Upgrading and Repairing PCs

Chipsets

We can't talk about modern motherboards without discussing chipsets. The chipset is the motherboard; therefore, any two boards with the same chipsets are functionally identical. The chipset contains the processor bus interface (called front-side bus, or FSB), memory controllers, bus controllers, I/O controllers, and more. All the circuits of the motherboard are contained within the chipset. If the processor in your PC is like the engine in your car, the chipset represents the chassis. It is the framework in which the engine rests and is its connection to the outside world. The chipset is the frame, suspension, steering, wheels and tires, transmission, driveshaft, differential, and brakes. The chassis in your car is what gets the power to the ground, allowing the vehicle to start, stop, and corner. In the PC, the chipset represents the connection between the processor and everything else. The processor can't talk to the memory, adapter boards, devices, and so on without going through the chipset. The chipset is the main hub and central nervous system of the PC. If you think of the processor as the brain, the chipset is the spine and central nervous system.

Because the chipset controls the interface or connections between the processor and everything else, the chipset ends up dictating which type of processor you have; how fast it will run; how fast the buses will run; the speed, type, and amount of memory you can use; and more. In fact, the chipset might be the single most important component in your system, possibly even more important than the processor. I've seen systems with faster processors be outperformed by systems with slower processor but a better chipset, much like how a car with less power might win a race through better cornering and braking. When deciding on a system, I start by choosing the chipset first because the chipset decision then dictates the processor, memory, I/O, and expansion capabilities.

Chipset Evolution

When IBM created the first PC motherboards, it used several discrete (separate) chips to complete the design. Besides the processor and optional math coprocessor, many other components were required to complete the system. These other components included items such as the clock generator, bus controller, system timer, interrupt and DMA controllers, CMOS RAM and clock, and keyboard controller. Additionally, many other simple logic chips were used to complete the entire motherboard circuit, plus, of course, things such as the actual processor, math coprocessor (floating-point unit), memory, and other parts. Table 4.8 lists all the primary chip components used on the original PC/XT and AT motherboards.

Table 4.8. Primary Chip Components on PC/XT and AT Motherboards

Chip Function	PC/XT Version	AT Version
Processor	8088	80286
Math Coprocessor (Floating-Point Unit)	8087	80287
Clock Generator	8284	82284
Bus Controller	8288	82288
System Timer	8253	8254
Low-order Interrupt Controller	8259	8259
High-order Interrupt Controller	—	8259
Low-order DMA Controller	8237	8237
High-order DMA Controller	—	8237
CMOS RAM/Real-Time Clock	—	MC146818
Keyboard Controller	8255	8042

In addition to the processor/coprocessor, a six-chip set was used to implement the primary motherboard circuit in the original PC and XT systems. IBM later upgraded this to a nine-chip design in the AT and later systems, mainly by adding more interrupt and DMA controller chips and the nonvolatile CMOS RAM/Real-time Clock chip. All these motherboard chip components came from Intel or an Intel-licensed manufacturer, except the CMOS/Clock chip, which came from Motorola. To build a clone or copy of one of these IBM systems required all these chips plus many smaller discrete logic chips to glue the design together, totaling 100 or more individual chips. This kept the price of a motherboard high and left little room on the board to integrate other functions.

In 1986, a company called Chips and Technologies introduced a revolutionary component called the 82C206—the main part of the first PC motherboard chipset. This was a single chip that integrated into it all the functions of the main motherboard chips in an AT-compatible system. This chip included the functions of the 82284 Clock Generator, 82288 Bus Controller, 8254 System Timer, dual 8259 Interrupt Controllers, dual 8237 DMA Controllers, and even the MC146818 CMOS/Clock chip. Besides the processor, virtually all the major chip components on a PC motherboard could now be replaced by a single chip. Four other chips augmented the 82C206 acting as buffers and memory controllers, thus completing virtually the entire motherboard circuit with five total chips. This first chipset was called the CS8220 chipset by Chips and Technologies. Needless to say, this was a revolutionary concept in PC motherboard manufacturing. Not only did it greatly reduce the cost of building a PC motherboard, but it also made designing a motherboard much easier. The reduced component count meant the boards had more room for integrating other items formerly found on expansion cards. Later, the four chips augmenting the 82C206 were replaced by a new set of only three chips, and the entire set was called the New Enhanced AT (NEAT) CS8221 chipset. This was later followed by the 82C836 Single Chip AT (SCAT) chipset, which finally condensed all the chips in the set down to a single chip.

The chipset idea was rapidly copied by other chip manufacturers. Companies such as Acer, Erso, Opti, Suntac, Symphony, UMC, and VLSI each gained an important share of this market. Unfortunately for many of them, the chipset market has been a volatile one, and many of them have long since gone out of business. In 1993, VLSI had become the dominant force in the chipset market and had the vast majority of the market share; by the next year, VLSI (which later was merged into Philips Semiconductors), along with virtually everybody else in the chipset market, was fighting to stay alive. This is because a new chipset manufacturer had come on the scene, and within a year or so of getting serious, it was totally dominating the chipset market. That company was Intel, and after 1994, it had a virtual lock on the chipset market. If you have a motherboard built since 1994 that uses or accepts an Intel processor, chances are good that it has an Intel chipset on it as well.

More recently, Intel has struggled somewhat with chipsets because of its reliance on RDRAM memory. Intel originally signed a contract with Rambus back in 1996 declaring it would support this memory as its primary focus for desktop PC chipsets through 2001. I suspect this has turned out to be something Intel regrets (the contract has since expired). RDRAM memory has had a significantly higher price than SDRAM memory—although the prices have recently come down a bit—and it does have some performance advantages when used in a dual-channel mode. There is a lot of momentum in the market for supporting double data rate (DDR) SDRAM. Consequently, Intel introduced the 845 chipset (code named Brookdale), which supports DDR-SDRAM with the Pentium 4. Intel's latest chipsets for Pentium 4, the i865 (code named Springdale) and i875P (code named Canterwood), continue the trend of support for faster DDR SDRAM memory and FSB bus speeds. Intel is not alone in the Pentium 4 chipset business: Silicon Integrated Systems (SiS), ATI, and ALi Corporation all make licensed chipsets for the Pentium 4. VIA Technologies also makes Pentium 4–compatible chipsets, but without a license, which has greatly limited their popularity with motherboard makers.

Although AMD has developed its own chipsets for the K6 and Athlon family of processors, it now emphasizes encouraging third-party chipset developers to support its products. Today, VIA Technologies is the leading developer of AMD Athlon/Athlon XP/Duron chipsets for both discrete and integrated uses. The popularity of AMD processors has also encouraged SiS, ATI, NVIDIA, and ALi Corporation to develop chipsets for both Intel- and AMD-based systems.

It is interesting to note that the original PC chipset maker, Chips and Technologies, survived by changing course to design and manufacture video chips and found a niche in that market specifically for laptop and notebook video chipsets. Chips and Technologies was subsequently bought out by Intel in 1998 as a part of Intel's video strategy.

Intel Chipsets

You can't talk about chipsets today without discussing Intel because it currently owns the vast majority of the chipset market. It is interesting to note that we probably have Compaq to thank for forcing Intel into the chipset business in the first place!

The thing that really started it all was the introduction of the EISA bus designed by Compaq in 1989. At that time, it had shared the bus with other manufacturers in an attempt to make it a market standard. However, Compaq refused to share its EISA bus chipset—a set of custom chips necessary to implement this bus on a motherboard.

Enter Intel, who decided to fill the chipset void for the rest of the PC manufacturers wanting to build EISA bus motherboards. As is well known today, the EISA bus failed to become a market success except for a short-term niche server business, but Intel now had a taste of the chipset business and this it apparently wouldn't forget. With the introduction of the 286 and 386 processors, Intel became impatient with how long it took the other chipset companies to create chipsets around its new processor designs; this delayed the introduction of motherboards that supported the new processors. For example, it took more than two years after the 286 processor was introduced for the first 286 motherboards to appear and just over a year for the first 386 motherboards to appear after the 386 had been introduced. Intel couldn't sell its processors in volume until other manufacturers made motherboards that would support them, so it thought that by developing motherboard chipsets for a new processor in parallel with the new processor, it could jumpstart the motherboard business by providing ready-made chipsets for the motherboard manufacturers to use.

Intel tested this by introducing the 420 series chipsets along with its 486 processor in April 1989. This enabled the motherboard companies to get busy right away, and in only a few months the first 486 motherboards appeared. Of course, the other chipset manufacturers weren't happy; now they had Intel as a competitor, and Intel would always have chipsets for new processors on the market first!

Intel then realized that it made both processors and chipsets, which were 90% of the components on a typical motherboard. What better way to ensure that motherboards were available for its Pentium processor when it was introduced than by making its own motherboards as well and having these boards ready on the new processor's introduction date. When the first Pentium processor debuted in 1993, Intel also debuted the 430LX chipset as well as a fully finished motherboard. Now, besides the chipset companies being upset, the motherboard companies weren't too happy, either. Intel was not only the major supplier of parts needed to build finished boards (processors and chipsets), but was now building and selling the finished boards as well. By 1994, Intel dominated the processor and chipset markets and had cornered the motherboard market as well.

Now as Intel develops new processors, it develops chipsets and motherboards simultaneously, which means they can be announced and shipped in unison. This eliminates the delay between introducing new processors and waiting for motherboards and systems capable of using them, which was common in the industry's early days. For the consumer, this means no waiting for new systems. Since the original Pentium processor in 1993, we have been able to purchase ready-made systems on the same day a new processor is released.

In my seminars, I ask how many people in the class have Intel-brand PCs. Of course, Intel does not sell or market a PC under its own name, so nobody thinks they have an "Intel-brand" PC. But, if your motherboard was made by Intel, for all intents and purposes you sure seem to have an Intel-brand PC, at least as far as the components are concerned. Does it really matter whether Dell, Gateway, or Micron put that same Intel motherboard into a slightly different looking case with their name on it?

If you look under the covers, you'll find that many, if not most, of the systems from the major manufacturers are really the same because they basically use the same parts. Although more and more major manufacturers are offering AMD Athlon- and Duron-based systems as alternatives to Intel's, no manufacturer dominates AMD motherboard sales the way Intel has dominated OEM sales to major system manufacturers.

To hold down pricing, many low-cost retail systems based on micro-ATX motherboards use non-Intel motherboards (albeit with Intel chipsets in most cases). But, even though many companies make Intel-compatible motherboards for aftermarket upgrades or local computer assemblers, Intel still dominates the major vendor OEM market for midrange and high-end systems.

Intel Chipset Model Numbers

Starting with the 486 in 1989, Intel began a pattern of numbering its chipsets as follows:

Chipset Number	Processor Family
420xx	P4 (486)
430xx	P5 (Pentium)
440xx	P6 (Pentium Pro/PII/PIII)
8xx	P6/P7 (PII/PIII/P4) with hub architecture
450xx	P6 server (Pentium Pro/PII/PIII Xeon)
E72xx	Xeon workstation with hub architecture
E75xx	Xeon server with hub architecture
460xx	Itanium processor
E88xx	Itanium 2 processor with hub architecture

The chipset numbers listed here are abbreviations of the actual chipset numbers stamped on the individual chips. For example, one of the popular Pentium II/III chipsets was the Intel 440BX chipset, which consisted of two components: the 82443BX North Bridge and the 82371EX South Bridge. Likewise, the 845 chipset supports the Pentium 4 and consists of two main parts, including the 82845 Memory Controller Hub (MCH; replaces the North Bridge) and an 82801BA I/O Controller Hub (ICH2; replaces the South Bridge). By reading the logo (Intel or others) as well as the part number and letter combinations on the larger chips on your motherboard, you can quickly identify the chipset your motherboard uses.

Intel has used two distinct chipset architectures: a North/South Bridge architecture and a newer hub architecture. All its more recent 800 series chipsets use the hub architecture.

AMD Athlon/Duron Chipsets

AMD took a gamble with its Athlon family of processors (Athlon, Athlon XP, Athlon MP, and the now-discontinued Duron). With these processors, AMD decided for the first time to create a chip that was Intel compatible with regards to software but not directly hardware or pin compatible. Whereas the K6 series would plug into the same Socket 7 that Intel designed for the Pentium processor line, the AMD Athlon and Duron would not be pin compatible with the Pentium II/III and Celeron chips. This also meant that AMD could not take advantage of the previously existing chipsets and motherboards when the Athlon and Duron were introduced; instead, AMD would have to either create its own chipsets and motherboards or find other companies who would.

The gamble seems to have paid off. AMD bootstrapped the market by introducing its own chipset, referred to as the AMD-750 chipset (code named Irongate). The AMD 750 chipset consists of the 751 System Controller (North Bridge) and the 756 Peripheral Bus Controller (South Bridge). More recently, AMD introduced the AMD-760 chipset for the Athlon/Duron processors, which is the first major chipset on the market supporting DDR SDRAM for memory. It consists of two chips—the AMD-761 System Bus Controller (North Bridge) and the AMD-766 Peripheral Bus Controller (South Bridge). Although AMD no longer puts much emphasis on chipset sales, its pioneering efforts have inspired other companies, such as VIA Technologies, NVIDIA, and SiS, to develop chipsets specifically designed for the Slot A and current Socket A and Socket 754 processors from AMD. This has enabled the motherboard companies to make a variety of boards supporting these chips and the Athlon processors to take a fair amount of market share away from Intel in the process.

North/South Bridge Architecture

Most of Intel's earlier chipsets (and virtually all non-Intel chipsets) are broken into a multi-tiered architecture incorporating what are referred to as North and South Bridge components, as well as a Super I/O chip:

• The North Bridge. So named because it is the connection between the high-speed processor bus (400/266/200/133/100/66MHz) and the slower AGP (533/266/133/66MHz) and PCI (33MHz) buses. The North Bridge is what the chipset is named after, meaning that, for example, what we call the 440BX chipset is derived from the fact that the actual North Bridge chip part number for that set is 82443BX.

• The South Bridge. So named because it is the bridge between the PCI bus (66/33MHz) and the even slower ISA bus (8MHz).

• The Super I/O chip. It's a separate chip attached to the ISA bus that is not really considered part of the chipset and often comes from a third party, such as National Semiconductor or Standard MicroSystems Corp. (SMSC). The Super I/O chip contains commonly used peripheral items all combined into a single chip. Note that most recent South Bridge chips now include Super I/O functions, so that most recent motherboards no longer include a separate Super I/O chip.

Chipsets have evolved over the years to support various processors, bus speeds, peripheral connections, and features.

Figure 4.27 shows a typical AMD Socket A motherboard using North/South Bridge architecture with the locations of all chips and components.

The North Bridge is sometimes referred to as the PAC (PCI/AGP Controller). It is essentially the main component of the motherboard and is the only motherboard circuit besides the processor that normally runs at full motherboard (processor bus) speed. Most modern chipsets use a single-chip North Bridge; however, some of the older ones actually consisted of up to three individual chips to make up the complete North Bridge circuit.

The South Bridge is the lower-speed component in the chipset and has always been a single individual chip. The South Bridge is a somewhat interchangeable component in that different chipsets (North Bridge chips) often are designed to use the same South Bridge component. This modular design of the chipset allows for lower cost and greater flexibility for motherboard manufacturers. The South Bridge connects to the 33MHz PCI bus and contains the interface or bridge to the 8MHz ISA bus. It also typically contains dual IDE hard disk controller interfaces, one or two USB interfaces, and in later designs even the CMOS RAM and real-time clock functions. The South Bridge contains all the components that make up the ISA bus, including the interrupt and DMA controllers.

The third motherboard component, the Super I/O chip, is connected to the 8MHz ISA bus and contains all the standard peripherals that are built in to a motherboard. For example, most Super I/O chips contain the serial ports, parallel port, floppy controller, and keyboard/mouse interface. Optionally, they might contain the CMOS RAM/Clock, IDE controllers, and game port interface as well. Systems that integrate IEEE-1394 and SCSI ports use separate chips for these port types.

Most recent motherboards that use North/South Bridge chipset designs incorporate a Super-South Bridge, which incorporates the South Bridge and Super I/O functions into a single chip.

Hub Architecture

The newer 800 series chips from Intel use a hub architecture in which the former North Bridge chip is now called a Memory Controller Hub (MCH) and the former South Bridge is called an I/O Controller Hub (ICH). Rather than connect them through the PCI bus as in a standard North/South Bridge design, they are connected via a dedicated hub interface that is twice as fast as PCI. The hub design offers several advantages over the conventional North/South Bridge design:

• It's faster. The hub interface is a 4X (quad-clocked) 66MHz 8-bit (4x66MHzx1 byte = 266MBps) interface, which has twice the throughput of PCI (33MHzx32 bits = 133MBps).

• Reduced PCI loading. The hub interface is independent of PCI and doesn't share or steal PCI bus bandwidth for chipset or Super I/O traffic. This improves performance of all other PCI bus connected devices because the PCI bus is not involved in these transactions.

• Reduced board wiring. Although twice as fast as PCI, the hub interface is only 8 bits wide and requires only 15 signals to be routed on the motherboard. By comparison, PCI requires no less than 64 signals be routed on the board, causing increased electromagnetic interference (EMI) generation, greater susceptibility to signal degradation and noise, and increased board manufacturing costs.

This hub interface design allows for a much greater throughput for PCI devices because there is no South Bridge chip (also carrying traffic from the Super I/O chip) hogging the PCI bus. Due to bypassing PCI, hub architecture also enables greater throughput for devices directly connected to the I/O Controller Hub (formerly the South Bridge), such as the new higher-speed ATA-100 and USB 2.0 interfaces.

The hub interface design is also very economical, being only 8 bits wide. Although this seems too narrow to be useful, there is a reason for the design. By making the interface only 8 bits wide, it uses only 15 signals, compared to the 64 signals required by the 32-bit-wide PCI bus interface used by North/South Bridge chip designs. The lower pin count means less circuit routing exists on the board, less signal noise and jitter occur, and the chips themselves have many fewer pins, making them smaller and more economical to produce.

Although it transfers only 8 bits at a time, the hub interface executes four transfers per cycle and cycles at 66MHz. This gives it an effective throughput of 4x66MHzx1 byte = 266MB per second (MBps). This is twice the bandwidth of PCI, which is 32 bits wide but runs only one transfer per 33MHz cycles for a total bandwidth of 133MBps. So, by virtue of a very narrow—but very fast—design, the hub interface achieves high performance with less cost and more signal integrity than with the previous North/South Bridge design.

The MCH interfaces between the high-speed processor bus (533/400/133/100/66MHz) and the hub interface (66MHz) and AGP bus (533/266/133/66MHz), whereas the ICH interfaces between the hub interface (66MHz) and the ATA (IDE) ports (66/100MHz) and PCI bus (33MHz).

The ICH also includes a new low-pin-count (LPC) bus, consisting basically of a stripped 4-bit wide version of PCI designed primarily to support the motherboard ROM BIOS and Super I/O chips. By using the same 4 signals for data, address, and command functions, only nine other signals are necessary to implement the bus, for a total of only 13 signals. This dramatically reduces the number of traces connecting the ROM BIOS chip and Super I/O chips in a system as compared to the 96 ISA bus signals necessary for older North/South Bridge chipsets that used ISA as the interface to those devices. The LPC bus has a maximum bandwidth of 6.67MBps, which is close to ISA and more than enough to support devices such as ROM BIOS and Super I/O chips.

Note Although other chipset makers typically use the North Bridge/South Bridge nomenclature for their chipsets, several have developed high-speed connections similar to Intel's hub architecture. For example, most of VIA's recent chipsets use the V-Link Hub Architecture, which provides a dedicated 266MHz bus between the North and South Bridge chips. The high-speed HyperTransport bus between the North and South Bridges originally developed by AMD has been licensed by chipset vendors such as NVIDIA, VIA, and ALi Corporation, and SiS's MuTIOL Connect is used by recent SiS chipsets.

Figure 4.28 shows a typical Intel motherboard that uses bus architecture—the Intel D845PEBT2, which supports the Intel Pentium 4 processor. Unlike some of Intel's less-expensive hub-based motherboards, the 845PEBT2's Intel 845 chipset doesn't incorporate video.

Let's examine the popular chipsets, starting with those used in 486 motherboards and working all the way through to the latest Pentium III/Celeron, Pentium 4, Athlon XP, and Athlon 64 chipsets.

Intel's Early 386/486 Chipsets

Intel's first real PC motherboard chipset was the 82350 chipset for the 386DX and 486 processors. This chipset was not very successful, mainly because the EISA bus was not very popular and many other manufacturers were making standard 386 and 486 motherboard chipsets at the time. The market changed very quickly, and Intel dropped the EISA bus support and introduced follow-up 486 chipsets that were much more successful.

Table 4.9 shows the Intel 486 chipsets.

Table 4.9. Intel 486 Motherboard Chipsets

Chipset	420TX	420EX	420ZX
Code name	Saturn	Aries	Saturn II
Date introduced	Nov. 1992	March 1994	March 1994
Processor	5V 486	5V/3.3V 486	5V/3.3V 486
Bus speed	Up to 33MHz	Up to 50MHz	Up to 33MHz
SMP (dual CPUs)	No	No	No
Memory types	FPM	FPM	FPM
Parity/ECC	Parity	Parity	Parity
Max. memory	128MB	128MB	160MB
L2 cache type	Async	Async	Async
PCI support	2.0	2.0	2.1
AGP support	No	No	No
AGP = Accelerated graphics port FPM = Fast page mode PCI = Peripheral component interconnect SMP = Symmetric multiprocessing (dual processors) Note: PCI 2.1 supports concurrent PCI operations.

The 420 series chipsets were the first to introduce the North/South Bridge design that is still used in many chipsets today.