skann@gnh-applesauce.cts.com (Steve Kann) (04/07/90)
I disagree.. A couple of years ago, your assumptions about using the ac
line as a timer would have been correct, but today, almopst all time sources
are based on a quartz crystal. It is obvious that the GS would be for two
reasons 1) when the power is off, the clock is run off the battery, not the AC
line, and 2) Unless I'm wrong, the clock uses the same timer as the CPU.
Proline:skann@gnh-applesauce <--- Preferred
America online: NYSteve
CI$:70265,1141
************************
UUCP: !crash!pnet01!gnh-applesauce!skann This is just my opinion!
INET: skann@gnh-applesauce.cts.com ************************
ARPA: crash!pnet01!gnh-applesauce!skann@nosc.mil "Don't blame me!"
* Sent by AppleSauce at Fri, 6 Apr 90 13:01:38toddpw@tybalt.caltech.edu (Todd P. Whitesel) (04/08/90)
skann@gnh-applesauce.cts.com (Steve Kann) writes: > I disagree.. A couple of years ago, your assumptions about using the ac >line as a timer would have been correct, but today, almopst all time sources >are based on a quartz crystal. This is more due to the fact that most clocks are now digitally controlled and can actually use the crystals. There are still quite a few of the older clocks in use, but even after they are all gone the line frequency will be tightly regulated because a stable frequency is much easier to 'tune out' of a noise sensitive circuit -- Broadcast television all over the world uses a frame rate very close to the local line frequency, to reduce crossover noise from the power line. >It is obvious that the GS would be for two >reasons 1) when the power is off, the clock is run off the battery, not the AC >line, and 2) Unless I'm wrong, the clock uses the same timer as the CPU. The clock has its own internal 32768 hz crystal, and it runs independently of the CPU. Any access of the clock chip must go though a bizarre serial interface which is handled by a 6522 in Macs and by the VGC in the IIGS. Todd Whitesel toddpw @ tybalt.caltech.edu
cs4w+@andrew.cmu.edu (Charles William Swiger) (04/09/90)
Mr. Kann writes:
I disagree.. A couple of years ago, your assumptions about using the ac
line as a timer would have been correct, but today, almopst all time sources
are based on a quartz crystal. It is obvious that the GS would be for two
reasons 1) when the power is off, the clock is run off the battery, not the AC
line, and 2) Unless I'm wrong, the clock uses the same timer as the CPU.
I'm afraid that I wan't assuming anything. If you re-read my original
post, you'll note that I said: "I wonder about the //gs' clock, though?
Does it use the sixty cycle AC line for an accurate time source while
the computer's on?" These were questions, not assertations. You are
now the second person to misinterpret what I said. But yes, I was
implying that the //gs' clock might use the AC line for a time source.
Why did I dare to make such a dangerous assumption? Perhaps it's
because I like to take apart mechanical / electrical / electronic
gadgets in order to see how they work. Maybe I understand how some
things are designed from doing this. Or perhaps it's simply from
studying electronics and computers over the last four years.
All electrically operated mechanical clocks use the AC line. Almost
every digital clock that does not have a battery backup also uses the AC
line. (VCRs, TVs, and microwave ovens are some examples) Many, but not
all, clocks WITH battery backups still use the AC line WHILE they are
plugged in. It may surprise you to learn that there is a difference
between "turned off" and "unplugged." A computer could still use the AC
line for a time source without being turned on.
How about another example? My digital clock has a battery backup.
While it is unplugged, it does use a quartz crystal, but it isn't very
accurate. While it is plugged in, it does not use the crystal. It
relies on the AC line instead of the crystal for the simple reason that
the AC line is more accurate over a long period than a quartz crystal.
It would take no more than 5 minutes and an ocilloscope to prove this to
you. It does not seem too unreasonable for the //gs to be designed in a
similar manner. (Although I now know that it actually isn't. I checked
with a friend. (qv.)) I'm curious to know why "it is obvious [to you,
maybe] that the GS['s clock could not use anything but a quartz
crystal.]" Do you know what you're talking about, or were you
responding because you could disagree with someone?
As for #1, perhaps I should have said "while the computer's plugged in."
As for #2, you are wrong. According to a friend (I don't have either my
computer or the associated manuals here at CMU) with "The //gs Hardware
Reference Manual (right title?)", the clock in the //gs uses a 32.8 kHz
crystal accessed by the VGC (Video/Graphics Controller) chip.
-- Charles William Swiger
cs4w+@andrew.cmu.edu
PS: Before responding to this post, I suggest that you re-read all of
the other posts on the subject first. It would keep people from making
any more misinterpretations. If you want to argue further, this bboard
really isn't the appropriate place. I would be overjoyed to respond to
any remarks you care to make via email sent to me at the address above.