How To Assemble A Desktop PC/Choosing the parts/CPU

From Wikibooks, open books for an open world
Jump to: navigation, search

The Central Processing Unit (CPU), also called a processor, is the heart of your computer. It performs nearly all the actual computation that takes place as the computer is used. The choice of a CPU will affect the ultimate speed of the computer more than any other single component (gaming systems excepted; there the GPU is usually the bottleneck).

Vocabulary[edit]

Before we can explain the differences between CPUs, you must first be familiar with certain CPU terms.

An Intel Pentium CPU
Clock speed 
Clock speed, measured in gigahertz (GHz) or megahertz (MHz) (1 GHz = 1000 MHz), is the number of calculation cycles that your CPU can perform per second. Therefore, a higher clock speed generally indicates a faster processor. But not all CPUs perform an equal quantity of work per cycle, meaning two CPUs at the same clock speed can potentially perform at very different levels: Clock speeds should not be compared for CPUs from two different architectures (AMD Phenom versus Intel i7, etc.) - in such cases, use reviews and benchmarks to judge relative performance.
IPC 
IPC, or instructions per cycle, is the amount of work a CPU can do in a cycle. With modern processors this number is an average.
Front side bus speed (FSB) 
Front side bus speed is the rate at which the CPU communicates with the northbridge chipset component on your motherboard,and, through the northbridge with main memory, measured in MHz. A larger FSB value shows that your CPU is able to communicate with memory and other components on the motherboard at a higher speed. In an AMD processor the front side bus depends on the version of HyperTransport it has: HT 1.x 800MHz, 2.0 1400MHz, 3.0 2600MHz, and 3.1 3200MHz.
Interface (Socket/Slot) 
CPU’s plug into a socket on the motherboard. It is very important that your CPU is a COMPLETE MATCH to your motherboard CPU socket. Plugging a CPU into the wrong socket will never work and will usually break either the CPU, the socket, or both.
Bit width 
Modern processors are either 32-bit or 64-bit (this is a simplification; see http://en.wikipedia.org/wiki/64-bit for details). For our purposes a processor can be called 64-bit if it will support running a 64-bit operating system. The newer chips from Intel and AMD all support this for most 64-bit OS’s. If you have any doubts, check the support materials for the 64-bit OS you are interested in running.
Hyper-threading (HT) 
Hyper-threading is an Intel technology which allows a single core processor to simulate having two cores, giving a performance boost when running several programs at once. It requires motherboards and chipsets supporting Hyper-Threading technology. The advantages of the Hyper-Threading technology have never been breathtaking and Intel has gone back and forth on its support in multi-core processors.
Cache 
Cache is memory implemented directly on the CPU. Data which is being used in computations is stored in the cache as much as possible and can be retrieved from there much faster than the same data can be pulled through the northbridge from main memory. Generally, the larger the cache, the faster the system will run. Cache comes in (usually) three varieties, L1, L2, and L3. L1 being the smallest and fastest, and L3 being the largest and slowest. Usually only the L2 cache size will be shown, as L3 is rarely used in processor design, and L1 will often stay the same size throughout an entire processor product line.
Core 
The core is the heart of the CPU. Often several cores will be marketed under the same name, so look at what core you are buying. It’s a mistake to choose a processor based solely on it’s rated speed in hertz . This number, while easy to understand, does not tell the whole story. The number of cores and the way they communicate will have a big effect, as will the fundamental architecture of the core or cores themselves.

Slower processors are generally preferred for overclocking, as they can often achieve higher overclocking percentages. Retail CPU's come in a package containing a HSF (Heat Sink Fan), instructions, and a warranty, often 3 years. OEM CPUs do not include these.

The current CPU speeds and advantages change frequently, so for up-to-date comparisons, you may want to check a website that specializes in Hardware reviews, such as Tom's Hardware Guide or Anandtech. A current (as of 14 February 2006) beginner's explanation can be found at Behardware.



Clock speed 
the number of calculation cycles that the CPU can perform per second, measured in Gigahertz (GHz) or Megahertz (MHz). 1 GHz = 1000 MHz. A higher clock speed generally points toward a faster CPU and, therefore, system. However, not all CPUs perform an equal quantity of work per cycle, meaning two CPUs at the same clock speed can potentially perform at very different levels. Other factors may also limit the effectiveness of a fast CPU.
IPC 
instructions per cycle, the amount of work a CPU can do in a cycle. This varies with different properties of the CPU.
Front side bus speed (FSB) 
the rate (usually in MHz and/or MT/s) at which the CPU communicates with the northbridge chipset component on your motherboard. A larger FSB value shows that your CPU is able to communicate with other components on the motherboard (and thus your system) faster, provided the motherboard supports an equally high FSB speed. Note that MT/s (megatransfers per second) is a better gauge of FSB speed. For example, a "quad-pumped" 266 MHz FSB will do 1066 MT/s, whereas a 1000 MHz FSB that does two transfers per cycle will do 2000 MT/s.
Interface (Socket/Slot) 
the type of connector that the CPU uses to connect to the motherboard. It is VERY important that your CPU interface is a COMPLETE MATCH to your motherboard CPU socket otherwise you would be wasting money on a piece of spare silicon and may risk damage the the CPU and/or motherboard.
Bit-rate 
Most modern desktop CPUs are of the 32-bit x86 system, which works fine with most modern operating systems and hardware. Higher end models (and all currently manufactured AMD desktop CPUs) are of the x86-64 (a.k.a. AMD64/EM64T) 64-bit system which may allow faster CPU processing capabilities due to the larger bandwidth, but not all operating systems are compatible with the newer 64-bit format -- they will still run, but without a major performance boost. You will require 64-bit if you need to manage more than 4GB of RAM.
Cache 
The cache is memory that is on the processor dedicated for it to use. Cache is many times faster for the CPU to access than RAM. Generally, the larger the cache, the faster your system would run. However, cache uses a lot of transistors, and the larger the cache, the higher number of transistors needed, which consume more electricity, and output more heat. Cache comes in (usually) three varieties, L1, L2, and L3. L1 being the smallest and fastest, and L3 being the largest and slowest. Usually only the L2 cache size will be shown, as L3 is rarely used in processor design, and L1 will often stay the same size throughout an entire processor product line.
Core 
The core of the CPU is the heart of the CPU. Often several cores will be marketed under the same name, so look at what core you are buying.

Multiple cores[edit]

Advantages
Dual-core processors are a fairly new innovation built by both major processor manufacturers (Intel Core Duo / Core 2 Duo and AMD Athlon 64 X2).
  • Multitasking: Each processor has two processing centers (cores) for a theoretical maximum of twice the operating power and for better multitasking. Major advantages of dual core processors are evident when doing heavy multitasking, such as encoding video and playing video games at the same time.
  • Application Support: Newer applications are being written to take advantage of this technology by using a technique known as Multi-threading.
  • Power Saving: Dual core processors (especially Intel Core Duo) have the ability to turn off one of their cores when application demand is low to save power.
Disadvantages
  • Support: Older programs (with certain notable exceptions) do not support multi-threading and may run very slightly slower on dual core CPUs. Gamers should also take note of this as many games (typically those released before 2006, when the dual-core Intel and AMD processors were released) do not support processing with multiple cores that well, although some of the more bleeding-edge games will take advantage of using multiple cores.
  • Performance: A quad core isn't twice as fast as a dual core. Especially if you use a single, very demanding application (like games).
The future of Multi-Core Processing 
Quad core technology is finally here, and more and more programs are being built to take advantage of multiple cores. The large chip makers are discussing CPU's with eight, sixteen, even thirty two cores in the coming years. Only time and experience will tell how the advantages we are seeing with duo and quad core configurations will scale.

Descriptions[edit]

Often in advertisements (especially for pre-built systems) a processor will be described briefly, like this:

AMD FX-6300 at 3.5GHz

But there’s more to it than that, a more detailed specification would read as follows:

AMD FX-6300 six core processor at 3.5GHz/4.1GHz, AM3+ socket, 8MB L3 Cache, 6MB L2 Cache

Which parses out to:

Model: AMD FX-6300
Base Clock Speed: 3.5GHz (3500 MHz)
Turbo Clock Speed: 4.1 GHZ (4100 MHz)
Socket: AM3+
L3 Cache: 8MB, L2 Cache: 3x2MB

So we can see that while 3.5 GHz is a fast clock speed, this processor might not be as fast as, a higher end AMD FX-8320, which runs at the same 3.5GHz base clock speed, but has 2MB more L2 cache and an additional two processing cores.

Intel[edit]

Intel classifies its CPUs using a series of numbers. 3xx, 4xx, 5xx, 6xx and 7xx of which 7xx denominates the highest end products. Generally, the higher the number, the faster the CPU and the more expensive. Usually, models and ratings correspond. For instance: an Intel Core 2-Quad processor is at 2.4 while an AMD Phenom Quad-Core is at 2.2. An Intel Core 2-Duo's speed is not very different from an AMD Phenom Quad-Core's speed. Here is a list of most modern Intel Desktop CPUs:

  • Intel Celeron (Single and Dual core)

The Celeron series is a range of CPUs for budget computers and used to feature just one core. The last single core Celeron was released as part of the Sandy Bridge line-up, and since then all Celeron G-series processors have been dual-core.

  • Intel Pentium (Single and Dual core)

The Pentium series was part of Intel's most popular CPUs. Earlier Intels all featured just one core, although the newer Pentiums feature dual-core support, such as the Pentium D, the Pentium Dual-Core, and the Pentium G-series processors.

  • Intel Core 2 Duo (Dual core)

An extremely popular brand of Intel chips, the Core 2 Duo processors all support dual-core technology while fitting in to Intel's most common socket type, LGA 775 (Land Grid Array 775). Many older desktop computers from 2007-2010 will tend to use this CPU or another processor in the Core 2 series.

  • Intel Core 2 Quad (Quad core)

The Core 2 Quad range of processors feature 4 processing cores and have been made for both gaming (although gamers should be aware of the lack of multi-core support for games released before 2006) and professional 3D graphics design, video editing, etc. Keep in mind that the hardware level the Core 2 Quad processors implement may cause bottlenecking, but usually most Core 2 Quad users would not have to worry about that.

  • Intel Core 2 Extreme (Dual and Quad core)

This range of CPUs tends to be directed at enthusiasts and are basically improved versions of Core 2 Duo and Core 2 Quad processors. Be aware that these processors are quite expensive, even today, and typically aren't worth the buy (but if you really want it, nobody's going to stop you)

  • Intel Core i3 (Dual core)

These are dual core processors that have hyperthreading enabled, allowing them to make use of multi-threaded software more efficiently. This CPU series is made for LGA 1156, 1155, and 1150. This processor lineup is adequate for most applications, including the majority of games.

  • Intel Core i5 (Quad core)

These are quad core processors without hyperthreading. Like the rest of the 'i' lineup, they are made for LGA 1156,1155, and 1150. This series can take almost any game without bottlenecking the graphics card.

  • Intel Core i7 (Quad/Hexa core)

These are quad/hexa core processors with hyperthreading. Like the rest of the 'i' lineup, they are made for LGA 1156,1155, and 1150. This series can take any game without bottlenecking the graphics card.


  • Intel Xeon (Single, Dual, Quad, and Hexa core)

The Xeon brand was a brand of Intel x86 processors for workstations, servers, and embedded systems. The Xeon brand features single, dual, quad, and hexa (6) core processors (although don't get excited - remember that these are for servers and similar systems, not your "Ultra-awesome high-tech bleeding-edge owns-everybody-else's computer" with a 790i Ultra SLi motherboard and 4 graphics cards)

  • Intel Core i7 (Quad core)

Intel's newest line of consumer microprocessors, these CPUs are all have 4 cores and feature higher amounts of cache and Intel's new "Turbo Boost" technology, which allows all cores to automatically clock themselves to appropriate frequencies in intervals of 133 MHz without stressing the processor and risking overheating. The drawbacks? it's not compatible with Intel's de facto socket standard of LGA 775. Rather, it utilizes a newer socket - LGA 1366. It also is only compatible with DDR3 memory and does not use a FSB (Front Side Bus), but rather uses an Intel QuickPath interface (which, honestly, I have no idea how that makes it better from a FSB, but I do know that it only allows for one of these interfaces, so having multiple CPUs is out of the question). Basically, if you want to get this kind of processor, make sure the rest of your system is up to par. This line also includes the Core i3 and Core i5 processors. Core i3 is mainly used for low-power systems and it is usually single core. Core i5 is dual core and supports newer features like Turbo-Boost and Hyper Threading.

Other information about Intel's Classification System:

XD Technology: A CPU's model number followed by suffix J signifies XD (eXecute Disable) technology, which is an added security feature on Intel CPUs. Note that AMD processors have a similar technology dubbed EVP, or "Enhanced Virus Protection.
Intel's naming scheme: "Intel Pentium 4 3.0Ghz L2-1MB with HT" will easily convert to just "Intel Pentium 4 530J"

There is also the Core 2 Duo range out at the moment - it's naming scheme is relatively normal, except for the models ending with 50, they have 1333mhz FSB.

  • E6300/E6320: 1.86Ghz, 2MB L2 Cache (E6320 has 4MB), 1066MHz FSB
  • E6400/E6420: 2.13Ghz, same as above
  • (new) E6540: 2.33Ghz, 4MB L2 Cache, 1333MHZ FSB
  • (new) E6550: 2.33Ghz, same as above
  • E6600: 2.4Ghz, 4MB L2 Cache, 1066MHZ FSB
  • E6700: 2.66Ghz, same as above
  • (new) E6750: 2.66Ghz, 4MB L2 Cache, 1333MHZ FSB
  • (new) E6850: 3.0Ghz, same as above
  • X6800: 2.93Ghz, same as above except multiplier unlocked

The Nehalem/Lynnfield/Sandy Bridge/Ivy Bridge/Haswell series line-up uses a different naming scheme as follows. Celeron/Pentium indicates a basic dual core. Pentiums usually have more cache and higher clockspeed than celerons. i3 denotes dual core with hyperthreading. No Turbo functionality. i5 usually denotes a quad core without hyperthreading, and all feature turbo functionality which allows the processor to raise its clockspeed under load. However, the i5-6xx series was dual core hyperthreaded, and featured turbo unlike the almost identical i3-5xx series. i7 denotes quad/hexa cores with hyperthreading, and all feature turbo functionality

Examples follow:

To differentiate the families using the 'i' designation, it is necessary to look at the numbering of the processors. Any cpu that has the i designation and 3 digits is meant for LGA 1156 Any cpu that has the ix-2xxx/ix-3xxx type naming is for LGA 1155 Any cpu that has the ix-4xxx naming is for LGA 1150

Celeron G1101/Pentium G6950 is meant for LGA 1156 Celeron G4xx and G5xx/Pentium G6xx and G8xx series are for LGA 1155 Celeron G16xx/Pentium G2xxx are for LGA1155 Celeron G18xx/Pentium G3xxx are for LGA 1150

AMD[edit]

AMD CPUs have a different classification. The AMD Athlon CPU rating are not of the actual clock speed but rather the equivalence bench mark performance corresponding to a comparison to the AMD Athlon Thunderbird 1.0Ghz. The conversion Table is as follows:

AMD Athlon CPU


  • AMD Athlon 1500+ = Actually runs at 1.33 GHz
  • AMD Athlon 1600+ = Actually runs at 1.40 GHz
  • AMD Athlon 1700+ = Actually runs at 1.47 GHz
  • AMD Athlon 1800+ = Actually runs at 1.53 GHz
  • AMD Athlon 1900+ = Actually runs at 1.60 GHz
  • AMD Athlon 2000+ = Actually runs at 1.67 GHz
  • AMD Athlon 2100+ = Actually runs at 1.73 GHz
  • AMD Athlon 2200+ = Actually runs at 1.80 GHz
  • AMD Athlon 2400+ = Actually runs at 1.93 GHz
  • AMD Athlon 2500+ = Actually runs at 1.833 GHz
  • AMD Athlon 2600+ = Actually runs at 2.133 GHz
  • AMD Athlon 2700+ = Actually runs at 2.17 GHz
  • AMD Athlon 2800+ = Actually runs at 2.083 GHz
  • AMD Athlon 3000+ = Actually runs at 2.167 GHz
  • AMD Athlon 3200+ = Actually runs at 2.20 GHz
  • AMD Octeron = Low-Power Quad-Core
  • AMD Phenom x3 = Tri-Core
  • AMD Phenom x4 = Quad-Core
  • AMD Phenom II x4 = Quad-Core

You may wish to purchase a high end AMD64/EM64T (AMD and Intel, respectively) processor, which provides support for 64-bit operating systems (eg. Windows XP Professional 64-bit Edition). Most 64-bit processors are backwards-compatible with 32-bit operating systems and applications.. Given the huge expansion in addressable memory afforded by the transition from 32-bit to 64-bit (a 32-bit address space tops out at 4 Gigabytes while a 64-bit space encompasses 17,179,869,184 Gigabytes) 64 bit operating systems are the wave of the future. For the time being though, given the enormous quantity of 32-bit software out there, and the limited advantages of 64-bit as of yet, the move is likely to be a slow one.

CPU cooling[edit]

A processor heat sink with fan

CPU cooling is very important, a less than average CPU temperature prolongs CPU life (up to more than 10 years). On the other hand, high CPU temperatures can cause unreliable operation, such as computer freezes, or slow operation. Extremely high temperatures can cause immediate CPU destruction by melting the materials in the chip and changing the physical shape of the sensitive transistors on the CPU. Because of this, never switch on the computer if your CPU has no cooling at all, or if the cooling device power cable is not plugged into the motherboard. Turn your computer on "just to test whether my CPU works" and you'll likely find that the CPU fries in less than 2 seconds and you'll be off to buy a new one.

Most CPU installations use forced-air cooling, but convection cooling and water cooling and even oil submersion cooling are options. For traditional forced-air cooling, the heat sink and fan (HSF) included with most retail CPUs is usually sufficient to cool the CPU at stock speed. Overclockers might want to use a more powerful after-market fan, or even try water cooling to combat the increased heat engendered by overclocking.

Many retail HSF units have a thermal pad installed. These transfer heat from the CPU to the fan, helping to diffuse the heat created by the CPU. This pad is usable only once. If you wish to remove the fan from another CPU so that you can use it on your new one, or need to take it off for some reason, you will need to remove the pad and apply a thermal paste or another thermal pad. Note that some of the cheaper pads can melt in unexpected heat and may cause problems and potentially even damage if you are overclocking. In either case, thermal paste is usually more effective, just harder to apply. If you plan to do any high performance computing, or remove and replace the HSF often, thermal paste is suggested. If you are planning on a long term installation a thermal pad is often best. Using one or the other is essential to ensure consistent heat conduction between the CPU and heatsink.

Electrically insulating thermal pastes made up of silicon are the cheapest and safest, but silver-based thermal pastes sometimes perform better and carbon-based ones perform better still. When applied improperly both can be conductive, causing electrical shorts upon contact with the motherboard. A thin properly-applied layer will usually prevent this problem, although some pastes can become runny when they get hot. Users should also beware that many "silver" thermal pastes do not actually contain any silver metal.

For quiet operation, start with a low-heat (low wattage) CPU. You can also underclock your CPU, giving up some unneeded performance for some peace and quiet. Another option is to choose a large copper heat sink with an open fin pattern. However, true fanless operation is difficult to achieve in most case designs. You can position a case fan to blow across the heat sink, or mount a fan on the heat sink. With either choice, a large and slow fan will give better airflow and less noise than a small and fast fan.

Some low-noise CPU cooling fans require special mounting hardware on the motherboard. Be sure that the cooling systems you choose are compatible with your motherboard.






Multi-cores and Hyperthreading (HT)[edit]

Consumer multi-core processors are a fairly new innovation built by both major consumer processor manufactures, AMD's Athlon 64 X2 and Intel's Core Duo and Core 2 Duo. Intel also has a quad-core model, the Core 2 Quad, and AMD supports motherboard architectures that allow for two individual dual-core processors.

Multi-Core Advantages[edit]

  • Multitasking: Each processor has two processing centres (cores) for (a theoretical maximum of) twice the operating power and for better multitasking. Major advantages of dual core processors are evident when doing heavy multitasking, such as encoding video and playing video games at the same time.
  • Application Support: Newer applications are being written to take advantage of this technology by using a technique known as multithreading.
  • Power Saving: A dual core processor (specifically the Intel Core Duo) usually has the ability to turn off one of its cores when application demand is low to save power.

Disadvantages[edit]

  • Support: Most older programs do not support multithreading and will likely not run any faster on dual core CPUs than on a single-core of the same clock speed.
  • Price: Dual core processors are also significantly more expensive than their single core brethren, due to the fact that they are essentially two combined CPUs.

Multi-core Future?[edit]

The trend of CPU design is turning towards multi-core processors (one notable example is the Cell processor in the Sony Playstation 3, with 7 active cores), instead of higher clock speed, and more and more programs being built to take advantage of multiple cores, so prices are sure to decrease, and the availability and usability of multi-core processors is sure to continue to increase.

Hyper-threading (HT)[edit]

Hyper-threading is a technology which allows a single core processor to simulate having two cores, often giving a performance boost when running several programs at once. It requires motherboards, chipsets, and operating systems supporting Hyper-Threading technology. It also suffers similar disadvantages as Dual-Core Processors of support and price. In addition, under certain conditions, the additional overhead of Hyper-Threading may even decrease performance of some applications.

CPU Specifications[edit]

Now, one of the most common mistakes of choosing a CPU is ignoring the fine print of CPU specifications and relying completely on the clock speed. CPU specifications are written in full like this:

Intel Pentium 4 3.2GHz LGA775 FSB800 HT L2-2MB
Model: Intel Pentium 4
Clock Speed: 3.2GHz (=3200MHz)
Interface: Land Grid Array 775
L2-Cache: 2MB (=2048 kB)
Frontside Bus: 800 Mhz
Other Spec: HT technology

The consumer logic for processor speed may be misleading because many consumers think that clock speed, which is measured in gigahertz (GHz) or megahertz (MHz) is equal to system speed. While a higher clock speed will allow the CPU to do cycles more frequently and it does have a fundamental effect on speed, it is not the sole factor as the number of calculations (and data handled) per cycle is different for each different manufacturer and model. In addition, other factors, such as FSB speed will affect the CPU's ability to process data efficiently. If the FSB and RAM aren't able to get data to the CPU as quickly as the CPU can process it, then the CPU will speed some of the cycles doing nothing. A cycle during which the CPU does nothing is called a wasted cycle.

Intel classifies its x86 CPUs using a series of numbers. 3xx, 4xx, 5xx, 6xx and 7xx of which 7xx being the highest end products. Generally, the higher the number, the faster the CPU and the more expensive. Usually, models and ratings correspond.

  • 3xx Series: Intel Celeron (L2-128KB)
  • 4xx Series: Intel Celeron D (L2-512KB)
  • 5xx Series: Intel Pentium 4 / Celeron D (L2-1MB)
  • 6xx Series: Intel Pentium 4 / Pentium 4 XE (L2-2MB)
  • 7xx Series: Intel Pentium 4 XE
  • 8xx Series: Intel Pentium D
  • 9xx Series: Intel Pentium D

The number followed by suffix J signifies XD technology. E.g. Intel Pentium 4 3.0GHz L2-1MB with HT --> Intel Pentium 4 530J



You may wish to purchase a high end AMD64/EM64T (AMD and Intel, respectively) processor, which provides support for 64-bit operating system (eg. Windows XP Professional 64-bit Edition). A 64-bit system is very efficient in handling large amounts of RAM. In a 32-bit system, efficiency drops beyond about 512 to 864MB of RAM, and becomes significantly less efficient beyond 4GB of RAM. Most processors for gaming are about 2.8Ghz-3.2Ghz.

A 64-bit processor is not necessary yet as most applications run on the 32-bit system (which the 64-bit systems are backwards-compatible with). However, there is no doubt that the 32-bit system would gradually be replaced by the 64-bit system when the prices fall over a few years time. It is unlikely that the 64-bit system would completely replace the 32-bit system within 5 years but Linux users might find a great improvement in speed when a 64-bit processor is used. Note that there are processors that can run 32-bit code and 64-bit code; they are becoming very popular and may prove a very wise investment for the future of computing, which will invariably be 64-bit. They are more expensive than processors that only run 32-bit code; however, it should be noted that AMD has invested in this market faithfully, and therefore all of their processors, even their low-end Sempron line, runs both types of code. The same cannot be said for the Intel family, which only includes such technology in their high-end server processors and the new Core 2 series.

Smaller processors are generally preferred for overclocking, as they run cooler, and can achieve higher clock speeds. Retail CPU's come in a package containing a HSF (Heat Sink Fan), instructions, and a warranty, often 3 years. OEM CPUs do not include these.

The current CPU speeds and advantages change frequently, so for up-to-date comparisons, you may want to check a website that specializes in Hardware reviews, such as Tom's Hardware Guide or Anandtech.

CPU cooling[edit]

CPU cooling is very important and should not be overlooked. A less than average CPU temperature prolongs CPU life (up to more than 10 years). On the other hand, high CPU temperatures can cause unreliable operation, such as computer freezes, or slow operation. Extremely high temperatures can cause immediate CPU destruction by melting the materials in the chip and changing the physical shape of the sensitive transistors on the CPU. Because of this, never switch on the computer if your CPU has no cooling at all. It is an extremely stupid thing to do. Turn your computer on "just to test whether my CPU works" and you may find that the CPU fries in less than 5 seconds and you will be off to buy a new one.

Most CPU installations use forced-air cooling, but convection cooling and water cooling are also options. For traditional forced-air cooling, the heatsink and fan (HSF) included in most retail CPUs is usually sufficient to cool the CPU at stock speed. Overclockers might want to use a more powerful aftermarket fan, or even try water cooling because they need additional cooling ability given the increased heat of overclocking.

HSFs with decent performance are usually copper-based. The cooling effect is enhanced if the HSF has heatpipes. Silent (i.e. fanless) HSFs are there to provide users with nearly silent cooling.

Many retail HSF units have a thermal pad installed. These transfer heat from the CPU to the fan, helping to diffuse the heat created by the CPU. This pad is usable only once. If you wish to remove the fan from another CPU so that you can use it on your new one, or need to take it off for some reason, you will need to remove it, and apply a thermal paste or another thermal pad. Note that some of the cheaper pads can melt in unexpected heat and may cause problems and potentially even damage if you are overclocking. In either case, thermal paste is usually more effective, just harder to apply. If you plan to do any high performance computing, or removing and replacing the HSF, often thermal paste is suggested. If you are planning a long term installation a thermal pad is suggested. Non-conductive thermal pastes made up of silicon are the cheapest and safest.

Silver-based thermal pastes sometimes perform better than normal thermal pastes, and carbon-based ones perform better still. When applied improperly both can be conductive, causing electrical shorts upon contact with the motherboard. A thin properly-applied layer will usually prevent this problem, although some pastes can become runny when they get hot. Users should also beware that many "silver" thermal pastes do not actually contain any silver metal.

For quiet operation, start with a low-heat (low number of watts) CPU. Processors made by VIA, such as the VIA C3, tend to produce low amounts of heat. The Pentium M gives performance that is on par with many of the desktop processors, but gives off more heat than the Via processors. You can also underclock your CPU, giving up some unneeded performance for some peace and quiet. Another option is to choose a large copper heat sink with an open fin pattern. However, true fanless operation is difficult to achieve in most case designs. You can position a case fan to blow across the heat sink, or mount a fan on the heat sink. With either choice of fan placement, choose a large and slow fan over a small and fast fan to decrease noise and increase air flow.

Some low-noise CPU cooling fans require special mounting hardware on the motherboard. Be sure that the cooling fan you choose is compatible with your motherboard.