| Computer architecture bit widths |
|---|
| Bit |
| Application |
| Binary floating-pointprecision |
| Decimal floating-pointprecision |
Incomputer architecture,32-bit computing refers to computer systems with aprocessor,memory, and other major system components that operate on data in a maximum of 32-bit units.[1][2] Compared to smaller bit widths, 32-bit computers can perform large calculations more efficiently and process more data per clock cycle. Typical 32-bitpersonal computers also have a 32-bitaddress bus, permitting up to 4 GB ofRAM to be accessed, far more than previous generations of system architecture allowed.[3]
32-bit designs have been used since the earliest days of electronic computing, in experimental systems and then in largemainframe andminicomputer systems. The first hybrid 16/32-bitmicroprocessor, theMotorola 68000, was introduced in the late 1970s and used in systems such as the originalApple Macintosh. Fully 32-bit microprocessors such as theHP FOCUS,Motorola 68020 andIntel 80386 were launched in the early to mid 1980s and became dominant by the early 1990s. This generation of personal computers coincided with and enabled the firstmass-adoption of the World Wide Web. While 32-bit architectures are still widely-used in specific applications, the PC and server market has moved on to64 bits withx86-64 and other 64-bit architectures since the mid-2000s with installed memory often exceeding the 32-bit 4G RAM address limits on entry level computers. The latest generation ofsmartphones have also switched to 64 bits.
A 32-bit register can store 232 different values. Therange ofinteger values that can be stored in 32 bits depends on theinteger representation used. With the two most common representations, the range is 0 through4,294,967,295 (232 − 1) for representation as an (unsigned)binary number, and −2,147,483,648 (−231) through2,147,483,647 (231 − 1) for representation astwo's complement.
One important consequence is that a processor with 32-bitmemory addresses can directly access at most 4 GiB ofbyte-addressable memory (though in practice the limit may be lower).
The world's first stored-programelectronic computer, theManchester Baby, used a 32-bit architecture in 1948, although it was only aproof of concept and had little practical capacity. It held only 32 32-bit words of RAM on aWilliams tube, and had no addition operation, only subtraction.
Memory, as well as other digitalcircuits and wiring, was expensive during the first decades of 32-bit architectures (the 1960s to the 1980s).[4] Older 32-bit processor families (or simpler, cheaper variants thereof) could therefore have many compromises and limitations in order to cut costs. This could be a 16-bitALU, for instance, or external (or internal) buses narrower than 32 bits, limiting memory size or demanding more cycles for instruction fetch, execution or write back.
Despite this, such processors could be labeled32-bit, since they still had 32-bit registers and instructions able to manipulate 32-bit quantities. For example, theIBM System/360 Model 30 had an 8-bit ALU, 8-bit internal data paths, and an 8-bit path to memory,[5] and the originalMotorola 68000 had a 16-bit data ALU and a 16-bit external data bus, but had 32-bit registers and a 32-bit oriented instruction set. The 68000 design was sometimes referred to as16/32-bit.[6]
However, the opposite is often true for newer 32-bit designs. For example, thePentium Pro processor is a 32-bit machine, with 32-bit registers and instructions that manipulate 32-bit quantities, but the external address bus is 36 bits wide, giving a larger address space than 4 GB, and the external data bus is 64 bits wide, primarily in order to permit a more efficient prefetch of instructions and data.[7]
Prominent 32-bit instruction set architectures used in general-purpose computing include theIBM System/360,IBM System/370 (which had24-bit addressing),System/370-XA,ESA/370, andESA/390 (which had31-bit addressing), theDECVAX, theNS320xx, theMotorola 68000 family (the first two models of which had 24-bit addressing), theIntelIA-32 32-bit version of thex86 architecture, and the 32-bit versions of theARM,[8]SPARC,MIPS,PowerPC andPA-RISC architectures. 32-bit instruction set architectures used for embedded computing include the 68000 family andColdFire, x86, ARM, MIPS, PowerPC, andInfineon TriCore architectures.
On thex86 architecture, a 32-bit application normally meanssoftware that typically (not necessarily) uses the 32-bit linearaddress space (orflat memory model) possible with the80386 and later chips. In this context, the term came about becauseDOS,Microsoft Windows andOS/2[9] were originally written for the8088/8086 or80286,16-bit microprocessors with asegmented address space where programs had to switch between segments to reach more than 64kilobytes ofcode or data. As this is quite time-consuming in comparison to other machine operations, the performance may suffer. Furthermore,programming with segments tend to become complicated; specialfar andnear keywords ormemory models had to be used (with care), not only inassembly language but also in high level languages such asPascal, compiledBASIC,Fortran,C, etc.
The 80386 and its successors fully support the 16-bit segments of the 80286 but also segments for 32-bit address offsets (using the new 32-bit width of the main registers). If thebase address of all 32-bit segments is set to 0, and segment registers are not used explicitly, the segmentation can be forgotten and the processor appears as having a simple linear 32-bit address space.Operating systems like Windows or OS/2 provide the possibility to run 16-bit (segmented) programs as well as 32-bit programs. The former possibility exists forbackward compatibility and the latter is usually meant to be used for newsoftware development.[10]
In digital images/pictures, 32-bit usually refers toRGBA color space; that is, 24-bittruecolor images with an additional 8-bitalpha channel. Other image formats also specify 32 bits per pixel, such asRGBE.
In digital images, 32-bit sometimes refers tohigh-dynamic-range imaging (HDR) formats that use 32 bits per channel, a total of 96 bits per pixel. 32-bit-per-channel images are used to represent values brighter than whatsRGB color space allows (brighter than white); these values can then be used to more accurately retain bright highlights when either lowering the exposure of the image or when it is seen through a dark filter or dull reflection.
For example, a reflection in an oil slick is only a fraction of that seen in a mirror surface. HDR imagery allows for the reflection of highlights that can still be seen as bright white areas, instead of dullgrey shapes.
A 32-bit file format is abinary fileformat for which each elementary information is defined on 32 bits (or 4bytes). An example of such a format is theEnhanced Metafile Format.