The Processor

The processor (really a short form for microprocessor and also often called the CPU or central processing unit) is the central component of the PC. It is the brain that runs the show inside the PC. All work that you do on your computer is performed directly or indirectly by the processor. Obviously, it is one of the most important components of the PC, if not the most important. It is also, scientifically, not only one of the most amazing parts of the PC, but one of the most amazing devices in the world of technology. The processor plays a significant role in the following important aspects of your computer system:

Performance: The processor is probably the most important single determinant of system performance in the PC. While other components also play a key role in determining performance, the processor's capabilities dictate the maximum performance of a system. The other devices only allow the processor to reach its full potential. Software Support: Newer, faster processors enable the use of the latest software. In addition, new processors such as the Pentium with MMX Technology, enable the use of specialized software not usable on earlier machines. Reliability and Stability: The quality of the processor is one factor that determines how reliably your system will run. While most processors are very dependable, some are not. This also depends to some extent on the age of the processor and how much energy it consumes. Energy Consumption and Cooling: Originally processors consumed relatively little power compared to other system devices. Newer processors can consume a great deal of power. Power consumption has an impact on everything from cooling method selection to overall system reliability. Motherboard Support: The processor you decide to use in your system will be a major determining factor in what sort of chipset you must use, and hence what motherboard you buy. The motherboard in turn dictates many facets of your system's capabilities and performance.

This section discusses many different aspects of this important component in full detail, and concludes with a look at the major processor families used on the PC platform, from the original IBM PC processors to the latest technology. Note that the explanations of how the processor works and what its characteristics are, is kept mostly separate from the information on particular processors. This is done to keep information organized and easier to find. However, I also include summary tables in each of the description sections showing how the various processors fare in that particular area. This shows the evolution of the technology and lets you more eRoots

of the Processor: Digital Logic

and the Semiconductor

This section describes the technology that underlies the modern processor. CPUs are incredibly powerful, complicated devices, but their current state is the result of years of

evolution and improvement from rather humble beginnings. All processors work in the same basic way "deep down". This section describes the fundamentals of how digital logic and circuits work, which are the building blocks used to make modern processors (as well as most of the other chips in your PC, such as your chipset, controllers, etc.) If you are new to computers, you might want to read this introductory section, if you haven't already. It describes the basics of the binary numbering system, hardware and software, etc., and acts as a good backdrop for the contents below. Note: This is an advanced section containing in-depth information on processor and semiconductor technology. It is interesting background material but not strictly required reading in order to understand the practical aspects of your system processor. Note that you will find this section most helpful if you read all the subsections it contains in order. asily compare processors in various ways. Functions and Digital Logic PCs are digital devices; they deal in zeros and ones only. All modern digital computers, in fact all the circuits in the PC, work by manipulating these binary numbers using hardware functions. These functions take one or more inputs and produce from them a pre-defined set of outputs. There are several different types of functions: mathematical functions take two binary numbers and perform arithmetic on them such as addition or subtraction. Logical functions allow for comparing values and performing logical operations on them such as "AND" and "OR". There are many other kinds of functions as well. A simple digital function can be defined using a table that shows the inputs and the resulting outputs. For example, here is a table that defines a simple three-input "OR" function:
Input A 0 0 0 0 1 1 1 1 Input B 0 0 1 1 0 0 1 1 Input C 0 1 0 1 0 1 0 1 Output 0 1 1 1 1 1 1 1

As you can see, the output is 1 if input A is a 1 or input B is 1 or input C is 1. At its core, all computer hardware uses these types of fundamental building blocks. Simple logical functions like these are sometimes called gates (no, they weren't named after Bill. :^) ). Modern complicated processors don't actually use AND and OR gates to implement their logic (they integrate these functions into larger circuits), but the concepts are identical. Implementing Digital Logic: The Digital Switch Digital logic requires hardware that can take ones and zeros as input, and produce from them ones and zeros as output according to the formula required for the function. In digital computers the one and zero values are determined based on the voltage on the line at a given time; normally, a one is positive voltage and a zero is ground. For a +5V system then, a one is +5V and a zero is 0V. The hardware used to produce the output level from the input level according to a formula is a digital switch. Most functions are made up of many of these switches; the more complex the formula, the more switches required to implement it. A computer system is really a large group of interconnected sets of digital switches, all reacting to inputs of ones and zeros and producing outputs of ones and zeros in response. The very first computing devices were electromechanical; they used physical switches and relays. They were very slow, unreliable and noisy, because they had a mechanical component, and this meant something would move when a bit changed from a one to a zero. The next generation of devices used vacuum tubes. These were much better than the mechanical switches because they do not have moving parts. Vacuum tubes enabled the creation of the first electronic computers, but they had several major problems associated with them:

They were large; you have no doubt heard the old stories about computers filling whole rooms despite providing very limited functionality. They were slow; a computer's performance is directly related to how fast its switches can change state (one to zero or vice-versa). They were expensive, much of this being because of their size and poor reliability. They consumed enormous amounts of power; a single machine could consume tens of thousands of watts. Finally, and perhaps most importantly, they were notoriously unreliable. A large computer using thousands of these switches would often require replacements on a daily basis--or worse.

The Transistor: A Solid State Digital Switch One of the key inventions in the history of electronics, and in fact one of the most important inventions ever period, was the transistor. It was invented by Bell Laboratories in 1948. In short, a transistor is a device that conducts a variable amount of electricity through it, depending on how much electricity is input to it. In other words, it is a digital

switch. However, unlike the vacuum tube, it is solid state. This means that it doesn't change its physical form as it switches. There are no moving parts in a transistor. The advantages of the transistor over the vacuum tube were enormous. Compared to the old technology, transistors were much smaller, faster, and cheaper to manufacture. They were also far more reliable and used much less power. The transistor is what started the evolution of the modern computer industry in motion. Semiconductors and The Integrated Circuit The transistor was originally a single, discrete device, that you could place individually into a circuit much like any other. Today, some special-purpose transistors are still used that way. What allowed for the creation of modern processors was the invention of the integrated circuit, which is a group of transistors manufactured from a single piece of material and connected together internally, without extra wiring. Integrated circuits are also called ICs or chips. A special material is used to make these integrated circuits. While most materials either insulate from electrical flow (air, glass, wood) or conduct electricity readily (metals, water), there are some that only conduct electricity a small amount, or only under certain conditions. These are called semiconductors. The most commonly used semiconductor is of course silicon. By careful chemical composition and arrangement, it is possible to create a very small transistor directly on a layer of silicon, using various technologies to manipulate the material into the correct form. These transistors are small, fast and reliable, and use relatively little power. The first integrated circuit was invented in 1959 by Texas Instruments. It contained just six transistors on a single semiconductor surface. Miniaturization and Large Scale Integration After the invention of the integrated circuit, it took very little time to realize the tremendous benefits of miniaturizing and integrating larger numbers of transistors into the same integrated circuit. More transistors (switches) were required in order to implement more complicated functions. Miniaturization was the key to integrating together large numbers of transistors while increasing hardware speed and keeping power consumption and space requirements manageable. Large scale integration ("LSI") came to refer to the creation of integrated circuits that had previously been made from multiple discrete components. These devices typically contained hundreds of transistors. Early computers were made from many of these smaller ICs connected together on circuit boards. Very Large Scale Integration

As time progressed after the invention of LSI integrated circuits, the technology improved and chips became smaller, faster and cheaper. Building on the success of earlier integration efforts, engineers learned to pack more and more logic into a single circuit. This effort became known as very large scale integration (VLSI). VLSI circuits can contain millions of transistors. Originally, the functions performed by a processor were implemented using several different logic chips. Intel was the first company to incorporate all of these logic components into a single chip. This was the first microprocessor, the 4004, introduced by Intel in 1971. All of today's processors are (highly advanced!) descendants of this original 4-bit CPU. Hard-Wired vs. Programmable Logic There is one primary distinguishing feature of a processor that makes it more than just a highly-integrated regular hardware circuit. Ordinary electronic hardware implements a predefined set of functions; the logic is "hard wired" for a single purpose, and the logic circuits are customized to that task only. A different use would require changes to the circuitry. Processors are programmable, that is, they can perform different functions based on instructions read from a program. Instead of being customized to one single task, a processor is designed in a more general way to perform a broad range of functions. It does this by defining a set of subtasks, or instructions that controls how it works, and then letting users write programs from these instructions. This design introduces tremendous flexibility over the hard-wired design (although it comes at the small penalty of being slightly less efficient than a customized machine.) All modern computers use this "hardware-software" model, where hardware performs functions under the control of a program, which is of course called software. In fact, modern computers are made of many layers of software and hardware, each controlled by programs written at the next higher layer This section discusses the physical properties of processors. Microprocessors are solid-state integrated circuits, which means that they have no internal moving parts. Their physical characteristics are described in very different terms, and the manufacturing processes used to create them are an important factor in their reliability and performance. While most of the design factors in a processor have to do with their internal architecture, physical aspects such as packaging and cooling are very important as well.

Processor Manufacturing
Many people do not realize that when they pick up a small Pentium "chip" (for example) and look at it, they are not really seeing a Pentium chip at all. They are seeing the external packaging that is used to house the chip and allow it to interface with the rest of the PC. The actual chip is, amazingly, much smaller and more fragile. It is a square or rectangle of

chemically-altered silicon, thinner than a dime, and with a surface area less than a third of a square inch! Considering that packed into this space are literally millions of preciselyplaced electronic switches, processor design and manufacturing is truly one of the wonders of modern technology. This section discusses the steps followed to physically design and manufacture chips. Many of the descriptions in this section are applicable to most highly complex integrated circuits, not just processors. In fact, in a modern PC most of the chips you see on the motherboard or expansion cards are manufactured using concepts and techniques similar to those described here. Processors usually however are on the leading edge, as they are the largest and most complex circuits. The descriptions here are simplified of course; semiconductor manufacturing is a highly complex technical subject that I could never do justice to in only a few paragraphs. Note: This is an advanced section containing primarily background information. It's interesting, but not directly related to the matters of how a processor works, so you can skip it if you are looking for more practical information. Semiconductor Materials and Wafer Manufacture Processors are manufactured from semiconductor material. Semiconductors are materials that transmit electricity only under certain conditions, and therefore are ideal for making the ultra-small, high-speed transistors that implement the modern processor. Only certain materials are suitable for use in semiconductors. To be suitable, the material must be (at the very least):

Able to be made into a semiconductor. A chemical process called "doping" is used to introduce small impurities into a pure material, which enables the material to act electrically as a semiconductor. Able to be manufactured into large pieces of uniform composition and high quality. Of the appropriate hardness so that it can be cut into thin slices without being so brittle that it cracks. Readily available and relatively inexpensive, to ensure supply and keep costs down.

The most famous, and widely used material for semiconductors is of course silicon. It is used for processors, memory, and most other mainstream chips. Another popular material for use in semiconductors is gallium arsenide (GaAs), which is not nearly as widely encountered, and is generally used for specialty applications. Silicon has several great advantages that make it ideal for use for semiconductors. Obviously it is a semiconductive material. It also has several manufacturing advantages that make it the material of choice: it is plentiful and hence cheap in its raw form (regular sand is silicon dioxide). But most importantly, it can be grown into large, uniform crystals. The first step in the manufacture of a chip is growing these large crystals of silicon. This is similar in concept to how you can grow sugar crystals in a cup of water, although of course, special techniques are used.

These crystals are very large--the common current size is 8 inches in diameter, and this will probably actually increase to 12 inches very soon! The larger the crystal, the more chips that can be manufactured at the same time, and the less waste material, thus saving money. Despite being made from a cheap raw material, these crystals are expensive due to the precise technology used to make them, and the extremely high quality standards required to make a crystal suitable for use in chip making. A single, high-quality crystal will be used to make many thousands of processors. These large crystals are cut into wafers by high-precision saws. These thin slices of semiconductor material then undergo the process that transforms them into actual integrated circuits. Each crystal is cut into several hundred wafers, each less than 1 mm thick (one fortieth of an inch) and 8 inches in diameter. The thinner the wafers, the more that can be made from a single crystal, but the more fragile they are and the harder they are to manipulate. The wafers undergo several more steps before being ready for use, including precision polishing and chemical treatment. Logic Design The actual design of the microprocessor is done first at a logical level. The processor is defined in terms of its desired operation, what instructions it is meant to operate on, and how data is intended to flow between its various logical parts. This high-level design starts at a very conceptual level, and proceeds to more detailed levels where the various architectural features of the chip are defined in great detail, such as how instructions will be decoded, how the control unit(s) will function, etc. This is a very abbreviated description of the process, which involves hundreds or even thousands of engineers and takes months or years of time depending on how advanced the device is and also on how much is borrowed from previous designs. Once the processor has been fully defined in this manner it is laid out physically. The actual transistors necessary to implement the design are determined and a physical map is created showing how the chip must be fabricated. In real life there isn't just one design that is then mapped out in this way once; it is an iterative process with changes being made all the time in the design, due to bugs, enhancements, marketing issues, or manufacturing concerns. Process Technology There are several different types of semiconductors, which vary in terms of how they make transistors out of silicon. The differences between them are highly technical so I won't go into it in detail. The process used to manufacture the chips has an impact on the following:

Minimum circuit or feature size, which refers to how much you can miniaturize the processor and how many transistors you can pack into a given space. Maximum speed that you can run the chip at. Voltage requirements.

Heat generation and power consumption, which are a function to some extent of the previous three.

Most processors today are created using what is called a CMOS process. CMOS stands for "complementary metal oxide semiconductor" and refers to a particular method of making transistors in silicon. Older processors used older technology; the original 8088s for example using NMOS. The Pentium processor uses BiCMOS which is "bipolar CMOS" and has certain advantages and disadvantages over CMOS. Intel has returned to CMOS as the standard today because it allows for faster processor speeds at low voltages, to keep power use down to a manageable level. Circuit Size The circuit size or feature size refers to the level of miniaturization of the processor. To make more powerful processors, more transistors are needed. In order to pack more transistors into the same space, they must be continually made smaller and smaller. As processors get faster and denser, power consumption and heat generation become a big concern as well. Circuit size is a major limiting factor in processor speed, largely due to heat generation issues. It also has a major impact on die size. Technology advancements continue to allow circuit sizes to shrink and shrink. It was once considered "impossible" to shrink the circuit size below 1 micron; most processors today actually now use a 0.35 micron process, with 0.25 the current state of the art in the latest chips and 0.18 micron processors looming in the near future. It is now thought that current technology can eventually be shrunk to as low as 0.08 microns. This will make possible a whole new generation of very fast, low-power devices.

Die Size The die size of the processor refers to its physical surface area size on the wafer. It is typically measured in square millimeters (mm^2). In essence a "die" is really a chip, but it is only referred to in this way when discussing physical chip parameters and manufacturing issues. The importance of die size is rather obvious: the smaller the chip, the more of them that can be made from a single wafer. A larger die means fewer chips from the same wafer, and thus higher cost overall. A larger die also leads to increased power consumption. The three most important contributing factors to die size are the circuit size in microns, the process technology used, and of course, the design of the processor itself (newer processors are in general larger because they do a lot more). Reducing circuit size in particular is key to reducing the size of the chip. For example, the first generation Pentium used a 0.8 micron circuit size, and required 296 square milimeters per chip. The second generation chip had the circuit size reduced to 0.6 microns, and the die size dropped by a full 50% to 148 square milimeters.

Processor Speed The speed that the processor can run at is a function of many different factors. Some of these are related to design of the processor itself, which dictates the internal timing requirements that limit the maximum speed the processor can handle. Manufacturing factors relate to the process technology used, circuit size, die size and process quality. In general, the smaller the chip, the faster it can run. This is due in part to reduced power consumption and heat generation; heat is generated when transistors switch from a zero to a one or vice-versa, and the faster the chip runs, the more switching in a given unit of time, so the more heat that is produced. A chip that overheats locks up or causes computation errors. Designers move chips to smaller circuit sizes to keep heat down as they ratchet up CPU speed. In addition, there are manufacturing variabilities that allow some chips to run faster than others even though they were produced with the same process and even with the same wafer. This is called speed rating and is done during testing.

Processor Power and Voltage

In the early days of computers, there wasn't very much concern about how much power a processor used. There weren't as many of them, and we weren't doing nearly as much with them. We were just thrilled they existed at all! As time has gone on our demands on these machines have continued to increase and new uses have put power consumption in the spotlight. This has led to a confusing set of voltage specifications where before (up to the Intel 486DX2-66) everything ran on 5 volt power. The power usage and the voltage support of a processor are important for the following reasons:

Power consumption equates largely with heat generation, which is a primary enemy in achieving increased performance. Newer processors are larger and faster, and keeping them cool can be a major concern. With millions of PCs in use, and sometimes thousands located in the same company, the desire to conserve energy has grown from a non-issue to a real issue in the last five years. Reducing power usage is a primary objective for the designers of notebook computers, since they run on batteries with a limited life. (They also are more sensitive to heat problems since their components are crammed into such a small space).

Newer processors strive to add additional features and to run at faster speeds, which tends to increase power consumption. Processor designers compensate for this largely through technology, by using lower-power semiconductor processes, and shrinking the circuit size and die size. External and Internal Voltage Levels

Early processors had a single voltage level that was used by the motherboard and the processor, typically 5 volts. As processors have increased in speed and size the desire to use lower voltage levels has led designers to look at using lower voltage levels. The first step was to reduce the voltage level to 3.3 volts. Newer processors reduce voltage levels even more by using what is called a dual voltage, or split rail design. A split rail processor uses two different voltages. The external or I/O voltage is higher, typically 3.3V for compatibility with the other chips on the motherboard. The internal or core voltage is lower: usually 2.5 to 2.9 volts. This design allows these lower-voltage CPUs to be used without requiring wholesale changes to motherboards, chipsets etc. The voltage regulator on the motherboard is what must be changed to supply the correct voltages to the processor socket. Standard Voltage Levels and Motherboard Voltage Support There are several "industry standard" voltages in use in processors today. I put the phrase in quotes because it seems that the number of different voltages being used continues to increase, and the new market presence of AMD and Cyrix makes this even more confusing than when it is was just Intel we had to worry about. The table below shows the current standard voltages with their names and the typical range of voltages that is considered acceptable to run a processor that uses that nominal voltage level:
Name System +5V STD (Standard) VR (Voltage Reduced) VRE (Voltage Reduced Extended) 2.8V Nominal Voltage 5.000V 3.300V 3.380V 3.520V (or 3.500V) 2.800V Acceptable Range 4.750V to 5.250V 3.135V to 3.465V (or 3.135V to 3.600V) 3.300V to 3.465V 3.450V to 3.600V (or 3.400V to 3.600V) 2.700V to 2.900V

Note a couple of things about this table. First, the names are rather ridiculous; "Standard" isn't really standard at all, and "Voltage Reduced" is actually a higher voltage than "Standard"! (This is because the term refers to being "reduced" from the original +5V used on older processors). The newest low voltages don't appear to have names any more (which is a good thing, I think). Second, note that both VR and VRE have two slightly different definitions, as they were revised by Intel. The differences are pretty subtle and generally aren't anything to worry about. Most motherboards will have enough voltage settings to support a wide range of processors. Many will have additional voltages besides the ones in the table.

At the lower end of voltages (below 3V) the industry is still evolving, with 2.8V or 2.9V being the standard for core voltage for chips such as the Intel Pentium with MMX, the Cyrix 6x86L and the AMD K6 (the K6-233 is a bit of an oddball at 3.2V core). Socket 7 and later motherboards supply several voltages (such as 2.5V, 2.7V, 2.8V and 2.9V) for compatibility with future devices. The voltage regulator built into the motherboard does the work of converting the power supply voltage into the different levels required by the processors the motherboard supports. Pentium II systems take this a step further; the technology is designed to allow Intel to change the internal voltage required by the CPU without changes being necessary to the motherboard. Power Management Spurred on primarily by the goal of putting faster and more powerful processors in laptop computers, Intel has created power management circuitry to enable processors to conserve energy use and lengthen battery life. These were introduced initially in the Intel 486SL processor, which is an enhanced version of the 486DX processor. Subsequently, the power management features were made universalized and incorporated into all Pentium and later processors. This feature set is called SMM, which stands for "System Management Mode". SMM circuitry is integrated into the physical chip but operates independently to control the processor's power use based on its activity level. It allows the user to specify time intervals after which the CPU will be powered down partially or fully, and also supports the suspend/resume feature that allows for instant power on and power off, used mostly with laptop PCs. These settings are normally controlled via system BIOS settings.