Overclocking is not so mysterious as it sounds. To the average layman, however, the process of overclocking can sound as arcane as witchcraft. In this series we examine the basis and rationale of overclocking, the methods, the pitfalls and the benefits.

Why Overclock?

The first question that comes to the mind of any newbie is likely - why overclock? There are many reasons.

Some people overclock because they want more performance from their personal computers. Some people overclock for the thrill of it. It's like climbing a mountain or jumping out of a plane with a parachute. It's not anything anyone absolutely has to do, but if you can get some extra performance out of it and have some fun in the bargain, then well, why not?

Now, some say that those who don't know what they're doing shouldn't overclock. And they're right. The purpose of this article is to turn you from someone who doesn't know what he (or she) is doing, to someone who does.

In our opinion, there's no magic or mystery about overclocking. It's purely logical. It's not as if anyone needs a degree in overclocking or anything like that. Anyone (and we mean anyone) can overclock, and do so fairly safely. It's just a matter of knowing what to do, how to do and when to do it.

What's A MHz Or GHz?

If we're going to be overclocking, we're going to have to deal with MHz and GHz a lot. So, it pays to know what they are.

Hertz (usually abbreviated as Hz) is a measure of frequency. In the computer world, it denotes how many clock cycles occur in a single second. For example, a frequency of 100 clock cycles in a second is called 100 hertz (or 100 Hz), and a frequency of 1,000 clock cycles in a second is called 1,000 Hz.

As the frequency numbers get higher, they become quite tedious to write. Therefore, 1,000 Hz can be abbreviated as 1 kilohertz or 1 kHz. Not only is this easier on the writer, it's also easier on your eyes!

In that sense, you can easily deduce that a MHz is nothing more than a megahertz or one million hertz or a million clock cycles per second. A GHz, on the other hand, is a thousand million clock cycles per second or a thousand MHz. As you can see, it's easier to just write that as a gigahertz or 1 GHz.

Although the amount of work any processor can do in a single clock cycle varies in different processor models, the higher its frequency, the more work it can do in a single second. Therefore, the frequency at which a processor runs at is frequently used as a measure of its speed.

It might be simpler to equate a processor's MHz and GHz with the speed of a car. When it really comes down to it, a hertz isn't much different from the mph (miles per hour) or km/h (kilometers per hour) values we use to determine a car's speed. Just as a car at 100 mph is likely to reach its destination at half the time of a car at 50 mph, a processor running at 600 MHz is likely to complete a task at half the time of a processor at 300 MHz.

So, when you see people talking about MHz or GHz, all you need to remember is that they are merely the number of clock cycles a processor or chip or circuit goes through every second. The higher the value, the faster the processor/chip/circuit.

How Is Overclocking Possible?

Some people think that overclocking isn't possible. Why? Because if a processor can run at a higher clock speed, they would have sold it at that speed. After all, the cost of a processor increases with its clock speed rating.

However, computer chips aren't actually graded and sold according to the maximum speed at which they can safely run. No doubt chip makers would love to do that, but practicality and market forces dictate otherwise.

Speed Binning

All processors start out from the same line. Unlike what some may think, chip makers do not have dedicated lines for processors of different speed grades. Instead, they have a single line from which all processors of the same model are made. Even then, these chips will not turn out exactly the same.

At the end of the fabrication process, some chips will invariably be non-functioning or malfunctioning. Those that work though will not have the same characteristics. Some will be able to run faster while others can only run at a lower clock speed. However, it would take an extraordinarily long time to test each and every chip and discover their maximum potential.

To save time, chip makers use a method called "speed-binning". Instead of testing and selling chips at their maximum clock speed, they test and sell chips at certain speed grades - like 1GHz, 1.2GHz, etc. This is one of the reasons why overclocking is possible.

Suppose a chip maker sells a particular processor at 3 different speed grades - 1GHz, 1.2GHz and 1.4GHz. After every processor is made, it is first tested at 1.4GHz. If it passes, then it's labelled and sold as a 1.4GHz processor. If it fails at that clock speed, it's retested at 1.2GHz and labelled as a 1.2GHz processor if it passes. Otherwise, it's tested at 1GHz. If the processor fails at that speed, then it's discarded.

But no matter what clock speed a processors is rated at, some potential remains hidden thanks to speed-binning. For example, a processor may be speed-binned at 1.2GHz, but if you actually test its limits, it may run at 1.33GHz. That's one reason why overclocking is possible.

Market Considerations

In a perfect world, chip makers would first test each chip at the highest speed grade and work their way down until the chip passes the test. However, market forces often intervene and prevent them from doing so.

Even if a chip is capable of running at a very high clock speed, consumers may prefer a lower-speed grade. As such, demand for a lower-speed grade may far exceed the chip maker's production of slower-running chips. In such an event, chip makers often have to relabel faster chips and sell them as lower-speed chips. In other words, chip makers often grade chips according to market demand.

For example, a chip maker may be able to make a processor that runs at 2GHz about 90% of the time. However, only 10% of the market are demanding for that speed grade while 40% want a slightly slower 1.8GHz speed grade and the remaining 50% want the slowest 1.6GHz speed grade.

So, the chip maker will have no choice but to filter out and sell only 10% of their chips at 2GHz. Even though most of their remaining chips can potentially run at 2GHz, they will test and sell them at the slower speed grades.

This is another reason why overclocking is possible. Many chips out there are not running at their maximum potential. This is especially true for lower-speed grades of long-established chip production lines. The fabrication process may have been refined enough to produce many high-speed chips but market demand may force the chip maker to sell them at lower-speed grades.

Safety Margin

Like many things that we buy, computer chips are covered by warranty against malfunction during normal use. However, their behaviour is affected by different environmental conditions. For example, the hotter the ambient temperature, the higher the likelihood of a chip failing at a particular clock speed.

To pad against environmental variables like ambient temperature, chip makers add some margin to a processor's speed grade. For example, if they successfully certify that a processor can run at 1.4GHz, they will label and sell it as a 1.2GHz. The 200MHz margin is their way of ensuring that the processor will run properly even in a different environment from that of the test lab.

This margin is another reason why overclockability is not only possible, but a certainty with every chip. It's only a matter of how much a chip can be overclocked.

Heat

Chips produce heat. All else being equal, the higher the clock speed, the greater their thermal output. Unfortunately, the tremendous output of heat reduces a chip's ability to run at higher clock speeds. That's why processors often have active coolers which serve to whisk away heat from the processor.

The cooler's ability to remove heat from the processor is affected by ambient temperature. A processor that runs well at an ambient temperature of 25°C may not even boot up when used in an ambient temperature of 30°C.

Generally, a processor will be able to run faster in a cooler environment, especially if it has a good active cooler. That's why overclockers have tried using water and even liquid nitrogen to cool down the processor and allow it to run at much higher clock speeds.

This relationship with heat is also a reason why overclocking is possible. By reducing the thermal load of a processor, we can improve its ability to run at higher clock speeds. A cooler environment almost automatically improves the overclockability of a chip.

Which Type Of Overclocker Are You?

There are basically two categories of overclockers - casual overclockers and hardcore overclockers. Which one are you?

Casual overclockers push their processors to their limit but usually no more than that. Although it might be possible to push their processors further with by improving its cooling and even overvolting, they normally do not do so. Their main purpose of overclocking is likely to maximize their system's potential and/or to reduce the need for expensive upgrades.

Hardcore overclockers, on the other hand, spare no expense in their quest to push their system to the limit. In fact, they often strive to push their systems beyond what anyone can reasonably expect from them. To do that, they often augment their systems with better coolers and make use of techniques like overvolting to push the envelope.

Another way to distinguish a casual overclocker from a hardcore overclocker is to find out if they actually open the chassis of their system and mess around with the hardware itself. Most of the time, it's possible to do casual overclocking even without opening the chassis. But serious overclockers won't let a mere steel cover stop them from achieving the impossible.

It's common for them to mess around with the hardware, changing coolers, rearranging the cables and even modding the case itself for better cooling. Why? Simply because better cooling improves overclockability and that's their ultimate goal - to achieve ever greater clock speeds.

What Does A System Consists Of?

To understand the art and science of overclocking, one must first understand a little about system architecture. A computer system is basically made up of five main parts : a motherboard, the processor, memory modules, a hard drive and a display card.

The motherboard physically connects all these components together, with several system buses electrically and logically connecting the components to chipset and each other. These system buses function like highways transferring data between the components.

A key system bus is the front side bus (often abbreviated as the FSB) which connects the processor to the motherboard chipset. When people talk about overclocking the processor, this usually means either increasing the speed of this front side bus or increasing the processor's clock multiplier.

It's also possible to overclock other system buses as well as their components. Memory modules and graphics cards are components that are easily overclocked.

Know Your System

Before attempting to overclock, try to find out as much as possible about your system. There are quite a few ways to find out what's in the system case without actually opening it up. But the best software we have found is cpuid.com's "CPU-Z". It's small, free, and will tell you everything you need to know about your system.

Here's an explanation of the relevant terms you will find when you use CPU-Z :

Name : This is the name of the processor in the system.

Core Speed : This is the speed of the processor.

Multiplier : A processor runs at a clock speed that's many times faster than the front side bus. Using a built-in multiplier, the processor actually derives its clock speed from the front side bus. In other words, the core speed is derived by multiplying the front side bus (FSB) speed with the processor's multiplier.

FSB : The speed of the "Front Side Bus". "Front side bus" is another name for the processor-chipset bus.

Bus Speed : This is the "effective" processor bus speed. By multiplexing the signals of the front side bus, it's possible to greatly increase the bandwidth of the bus without actually increasing the clock rate. This bus speed is therefore the effective front side bus speed, as opposed to the true front side bus speed shown under FSB. For the purposes of overclocking the processor, you should not use this value. Use the "FSB" value instead, as that's the true front side bus speed.

This concludes the first part of this overclocking guide. In the next part, we'll delve further into the world of overclocking. Be sure to check back!