CPU

Machine Cycle

The core function of every CPU is made up of the following four simplified stages:

• • • • • • • • 1. Fetches program instructions from RAM;

                2. Decodes (or understands) them;

                3. Executes them (carries them out);

                4. Stores (in RAM) any results which are produced.

Comparing CPUs is not simple.  Many different factors affect the performance of a CPU.  We can look at the clock speed of the CPU (Usually measured in GHz), but we must also look at number of cores, amount of cache and, very importantly, the design or architecture of the CPU.

The main manufacturers of desktop, laptop and netbook CPUs are Intel and AMD.  ARM and Intel are the min manufacturers of cell-phone CPUs.

Architecture

There are many ways to compare CPUs.  One way is to compare the architecture or design of the CPU.

The internal architecture of a CPU has as much to do with a CPU's performance as the clock speed, so two CPUs with the same clock speed will not necessarily perform equally. Whereas an Intel 80286 microprocessor requires 20 cycles to multiply two numbers, an Intel 80486 or later processor can perform the same calculation in a single clock tick. (Note that clock tick here refers to the system's clock, which runs at 66 MHz for all PCs.) These newer processors, therefore, would be 20 times faster than the older processors even if their clock speeds were the same. In addition, some microprocessors are superscalar, which means that they can execute more than one instruction per clock cycle.[2]

Different architectures may also require more or less power than others.  The most well known modern architectures are those designed by Intel for desktop PCs. AMD is their largest competitor in the deskop space but ARM currently dominate the design of low power processors for mobile devices.

Here are a few terms which refer to the architecture of a CPU.

Celeron

A brand name used by Intel for several ranges of processors, targeted at lower-end PCs. Their performance is often significantly lower when compared to other ranges of Intel processors. Typically they have less cache memory or have advanced features purposefully disabled.

Core 2

Not to be confused with Dual Core, the Core 2 is a brand encompassing a range of Intel's consumer 64-bit x86-64 single-, dual-, and quad-core microprocessors based on the Core microarchitecture. The single- and dual-core models are single-die, whereas the quad-core models comprise two dies, each containing two cores, packaged in a multi-chip module. The introduction of Core 2 relegated the Pentium brand to the mid-range market, and reunified laptop and desktop CPU lines, which previously had been divided into the Pentium 4, Pentium D, and Pentium M brands.

The Core microarchitecture returned to lower clock rates and improved the usage of both available clock cycles and power when compared with the preceding NetBurst microarchitecture of the Pentium 4/D-branded CPUs.[2] The Core microarchitecture provides more efficient decoding stages, execution units, caches, and buses, reducing the power consumption of Core 2-branded CPUs while increasing their processing capacity. Intel's CPUs have varied widely in power consumption according to clock rate, architecture, and semiconductor process, shown in the CPU power dissipation tables.

Pentium D

An Intel processor family with a similar architecture to the Pentium 4 but it has 2 cores.

Pentium M

The Intel processor designed for use in laptops – including features such as variable speed, lower maximum speed, more cache, etc. to maximise a combination of processing power and minimise electrical power consumption (maximise battery life).

Turion

AMD’s mobile version of their 64 bit processor optimised for use in Laptops (minimise power usage).

Core

The core of a computer is the part of the CPU that does the processing  (i.e. not the cache).  A multi-core processor is therefore a processor which contains more than one core and can perform multiple tasks at the same time.  Dual Core (not to be confused with Core 2) means that the CPU contains 2 cores and Quad Core means that the CPU contains 4 cores.

Clockspeed

Also called clock rate, the speed at which a microprocessor executes instructions. Every computer contains an internal clock that regulates the rate at which instructions are executed and synchronizes all the various computer components. The CPU requires a fixed number of clock ticks (or clock cycles) to execute each instruction. The faster the clock, the more instructions the CPU can execute per second.

GHz

The clockspeed of the processor is usually measured in GHz. Hz refers to the unit Hertz – 1 Hz means once per second. 1 GHz is one billion times per second. The speed of microprocessors, called the clock speed, often is measured in gigahertz. For example, a microprocessor that runs at 200 GHz executes 200 billion cycles per second. Each computer instruction requires a fixed number of cycles, so the clock speed determines how many instructions per second the microprocessor can execute.

Overclocking

In the world of computers, a fast processor can be made even faster by overclocking it.  Overclocking means forcing a device to run at a higher speed than that for which it was originally designed.  This usually causes an increase in operating temperatures for the device and therefore it usually coupled with a non-standard cooling solution.  E.g. Overclocking involves increasing the clock speed of the computer's CPU past the rate at which it was originally designed to run.

Risks of overclocking:

• • You can void your warranty

  • The device can become unstable and crash

  • The extra heat generated can damage the device

Turbo boost (Dynamic overclocking)

Turbo boost is Intel’s version of dynamic overclocking which allows a CPU to increase or decrease the clock speed of its cores.   The increased clock rate is limited by the processor's power, current and thermal limits, as well as the number of cores currently in use and the maximum frequency of the active cores.  E.g. If a single threaded program required more processing power and was running on a dual core processor, then one core could have its clock speed increased.  This would lead to an increase in temperature which could be alleviated by reducing the clock speed of the other unused core.

Cache Memory

This term is pronounced like "cash".  There are many different types of caches but they all serve the same purpose. A cache stores recently-used information in a place where it can be accessed extremely fast. For example, a Web browser like Internet Explorer uses a cache to store the pages, images, and URLs of recently visited Web sites on your hard drive. With this neat strategy, when you visit a page you have recently been to, the pages and images don't have to be downloaded to your computer all over again. Because accessing your computer's hard disk is much faster than accessing the Internet, caching Web sites can speed up Web browsing significantly. Most Web browsers allow you to adjust the size of the cache in in browser preferences.

Another common type of cache is a disk cache. This stores information you have recently read from your hard disk in the computer's RAM, or memory. Since accessing RAM is much faster than reading data off the hard disk, this can help you access common files and folders on your hard drive much faster.

Another type of cache is a processor cache which stores small amounts of information right next to the processor. This helps make the processing of common instructions much more efficient, thereby speeding up computation time.

Caching is a technique whereby the number of accesses to a slower device is reduced by duplicating the data in a higher speed device and accessing it from there.  L1 and L2 Cache in the CPU are used to reduce the delay experienced when the CPU accesses RAM.  If some data needed by the CPU can be located in Cache then the system is speeded up overall.  Similarly, certain items from webpages can be cached on the users’ hard disc removing the necessity to continually download every component of the webpage from the internet on every occasion.

32-bit

In computer architecture 32-bit computers are those that have a processor whose architecture is based on working with 32-bits at a time.  This means that memory addresses, register sizes, address and data bus sizes and instruction sizes are all 32 bits in length.  I other words the computer is able to process at most 32 bits of data at once.

64-bit

In computer architecture 64-bit computers are those that have a processor whose architecture is based on working with 64-bits at a time.  This means that memory addresses, register sizes, address and data bus sizes and instruction sizes are all 64 bits in length.  I other words the computer is able to process at most 64 bits of data at once.

The advantages of a 64-bit computer (running a 64-bit operating system) over a 32-bit system are:

• • 32-bit processors can only address a theoretical maximum of 4GB of memory.  64-bit processors can address far larger amounts of memory.  More RAM would in turn result in computers needing to swap data between RAM and the relatively slow hard drive less often.

  • 64-bit processors have a larger amount of general purpose registers.

  • These registers can store larger values than 32-bit processors.  This means that in some cases a 64-bit can process large data or instructions more efficiently.  For example, when the computer is required to perform operations on numbers larger than 32-bit, the 32-bit processor would need to carry out the operation in more than one step.

To benefit from using a 64-bit processor, a 64-bit operating system must be installed and applications must be compiled to take advantage of the increased register size and higher number of general purpose registers.[1]

Multiprocessing

Multiprocessing is the use of two or more central processing units (CPUs) within a single computer system. The term also refers to the ability of a system to support more than one processor and/or the ability to allocate tasks between them. There are many variations on this basic theme, and the definition of multiprocessing can vary with context, mostly as a function of how CPUs are defined (multiple cores on one die, multiple chips in one package, multiple packages in one system unit, etc.).

Instruction Sets

Common Instruction Set Computing (CISC)

This is a type of microprocessor design. The CISC architecture contains a large set of computer instructions that range from very simple to very complex and specialized. Though the design was intended to compute complex instructions in the most efficient way, it was later found that many small, short instructions could compute complex instructions more efficiently. This led to a design called Reduced Instruction Set Computing (RISC), which is now the other major kind of microprocessor architecture. Intel Pentium processors are mainly CISC-based, with some RISC facilities built into them.

References 

[1] What advantages do 64-bit processors have over 32-bit ones? (n.d.) (Online) Available at: http://www.viva64.com/en/k/0002/ (Accessed 30 July 2012)

[2] What is clock speed? (n.d.) (Online) Available at: http://www.webopedia.com/TERM/C/clock_speed.html (Accessed 14 June 2013)