Re: evolution of bytes, The joy of FORTRAN

On 01/03/2025 01:48, John Levine wrote:

According to Peter Flass  <peter_flass@yahoo.com>:

A lot of older machines were character-addressable. The term “byte” hadn’t
been invented yet. The 1401 (etc), 1620, and many 70xx machines.

Oh sure but I am fairly sure that the 360 was the first machine that
was both character and word addressable with the words at power-of-two
addresses, and a design that allowed word operationw to work as a unit
rather than serially by character.

As so often, it rather depends what you mean by ....  Atlas had
24-bit addresses, which comprised [in the standard model] 21-bit word
addresses with the bottom three bits used to address a character within
the word.  Most of the machine instructions ignored the character part;
for example instructions were always taken from the word part.  AFAIR,
there were only two instructions that actually used the character part
/as an address/, one to copy a character into a register, the other to
write from a register into a character position.  That was out of a rich
supply of other basic instructions and extracodes, plenty of which used
the full 24-bits, but for arithmetic [24, 48 or 96 bits] or logic, not
as characters.  So you might regard the character addressing as somewhat
limited, but it was there.  Of course in those days [1959-ish] computers
were used for computation.  Characters were "only" for I/O, so once you
had enough to read/write numbers and captions, everything else was a
frippery and a waste of a valuable resource.  I got into trouble with my
supervisor for writing a program that wrote "poetry" [of a sort] when
I was supposed to be doing celestial mechanics.  Luckily, he never knew
what else I was doing.

As far as I know it was also the first character addressable binary
machine.  The earlier ones were decimal.

Atlas was definitely binary!
--
Andy Walker, Nottingham.
    Andy's music pages: www.cuboid.me.uk/andy/Music
    Composer of the day: www.cuboid.me.uk/andy/Music/Composers/Kontski