Revision as of 09:45, 3 July 2019

A microprocessor (µP) or a Microprocessing Unit (MPU) is a device that implements the core elements of a computer system on a single integrated circuit, or as a few integrated circuits operating as a cohesive unit.

Modern microprocessors typically incorporate the functionality of a clock, central processing unit (CPU), arithmetic logic unit (ALU), floating point unit (FPU), control unit (CU), memory management unit (MMU), interrupts, input/output interfaces, and cache. Specialized microprocessor may also serve as graphical processing units (GPUs), signal processing units (DSPs), neural processing unit (NPUs), microcontrollers, etc.

History

Main article: history of the microprocessor

The Intel 4004 is generally regarded as the first commercially available microprocessor.^[1][2] The "Busicom Project"^[3] that produced the 4004 originated in 1968, when Busicom engineer Masatoshi Shima designed a special-purpose LSI chipset, along with his supervisor Tadashi Tanba, for use in the Busicom 141-PF high-performance desktop calculator.^[4] Busicom's original design called for a programmable chip set consisting of seven different chips. Three of the chips were to make a special-purpose CPU with its program stored in ROM and its data stored in shift register read-write memory.^[3] Shima's initial design included arithmetic units (adders), multiplier units, registers, read-only memory (ROM), and a macro-instruction set to control a decimal computer system.^[4] Sharp engineer Tadashi Sasaki was also involved with its development, and conceived of a single-chip CPU in 1968, when he discussed the concept at a brainstorming meeting that was held in Japan. Sasaki attributes the basic invention to break the calculator chipset into four parts with ROM (4001), RAM (4002), shift registers (4003) and CPU (4004) to an unnamed woman, a software engineering researcher from Nara Women's College, who was present at the meeting. Sasaki then had his first meeting with Intel in 1968, and discussed the woman's four-division chipset concept with Busicom and Intel.^[5]

The initial Busicom design, which consisted of seven chips, including a three-chip CPU, was eventually simplified down to four chips, including a single-chip CPU.^[3] This simplified approach was initially conceived by Sasaki, influenced by the unnamed Nara Women's College reseacher in 1968,^[5] and then designed by Intel's Marcian "Ted" Hoff in 1969.^[3] Hoff, the Intel engineer assigned to evaluate the project, believed the Busicom design could be simplified by using dynamic RAM storage for data, rather than shift register memory, and a more traditional general-purpose CPU architecture. Hoff's design consisted of a four-chip architectural proposal: a ROM chip for storing the programs, a dynamic RAM chip for storing data, a simple I/O device and a 4-bit central processing unit (CPU). Although not a chip designer, he felt the CPU could be integrated into a single chip, but as he lacked the technical know-how the idea remained just a wish for the time being.

The architecture and specifications of the MCS-4 came from the interaction of an Intel team led by Hoff and Stanley Mazor, a software engineer reporting to him, and a Busicom team led by Shima, in 1969.^[3] However, Mazor and Hoff moved on to other projects. In April 1970, Intel hired Italian-born engineer Federico Faggin as project leader, a move that ultimately made the single-chip CPU final design a reality. Shima meanwhile designed the Busicom calculator firmware and assisted Faggin during the first six months of the implementation. Faggin had the right background to lead the project into what would become the first commercial general purpose microprocessor. Since SGT was his very own invention, Faggin also used it to create his new methodology for random logic design that made it possible to implement a single-chip CPU with the proper speed, power dissipation and cost. The manager of Intel's MOS Design Department was Leslie L. Vadász at the time of the MCS-4 development but Vadász's attention was completely focused on the mainstream business of semiconductor memories so he left the leadership and the management of the MCS-4 project to Faggin, who was ultimately responsible for leading the 4004 project to its realization. Production units of the 4004 were first delivered to Busicom in March 1971.

NEC released the μPD707 and μPD708, a two-chip 4-bit CPU, in 1971.^[6] They were followed by NEC's first single-chip microprocessor, the μPD700, in April 1972.^[7][8] It was a prototype for the μCOM-4 (μPD751), released in April 1973,^[7] combining the μPD707 and μPD708 into a single microprocessor.^[6]

The first 16-bit microprocessor was the NEC μCOM-16 in 1974.^[9] In 1975, Panafacom, a conglomerate formed by Japanese companies Fujitsu, Fuji Electric, and Matsushita, introduced the MN1610, a commercial 16-bit microprocessor.^{[10][11][12][11]}

Overview

Microprocessors are instruction set processors (ISPs), meaning they operate on a predefined set of instructions. In the broadest sense, their basic functionality is to continuously read in digital data consisting of instructions and possibly values; execute them by interpreting the instructions and performing a certain operation on the values; and finally outputs a result.

While the basic functionality is shared among all microprocessors, they vary greatly in the type and size of data they handle, the kind of operations they support, how they perform those operations, their intended purpose, and their performance characteristics. The functionality of a microprocessor is dependent on the characteristics of the instruction set it operates on. Every program that runs on that microprocessor is therefore bound by that instruction set architecture (ISA) and is encoded in that instruction set. The instruction set architecture is the specification of a microprocessor design while the realization of an ISA is known as the implementation. There can, and usually are, multiple implementations for a second ISA specification. Those implementations are known as microarchitectures. The kind of tradeoffs made in a microarchitecture ultimately determines the characteristics of the microprocessor (e.g., power and performance).

Variants

Microprocessors come in various flavors with Most microprocessors can be classified as one of the follow:

• general-purpose microprocessors - the most common form of microprocessors, not designed for any one specific task in mind. Instead they are designed to support a broad array of operations.
• bit-slice microprocessor (BSM) - a microprocessor designed as a module intended to be built up like Lego blocks into a desired word size and architecture as needed.
• system on chip (SoC) - a microprocessor that contains all the components of a computer system, including the extra functionality that would normally be provided by auxiliary chips, which could include things such as wireless, ethernet, SD card, ADC, DAC, LCD drivers, and FPGA. SoCs are capable of running full-fledged modern operating systems with all their features.
• microcontroller (MCU) - a microprocessor that contains a few additional components such as RAM, ROM, and programmable I/O ports primarily designed to control and drive other electronic equipment. MCUs are designed to be embedded, usually in a highly restrictive environment. They usually consume very little power, may run relatively slow, and typically execute individual task-specific programs.
• digital signal processor (DSP) - a microprocessor that specializes in the numerical manipulation of signals.
• floating point unit (FPU) - is a math microprocessor (or coprocessor) - a microprocessor that specializes in the creation and manipulation of floating point values.
• graphics processing unit (GPU) - is a graphics microprocessor - a microprocessor that specializes in the creation and manipulation of images through a set of optimized geometric operations. Modern graphic microprocessors tend to be highly parallelized, allowing large blocks of visual data to be processed efficiently.
• vector processor (VPU) - a vector processor is microprocessor that implements operations that manipulate one/multi-dimensional arrays of data known as vectors.
• physics processing unit (PPU) - a microprocessor that specializes in handling the calculations of physics. This includes things such as fluid dynamics and collision detection.
• neural processing unit (NPU) - is a neural microprocessor - a microprocessor that specializes in the manipulations of predictive models.
• coprocessor - a microprocessor that aides a master microprocessor by either offloading some of its work or by providing additional specialized processing operations, e.g. cryptography, math, graphics, string processing, or I/O interfacing. A coprocessor can act as an extension of the master microprocessor by extending the instruction set architecture or by acting like another peripheral on the main bus.

Some microprocessors can be a hybrid combinations of a few of the above. For example, a general-purpose microprocessor might come with an integrated GPU, implying an additional graphic processing unit has been added to the microprocessor to enable it to manipulate visual data more efficiently. Likewise almost all modern desktop microprocessors come with integrated floating point units.

Specifications

Main articles: central processing unit, architecture, microarchitecture, and instruction set architecture

The technical specifications of microprocessors are derived from the microarchitecture of the incorporated CPU, the semiconductor technology involved, and the properties of the overall system. Some common specifications are summarized below:

• technology - the semiconductor technology used to create the MPU (e.g. CMOS, BiCMOS, and TTL)
• process - the process used to manufacture the MPU - i.e the feature size and design rules. The feature size on its own (e.g. 10 µm) is usually taken as a synonym for the process.
• microarchitecture - the functional description of the underlying circuitry of the microprocessor.
• word size - the word size of a microprocessor usually refers specifically to the data word size used - i.e. highest operand width used to manipulate standard integer values. (this excludes special processing units such as SIMD and FPU)
• core count - the number of physical cores incorporated into the chip.
• endianness - the order of the bytes the microprocessor uses when operating on multi-byte values.
• base frequency - the internal operating frequency of CPU's core. It's one of many parameters that are used to assess the performance of a microprocessor.
• package - the physical casing of the microprocessor. This most often goes along with a socket, which is the interconnects that sits on the circuit board itself where the package is inserted into.

Note that not all the specifications listed above apply to all the types of microprocessors.

Components

This section is empty; you can help add the missing info by editing this page.

Design

A prototype of a RISC-V microprocessor with the heat spreader removed, showing the exposed die, January 2013

Main articles: integrated circuit and integrated circuit design

This section is empty; you can help add the missing info by editing this page.

Computational power

This section is empty; you can help add the missing info by editing this page.

Families

Main article: microprocessor family

This section is empty; you can help add the missing info by editing this page.

See also

• List of processor families

References