Central Processing Unit (CPU), or sometimes simply processor, is the component in a digital computer capable of executing a program. It interprets computer program instructions and processes data. CPUs provide the fundamental digital computer trait of programmability, and are one of the necessary components found in computers of any era, along with primary storage and input/output facilities. A CPU that is manufactured as a single integrated circuit is usually known as a microprocessor. Beginning in the mid-1970s, microprocessors of ever-increasing complexity and power gradually supplanted other designs, and today the term "CPU" is usually applied to some type of microprocessor.

The fundamental operation of most CPUs, regardless of the physical form they take, is to execute a sequence of stored instructions called a program. Discussed here are devices that conform to the common von Neumann architecture. The program is represented by a series of numbers that are kept in some kind of computer memory. There are four steps that nearly all von Neumann CPUs use in their operation: fetch, decode, execute, and writeback.

Diagram showing how one MIPS32 instruction is decoded. (MIPS Technologies 2005)

The first step, fetch, involves retrieving an instruction (which is represented by a number or sequence of numbers) from program memory. The location in program memory is determined by a program counter (PC), which stores a number that identifies the current position in the program. In other words, the program counter keeps track of the CPU's place in the current program. After an instruction is fetched, the PC is incremented by the length of the instruction word in terms of memory units. Often the instruction to be fetched must be retrieved from relatively slow memory, causing the CPU to stall while waiting for the instruction to be returned. This issue is largely addressed in modern processors by caches and pipeline architectures.

The instruction that the CPU fetches from memory is used to determine what the CPU is to do. In the decode step, the instruction is broken up into parts that have significance to other portions of the CPU. The way in which the numerical instruction value is interpreted is defined by the CPU's instruction set architecture (ISA). Often, one group of numbers in the instruction, called the opcode, indicates which operation to perform. The remaining parts of the number usually provide information required for that instruction, such as operands for an additional operation. Such operands may be given as a constant value (called an immediate value), or as a place to locate a value: a register or a memory address, as determined by some addressing mode. In older designs the portions of the CPU responsible for instruction decoding were unchangeable hardware devices. However, in more abstract and complicated CPUs and ISAs, a microprogram is often used to assist in translating instructions into various configuration signals for the CPU. This microprogram is sometimes rewritable so that it can be modified to change the way the CPU decodes instructions even after it has been manufactured.

Block diagram of a simple CPU

After the fetch and decode steps, the execute step is performed. During this step, various portions of the CPU are connected so they can perform the desired operation. If, for instance, an addition operation was requested, an arithmetic logic unit (ALU) will be connected to a set of inputs and a set of outputs. The inputs provide the numbers to be added, and the outputs will contain the final sum. The ALU contains the circuitry to perform simple arithmetic and logical operations on the inputs (like addition and bitwise operations). If the addition operation produces a result too large for the CPU to handle, an arithmetic overflow flag in a flags register may also be set (see the discussion of integer range below).

The final step, writeback, simply "writes back" the results of the execute step to some form of memory. Very often the results are written to some internal CPU register for quick access by subsequent instructions. In other cases results may be written to slower, but cheaper and larger, main memory. Some types of instructions manipulate the program counter rather than directly produce result data. These are generally called "jumps" and facilitate behavior like loops, conditional program execution (through the use of a conditional jump), and functions in programs. Many instructions will also change the state of digits in a "flags" register. These flags can be used to influence how a program behaves, since they often indicate the outcome of various operations. For example, one type of "compare" instruction considers two values and sets a number in the flags register according to which one is greater. This flag could then be used by a later jump instruction to determine program flow.

After the execution of the instruction and writeback of the resulting data, the entire process repeats, with the next instruction cycle normally fetching the next-in-sequence instruction because of the incremented value in the program counter. If the completed instruction was a jump, the program counter will be modified to contain the address of the instruction that was jumped to, and program execution continues normally. In more complex CPUs than the one described here, multiple instructions can be fetched, decoded, and executed simultaneously. This section describes what is generally referred to as the "Classic RISC pipeline," which in fact is quite common among the simple CPUs used in many electronic devices (often called microcontrollers).

Clock signal

As CPU's advanced, they became faster and the motherboards could not keep up. This resulted in the development of clock doubling. The 486DX2 66 has a clock rate of 33MHz but by using the clock multiplier you can adjust the clock signal to higher or lower than what it is designed for. The term used for increasing the voltage beyond the designed rating is Overclocking. Overclocking is generally used in CPU's and RAM. By changing the multiplier on the 486DX2 33MHz CPU to a multiplier of 2.0x, you increase the processor speed to 66MHz.

Most CPUs, and indeed most sequential logic devices, are synchronous in nature. That is, they are designed and operate on assumptions about a synchronization signal. This signal, known as a clock signal, usually takes the form of a periodic square wave. By calculating the maximum time that electrical signals can move in various branches of a CPU's many circuits, the designers can select an appropriate period for the clock signal.

The clock rate is the fundamental rate in cycles per second (measured in hertz) at which a computer performs its most basic operations such as adding two numbers or transferring a value from one processor register to another. Different chips on the motherboard may have different clock rates. Usually when referring to a computer, the term "clock rate" is used to refer to the speed of the CPU.

The clock rate of a CPU is normally determined by the frequency of an oscillator crystal. The first commercial PC, the Altair (by MITS), used an Intel 8080 CPU with a clock rate of 2 MHz. The original IBM PC (c. 1981) had a clock rate of 4.77 MHz (4,770,000 cycles/second). In 1995, Intel's Pentium chip ran at 100 MHz (100 million cycles/second), and in 2002, an Intel Pentium 4 model was introduced as the first CPU with a clock rate of 3 GHz (three billion cycles/second).

Processor Registers

In computer architecture, a processor register is a small amount of very fast computer memory used to speed the execution of computer programs by providing quick access to frequently used values—typically, these values are involved in multiple expression evaluations occurring within a small region on the program. Most, but not all, modern computer architectures operate on the principle of moving data from main memory into registers, operating on them, then moving the result back into main memory—a so-called load-store architecture.

Processor registers are at the top of the memory hierarchy, and provide the fastest way for a CPU to access data. The term is often used to refer only to the group of registers that are directly encoded as part of an instruction, as defined by the instruction set. More properly, these are called the "architectural registers". For instance, the x86 instruction set defines a set of eight 32-bit registers, but a CPU that implements the x86 instruction set will contain many more registers than just these eight.

Address Bus

An address bus is a computer bus, used by CPUs or DMA-capable units for communicating the physical addresses of computer memory elements/locations that the requesting unit wants to access (read/write).

The width of an address bus, along with the size of addressable memory elements, determines how much memory can be accessed. For example, a 16-bit wide address bus (commonly used in the 8-bit processors of the 1970s and early 1980s) reaches across 2¹⁶ = 65,536 = 64Ki memory locations, whereas a 32-bit address bus (common in today's PC processors) can address 2³² = 4,294,967,296 = 4Gi locations.

In most microcomputers the addressable elements are 8-bit bytes (so a "Ki" in that case is equal to a "KiB", i.e. a kibibyte), while there are also many examples of computers with larger "chunks" of data as their minimum physically addressable elements, notably some mainframes and some microprocessors.

Since the very first CPU both Intel and AMD have been creating several different sockets and slots to be used by their processors. In this tutorial we will list all socket and slot types released to date with their respective pinouts, also giving examples of compatible CPUs.

In the beginning, a CPU socket was compatible with just one kind of processor. This story changed with the launching of the 486 processor and the massive use of ZIF (Zero Insertion Force) socket, also known as LIF (Low Insertion Force), which has a lever that installs and removes the CPU from the socket without the need of the user or the technician to press the CPU down in order to be installed on the socket. The use of this socket lowered a lot the chances of breaking or bending the CPU pins during its installation or removal. The use of the same pinout by more than one processor allowed the user or the technician to install different processor models on the same motherboard by just removing the old CPU and installing the new one. Of course the motherboard needed to be compatible with the new CPU being installed and also properly configured.

Since then both Intel and AMD have been developing a series of sockets and slots to be used by their CPUs.

The socket created to be used together with the very first 486 processor wasn’t ZIF and didn’t allow you to replace the CPU with a different processor model. Even though this socket didn’t have an official name, let’s call it socket 0. After socket 0 Intel released socket 1, which had the same pinout of socket 0 with the addition of a key pin. It also adopted ZIF standard, allowing the installation of several different processor types on the same socket (i.e. on the same motherboard). Other socket standards were released for the 486 family after socket 1 – socket 2, socket 3 and socket 6 – in order to increase the number of CPU models that could be installed on the CPU socket. Thus socket 2 accepts the same CPUs accepted by socket 1 plus some more models, and so on. Even though socket 6 was designed, it was never used. Thus we usually call the pinout used by 486-class processors as “socket 3”. Intel called “overdrive” the possibility of a socket to accept more than one CPU model. Intel also adopted this name on CPUs that used a pinout from an older CPU, in order to allow it to be installed on an older motherboard.

The first Pentium processors (60 MHz and 66 MHz) used a pinout standard called socket 4, which was fed with 5 V. Pentium processors from 75 MHz on were fed with 3.3 V and thus required a new socket, called socket 5, which was incompatible with socket 4 (a Pentium-60 couldn’t be installed on socket 5 and a Pentium-100 couldn’t be installed on socket 4, for example). Socket 7 uses the same pinout as socket 5 with the addition of one key pin, accepting the same processors accepted by socket 5 plus new CPUs, especially CPUs designed by competing companies (the real difference between socket 5 and socket 7 is that while socket 5 always fed the CPU with 3.3 V, socket 7 allowed the CPU to be fed with a different voltage level, like 3.5 V or 2.8 V, for example). Super 7 socket is a socket 7 capable of running up to 100 MHz, used by AMD CPUs. We usually call the Pentium Classic and compatible CPUs pinout as “socket 7”.

As you may notice, sockets and pinouts at this stage were very confusing, as a given processor could be installed on different socket types. A 486DX-33 could be installed on sockets 0, 1, 2, 3 and, if it were released, 6.

For the next CPUs manufacturers followed a simpler scheme, where each CPU could be installed only on just one socket type.

Intel Processors

CPU Packages
DIPP - 8088, 8086, 80286
PGA - 80286, 386, 486
SPGA (Staggered Pin Grid Array) Pentium, Pentium Pro, K7, 686 Diagonal Pattern
SEC (Single Edge Contact)
FCPGA (Flip Chip Pin Grid Array) PPGA is FCPGA upside down
Micro FCPGA - P4

Intel 4004 - 16 pin DIP (Dual In-Line Pin) 1971-1974

The Intel 4004 is a 4-bit central processing unit (CPU) released by Intel Corporation in 1971. Intel purports it is the world's first commercial microprocessor. The 4004 employed a 10 μm silicon-gate PMOS technology and could execute approximately 60,000 instructions per second.

Intel 8086 40 pin DIP (Dual In-Line Pin) 1978-Present

The 8086 is a 16-bit microprocessor chip designed by Intel in 1978, which gave rise to the x86 architecture. Intel 8088, released in 1979, was essentially the same chip, but with an external 8-bit data bus (allowing the use of cheaper and fewer supporting logic chips), and is notable as the processor used in the original IBM PC.

Intel 80186 68 pin PLCC (Plastic Leader Chip Carrier) 1982-Present

The 80186 is a microprocessor that was developed by Intel circa 1982. The 80186 was an improvement on the Intel 8086 and Intel 8088. As with the 8086, it had a 16-bit external bus and was also available as the Intel 80188, with an 8-bit external data bus. The initial clock rate of the 80186 and 80188 was 6 MHz, but due to more hardware (in place of microcode) some of the individual instructions ran 10-20 times faster than on an 8086 at the same clock frequency. On the average, it ran at 1 million instructions per second. They were generally used as embedded processors (roughly comparable to microcontrollers). They were not used in many personal computers, but there were some notable exceptions: the Mindset, the Siemens PC-D (the first DOS PC line of Siemens, with MSDOS v2.11), the Compis (a Swedish school computer), the RM Nimbus (a British school computer), the Unisys ICON (a Canadian school computer), the HP 200lx, the Tandy 2000 desktop (a somewhat PC-compatible workstation featuring particularly sharp graphics for its day) and the Philips "Yes". Another British computer manufacturer, Acorn, created a plug-in second processor that contained the 80186 chip along with assorted support chips and 512 KiB of RAM – hence the Master 512 system. One major function of the 80186/80188 series was to reduce the number of chips required by including features such as a DMA controller, interrupt controller, timers, and chip select logic.

Intel 80286 68 pin PLCC (Plastic Leader Chip Carrier) 1982-1986

The Intel's 286, introduced on February 1, 1982, (originally named 80286, and also called iAPX 286 in the programmer's manual) was an x86 16-bit microprocessor with 134,000 transistors. After the 6 and 8 MHz initial releases, it was subsequently scaled up to 12.5 MHz. (AMD and Harris later pushed the architecture to speeds as high as 20 MHz and 25 MHz, respectively.) It was widely used in IBM PC compatible computers during the mid 1980s to early 1990s. On average, the 80286 had a speed of about 0.21 million instructions per clock. The 6 MHz model operated at 0.9 MIPS, the 10 MHz model at 1.5 MIPS, and the 12 MHZ model at 2.66 MIPs.

Intel 80386 68 pin PLCC (Plastic Leader Chip Carrier) 1986-Sep. 2007

The Intel386 is a microprocessor which was used as the central processing unit (CPU) of many personal computers from 1986 until 2007. During its design phase the processor was code-named simply "P3", the third-generation processor in the x86 line, but is normally referred to as either i386 or just 386. The 80386 operated at about 5 million instructions per second (MIPS) to 11.4 MIPS for the 33 MHz model. It was the first x86 processor to have a 32-bit architecture, with a basic programming model that has remained virtually unchanged for over twenty years and remains completely backward compatible. Successively newer implementations of this same architecture have become literally several hundred times faster than the original i386 chip during these years.

Socket 1 - 80486

Socket 1 was the second of a series of standard sockets created by Intel into which various x86 microprocessors were plugged. It was an upgrade to Intel's first standard PGA socket and the first with an official designation. Socket 1 was intended as a 486 upgrade socket, and added one extra pin to prevent upgrade chips from being inserted into the older socket.

Socket 1 was a 169 pin LIF/ZIF PGA (17x17) socket suitable for the 5v, 16-33 MHz 486 SX, 486 DX, 486 DX2 and DX4 Overdrive processors.

Socket 2 - 80486

Socket 2 was one of the series of standard sockets into which various x86 microprocessors were plugged. It was an updated Socket 1 with added support for Pentium Overdrive processors.

It was a 238 pin LIF/ZIF PGA (19x19) socket suitable for the 5v, 25-50MHz 486 SX, 486 DX, 486 DX2, 486 DX4, DX4 OverDrive and 63/83MHz Pentium OverDrive processors.

Socket 3 - 80486 (3.3 V and 5 V) and compatibles

Intel's Socket 3 was a type of CPU socket into which various x86 microprocessors were inserted. It was commonly found alongside a secondary socket designed for a math coprocessor chip, in this case the 487. Socket 3 resulted from Intel's creation of lower voltage microprocessors. An upgrade to Socket 2, it rearranged the pin layout and omitted one pin so that 3V processors could not be plugged into older 5V only sockets.

Socket 3 was a 237 pin LIF/ZIF PGA (19x19) socket suitable for the 3.3V and 5V, 25-50 MHz 486 SX, 486 DX, 486 DX2, 486 DX4, 486 OverDrive and Pentium OverDrive processors.

Socket 4 - Intel Pentium 60/66 MHz

Socket 4, presented in 1993, was the first CPU socket designed for the early Pentium processors. Socket 4 was the only 5 volt Pentium socket. After socket 4 Intel switched to the 3.3 volt powered socket 5. Socket 4 does support a special Pentium OverDrive, which allows running at 120 MHz (for the 60 MHz Pentium) or 133 MHz (for the 66 MHz Pentium).

Socket 5 - Intel Pentium 75-133 MHz; AMD K5; IDT WinChip C6, WinChip 2

Socket 5 was created for the second generation of Intel Pentium processors operating at speeds from 75 to 133 MHz as well as certain Pentium OverDrive processors. Pentium MMX processors are not compatible with Socket 5. Consisting of 320 pins,Core Voltage 3.3V ,this was the first socket to use a staggered pin grid array, or SPGA. This design allowed the chip's pins to be spaced closer together. Socket 5 was replaced by Socket 7.

Socket 6 - Designed but not used

Socket 6 was a 486-generation CPU socket, a slightly modified version of the more common Socket 3.

Intel designed this new standard near the end of the 80486's market life,so few motherboards were produced that used it, especially as the Socket 3 standard was already sufficient.

Specifications:

• 3.3volt electrical output

• supported the Intel 80486DX4 and Pentium OverDrive

• 235 pin ZIF socket

Socket 7 - Intel Pentium, Pentium MMX; AMD K6

Socket 7 is a physical and electrical specification for an x86-style CPU socket on a personal computer motherboard. The socket supersedes the earlier Socket 5, and accepts Pentium microprocessors manufactured by Intel, as well as compatibles made by Cyrix/IBM, AMD, IDT and others.

Socket 7 was the only socket that accepted such a wide range of CPUs, (range of speeds and CPUs from different manufacturers).

Differences between Socket 5 and Socket 7 are that Socket 7 has an extra pin and is designed to provide dual split rail voltage, as opposed to Socket 5's single voltage. (However, not all motherboard manufacturers supported the dual voltage on their boards initially.) Socket 7 is backwards compatible; a Socket 5 CPU can be placed in a Socket 7 motherboard.

Supported processors include: 2.5V - 3.5V Pentium 75-200 MHz and Pentium MMX 166-233 MHz, AMD K5 through K6, Cyrix 6x86 (and MX) P120 - P233, the entire IDT WinChip series (180-250 MHz) and the Rise Technology MP6 family.

Socket 7 uses a SPGA socket, either a 296 pin LIF arranged at 37x37 (which is very rare), or the much more common 321 pin ZIF arranged at 19x19.

An extension, Super Socket 7, was designed to allow the AMD K6-2 and AMD K6-III chips to operate at a higher clock rate and use AGP.

Super Socket 7

The Super Socket 7, also referred to as Super 7, is an extension of the Socket 7 ZIF socket specification. It features a 100 MHz front-side bus, support for the AGP bus, and a SPGA package. Super Socket was used by AMD K6-2 and K6-III processors, and some of the final Cyrix M-II processors. It is backward compatible with Socket 7, meaning a Socket 7 CPU can be used with a Super Socket 7 motherboard, but a Super Socket 7 CPU cannot operate at full speed in a Socket 7 motherboard. Socket 5 CPUs are pin-compatible with Super Socket 7, but not all motherboards designed for Super Socket 7 supported the voltages needed for Socket 5 CPUs.

While AMD had previously always used Intel sockets for their processors, Socket 7 was the last one for which AMD retained legal rights. Intel had hoped by discontinuing Socket 7 development and moving to Slot 1 that AMD would be left with an outdated platform, making their processors non-competitive. By extending the FSB from 66 to 100 MHz, Super Socket 7 gave AMD the stopgap solution they needed while developing their own independent motherboard infrastructure, Slot A.

While the architecture was cheap, and served the intended purpose, many of the third party chipsets provided by VIA, SiS, and others, were of low quality, especially in regard to the AGP implementations. The reputation AMD gained for buggy incompatible motherboards lingered, and though an internal program in VIA to raise standards for their chipsets had begun to show results in the K6-III and Pentium III era, AMD put in place a quality assurance program for the Athlon processor.

Socket 8 - Intel Pentium Pro

Socket 8 CPU socket was used exclusively with the Intel Pentium Pro and Pentium II OverDrive computer processors. Intel discontinued socket 8 in favor of Slot 1 with the introduction of the Pentium II.

Socket 8 has a unique rectangular shape with 387 pins, a bus speed ranging from 66-75Mhz, voltage from 2.1 - 3.5V, and support for Pentium Pro chips from 150-200Mhz, Pentium Pro OverDrive CPUs, and Pentium II OverDrive CPUs. This unique shape is the only currently known arrangement of two geometries for pins. On one half of the socket, it was in the common rectangular grid, and on the other half had an offset arrangement like the socket 7.

Socket 370 - Intel Pentium III, Celeron; Cyrix III; VIA C3

Socket 370 is a common format of CPU socket first used by Intel for Pentium III and Celeron processors to replace the older Slot 1 CPU interface on personal computers. The "370" refers to the number of holes in the socket for CPU pins. Modern Socket 370 fittings are usually found on Mini-ITX motherboards and embedded systems.

Socket 370 was originally used for the Intel Celeron, but later became the socket/platform for the Coppermine and Tualatin Pentium III processors, as well as the Via-Cyrix Cyrix III, later renamed the VIA C3. Some motherboards that used Socket 370 support Intel processors in dual CPU configurations. Others allowed the use of a Socket 370 or Slot 1 CPU, although not at the same time.

The weight of a Socket 370 CPU cooler should not exceed 180 gram. Heavier coolers may result in damage to the die when the system is not properly handled.

Socket 423 - Intel Pentium 4¹ Willamette core

Socket 423 was a CPU socket used for the first Pentium 4 processors, based on the Willamette core. The socket was short-lived, as it became apparent that its electrical design proved inadequate for raising clock speed beyond 2.0 GHz. Intel produced chips using this socket for less than a year, from November 2000 to August 2001. It was replaced by Socket 478.

The "PowerLeap PL-P4/N" is a device developed in the form of a socket adapter allowing the use of socket 478 processors on the socket 423.

Along with the socket these CPUs use (and therefore the motherboards), there is another short lived and odd piece of hardware: the RAM. The type of RAM used on some of these motherboards is RDRAM. This type of RAM is now very expensive, ranging from $US54 for 128MB to $US214 for 512MB. These sticks of RAM also must be used in pairs similar to modern dual channel memory.

Socket 478 - Intel Pentium 4, Celeron, Pentium 4 Extreme Edition¹, Pentium M Socket N

In computing, Socket 478 is a type of CPU socket used for Intel's Pentium 4 and Celeron series CPUs. Socket 478 was phased out with the launch of LGA775.

Socket 478 has been used for all of the Northwood Pentium 4s and Celerons, the first Prescott Pentium 4s, and some Willamette Celerons and Pentium 4s. Socket 478 also supports newer Prescott Celeron Ds, and early Pentium 4 Extreme Edition processors with 2MB of L3 cache and some Core Duos. The socket was launched with the Northwood core to compete with AMD's 462-pin Socket A and their Athlon XP processors. Socket 478, which accommodates high and low-end processors, was also the replacement for Socket 423, a Willamette processor socket which remained in the market for only a short time.

Motherboards that use this socket support DDR, RDRAM, and in some cases SDRAM. However, the majority of boards are DDR based. Initial motherboards only supported RDRAM, however RDRAM is quite expensive, compared to DDR and SDRAM, and consumers demanded an alternative, thus DDR and SDRAM boards were made. Later revisions to chipsets that support Socket 478 added higher FSB speeds, higher DDR speeds, and support for dual channel DDR.

Like the previous Socket 423, Socket 478 is based on Intel's Quad Data Rate technology, with data transferring at four times the clock rate of its Front Side Bus. As such, the 400 MT/s bus was based on a 100 MHz clock signal, but was still able to provide 3.2GB/s of data to the chipset. At its release, no SDRAM product was capable of supporting so high a data rate, so Intel pushed forward RDRAM technology, with two channels of PC800 providing synchronous data capability. Poor consumer acceptance of expensive RDRAM lead Intel to release low-performance PC133-supporting chipsets, and finally DDR chipsets.

While the original 400 MT/s bus matched the data rate of PC3200, this bus speed was already outdated by the time PC3200 became available. Dual-channel memory was introduced on later chipsets, so that a matched pair of PC3200 modules was able to match the final 800 MT/s FSB.

The Celeron D is also available for Socket 478 and they are now the only CPU's still made for the socket. They use a quad-pumped 133MHz Bus Giving A 533MT/s FSB. They are available with 256KB L2 Cache and are built on the 90nm manufacturing process, using the Prescott Core.

Socket 478 is officially known by Intel as FC-PGA2 [1]. While the Intel mobile CPU's are available in 478-pin packages, they in fact only operate in a range of slightly differing sockets, Socket 479, Socket M, and Socket P, which are also incompatible with each other.

Socket 479 - Intel Pentium M and Celeron M

Socket 479 is the CPU socket for the Intel Pentium M, a mobile processor, normally used in laptops, as well as Tualatin-M Pentium III processors. The official naming by Intel is mFCPGA and mPGA479M. Despite the 479 in the name the Pentium M Processors for this socket use only 478 Pins.

It uses a different electrical pin-arrangement from socket 478, making it impossible to use a Pentium M in a normal 478 board, but yet the Pentium M fits mechanical in a Socket 478. For this reason Asus makes a drop-in board (CT-479) which lets you use socket 479 processors in selected Asus boards. Currently, the only chipsets for the Pentium M are the Intel 855GME, Intel 915GM and Intel 6300ESB.

Recently, Intel has released a new socket 479 with a revised pinout for its Core processor, called Socket M. This socket has the placement of one pin changed from the original Socket 479 in order to make the different processors incompatible in the incorrect socket. Socket M supports a 667 MT/s front side bus with the Intel 945GM chipset.

Micro-FCBGA - Intel Mobile Celeron, Core 2 Duo (mobile), Core Duo, Core Solo, Celeron M, Pentium III (mobile), Mobile Celeron

Micro-FCBGA is Intel's current BGA mounting method for mobile processors. It was introduced with the Coppermine Mobile Celeron and replaces the older BGA2 ball-grid-array mounting method used in the Coppermine Pentium III mobile CPUs. Micro-FCBGA has 479 balls that are 0.78 mm in diameter and arranged similarly to the pins in a pin grid array. The processor is affixed to the motherboard by soldering the balls to the motherboard. This allows for a much thinner CPU/interface profile than a pin grid array socket arrangement, but the micro-FCBGA chip is not removable from the motherboard. Micro-FCBGA is commonly used to mount ultra-low-voltage and low-voltage versions of mobile CPUs to the motherboard, likely due to its much thinner profile fitting better in the very thin sub-notebooks that those CPUs are usually used in. However, standard-voltage versions of the CPUs also occasionally use a micro-FCBGA interface instead of a pin grid array socket.

LGA 771 (also known as Socket 771 or Socket J) - Intel Xeon

Socket J, also known as LGA 771, is a CPU interface introduced by Intel in 2006. It is used in Intel's most recent DP-capable server processors, the Dual-Core Xeons codenamed "Dempsey" and "Woodcrest" and the Quad-Core Clovertown.

As its name implies, it is a land grid array with 771 contacts. The "J" in "Socket J" refers to the now-cancelled processor codenamed "Jayhawk", which was expected to debut alongside this interface. It is intended as a successor to Socket 604 and takes much of its design from LGA 775.

LGA 775 (also known as Socket 775 or Socket T) - Intel Pentium 4, Pentium D, Celeron D, Pentium Extreme Edition, Core 2 Duo, Core 2 Extreme, Celeron¹, Xeon 3000 series, Core 2 Quad

Socket T, also known as LGA775, is Intel's latest desktop CPU socket. LGA stands for Land Grid Array. The word "socket" is now a misnomer, because an LGA775 motherboard has no socket holes, instead it has 775 protruding pins which touch contact points on the underside of the processor (CPU).^[1]

The Prescott and Cedar Mill Pentium 4 cores, as well as the Smithfield and Presler Pentium D cores, currently use the LGA775 socket type. In July 2006, Intel released the desktop version of the Core 2 Duo (codenamed Conroe), which also uses this socket, as does the subsequent Core 2 Quad. Intel changed from Socket 478 to LGA775 because the new pin type offers better power distribution to the processor, allowing the front side bus to be raised to 1333 MT/s. The 'T' in Socket T was derived from the now cancelled Tejas core, which was to replace the Prescott core. As it is now the motherboard which has the pins, rather than the CPU, the risk of pins being bent is transferred from the CPU to the motherboard. The risk of bent pins is reduced because the pins are spring-loaded and locate onto a surface, rather than into a hole. Also, the CPU is pressed into place by a "load plate", rather than human fingers directly. The installing technician lifts the hinged "load plate", inserts the processor, closes the load plate over the top of the processor, and pushes down a locking lever. The pressure of the locking lever on the load plate clamps the processor's 775 gold contact points firmly down onto the motherboard's 775 pins, ensuring a good connection. The load plate only covers the edges of the top surface of the CPU; the center is free to make contact with the cooling mechanism placed on top of the CPU.

The force from the load plate ensures that the processor is completely level, giving the CPU's upper surface optimal contact with the heat sink or cold-water block fixed onto the top of the CPU to carry away the heat generated by the CPU. This socket also introduces a new method of connecting the heat dissipation interface to the chip surface and motherboard. With Socket T, the heat dissipation interface is connected directly to the motherboard on four points, compared with the two connections of the Socket 370 and the "clamshell" four-point connection of the Socket 478. This was done to avoid the reputed danger of the heatsinks/fans of pre-built computers falling off in transit. LGA775 was announced to have better heat dissipation properties than the Socket 478 it was designed to replace; but the Prescott core CPUs (in their early incarnations) ran much hotter than the previous Northwood-core Pentium 4 CPUs, and this initially neutralized the benefits of better heat transfer. However, modern Core 2 Duo processors run at lower temperatures than the Prescott CPUs they replace.

Socket 603 - Intel Xeon

Socket 603 is a motherboard socket for Intel's Xeon processor.

Socket 603 was designed by Intel as a Zero Insertion Force socket intended for workstations and server platforms. It contains 603 contacts arrayed about the center of the socket, each contact has a 1.96mm pitch with regular pin array, to mate with a 603-pin processor package. Intel's design notes distinguish Socket 603 from Socket 604 as low cost, low risk, robust, high volume manufacturable, and multi-sourceable.^[1]

All Socket 603 processors utilize a bus speed of 400 MHz and were manufactured in either a 180 nm process, or 130 nm process. Socket 603 processors can be inserted into Socket 604 designed motherboards, but Socket 604 processors cannot be inserted into Socket 603 designed motherboards due to one additional pin being present. Currently, Socket 603 processors range from 1.4 GHz, to 3 GHz.

Currently, no Socket 604 processors are produced with Intel's "MP" designation, while some Socket 603 processors have receive the "MP" designation. The "MP" next to the processors speed is the inclusion of a L3 cache to boost performance in multi-processor computers, however, some socket 604 processors are now being manufactured with the addition of an L3 cache up to 16 mb.

Socket 604 - Intel Xeon

Socket 604 is a motherboard socket for Intel's Xeon processor.

Socket 604 was designed by Intel as a Zero Insertion Force socket intended for workstations and server platforms. It contains 604 contacts arrayed about the center of the socket, each contact has a 1.27mm pitch with regular pin array, to mate with a 604-pin processor package.[1]

Socket 604 processors utilize a bus speed of either 400, 533, 667, or 800 MHz and were manufactured in either a 130 nm process, 90 nm process or 65 nm process. Socket 604 processors cannot be inserted into Socket 603 designed motherboards due to one additional pin being present, but Socket 603 processors can be inserted into Socket 604 designed motherboards. Currently, Socket 604 processors range from 2 GHz, to 3.80 GHz.

Recently, Intel has released more powerful Socket 604 Processors. Features included 65 nm manufacturing, dual cores, and the addition of an L3 processor cache up to 16 mb.

PAC418 - Intel Itanium

PAC611 - Intel Itanium 2, HP PA-RISC 8800 and 8900

Socket B (LGA 1366) - a new socket for future Intel CPUs incorporating the integrated memory controller and Intel QuickPath Interconnect.

Socket H (LGA 715) - a future replacement for the current Socket T (LGA 775), without integrated memory controller and newer point-to-point processor interconnect.

Socket M - Intel Core Solo, Intel Core Duo and Intel Core 2 Duo

Socket M (sometimes referred to as FCPGA6) is a CPU interface introduced by Intel in 2006 for the Intel Core line of mobile processors [1]. It is used in all Intel Core products, as well as the Core-derived Dual-Core Xeon codenamed "Sossaman". It was also used in the first generation of the mobile version of Intel's Core 2 Duo, specifically, the T5x00 and T7x00 ("Merom") lines, though that line switched to Socket P in 2007.

Socket M is not pin-compatible with Socket 479 or the desktop Socket 478 made for Pentium 4 processors.

Socket N - Intel Dual-Core Xeon LV

Socket P - Intel-based; replaces Socket 479 and Socket M. Released May 9th, 2007.

The Intel Socket P is the mobile processor socket replacement for the new Intel Core 2 chips. It has an 800 MT/s FSB, that is switchable on the fly to 400MT/s to save power. It launched on May 9, 2007, as part of the Santa Rosa platform. Socket P has 478 pins, but is not pin-compatible with Socket M.

On the table below we list all socket and slot types created by Intel and AMD since the 486 CPU and examples of CPUs compatible with them.