The “brain” of your computer, the CPU is responsible for three main things: calculating, writing/reading to memory, and decision making based on instruction sets and thus moving from one instruction set to another based on these decisions. I’m not going into the technical part of CPUs, as they are extremely complex and would be useless information you don’t need to know. Just to say that the CPU is what brings all the other parts of the system together to create the system with tangible results :)

The FSB of a CPU is one part of a two-part step for calculating a CPUs clock frequency (speed). The other is the multiplier, and you get the clock speed simply by multiplying the FSB by the multiplier. So, on my XP1800+ which has a FSB of 133Mhz, and a multiplier of 11.5, the clock speed is 133 x 11.5 = 1533 MHz. Now, this is just a core value.

On AthlonXP processors, the FSB is double-pumped, so my XP1800+ has 133 dual-pumped to to give effectivly a 266FSB. The FSB found on the Athlons XP’s ideally suits DDR RAM.

AMD no longer sell processors which use the FSB per se, instead they have progressed to the Hyper-Transport (HT) bus instead.

The HTT bus is technically different to the FSB Bus, however as they complete essentially the same task, this article will refer to both as the FSB to avoid confusion.

Intel’s latest Celeron and Pentium 4 processors (and also newer Pentium M, Core and Core 2 processors) are quad-pumped. So, I also own a Pentium 4 1.8 GHz. This has a stock FSB of 100Mhz, but is quad-pumped to give an effective FSB of 400. This FSB is best matched with Rambus RAM or Dual Channel DDR RAM.

Intel currently produces 133Mhz and 200Mhz FSB CPUs (effective 533 and 800 FSB respectively).

Specifically L2 cache. Cache is where commonly-used instructions and data are held, and L2 cache is where instructions that may be used again shortly are held. CPU manufacturers have to hold the balance between size of cache and price of processor well. Cache speed goes in this order, fastest to slowest: L1 – L2 – L3 – System memory. L2 cache sizes range from 64KB on Celerons to several megabytes on Intel’s Itanium. The higher the cache, generally the faster a CPU will be at a given speed. The cost of having on-die cache increases as you increase the amount as well, one reason why XeonMP CPUs are so costly, and also the price difference between 512kb Pentium 4’s and 128kb Celerons.

An instruction set, stands for Streaming SIMD Extensions. Streaming meaning graphics and sound media. This is a set of instructions that help with graphics “projects”, increasing performance a lot like MMX did a few years ago. SSE2 is an extension of the original SSE instruction set. Again SSE3 is an extension of the preceeding two versions, however SSE3 was mainly invented to make up for the lost impact which SSE2 initially failed to create.

The number of pins on the bottom of the CPU, which are the things which make physical contact with the motherboard, in the form of a socket, or a slot on some older CPUs.

The actual clockspeed (e.g 1.8GHz in the aformention Pentium 4) is NOT representative of a modern processor’s performance. Due to underlying techinical differences, different processors often perform different amounts of work per cycle, and thus clockspeed is not a viable method of comparing processor performance apart from within a particular processor model.

Three major manufacturers: AMD, Intel and VIA

lower-clock speed than Intel, thanks to the higher instruction per clock cycle count on their Athlon XP range of CPUs, typically a bit lower tech than Intel, trailing marginally behind in terms of technology (eg. multiple threads on die, lower onboard L2 cache until recently, lower Front Side Bus (FSB) thanks to dual-pumped FSB compared to Intel’s quad-pumped FSB). Makes up for it in pricing of processors, typically at the low-mid range about half the price of Intel. AMD’s new 64-bit processors look very promising in terms of performance. AMD is the bang-for-buck king.

However due to the quickly changing nature of the PC hardware landscape, the recent Intel Core and Core 2 processors have been able to feature even higher performance than AMD at lower clock speeds at the time of writing (19/11/06)

Athlon 64 Dual Core up to 5200+. The Dual Core (X2) comes in both the 939 Socket, and Socket 940 (AM2).
The FX Series, in both 939 and AM2 are targeted at the higher end, gamers, overclockers etc, and contain unlocked multipliers. The FX range at release always have the highest speed available, starting at the FX57, through to the FX62.
939: Utilises standard DDR400 ram
AM2: Utilises the faster and newer DDR2, which runs at the maximum speed of DDR2-800

Quad Core

Wikipedia Article

Intel prides itself on having the fastest, and the fastest they do have. The latest Core and Core 2 processors have peformed well above their AMD counterparts at the time of writing (19/11/06). Intel also incorporates the latest technology into their processors (usually invented by themselves), such as Hyper-Threading (HTT) and iSSE2/3, but this of course comes at a price - the CPUs are a lot more expensive than their AMD counterparts.

Core 2: The Main foray into the competitive Dual Core game with its incredibly powerful Conroe series. Built on a 65Nm Process, and start at the e6300 series with only 2MB of the 4MB/L2 Cache enabled. The Upper range of the Conroes the e6600 and the e6700 utilise the full 4MB/ of Level 2 Cache.

Popular with mainstream overclockers and power users for its low heat output and although locked multipliers, handle high frequency fsb modification.

Kentsfield- Quad Core Processor

Wikipedia Article

Lagging behind AMD and Intel in terms of performance, however their main target is for power efficiency and low heat output, and is favoured in some areas by system integrators due to this fact (such as in-car computing). Due to this, their processors also include a hardware accellerated cryptography model which increases the security of the system when the software leverages the avaliable cryptographic power.

C3/700 MHz – 1.0GHz (super silent, requiring a heatsink but no fan to operate [1.0GHz model comes with a small fan], 64kb L2 cache)

C7/1.5 - 2GHz (Also low power usage and output. Higher heat output than the C3, however performance is vastly increased from the C3. Supports iSSE2 and iSSE3)