Re: The joy of FORTRAN

Lawrence D'Oliveiro <ldo@nz.invalid> writes:

On Tue, 25 Feb 2025 15:19:41 -0800, John Ames wrote:

On Tue, 25 Feb 2025 23:02:32 -0000 (UTC)
Lawrence D'Oliveiro <ldo@nz.invalid> wrote:

I wonder why DEC's 18-bit range weren't more popular; though I think they
had less consistency between members of the range than DEC's other product
families.

I suspect that, in the computer market of the early '60s, they ended up
as the awkward middle child  ...

Mr. Ames's suspicion is unfounded.  The 18 bit systems were neither awkward nor
unsuccessful.

Still, they were the product line that launched DEC's computer career, with
the PDP-1.

Since DEC's original intent was to build small computers (at their founding in
1957), that's rather a fatuous statement.

The choice of 18 bits was related to the scientific computing standard of the
day, in which large computers used a 36 bit word.

See the relationship?

... small labs/offices/embedded applications
could get a basic computer cheaper with the -12 ...

Cost was certainly a factor in putting up with the compromises. The 12-bit
line was clearly a cut-down derivative of the 18-bit line. In those days, all
hardware was expensive, but the 12-bit machines less so.

In fact, the PDP-5 (DEC's third computer and first 12 bit system) was
originally designed as a data capture front end for the 18 bit PDP-4, in a
nuclear control application for the Canadians.  Cf. Bell's _Computer Engineering_.

--
Rich Alderson   news@alderson.users.panix.com
      Audendum est, et veritas investiganda; quam etiamsi non assequamur,
  omnino tamen proprius, quam nunc sumus, ad eam perveniemus.
--Galen