30-01-2015, 13:28PrintHardware (abbreviation: HW) the upper concept is for the mechanical and electronic equipment of a system, e.g., of a computer system. However, he must not refer exclusively to systems with a processor. These can also be purely electromechanical devices as for example a stairwell light machine. The English hardware about synonymous with "ironwares" is original and today is used in the English-speaking space still in this sense – not only for computer hardware.
Demarcation hardware and software
Very easy systems can be implemented directly in invariable hardware. Then the function of these systems is firmly given by the structure of the hardware. However, more complicated hardware systems contain mostly also programmable elements, e.g., processors. These, in its structure to also agreed construction elements, execute a sequence of instructions which can be manipulated. In addition, the components which can be also defined in its structure (see PLD) step. Therefore the function of the overall system can be easily customised. The configuration and instruction data are designated in general software - by less complicated devices where merely structures and easy executions are fixed it is mostly called firmware. However, assemblies (components count to the hardware: Processor, main memory etc.) and peripheral devices. If everything simplifies said heard what can be touched, to the hardware. Computer hardware is usable exclusively with suitable software.
Software designates by contrast programmes and data which one cannot touch.
The data carriers on which the software is, e.g., disk, CD/DVD/BR D/HD DVD, RAM, Flash memory, hard disk etc., are against it hardware.
Motherboard (English mainboard)
A possible realisation or partitioning of the hardware is in 1945 introduced from Neumann's architecture. It makes a distinction abstractly between tax work, arithmetic work ("ALUMINIUM"), memory unit and input and output work (peripheral device). Nowadays such a partitioning still makes sense to the understanding of the structure of a computer and is necessary, although meanwhile with modern processors many functions are integrated in a hardware chip, possibly:
Tax work to the ALUMINIUM and to the command coding immediately several times to the parallel processing;
Tax work MMU to the memory management;
the cache as a part of the memory unit,
the control for a bus system which connects internal and external components with each other.
With so-called embedded processors (for e.g. PDAs or washing machines) one finds in the same cabinet still an one and output work in the form of serial interfaces (e.g., USB), to digital I/O (input / Ouput), e.g., for lamp and analogous I/O for an e.g. touch screen.
Even stronger so-called SoC (system on a chip) are integrated for e.g. Smartphones which integrate other components up to main memory (RAM) and Flash memory.
The PC components belong to the computer hardware:
The basic elements of the computer architecture: Motherboard (also Motherboard or mainboard called) with chipset (for IO), processor and main memory (RAM)
Bulk storage: Disk drives (hard disk drive, Flashspeicher, CD-ROM disk drive, DVD disk drive, Zip disk drive, Jaz disk drive, …) and storage media
Extension cards (graphics card, sound card, network card, TV card, ISDN card, USB card, PhysX card, …)
Power supply unit, cabinet, fan
Output devices (printer, display, video projector, loudspeaker, …)
Input devices (keyboard, mouse, joystick, …)
Einlesegeräte (different kinds of scanners, microphones, map reading devices, …)
All these peripheral devices and assemblies of a computer are built up for the most part with logical circuits.
The hardware of a computer is controled by the firmware (e.g., BIOS or EFI) and the software (among other things operating systems and the matching device drivers) and is administered.
Often hardware is provided with a FCC number which permits an unequivocal identification of the manufacturer.
Originally there comes the concept Firmware from the range of programmable logic. In the 1980s boolean functions as for example Adressdekodierung with individually available components were realised hardwired. (And gates, Or gates, inverters, flip flops, etc.) with increasing complexity were seated programmable components as for example PLDs. This has the advantage to be able to react adaptably to errors found anew by new programming instead of having to unsolder parts. In spite of the programming of the hardware with Hochsprachen similar to software (e.g., Abel, VHDL, etc.) one speaks of the firmware which one mostly associates to the hardware or the microelectronics.
The hard error is the name for the failure of hardware of an electronic data-processing system in contrast to the program error. These can be caused by error in the design or the production, however, also by additional changes or improper use. The most frequent source explain faulty installation, directly followed by additional changes as for example Übertaktung. While production and design errors are irreversible as a rule and can be repaired only with the following product, hard errors are as a result of additional changes, provided that still no physical damages to hardware components have appeared, reversible, while the changes are cancelled.
One of the best known errors in the design is the Pentium FDIV bug: A processor of the mark of Pentium has delivered wrong results with floating point divisions what led to a big whirl in public, but had no other spectacular effects.
An increasingly more frequent cause for hard error originates from the reduction of the hardware structures. Thus low manufacturing divergences can already lead to immediate as well as creeping hard errors. Thus the normal company can possibly already lead, for example, by temperature change charges or electric migration (see SNDS) to changes to be remained (ageing) of the hardware which can release hard error with the excess of certain tolerance borders. Factors like insufficient cooling or Übertakten can accelerate these processes strongly.
Hard errors mostly lead to the fact that the programmes running on the hardware and the system do not behave as specified. In such a case is not known first whether a hard or software error is given. If a hard error cannot be excluded, very perfect and simply manageable software which can be launched mainly completely by change media (disk, CD-ROM, USB-Stick) from is required to his localisation.
Because hard errors are never to be excluded, error-tolerant systems are used in ranges relevant for security.
Damaged sectors on hard disks and defective storage areas belong to the most frequent hard errors which can appear ex works as well as only late in the company, in the banks of memory which are usually seated as main memory (RAM). Because of the frequency of these hard errors easy diagnostic programs which can be also launched from change media exist for both cases.
On account of similar symptoms hard errors are often distinguishable hard from driver problems.
Also one may not mistake hard error for so-called Softly Errors which cause though the same symptoms like hard error, are due, however, not to defective hardware, but to accidental events as for example cosmic radiation.