Microprocessor: the central processing unit on a single chip

A microprocessor is a processor in which all the building blocks of the processor are combined on one microchip. Due to their different areas of application, microprocessors are adapted to the respective field of application. For example, special versions for aerospace applications must be able to withstand particularly high temperatures and radiation exposure during operation without errors, while mobile processors must have a high IPC rate, low leakage currents and low power consumption. These needs are taken into account in various ways: a fundamental design decision is already made with the selection of the instruction set (CISC or RISC), the implications of which are explained in more detail in the respective special articles. Subsequently, the most efficient microcode possible is developed, which is optimally adapted to boundary conditions such as cache sizes, memory bandwidth and latencies as well as the internal functional units.

The logical design of the microprocessor, which is available in a hardware description language, is then passed on to a high-performance computer, which "routes" the conductor paths, i.e. seeks to determine an optimum arrangement with as few transistors as possible and minimum power dissipation (so-called technology binding or technology mapping). Since these routing problems are NP-complete, usually only approximate solutions are found, which can be improved considerably in detail. From these routing calculations, masks are created which are used by photolithography to expose wafers which are then etched. The production of a microprocessor today comprises well over 100 individual steps, in the course of which even a single error can render the entire processor unusable.

In the final inspection, the processors are finally classified in terms of their clock stability, with physical properties such as signal levels at different clocks being checked using a test program developed individually for each processor type. Particular attention is paid to runtime-critical signal paths on the CPU die in order to prevent speed paths (errors caused by signal delays).

In general it can be stated that the validation effort of modern processors has taken on enormous proportions, and despite all efforts not all error situations can be checked before delivery. The last x86 processor to be completely verified in all functions (and errors!) was the 80286, which is why all manufacturers supply so-called errata lists in which discovered errors are recorded. Intel, for example, had to admit the notorious FDIV bug in early Pentium CPUs, which was caused by several missing entries in an internal lookup table of the FPU.

Over time, the number of instructions supported by the processor increased due to the ever-improving technology. Today, 32 and 64 bit processors are predominantly found, whereby the most common operating systems for the user support a maximum of 64, but mostly only 32 bits. This already shows that software lags behind hardware in the case of processors. The 386s developed in the 1980s were the first 32-bit processors of the Intel 80x86 family.

In 2006, ARM introduced the first commercial unclocked asynchronous processor, the ARM996HS. Since it does not require clocking, an asynchronous processor has a lower and much less concise radiation in the high frequency range and does not consume any significant power during process pauses.

As semiconductor processes have become more integrated, CPU designers have added more functions to the hardware. Units that previously had to be connected as separate chips and could be integrated into the CPU itself over time include:

• the Memory Management Unit for memory management;
• the numerical coprocessor for faster arithmetic operations with floating point numbers;
• Vector arithmetic units, especially for fast graphics processing - called MMX, SSE and successors at Intel, AltiVec at PowerPC;
• Cache memory, first only level 1, today additionally level 2 and also already level 3;
• sometimes the chipset (or parts of it) to control the RAM;
• sometimes a graphics chip for display control;
• up to 100 processor cores on one chip (multi-core processor, Terascale processor).

Microcontrollers, on the other hand, often have only a few registers and a limited instruction set, where addition and subtraction are often already the most complex operations. However, for simple applications, such as controlling a simple machine, this functionality is sufficient, especially since higher functions can be implemented by a few basic operations, for example multiplication by shifting and adding (see Russian Pawn Multiplication). For this purpose, microcontrollers integrate peripheral functions and often also main memory on the chip.