First computer with real-time clock?

From: Scott Stevens <sastevens_at_earthlink.net>
Date: Tue Aug 3 18:29:12 2004

On Mon, 02 Aug 2004 22:30:30 -0400
"David V. Corbin" <dvcorbin_at_optonline.net> wrote:

> >>> Of course, it is also quite possible that the software just
> >>> asked the user what date/time to put on the printout :-).
> >>>
>
> I was thinking the same thing. There were some programs I worked on
> during the early seventies where the computer did not have a clue
> about the date or time. The input "deck" contained this information
> which was basically just echoed on to the printouts. I remember
> running batch jobs [large decks] where the only cards that would get
> changed were a few parameters and the leading "identity" card....
>
>

IBM PC-DOS 1.0, for those who have never booted it, presents the user
with a Time and Date prompt that can not be bypassed by simply hitting
carriage return. The user can enter fictitious time and date if so
desired, but not null.

This was changed with later versions of PC-DOS and the parallel MS-DOS
versions. You could simply hit return and default time/date values were
assumed. I'm not sure if PC-DOS 1.0 supported autoexec.bat, which
bypasses the Time/Date prompt on later versions (i.e. the Time/Date
prompts are invisibly bypassed and a call to a small program that
accessed the 'clock' hardware was not included (did anybody else back
then feel like their machine was stunted if there wasn't a clock card
plugged into an ISA slot on the machine?))

My point in bringing this up is to note that even though it didn't have
a real-timekeeping function built in, an original IBM PC running PC-DOS
1.0 did have an inherent 'timekeeping' fuction which was often somewhat
accurate. It didn't cost any more or less to enter the real time/date
than a fictitious one, since some input with the right format was forced
on system bootup.

Granted, not many people at all ran IBM PCs on DOS 1.0, which was
crippled in many other ways, and wasn't really that viable an 'OS' to
run for several further releases.

But back 'with the thread' being discussed, when I was in a programming
class in college where we had to submit our FORTRAN decks to a batch
terminal, there was definitely a high quality time and datestamp placed
on the resulting printouts. This was on CDC Cyber hardware in the late
70's.
Received on Tue Aug 03 2004 - 18:29:12 BST