Why aren't computer clocks as accurate as cheap quartz watches?

Most older motherboards used the Dallas real-time clock chip, which had
a cheapo oscillator built in, and could use an external crystal for
precision timekeeping. A good crystal will add a couple of dollars to
the price of the motherboard, but the clock ic itself was over ten
bucks. I use these chips in some other pieces of equipment and they are
very accurate when given a good crystal for a timebase.

I don't know what they put on modern motherboards. I throw computers
out when they stop working now.

However, the datasheet for the Dallas chip gives some hints that may be
useful here. If the chip is put in an electrically noisy device like a
computer, it can be accelerated by stray signals picked up by the clock
circuit. I believe that NTP machines (usually) set the RTC on shutdown,
as the cpu is more likely to be on time if it's been syncing itself
with an atomic standard.

wrote:
James Sweet wrote:


I have one PC that will lose or gain hours at a time if the machine goes
into standby, another that does ok but still drifts enough that I have
to use a utility to keep it in sync with the atomic clock. On the other
hand a Sun workstation that I fire up occasionally to play with keeps
excellent time, even after sitting unplugged for 6 months it's usually
within a few seconds but then it was originally $25K.


The only PC motherboards I've seen that consistently kept accurate time
all had Dallas clock chips in them, and some of those chips were still
running properly from their internal lithium cells (permanently
encapsulated) 10 years later.



Yeah the Sun has a Dallas in it, in fact I had to hack in a new battery
in one of them as the original went dead and modern chips won't work in
it. Come to think of it I had a 486 PC with a Dallas, I think that was
the most recent though.

Watches have an internal capacitor to adjust for each
crystal. I have never seen that capacitor on motherboards
since (I believe it was) the IBM AT. Furthermore, the PC
clock operates at two significantly different voltages that
will change crystal frequency. Battery voltage and voltage
when PC is powered will cause additional fluctuation. Which
voltage should they adjust the capacitor to? Just easier to
not install and adjust the capacitor.

wrote:
I'm looking for a techical explanation.

On Wed, 26 Oct 2005 17:46:28 -0500 jakdedert
wrote:

wrote:
Why do the battery powered clocks in personal computers tend to keep
worse time than quartz watches, even the $1 ones?

I've often wondered the same thing...also why a $9 alarm clock will have
provision for battery backup, but $200 VCR (back when VCRs were $200)
needed to be rweset with each blip in the mains voltage.

I recently discovered that early Sony Beta VCRs (at least) actually
had back-up NiCad batteries which would keep the clock going for
seveeral minutes of blackout. I just replaced one today, with a NiMH,
and will replace a couple more in the next few days.

I'm sure I knew about these at one time, but I just forgot about them
until other things got me lnside that VCR again.

-
-----------------------------------------------
Jim Adney
Madison, WI 53711 USA
-----------------------------------------------

On 26 Oct 2005 15:03:04 -0700 wrote:

Why do the battery powered clocks in personal computers tend to keep
worse time than quartz watches, even the $1 ones?

There are 2 causes.

The first cause is that the internal clock (real time clock) on the
motherboard is a device which the motherboard manufacturer much buy
from someone else. They can buy this item in various degrees of
precision, but the cost goes up as you request more precision. Since
there is lots of pressure to keep cost down, but not so much pressure
to keep good time, the motherboard manufacturers don't press for
highly accurate real time clocks.

The second reason was explained by someone else. It's the fact that
the OS clock can fall behind the real time clock due to the way the OS
is written. This error is minimal when your computer is first booted
up, but it can become significant if it is left running for days. This
error is reset to zero each time you reboot.

-
-----------------------------------------------
Jim Adney
Madison, WI 53711 USA
-----------------------------------------------

Well I can't answer your question but for those who don't know, XP has
an automtic time synchronization feature which operates weekly. You
have to be on line of course. The synchronization can also be done
manually. See: StartControl PanelDate and Time. My pc keeps time
very well.

Glenn

On 26 Oct 2005 15:03:04 -0700, wrote:

Why do the battery powered clocks in personal computers tend to keep
worse time than quartz watches, even the $1 ones?

The computer batteries measure fine, at least 3.15V.

I thought that the problem was temperature swings in the computers
(25-38C), but a couple of cheapo watches taped inside the computers
kept better time.

On Wed, 26 Oct 2005 23:25:49 GMT, Don Bruder wrote:



Dunno if it's still true in PC-land - I've been living in a Mac world
for a LONG time now - but when I was playing with them years ago, the

I've always wondered why the batteries in macs run down so quickly. I
rarely see a PC newer than 10 years old with a bad battery, but I
consistently see 3-5 year old macs with totally dead clock batteries.
Macs even have a much larger (and more expensive) lithium cell than
most PCs. I've seen a few 15 year old 486's with the same type and
brand battery used by apple that still measures full voltage.

I think the answer to the original question is just no quality control
for clock accuracy. My current PC is pretty accurate, but it's still
not as good a cheap watch.
Andy Cuffe

-- Use this address until 12/31/2005

-- Use this address after 12/31/2005

In message Jim Adney
wrote:

On 26 Oct 2005 15:03:04 -0700 wrote:

Why do the battery powered clocks in personal computers tend to keep
worse time than quartz watches, even the $1 ones?

There are 2 causes.

The first cause is that the internal clock (real time clock) on the
motherboard is a device which the motherboard manufacturer much buy
from someone else. They can buy this item in various degrees of
precision, but the cost goes up as you request more precision. Since
there is lots of pressure to keep cost down, but not so much pressure
to keep good time, the motherboard manufacturers don't press for
highly accurate real time clocks.

And at the end of the day, most users have internet access these days,
so it's less critical since the clock will be updated by most modern
operating systems automatically.

The second reason was explained by someone else. It's the fact that
the OS clock can fall behind the real time clock due to the way the OS
is written. This error is minimal when your computer is first booted
up, but it can become significant if it is left running for days. This
error is reset to zero each time you reboot.

It's worth noting that an OS can correct for this too, either by using
an external time source when available, or periodically resynchronizing
from the BIOS clock if no reliable external source is available.

--
Is it my imagination, or do buffalo wings taste like chicken?

On Wed, 26 Oct 2005 23:25:49 GMT, Don Bruder put
finger to keyboard and composed:

In article . com,
wrote:

Why do the battery powered clocks in personal computers tend to keep
worse time than quartz watches, even the $1 ones?

The computer batteries measure fine, at least 3.15V.

I thought that the problem was temperature swings in the computers
(25-38C), but a couple of cheapo watches taped inside the computers
kept better time.


Dunno if it's still true in PC-land - I've been living in a Mac world
for a LONG time now - but when I was playing with them years ago, the
battery-backed real-time clock was read once at startup to set the
computer's software clock, which then kept time by counting clock
interrupts generated by the motherboard timing circuitry.

It's true for Win98, but I don't know about XP.

Just for fun, here's something interesting that I discovered recently:

http://groups.google.com/group/comp....e=source&hl=en

I could make time run backwards on a Win98SE machine by doing
something innocuous.

-- Franc Zabkar

Please remove one 'i' from my address when replying by email.

wrote:
mike wrote:

Supply and demand.
If your watch kept bad time, you'd send it back.
If your computer keeps bad time, you'll reset the clock
and bitch about it on the internet.


I'm looking for a techical explanation.


Very often, the economic or political considerations
dwarf any technical consideration.
The technical explanation is that they do what's cheap.
Wishing for a grand technical reason won't make it so.
mike

--
Wanted, Serial cable for Dell Axim X5 PDA.
Return address is VALID but some sites block emails
with links. Delete this sig when replying.
FS 500MHz Tek DSOscilloscope TDS540 Make Offer
Bunch of stuff For Sale and Wanted at the link below.
MAKE THE OBVIOUS CHANGES TO THE LINK
htremovethistp://www.geocities.com/SiliconValley/Monitor/4710/