English I CPU

Table of Contents

1. Where Did CPUs Come From?..........................................................................................................1

2. How Does a CPU Work?.....................................................................................................................1

3. The Anatomy........................................................................................................................................4

4. The First Microprocessor...................................................................................................................5

5. 4 Bits Isnt Enough..............................................................................................................................5

6. The First Step Up, 8 Bits.....................................................................................................................6

7. 16 Bits Enables the IBM PC...............................................................................................................6

8. Cache RAM, Catching Up With the CPU.........................................................................................7

9. Cache Grows Up..................................................................................................................................8

10. Cache Splits Up.................................................................................................................................8

Nikola Sve njak Page 0

English I CPU

The microprocessor, or CPU, as some people call it, is the brains of our personal computer. Im getting
into this history lesson to go through the development step-by-step to explain how they work.

Well, not everything about how they work, but enough to understand the importance of the latest
features and what they do for you. I hope its an interesting for you and helps you recognize computer
buzzwords when youre making your next computer purchase .

1. Where Did CPUs Come From?

When the 1970s dawned, computers were still monster machines hidden in air-conditioned rooms and
attended to by technicians in white lab coats. One component of a mainframe computer, as they were
known, was the CPU, or Central Processing Unit. This was a steel cabinet bigger than a refrigerator
full of circuit boards crowded with transistors.

Computers had only recently been converted from vacuum tubes to transistors and only the very latest
machines used primitive integrated circuits where a few transistors were gathered in one package. That
means the CPU was a big pile of equipment. The thought that the CPU could be reduced to a chip of
silicon the size of your fingernail was the stuff of science fiction.

2. How Does a CPU Work?

In the '40s, mathematicians John Von Neumann, J. Presper Eckert and John Mauchly
came up with the concept of the stored instruction digital computer. Before then,
computers were programmed by rewiring their circuits to perform a certain calculation
over and over. By having a memory and storing a set of instructions that can be
performed over and over, as well as logic to vary the path of instruction, execution
programmable computers were possible.

The component of the computer that fetches the instructions and data from the memory
and carries out the instructions in the form of data manipulation and numerical
calculations is called the CPU. Its central because all the memory and the input/output devices must

Nikola Sve njak Page 1

English I CPU

connect to the CPU, so its only natural to keep the cables short to put the CPU in the middle. It does
all the instruction execution and number calculations so its called the Processing Unit.

The CPU has a program counter that points to the next instruction to be executed. It goes through a
cycle where it retrieves, from memory, the instructions in the program counter. It then retrieves the
required data from memory, performs the calculation indicated by the instruction and stores the result.
The program counter is incremented to point to the next instruction and the cycle starts all over.

3. The Anatomy

Nikola Sve njak Page 2

English I CPU

There are two major parts to every CPU, the silicon that actually contains the "processor" and the
packaging that connects the core to the rest of the outside world. The packaging of a processor can
control how high of a clock speed the CPU will reach, and it can also control factors such as FSB
frequencies. How does something as simple as the packaging of a CPU control things like clock
speeds?

Let's take a look at a present day CPU; for this example we'll use the AMD Duron running at 1GHz.

In the center of the CPU you have what we always refer to as the "core." This is the silicon where it all
takes place; all of the calculations, load/store requests, branches, everything happens right here.

In the case of this Duron the silicon is connected to the rest of the chip (known generically as the
package) using what is known as a flip-chip package. The reason it is called "flip-chip" is because the
exposed part of the core you are looking at is actually the "bottom" of the core; it is flipped so that the
core can have direct contact with your heatsink to improve cooling which is vital when looking at the
heat production of today's processors.

On the other (non-exposed) side of the silicon we actually have the "interface" or interconnect between
the silicon and the package. This Duron CPU has 25.18 million transistors in that tiny piece of silicon;
groups of those transistors have to be connected to the underside of the package where the pins are
located so the CPU can actually communicate with the outside world. The connection between the
silicon and the packaging itself is, in this case, made using very small solder bumps that are simply
referred to as flip chip bumps. For this particular CPU there are about 3,000 flip chip bumps on the
underside of that piece of silicon. For comparison purposes the Pentium 4 has about 5,000 flip chip
bumps and the Intel Itanium has around 7,500. As you can probably guess, these solder bumps have to
be very small and very carefully placed. As CPUs increase in complexity, the placement of these
bumps will become exponentially more difficult. Keep this in mind as we'll come back to this problem
later on.

Nikola Sve njak Page 3

English I CPU

Since all of the processing work gets done on the silicon, the silicon generates quite a bit of heat. If
you touch the package of a CPU while it's operational it's not usually hot at all; if you were to stick
your finger on the core of a CPU however, you'd quickly be filing a lawsuit against one of our beloved
CPU manufacturers. This temperature difference can actually be pretty problematic. As you know,
things can expand and contract according to their temperature. So if we have a core running at 50
degrees Celsius and a package that is running at 27, we're going to end up with some problems where
the core is expanding at one rate while the package expands at another. If this goes untouched,
something is going to crack unless there's a stress reliever.

This is where that bluish/purplish material surrounding the core comes into place. You'll notice that it
covers the sides of the core and even seems to extend out past the core. This material is known as the
underfill. The underfill acts as a stress reliever and it is present between the silicon and the package, as
well as around the base of the silicon-package interconnect to compensate for temperature differences.
Since the underfill plays such a key role in connecting the silicon to the package, problems with the
underfill can quickly result in defective CPUs. This is one of many cases where you could have a
perfectly fine and high yield core, but the packaging holds you back.

The final part of packaging anatomy is the material that the package or substrate is made out of. In the
case of the Duron the substrate is ceramic, while in the case of all Intel processors since the P54C
(later versions of the Pentium Classic) the substrate is made out of an organic material that's lighter and
has other beneficial characteristics.

4. The First Microprocessor

Nikola Sve njak Page 4

English I CPU

In 1971 when the heavy iron mainframe computers still ruled, a small Silicon Valley company was
contracted to design an integrated circuit for a business calculator for Busicom. Instead of hardwired
calculations like other calculator chips of the day, this one was designed as a tiny CPU that could be
programmed to perform almost any calculation.

The expensive and time-consuming work of designing a custom wired chip was replaced by the
flexible 4004 microprocessor and the instructions stored in a separate ROM (Read Only Memory)
chip. A new calculator with entirely new features can be created simply by programming a new ROM
chip. The company that started this revolution was Intel Corporation. The concept of a general purpose
CPU chip grew up to be the microprocessor that is the heart of your powerful PC.

5. 4 Bits Isnt Enough

The original 4004 microprocessor chip handled data in four bit chunks. Four bits gives you sixteen
possible numbers, enough to handle standard decimal arithmetic for a calculator. If it were only the
size of the numbers we calculate with, we might still be using four bit microprocessors.

The problem is that there is another form of calculation a stored

instruction computer needs to do. That is it has to figure out where in
memory instructions are. In other words, it has to calculate memory
locations to process program branch instructions or to index into tables
of data.

Like I said, four bits only gets you sixteen possibilities and even the
4004 needed to address 640 bytes of memory to handle calculator
functions. Modern microprocessor chips like the Intel Pentium 4 can
address 18,446,744,073,709,551,616 bytes of memory, though the
motherboard is limited to less than this total. This led to the push for
more bits in our microprocessors. We are now on the fence between 32
bit microprocessors and 64 bit monsters.

6. The First Step Up, 8 Bits

Nikola Sve njak Page 5

English I CPU

With a total memory address space of 640 bytes, the Intel 4004 chip was not the first microprocessor to
be the starting point for a personal computer. In 1972, Intel delivered the 8008, a scaled up 4004. The
8008 was the first of many 8- bit microprocessors to fuel the home computer revolution. It was limited
to only 16 Kilobytes of address space, but in those days no one could afford that much RAM.

Two years later, Intel introduced the 8080 microprocessor with 64 Kilobytes of memory space and
increased the rate of execution by a factor of ten over the 8008. About this time, Motorola brought out
the 6800 with similar performance. The 8080 became the core of serious microcomputers that led to
the Intel 8088 used in the IBM PC, while the 6800 family headed in the direction of the Apple II
personal computer.

7. 16 Bits Enables the IBM PC

By the late '70s, the personal computer was bursting at the seams of the 8 bit microprocessor
performance. In 1979, Intel delivered the 8088 and IBM engineers used it for the first PC. The
combination of the new 16 bit microprocessor and the name IBM shifted the personal computer from a
techie toy in the garage to a mainstream business tool.

In 1978 Intel introduced the 8086 processor, which even now is the basis for all x86-compatible CPUs.
A European headquarters for the chip manufacturer, now located in Feldkirchen, Germany, near
Munich, was not even on the drawing board. The "XT" PC, as it was then called, clocked 4.77 and
later 8 MHz and could access maximum memory of 1 MB.

At this time, RAM was still in the range of kiloBytes. Well-equipped systems had 256 kB comprising
individual memory chips. The crowning touch: If it was equipped with 320 kB, the computer could
start Windows 1.0. On the other hand, hard drives were expensive and rare commodities at the time.
Those well-heeled people who could afford them equipped their personal computers generously with
two disk drives. Some users may even remember the big floppy disks in 8 inch format - the precursors
of the 5.25 inch media. In 1982 Intel introduced its 286, which included 16 bit ISA slots in its
platform. The maximum memory expansion using SIMM memory modules rose to 1 MB. Three years
later the 386 followed (in 1985), which theoretically could manage a gigantic 4GB of RAM. The
motherboards needed for this (and the technology for them) did not exist.

Nikola Sve njak Page 6

English I CPU

The major advantage of the 8086 was up to 1 Megabyte of memory addressing. Now, large
spreadsheets or large documents could be read in from the disk and held in RAM memory for fast
access and manipulation. These days, its not uncommon to have a thousand times more than that in a
single 1 Gigabyte RAM Module, but back in that time it put the IBM PC in the same league with
minicomputers the size of a refrigerator.

8. Cache RAM, Catching Up With the CPU

Well have to continue the march through the lineup of microprocessors in the next installment to
make way for the first of the enhancements that you should understand. With memory space expanding
and the speed of microprocessor cores going ever faster, there was a problem of the memory keeping
up.

Large low-powered memories cannot go as fast as smaller higher power RAM chips. To keep the
fastest CPUs running full speed, microprocessor engineers started inserting a few of the fast and small
memories between the main large RAM and the microprocessor. The purpose of this smaller memory
is to hold instructions that get repeatedly executed or data that is accessed often.

Nikola Sve njak Page 7

English I CPU

This smaller memory is called cache RAM and allows the microprocessor to execute at full speed.
Naturally, the larger the cache RAM the higher percentage of cache hits and the microprocessor can
continue running full speed. When the program execution leads to instructions not in the cache, then
the instructions need to be fetched from the main memory and the microprocessor has to stop and wait.

9. Cache Grows Up

The idea of cache RAM has grown along with the size and complexity of microprocessor chips. A
high-end Pentium 4 has 2 Megabytes of cache RAM built into the chip. Thats more than twice the
entire memory address space of the original 8088 chip used in the first PC and clones. Putting the
cache right on the microprocessor itself removes the slowdown of the wires between chips. You know
you are going fast when the speed of light for a few inches makes a difference!

10. Cache Splits Up

As I mentioned above, smaller memories can be addressed faster. Even the physical size of a large
memory can slow it down. Microprocessor engineers decided to give the cache memory a cache. Now
we have what is known as L1 and L2 cache for level one and level two. The larger and slower cache is
L2 and is the usual size quoted in specifications for cache capacity. A few really high-end chips like the
Intel Itanium II had three levels of cache RAM.

Nikola Sve njak Page 8

English I CPU

Beware that the sheer size of cache RAM or the number of layers are not good indications of cache
performance. Different microprocessor architectures between Intel and AMD make it especially hard
to compare their cache specifications. Just like Intels super high clock rates dont translate into
proportionately more performance, doubling of cache size certainly doesnt double the performance of
a microprocessor. Benchmark tests are not perfect, but are a better indicator of microprocessor speed
than clock rate or cache size specifications.

Nikola Sve njak Page 9