From: Jeff Barber <jeffb@sware.com>
Date: Fri, 12 Apr 1996 07:37:55 +0800
To: rmartin@aw.sgi.com (Richard Martin)
Subject: Re: questions about bits and bytes [NOISE]
In-Reply-To: <9604110950.ZM8850@glacius.alias.com>
Message-ID: <199604111452.KAA24457@jafar.sware.com>
MIME-Version: 1.0
Content-Type: text/plain


Richard Martin writes:

> On Apr 10,  6:57pm, jim bell wrote:
> > At 06:29 PM 4/10/96 -0700, Simon Spero wrote:
> > >No, bytes are no always 8 bits - some machines use(d) 9-bit bytes.
> > I notice you gave no examples.  Why is that?
> Perhaps he thought that most people who were interested could go look
> it up themselves.
> 
> - From a really quick web search, we find that the SGI Impact jams 9-bit
> bytes [that's what it says] across the Rambus internally. I'm not sure
> if the memory itself is 9-bit.

[I told myself I was going to stay out of this, but Jim Bell's dogmatic
stance irks me... ]  Here's a citation from "Portability of C Programs
and the Unix System" by S.C. Johnson and D.M. Ritchie (yes, that Richie)
in the Bell System Technical Journal volume 57, Number 6, July-August 1978.

    "A representation of characters (bytes) must be provided with at
    least 8 bits per byte. ...  Most programs make no explicit use of
    this fact, but the I/O system uses it heavily.  (This tends to rule
    out one plausible representation of characters on the DEC PDP-10,
    which is able to access 5 7-bit characters in a 36-bit word with 
    one bit left over.  Fortunately, that machine can access four 9-bit
    characters equally well.) ..."

The clear implication is that "byte" means the number of bits used or
needed to represent a single character.


-- Jeff