| Revision as of 18:21, 16 April 2005 editWtshymanski (talk | contribs)Extended confirmed users76,122 edits Link← Previous edit | Revision as of 02:21, 17 May 2005 edit undoWtshymanski (talk | contribs)Extended confirmed users76,122 edits Link to 46-bit wordsNext edit → | ||
| Line 1: | Line 1: | ||
| In ] terminology, '''word size''' ('''word length''') is the number of ]s that a ] can process at one time (the ]). Processors with many different word sizes have existed, where the word size was selected as a trade off between precision and range of numeric results and the cost of the processing hardware. A 12-bit data word was used by some ]s such as the ] ] and is still used on special-purpose ]s such as ]s and ]s. Some early ] systems used 35, 36, 40, 44 or 45-bit words, or words that contained between 6 and 16 decimal digits. Word lengths in powers of two (8, 16, 32, 64) have predominated for many years, roughly since the introducionn of the ] System 360. See ], ], ], ], ]. | In ] terminology, '''word size''' ('''word length''') is the number of ]s that a ] can process at one time (the ]). Processors with many different word sizes have existed, where the word size was selected as a trade off between precision and range of numeric results and the cost of the processing hardware. A 12-bit data word was used by some ]s such as the ] ] and is still used on special-purpose ]s such as ]s and ]s. Some early ] systems used 35, ], 40, 44 or 45-bit words, or words that contained between 6 and 16 decimal digits. Word lengths in powers of two (8, 16, 32, 64) have predominated for many years, roughly since the introducionn of the ] System 360. See ], ], ], ], ]. | ||
| A processor's word size for arithmetic is often equal to the width of its external ] though sometimes the bus is made narrower than the CPU (often half as many bits) to economise on packaging and ] costs, space, and power consumption. For machines of the ] type, a word size equal or greater than the logical address size is convenient, though the physical address range of a processor may be greater or less than the word size for arithmetic. | A processor's word size for arithmetic is often equal to the width of its external ] though sometimes the bus is made narrower than the CPU (often half as many bits) to economise on packaging and ] costs, space, and power consumption. For machines of the ] type, a word size equal or greater than the logical address size is convenient, though the physical address range of a processor may be greater or less than the word size for arithmetic. | ||
Revision as of 02:21, 17 May 2005
In computer hardware terminology, word size (word length) is the number of bits that a CPU can process at one time (the word). Processors with many different word sizes have existed, where the word size was selected as a trade off between precision and range of numeric results and the cost of the processing hardware. A 12-bit data word was used by some minicomputers such as the Digital Equipment Corporation PDP-8 and is still used on special-purpose microprocessors such as microcontrollers and digital signal processors. Some early mainframe systems used 35, 36, 40, 44 or 45-bit words, or words that contained between 6 and 16 decimal digits. Word lengths in powers of two (8, 16, 32, 64) have predominated for many years, roughly since the introducionn of the IBM System 360. See 8-bit, 16-bit, 32-bit, 64-bit, 128-bit.
A processor's word size for arithmetic is often equal to the width of its external data bus though sometimes the bus is made narrower than the CPU (often half as many bits) to economise on packaging and circuit board costs, space, and power consumption. For machines of the Von Neumann type, a word size equal or greater than the logical address size is convenient, though the physical address range of a processor may be greater or less than the word size for arithmetic.
A word n bits long can hold 2 distinct values. If these values represent integers, by far the most common ranges are −2 to 2−1 inclusive, or 0 to 2−1 inclusive.
This article is based on material taken from the Free On-line Dictionary of Computing prior to 1 November 2008 and incorporated under the "relicensing" terms of the GFDL, version 1.3 or later.
Reference
Anthony Ralston and Edwin D. Reilly, Encyclopedia of Computer Science Third Edition, Van Nostrand Reinhold, New York, 1993 ISBN 0442276795
Categories: