dricej@drilex.UUCP (Craig Jackson) (12/16/86)
In article <9200002@rmi.UUCP> wsiebeck@rmi.UUCP writes: >>/***** rmi:comp.arch / bnr-vpa!pdbain / 1:59 pm Dec 5, 1986*/ >>Contrary to popular belief, a byte is not necessarily 8 bits - >>it just frequently happens to be that size. > >A very easy definition of a byte is as follows: > > A byte is the smallest addressable unit of memory in a given machine. > >This should do it ?? Most of the machines with byte sizes of other than eight bits are also word-addressable machines. The idea that address bits could be wasted on addressing to the character came along about the same time as eight-bit characters. Before that, most machines were word addressable (aside from a few like the IBM 1401). Some of those machines had useful subdivisions of the word, and those were called bytes. (In my case, the CDC 6400 had 60 bit words, but its peripheral processors only had 12-bit words. Since the operating system resided in the PPs, there were many 12-bit fields in memory (packed five to a word). These fields were typically called bytes. Incidently, each such byte could hold two six-bit characters. The definitions I've always worked with are: A byte is a collection of two or more bits, smaller than a word. A word is the 'natural' unit of operations on a machine. (Which means that 32 bits on a vax should really be called a word, not a longword.) -- Craig Jackson UUCP: {harvard!axiom,linus!axiom,ll-xn}!drilex!dricej BIX: cjackson