard are the expansionslots . These look like small plastic slots, usually from 3 to 11inches long and approximately 1⁄2 inch wide. As their namesuggests, these slots are used to install various devices in thecomputer to expand its capabilities. Some expansion devicesthat might be installed in these slots include video, network,sound, and disk interface cards.If you look at the motherboard in your computer, you will morethan likely see one of the main types of expansion slots used incomputers today: • ISA • PCI • PCIe • AGP • AMR • CNR Each type differs in appearance and function. In this section,we will cover how to visually identify the different expansionslots on the motherboard. ISA Expansion Slots: If you have a computer made before 1997, chances are themotherboard has a few Industry Standard Architecture (ISA)expansion slots . They‘re easily recognizable because they areusually black and have two parts: one shorter and one longer.Computers made after 1997
23
generally include a few ISA slotsfor backward compatibility with old expansion cards
(althoughmost computers are phasing them out in favor of PCI).
ISA expansion slotsPCI Expansion Slots: Most computers made today contain primarily PeripheralComponent Interconnect (PCI) slots. They are easilyrecognizable because they are short (around 3 inches long) andusually white. PCI slots can usually be found in any computerthat has a Pentium-class processor or higher.
24
PCI Expansion SlotsPCIe Expansion Slots: The newest expansion slot architecture that is being used
bymotherboards is PCI Express (PCIe)
25
. It was designed to be areplacement for AGP and PCI. It has the capability of beingfaster than AGP while maintaining the flexibility of PCI. Andmotherboards with PCIe will have regular PCI slots for backwardcompatibility with PCI. PCIe Expansion Slots AGP Expansion Slots: Accelerated Graphics Port (AGP) slots are very popular for videocard use. In the past, if you wanted to use a high-speed,accelerated 3D graphics video card, you had to install the cardinto an existing PCI or ISA slot. AGP slots were designed to be adirect connection between the video circuitry and the PC‘smemory. They are also easily recognizable because they areusually brown, are located right next to the PCI slots on themotherboard, and are shorter than the PCI slots
AGP Expansion Slots There are seven different speed levels for PCIe, and theyare designated 1X, 2X, 4X, 8X, 12X, 16X, and 32X. Thesedesignations roughly correspond to similarly designated AGPspeeds. The slots for PCIe are a
26
bit harder to identify than other expansion slot types because the slot size corresponds to its speed. For example, the 1X slot is extremely short (less than an inch). The slots get longer in proportion to the speed; the longer the slot, the higher the speed. The reason for this stems from the PCIe concept of lanes, which are the multiplied units of communication between any two PCIe components and are directly related to physical wiring on the bus. Because all PCIe communications are made up of unidirectional coupling between devices, each PCIe card negotiates for the best mutually supported number of lanes with each communications partner. AMR Expansion Slots: As is always the case, Intel and other manufacturers are constantly looking for ways to improve the production process. One lengthy process that would often slow down the production of motherboards with integrated analog I/O functions was FCC certification. The manufacturers developed a way of separating the analog circuitry, for example, modem and analog audio, onto its own card. This allowed the analog circuitry to be separately certified (it was its own expansion card), thus reducing time for FCC certification. This slot and riser card technology was known as the Audio Modem Riser , or AMR . AMR‘s 46-pin slots were once fairlycommon on many Intel motherboards, but technologiesincluding CNR and Advanced
Communications Riser (ACR) areedging out AMR. In addition and
27
despite FCC concerns,integrated components still appear to be enjoying the mostsuccess comparatively
An AMR slotCNR Expansion Slots: The Communications and Networking Riser (CNR) slots that canbe found on some Intel motherboards are a replacement forIntel‘s AMR slots. Essentially, these 60-pin slots allow amotherboard manufacturer to implement a motherboardchipset with certain integrated features. Then, if the built-infeatures of that chipset need to be enhanced (by adding DolbyDigital Surround to a standard sound chipset, for example), a CNR riser card could be added to enhance the onboardcapabilities. Additional advantages of CNR over AMR includenetworking support, Plug and Play compatibility, support forhardware acceleration (as opposed to CPU control only), and noneed to lose a competing PCI slot unless the CNR slot is in use.
28
29
PROCESSOR
A processor is the logic circuitry that responds to andprocesses the basic instructions that drive a computer. Theterm processor has generally replaced the term centralprocessing unit (CPU). The processor in a personal computer orembedded in small devices is often called a microprocessor. The ―brain‖ of any computer is the central processing unit (CPU) . This component does all the calculations and performs90 percent of all the functions of a computer. • From the 2MHz Intel 4004 launched in 1971 to the mindboggling 2 GHz Pentium 4 in 2002 from the samecompany-the Microprocessor technology has comeacross a long way over these thirty years. • A microprocessor is an integrated circuit (IC) thatcontains a complete CPU on a single chip. • All the processors are backward compatible with 8086,and therefore appropriately called as x86 processors. • The original IBM PC design was based on CPUsincorporating CISC architecture while its nearest rival,the Apple Macs were designed on Motorola 680x0 andIBM PowerPC processors features featuring RISC architecture • Though initially all x86 processors came with CISCarchitecture, present day x86 processor architecture isactually a perfect blending of RISC and CISC.
30
WORKING OF CPU
The CPU is centrally located on the motherboard. Since the CPUcarries out a large share of the work in the computer, data pass continually through it. The data come from the RAM and the units (keyboard, drives etc.). After processing, the data is sendback to RAM and the units. The CPU continually receives instructions to be executed. Each instruction is a data processing order. The work itself consists mostly of calculations and data transport:
Data have a path to the CPU. It is kind of a data expresswaycalled the system bus.
CISC and RISC instructions CISC (Complex Instruction Set Computer):
31
The first CPUs had a so called Complex Instruction Set Computer (CISC) .This means that the computer canunderstand many and complex instructions. The X86 instruction set, with its varying length from 8 to 120 bit, was originallydeveloped for the 8086 with its mere 29000 transistors. RISC (Reduced Instruction Set Computer) : The RISC instructions are brief and the same length (forexample 32 bit long, as in Pentium Pro), and they process muchfaster than CISC instructions. Therefore, RISC is used in allnewer CPUs. However, the problem is that the instructionsarrive at the CPU in 8086 format. Thus, they must be decodedfor every new CPU generation; the instruction set has beenexpanded. The 386 came with 26 new instructions, the 486with 6 new instructions, and Pentium with 8 new instructions. These changes mean that some programs require at least a386 or a Pentium processor to work.
32
PROCESSOR GENERATION AND FAMILIES
Inside a Modern Processor • BIU • Internal Registers • L1 Cache • Control Unit • ALU • Integer Execution Unit • FPU • Branch Prediction Unit • Virtual Memory Support Processor Data Bus • There are data busses both inside and outside theprocessor. • Usually by the term Data Bus we mean External DataBus.
33
• Internal data bus width is largely determined by thewidths of the CPU Internal Registers. • Each wire in the data bus carries one bit information. Processor Address Bus • The Front Side Bus: FSB is also known as the Processor Bus, Memory Bus , or System Bus and
connects the CPU with the main memory and is used toconnect to other components within the computer. • The FSB can range from speeds of 66 MHz, 133 MHz, 100MHz, 266 MHz, 400 MHz, and up. • The FSB speed can generally be set either using the system BIOS or with jumpers located on the computer motherboard. • The address bus part of a host processor bus (FSB) or memory bus physically consists of a set of wires that carry the addressing
34
information used to select a memory or information used to select a memory or I/O port location to which the data is being retrieved by the CPU. • As with the data bus, each wire in the Address bus carries a single bit of address information. Backside Bus • The back side bus is a special bus that allowscommunication between the CPU and the l2 cache, whichis a device that offloads some specialized computing tasksto make the CPU operate more quickly. Nothing besidesthe cache and CPU are connected to the back side bus. Processor Control Bus • The control bus represents the different control signals such as Memory and I/O Device Read, Memory Write, DMA Hold, Interrupt, etc. • Some Lines are output only, some lines are input only. Processor numbers are based on a variety of features that may include the processor's underlying architecture, cache front side bus, and clock speed • Architecture: Basic design of a microprocessor. May include process technology and/or other architectural enhancements. • Cache (MB/KB): A temporary storage area for frequently accessed or recently accessed data. Having certain data stored in a cache speeds up the operation of
35
the computer. Cache size is measured in megabytes (MB) or kilobytes (KB). • Front Side Bus: The connecting path between the processor and other key components such as the memory controller hub. FSB speed is measured in GHz or MHz • Clock Speed: Speed of the processor's internal clock, which dictates how fast the processor can process data. Clock speed is usually measured in GHz (gigahertz, or billions of pulses per second). Hyper Threading (HT) Hyper Threading is a new technology in Intel P4 processor to increase the CPU performance to immense level. Hyper Threading means1 physical processor and 2 logical processor. Normally More than one processor systems will give a high performance than single processor. But introducing multiple processors in a system involves high cost. HT is a solution which is cost-effective and High-Performance. See the video of HYPER-THREADING to understand about this topic.
PROCESSOR SOCKETS
36
37
Processor Packaging • DIP • PGA • CPGA • SPGA • FC-PGA DIP • MCM • LCC • PLCC • PQFP • BGA • SEC
38
Standardized Sockets and Slots • Originally the purpose of providing a CPU socket on the motherboard was just to provide a place to insert a processor onto the motherboard.
39
• However over the last few years Intel and AMD, the two major processor makers in the PC world, have defined several socket interface standards for PC motherboards. • These are standardized Socket and Slot specificationsto be used
with matching processors that arespecifically designed for them. Inserting a CPU There are several types of CPU sockets available. Todayvirtually all desktop PCs come with some variation of the SECpackaging. Other CPUs are generally not worth upgrading andmay be one of two common types of package: • Low-insertion-force (LIF) • Zero-insertion-force (ZIF)
40
COMPUTER MEMORY
The term computer memory refers to any form of electrical storage device inside a computer. However, most often the term refers to fast, temporary forms of storage. If the processor needs to retrieve each and every piece of data from the hard disk drive, the speed of the processor will become considerable slow. But when the same piece of data is stored in the computer memory, the processor can access it more quickly. Most forms of memory are intended to store data temporarily.
Types of Computer Memory • Cache and registers • RAM • ROM • FLASH Memory
41
• EMBEDED Memory • OTPICAL Memory Cache: • To cache is to set something aside, or to store foranticipated use. • Caching, in PC terms, is the holding of a recently orfrequently used code or data in a special memorylocation for rapid retrieval. • Speed is everything when it comes to computers. • While working, the CPU is constantly requesting andusing information and • Executing code. The closer the necessary data is to theCPU, the faster the system can locate it and executethe operation. • High-speed memory chip generally used for caching is called static RAM (SRAM) L1 Cache: • Starting with the 486 chips, a cache has been includedon every CPU. This original on-board cache is known as the L1 (level 1) or internal cache. • All commands for the processor go through the cache. The cache stores a backlog of commands so that if await state is encountered, the CPU can continue to process using commands from the cache. • Caching will store any code that has been read and keep it available for • The CPU to use. This eliminates the need to wait for fetching of the data from DRAM.
42
• Registers: These are memory cells built right into the CPU thatcontain specific data needed by the CPU, particularly thearithmetic and logic unit (ALU).
L2 Cache: • Additional cache can be added to most computers, depending on the motherboard. This cache is mounted directly on the motherboard, outside the CPU. • The external cache is also called L2 (level 2) and is the same as L1, but larger. L2 can also (on some motherboards) be added or expanded. L3 Cache : • Tertiary (third) caches originated out of server technology, where high-end systems use more than a single processor. • Initially an L-3 cache memory chips were built directly into the North Bridge but now it comes built-in with the processor.
43
44
RAM • The capacitor can retain charge for a fraction of a second, so DRAM requires an entire circuitry to keep the capacitors charged. The process of recharging the capacitors is known as refreshing. • Each transaction between the CPU and memory is called a bus cycle • Memory controller handles the movement of data to and from the CPU and the system memory banks. The memory controller is also responsible for the integrity of the data as it is swapped in and out. o Parity o ECC Types of RAM • Static RAM (SRAM): • Dynamic RAM (DRAM): • FPM DRAM: • EDO DRAM: • SDRAM: • DDR SDRAM: • RDRAM: • DDR2: • VRAM: SRAM: SRAM memory requires no refresh at all, it will maintain its information so long as it has sufficient power to keep it. This is due to the fact that internally, the SRAM component is made up of flip-flop circuitry, which does not depend on refreshing.
45
DRAM: The Dynamic RAM like SDRAM too consists of transistors, but it has only a pair of transistor unlike four to six transistors in SRAM. Another difference is that, DRAM stores each bit of data in a separate capacitor within an integrated circuit. Now a capacitor leaks charge, because of which data is stored in a DRAM for on it a tiny fractions of seconds before getting lost. To overcome this problem, the DRAM needs to be refreshed periodically. Because of periodic refreshing it gets its name as ―dynamic‖ and hence called Dynamic RAM (DRAM). The DRAM is slower than SRAM, but the advantage is that it requires less power is also inexpensive. FPM DRAM: The Fast Page Mode DRAM (FPM DRAM) is slightly faster than conventional DRAM. This is because of the fact that FPM DRAM works by eliminating the need for a row address if data is located in the row previously accessed. It is sometimes called page mode memory.
46
EDO DRAM: Extended data-out DRAM (EDO RAM) is much faster version of DRAM. Unlike conventional DRAM which can only access one block of data at a time, EDO RAM usually start fetching the next block of memory as soon as it sends the previous block to the processor. It is about five percent faster than FPM. Maximum transfer rate to L2 cache is approximately 264Mbps.
SDRAM: This type of memory synchronizes its input and output signals with the incoming clock that is used in the system board. By doing so, data transactions can continually take place with each successive rising edge of the clock. SDRAM is about five percent faster than EDO RAM and its transfer rate to L2 cache is approximately 528 Mbps.
47
DDR RAM: Double Data Rate SDRAM also known as DDR RAM is just like SDRAM except that is higher bandwidth, meaning greater speed. This is achieved by transferring data on the up and down tick of clock cycle. Maximum transfer rate to L2 cache is approximately 1064 Mbps for 133 MHz
RDRAM: Rambus DRAM is a radical departure from the previous DRAM architecture. It was designed by Rambus and uses a special high-speed data bus called Rambus channel to transfer data between memory and processor. RDRAM memory chips works in parallel to achieve a data rate of 800 MHz or 1600 Mbps.
48
DDR2: DDR2 is the recent version of DDR SDRAM. It offers newfeatures and functions that enable higher clock and data rate operations. DDR2 transfer 64 bits of data twice every clock cycle. However, a drawback of DDR2 is that it is not compatible with the DDR SDRAM memory slots.
VRAM: Video RAM is a type of RAM which can be read from and written to at the same time. This is called dual-ported memory. On the other hand DRAM is single ported, which means that the memory can be written to and read from, but one at a time and not simultaneously. VRAM is most commonly used on video accelerator because it outperforms the other memory type by being dual ported. Flash Memory: Flash Memory is used to quickly store data from electronic devices such as digital cameras and MP3 players. Unlike DRAM or SRAM, data written to flash memory doesn't require power to maintain the stored contents. Embedded Memory: Embedded Memory is a small, dense format frequently used on electronic devices with small form factors, such as PDAs and cell phones. While limited in storage capacity when compared with DRAM modules used on personal computers, embedded memory plays a crucial role in many electronic devices due to its small size.
49
Optical Memory: In Optical Memory, data is stored on an optical medium (i.e., CD-ROM or DVD), and read with a laser beam. While not currently practical for use in computer processing, optical memory is an ideal solution for storing large quantities of data very inexpensively, and more importantly, transporting that data between computer devices. ROM • ROM is a nonvolatile memory that is installed by the vendor of the computer during the process of manufacturing the motherboard or secondary components that need to retain the code when the machine is turned off. • With the use of ROM, information that is required to start and run the computer cannot be lost or changed. • ROM is used extensively to program operation of computers, as well as devices like cameras and controls for fuel injectors in modern cars. • In Computers ROM generally holds instructions for performing owner on Self Test (POST) routine, and the BIOS information used to describe system configuration Types of ROMs • PROM • EPROM • EEPROM
50
PROM: A programmable read-only memory ( PROM) or field programmable read-only memory ( FPROM) is a form of digital memory. Such PROMs are used to store programs permanently. They are frequently seen in computer games or such products as electronic dictionaries, where PROMs for different languages can be substituted
EPROM: An EPROM , or erasable programmable read-only memory , is a type of computer memory chip that retains its data when its power supply is switched off. In other words, it is non-volatile normally used in electronic circuits. Once programmed, an EPROM can be erased only by exposing it to strong ultraviolet light.
51
EEPROM: EEPROM such as Flash memory (Electrically ErasableProgrammable Read-Only Memory) allow the entire ROM (or selected banks of the ROM) to be electrically erased (flashed back to zero) then written to without taking them out of the computer (camera, MP3 player, etc.).Flashing is much slower (milliseconds) than writing to RAM(nanoseconds) (or reading from any ROM).EPROMs are easily recognizable by the transparent window in the top of the package, through which the silicon chip can be seen, and which admits UV light during erasing.
Memory Packaging Memory is available in various physical packaging. Roughly in order of their appearance, the major types of packaging include: • DIP • SIPP • SIMM • DIMM • SODIMM • RIMM
52
Dual Inline Pin Package (DIP): • Early versions of RAM were installed as single chips,usually 1-bitwide DIP (dual inline package) • To upgrade or add memory, new chips had to beindividually installed on the motherboard (eight or ninechips per row—nine chips if using parity). This could bechallenging, because each chip had 16 wires that neededto be perfectly aligned before insertion into the base.
Single in-line Pin Package (SIPP): • One of the first module forms of DRAM, the SIPP is a printed circuit board with individual DRAM chips mounted on it. • A few 80286-based computers used (often non-standard) memory modules like SIPP memory (single in-line pin package). • SIPP's 30 pins often bent or broke during installation,which is why they were quickly replaced by SIMMswhich used contact plates rather than pins.
53
Single Inline Memory Module (SIMM): • Used in personal computers. • A 30-pin. The first variant of SIMMs has 30 pins and provides 8 bits of data (9 bits in parity versions). • SIMMs have 30 contacts in a single row along the lower edge • A 30-pin SIMM can have as few as two or as many asnine individual DRAM chips. • The second variant of SIMMs—also called PS/2 has 72pins and provides 32 bits of data (36 bits in parity versions).
54
Dual Inline Memory Module (DIMM): • A DIMM or dual in-line memory module comprises a series of random access memory integrated circuits. • These modules are mounted on a printed circuit board and designed for use in personal computers. •DIMMs began to replace SIMMs (single in-line memory modules) as the predominant type of memory module as Intel's Pentium processors began to control the market. • The main difference between SIMMs and DIMMs is that SIMMs has a 32-bit data path, while DIMMs have a 64-bitdata path.
55
Small Outline DIMM (SODIMM): • Small outline DIMM (SO-DIMM). Smaller version of theDIMM, used in laptops. Comes in versions with 72 (32 bit),144 (64 bit), 200 (72 bit) pins. • SO-DIMMs are a smaller alternative to a DIMM, beingroughly half the size of regular DIMMs. As a result SO-DIMMs are mainly utilized in laptops.
Rambus Inline Memory Module (RIMM): • Direct Rambus DRAM or DRDRAM (sometimes justcalled
Rambus DRAM
or RDRAM ) is a type of synchronous dynamic
RAM, created by the RambusCorporation. • The first PC motherboards with support for RDRAMdebuted in 1999. They supported PC800 RDRAM, whichoperated at 800 MHz over a 16 bit bus using a 184 pinRIMM form factor. • High speed 1066, 800, 711 and 600 MHz RDRAM storage • Overheating causes the problem of hanging.
56
Memory Allocation • How the memory is allocated for use by the CPU is called memory mapping • It uses hexadecimal addresses to define ranges of memory. • The original processors developed by Intel were unable to use more than 1 MB of RAM, and the original IBM PC allowed only the first 640 KB of memory for direct use. • MS-DOS applications were written to conform to this limitation. • The first 640 KB was reserved for the operating system and applications (designated as conventional memory). The remaining 384 KB of RAM (designated as upper memory) was earmarked for running the computer'sown housekeeping needs (BIOS, video RAM, ROM, and so on) Memory Mapping: • Conventional (base) memory • Upper memory area (UMA) • Expanded memory( become obsolete nowadays) • High memory area (HMA)
57
• Extended memory (XMS)
Conventional Memory: • Conventional memory is the amount of RAM, typically640 KB, addressable by an IBM PC or compatiblemachine operating in real mode.
58
• Conventional memory is located in the area between 0and 640 KB. Without the use of special techniques,conventional memory • is the only kind of RAM accessible in DOS mode andDOS mode programs. Upper Memory Area: • The upper memory area (UMA), the memory block from640 KB to 1024 KB, is designated for hardware use, such as video RAM, BIOS, and memory-mapped hardware drivers that are loaded into high memory. Expanded Memory: • Expanded memory can provide up to 32 MB of additional memory, and because it is loaded from a 64-KB section, it is below the 1-MB limit and therefore recognizable by MS-DOS. • MS-DOS applications must be specifically written to take advantage of expanded memory. • 80386 and newer processors can emulate expanded memory by using memory managers such asEMM386.EXE and HIMEM.SYS. High Memory Area: • An irregularity was found in the Intel chip architecture that allowed MS-DOS to address the first 64 KB of extended memory on machines with 80286 or faster processors. This special area is called the high memory area (HMA). • A software driver called an A20 handler must be run toallow the processor to access the HMA.
59
Extended Memory Specification (XMS): • RAM above the 1-MB address is called extended memory • With the introduction of the 80286 processor, memorywas addressable up to 16 MB. Starting with the memory was addressable up to 4GB. • Extended memory is Accessed through an extended memory manager(HIMEM.SYS). Shadow RAM: • Many high-speed expansion boards use shadow RAM toimprove the performance of a computer. Shadow RAM rewrites (or shadows) the contents of the ROM BIOSand/or video BIOS into extended RAM (between the640-KB boundary and 1 MB). • This allows systems to operate faster when applicationsoftware calls any BIOS routines. In some cases, systemspeed can be increased up to 400 percent. 80386DX processor,
60
Memory Errors: Detection and Correction Almost all computers check every memory bit at systemstartup to determine whether there are any memory errors. There are two methods available to detect and correct thememory errors. These are: • Parity checking • Error-correction code (ECC) Parity Checking: In parity checking, the computer manufactures add an extra bitto each byte of memory. This additional bit is known as a paritybit and is used to check the integrity of data contained in theRAM. A parity bit can be either odd or even.An even parity bit is set to 1 if the total count of 1s in the givenset of bits is odd (making total count of 1s including parity bitto even). If the total count of 1s is even, the even parity bit isset to 0. On the other hand, the odd parity bit is set to 1 if thecount of 1s in the
61
given set of bit is even(making total count of 1s including parity bit to odd). If the total count of 1s is odd, theodd parity bit is set to 0.Whether it is an odd or even parity bit, the algorithm todetermine the memory error is same. The parity bit checks thatif a given byte of memory has the right number of binary zerosand ones in it. If the count changes, your computer knows anerror occurred.However, remember that parity checking is only an errordetection mechanism and not an error correction mechanism.And because of not having any error correction mechanism, theparity checking is rarely used in modern day memory modules. Error-Correction Code (ECC): A more sophisticated mechanism of error detection and also to remove such memory errors is the error-correction code (ECC)mechanism. The ECC mechanism works in conjunction with the memory controller. The ECC mechanism functions in the following way: • When a data is stored in memory, the controller using ECC mechanism calculates a code that describes the bit sequence of the data to be stored in the memory. This code is stored along with the unit of data. • When the unit of data is requested for reading, the memory controller again calculates the code for about-to-be-read data using the original algorithm. • The memory controller then compares the newly generated code with the code generated when the data was stored. • If the codes match, the data is free of errors and is sent.
62
• If the codes don‘t match, there is an error in the data that is being read. The memory controller then determines the missing or erroneous bits through the code comparison. • The memory controller then corrects the erroneous data using some kind of error correction algorithm built into ECC. The ECC scheme requires extra bits per byte of storage to store the code that contains the bit sequence of data to be stored or to be read. ECC memory requires five extra bits to protect an 8-bit byte, six to protect a 16-bit word, seven to protect a 32-bitword, and eight to protect a 64-bit word.
63
BIBLIOGRAPHY WEBSITES : www.google.com www.en.wikipedia.org www.motherboards.com www.rams.com
---- THANKYOU
64
