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

Hi!

Hmmm...not all of them do. I've got a Panasonic VCR that will hold the time for a couple of hours after a power failure and a Sony that seems to be good for a few days. (Of course, both are 'mid-range' machines and the Sony might qualify as lower high end equipment.)

I also have a Panasonic time-lapse VCR that has a NiCad battery in it. I know it will the clock for at least two weeks. The manual says it could last for a month.

William

Hi!

Is that a DS1387 or something else?

I've got some microchannel-based computers that use that module. None have died yet but I know it is a matter of time.

How did you get the module open? Do you have pictures?

William

Hi!

Yep, I've seen the same thing here. I have a number of old Macs (6100,

6300, LC, 9600, SE/30) that still keep decent time on the factory-installed batteries. Some of these have gotten to the point where the machines need to be powered up somewhat often to keep the settings intact.

By compare I have many a G3 CRT-type iMac around (the 2001 models,

350~500MHz) and almost all of them have had to have their batteries replaced.

Looking at things I can see one difference. Most PCs new and old power their CMOS RAM chips from the power supply when they're running. A great many ATX systems seem to keep the CMOS RAM and clock running from the ATX standby supply as long as it is running. Some older PCs also have NiCad or NiMH batteries onboard along with the circuitry to charge them when powered up.

I've investigated the Macintosh a little bit and it looks like the clock/NVRAM battery is constantly pulled upon, even when the computer is on. This could also shorten battery life.

William

Hi!

Okay then, I'll jump in here. I was just waiting for someone to say it. :-)

I am a collector and operator of the IBM PS/2 line of computers. I have a lot of them and most are powered up and running fairly often. Some run 24/7 as servers.

The clocks on these things are--for the most part--extremely accurate. I synchronize the clocks on my computers by way of the 'net to one of the many network time protocol servers in the world. The synchronization happens at least once a week, sometimes more often. I don't think I've ever seen one drift more than a second or three between synchronizations. Of course, it does depend upon the health of the clock battery. Most people never changed them during the lives of these computers...so they still run, but are rather weak. I usually replace them immediately just to avoid a leaking episode.

You can also use these computers as rather large and heavy digital clocks--they have a flourescent display panel near the power switch that is normally used for power on self test codes. Fortunately, it is is user/software-addressable and can be misused in a variety of fun ways.

The only bad thing about these is that these PS/2s make the clocks in almost all of my other machines look like a sad joke. :-)

William

You seem to be saying curiosity is a worthless trait.

I bet you're a republican.

I'm not absolutely certain on that myself, though many explanations have been put forth over the years, with the one I think is probably "the real situation" being that Macs (A) Don't cut the battery out of the circuit when powered up and (B) the battery isn't just keeping the RTC running, but also keeping a chunk of memory (which we call "PRAM" here in Mac-land - holds various fairly-to-really critical information) alive.

I don't know for certain about measured voltage, but I've only actually

*NEEDED* (as opposed to "shotgunning" a startup issue) to replace one battery in my Hmmm... I guess that would be about 8 Macs over the last 15 years or so. That was in a Performa 637CD that I picked up at a thrift store for ten bucks. The machine I'm typing on, a PowerMac 7500, came to me secondhand also, and as far as I have any way to know, it's still running on the factory-installed battery - 10+ years since it came off the line.

--
Don Bruder - dakidd@sonic.net - If your "From:" address isn\'t on my whitelist,
or the subject of the message doesn\'t contain the exact text "PopperAndShadow"
somewhere, any message sent to this address will go in the garbage without my
ever knowing it arrived. Sorry...  for more info

I have an ancient Seiko quartz with a trimmer inside, and by adjusting it I was able to make it accurate to 30 seconds a year. But few cheap watches have them, including none of those I tried in this test.

File that one under D for DUH.

The OP was asking for a technical explanation, and so far the responses have been enlightening. Well, except for yours.

They call themselves NIST these days, and they also still have a dialup service.

There are probably more PC with good clocks than cheap inaccurate clocks. The reason we rarely saw such posts are because people usually complain loudly if their product is a bit "off" for any reason and not many post positive comment on their product.

If everyone were to post every reviews, the bad one would get buried in a hurry.

--
When you hear the toilet flush, and hear the words "uh oh", it\'s already
too late.    - by anonymous Mother in Austin, TX
To reply, replace digi.mon with phreaker.net

Because the processor in your computer might hang or busy itself with other things besides keeping time. It may also have something to do with the clock pulses your computer uses not being exactly divisible into real time.

look for a program called D4. It is a free download and will keep your clock synced to universal time. Also, Widows XP can sync to the same time servers that D4 uses. Both work great!

Not one single person has mentioned that their computer keeps great time. My first personal computer was back in '81 and it didn't even sport a clock. Today I have collected about 16 computers (most of them are in the closet). But the majority of them keeps very good time. Some haven't been fired up in years, and it is really rare to find one off more than 5 minutes. So I am at least one user who has been quite happy with my computer clocks.

______________________________________________ Bill (using a Toshiba 2595XDVD & Windows 2000)

-- written and edited within Word 2000

And your answer was? or were you just posting to post.

In this environment speculation is just about all you have as the machine (subject) is rarely in your hands. Since then a phrase has been invented to take the place of that....and that would be...YMMV

But even worse, he posted

time

There are two ways to do as suggested. The first is to make 'Benjamins' part of the technical facts during design. The second is to do the design, then let bean counters change the design per what they 'feel' is not worth the bucks. The latter is too often how GM cars are designed. Which is why a GM car needs two extra pistons to get the same horsepower as the competition. Which is why GM cars even in the 1990s required annual wheel alignment. Which is why GM cars would have what appeared to be computer failures when failure was really due to cheap connectors. Classic examples of failures when the design is modified after the design.

Two examples: how 'Benjamin' decisions become part of a successful design verses how 'Benjamin' decisions after application of technical facts makes bankruptcy.

Meanwhile, the technical reas> There is no technical explanation except that the technology that

Who the hell made you the NG God who should say who should and who should not post a reply. In short, Fuck You!

Don't tell me that the time kept by your computer does not require a processor and that it does not ever hang because that is bullshit.

The fact that the OP asked the question leads me to assume that his clock on his computer not keeping accurate time is annoying him. Therefore, I recommended the Program called D4. It is a solution to the problem and it works.

IBM did a lot of things wrong with their PCs in terms of performance, upgradability and user friendliness, but their quality was second to none un until the end of the PS/2 line. Andy Cuffe

snipped-for-privacy@psu.edu

Why so many "it might be this" or "it might be that" or "time is updated from the internet"? Every posts says nothing useful AND does not answer the OPs question. OP even clarified the question when some replies were rubbish.

The answer -- technically -- was posted without speculation. Processor hangs obviously do not affect that clock operation - it one first learned how something works before posting. The OP posted this - a technical question that required technical knowledge before replying:

See that word "might" ? That word "might" means the poster does not know the answer and therefore should not have posted. Any> Because the processor in your computer might hang or busy itself

What Happens and Why There is a "CMOS clock" in your computer which is powered by a tiny battery. As long as the battery is good, this clock keeps the correct time, and each time your computer is restarted, Windows98 reads its initial time from the CMOS clock. However, while Windows98 is running, it keeps track of the time on its own without continuing to check the CMOS clock, and keeping track of the time is not the only thing Windows has to do. The busier your system gets, the more likely it is to lose time. Generally, the longer you use your computer, the further behind it gets. When you leave your computer on for an extended amount of time, the Windows clock (displayed on the taskbar) may lose from two minutes to an hour per day.

Likely culprits Anything that makes your computer especially "busy" can take Windows' attention away from its time-keeping function and lead to this "losing time" symptom. If you're running lots of programs, or even just one or two very demanding programs, you may see the computer clock losing time. Furthermore, anything you are running which causes the computer to have to spend time "watching" for something to happen can also lead to a slow clock. Here are the most common culprits:

a.. Games and other video-intensive programs

b.. Screen savers and "scheduling" programs

c.. Internet chat programs (ICQ, IM, etc.)

d.. Playing MP3 files, CDs, or internet audio

e.. Anti-virus programs

f.. Processor-intensive applications