Tom,
I have to disagree with you completely!
You are NOT being pedantic!
An RTC is indeed any source of information which can be used to monitor the
passage if time. It does not need to have any synchronization with "wall
time" at all.
The RTC may just provide an periodic norification that a certain amount of
time has elapsed [16.666mS for a 60Hz LTC] or may have register(s) for
accumulating a number of these events. If the processor contains built in
timers that can divide a clock signal, then any of these can be considered
an RTC.
Once any type of RTC exists, a computer can be programmed with a value which
matches the current "wall time" and use the information from the RTC to
update the wall time.
The biggest limitation with the minimal implementation of this is the
duration which a a computer may be non-responsive [or even "off"] and still
have the ability to correlate the information from an RTC with "wall time".
In general LTC's are totally useless for this purpose [you have to manually
enter the "base correlation" value at boot].
At the high end, the RTC may accumulate a significant number of "ticks" in
hardware [usually with battery backup[ and have the ability to be "preset"
in order to correlate with "wall time". In many of these cases the RTC also
has decoding circuitary to represent the value in Seconds, Minutes, Hours,
Days, etc... At this point the RTC has the functionallity of a
"Clock/Calendar".
If we are in agreement the the original question really needs to be broken
down further...
A) What was the first computer to make some use of a periodic signal to keep
a time/date that was synchronized with "wall time" [either in hardware or in
system software]?
Or
B) What was the first computer to make use of a device which could remain
synchronized to "wall time" even if the computer was not operational?
>> -----Original Message-----
>> From: cctalk-bounces(a)classiccmp.org
>> [mailto:cctalk-bounces@classiccmp.org] On Behalf Of Tom Jennings
>> Sent: Sunday, August 01, 2004 5:10 PM
>> To: General Discussion: On-Topic Posts Only
>> Cc: cctalk(a)classiccmp.org
>> Subject: Re: First computer with real-time clock?
>>
>> On Thu, 2004-07-29 at 16:18, Tony Duell wrote:
>>
>> > RTCs keep time in hours/minutes/seconds ...
>>
>> To be pedantic, "real time" does not mean wall time, clock
>> time, or calendars, it means
>> isochronous-external-to-the-computer. Things that keep time
>> in traditional wall time are clock/calendars. The
>> terminology for Y2K-problem-capable thingies is not as concise.
>>