View source for 48-bit computing

{{More references|date=August 2007}}
{{Computer architecture bit widths}}
In [[computer&nbsp;architecture]], '''48-bit''' [[integer (computer science)|integer]]s can represent 281,474,976,710,656 (2<sup>48</sup> or 2.814749767×10<sup>14</sup>) discrete values. This allows an [[Unsigned integer|unsigned]] binary integer range of 0 through 281,474,976,710,655 (2<sup>48</sup> − 1) or a [[Signed number representations|signed]] [[two's complement]] range of −140,737,488,355,328 (−2<sup>47</sup>) through 140,737,488,355,327 (2<sup>47</sup> − 1). A 48-bit [[memory address]] can directly address every byte of 256 [[terabyte]]s of storage. 48-bit can refer to any other [[Data (computing)|data]] unit that consumes 48 [[bit]]s (6 [[octet (computing)|octets]]) in width.  Examples include 48-bit [[Central processing unit|CPU]] and [[Arithmetic logic unit|ALU]] [[computer architecture|architecture]]s that are based on [[processor register|register]]s, [[address bus]]es, or [[Bus (computing)|data bus]]es of that size.

==Word size==
Computers with 48-bit [[Word (data type)|words]] include the [[AN/FSQ-32]], [[Control Data Corporation|CDC]] 1604/[[CDC 3000|upper-3000 series]], [[BESM-6]], [[Ferranti]] [[Atlas (computer)|Atlas]], [[Philco]] [[Philco Transac S-2000|TRANSAC S-2000]] and [[Burroughs large systems]].{{efn|The B5000, B5500 and B5700 took 3 bits in control words and numeric data for use as a tag; alphanumeric data and instruction syllables were stored in the full 48 bits and had no tags.}}{{efn|The B5900-B8xxx additionally had a 3- or 4-bit type tag.}}

The Honeywell [[DATAmatic 1000]],
[[Honeywell 800|H-800]],
the [[MANIAC II]],
the [[MANIAC III]],
the [[Brookhaven National Laboratory]] Merlin,<ref>
[http://ed-thelen.org/comp-hist/BRL61-m.html#MERLIN "MERLIN"].
</ref>
the [[Philco computers#CXPQ|Philco CXPQ]],
the [[Ferranti Orion]],
the [[Telefunken Rechner TR 440]],
the [[ICT 1301]],
and many other early transistor-based and vacuum tube computers<ref>
[https://retrocomputingforum.com/t/in-search-of-48-bit-computers/898 "In search of 48 bit computers"].
</ref>
used 48-bit words.

==Addressing==
The [[IBM System/38]], and the [[IBM AS/400]] in its [[complex instruction set computer|CISC]] variants, use 48-bit addresses.

The address size used in [[logical block addressing]] was increased to 48 bits with the introduction of [[AT Attachment#ATA standards versions, transfer rates, and features|ATA-6]]. The [[Ext4]] file system physically limits the file block count to 48 bits.

The minimal implementation of the [[x86-64]] architecture provides 48-bit addressing encoded into 64 bits; future versions of the architecture can expand this without breaking properly written applications.

The media access control address ([[MAC address]]) of a [[network interface controller]] uses a 48-bit address space.

==Images==
{{See also|Deep color}}
In digital images, 48 bits per pixel, or 16 bits per each color channel (red, green and blue), is used for accurate processing. For the human eye, it is almost impossible to see any difference between such an image and a 24-bit image,{{Citation needed|date=June 2013}} but the existence of more shades of each of the three primary colors (65,536 as opposed to 256) means that more operations can be performed on the image without risk of noticeable [[Colour banding|banding]] or [[posterization]].

==Notes==
{{Notelist}}

==References==
{{Reflist}}

{{CPU technologies}}

[[Category:Data unit]]