OT: System Clock Loses Hour

[Fyron]

I've got a winXP box that loses an hour consistently from its system clock. It is basically setting itself an hour earlier than what the real time is, some time after I fix it. I've fiddled with DST settings and such, but every combination still results in losing an hour after I adjust it. The machine has also now lost the ability to sync with time.windows.com or nist.time.gov, in spite of taking efforts to reset and reinitialize w32time service. I even tried disabling the "Windows Time Service" entirely, but it still sets it back an hour.

I thought it might have been a bug with the recent DST update, but the problem persists after uninstalling it.

I don't think this is a CMOS battery issue, because the time is not randomly losing minutes. It also does not lose multiple extra hours; it only sets it back an hour before the real time.

Has anyone encountered this before? Any idea on how to resolve it, short of reinstalling windohs?

For reference, I have attached a dump of the w32time registry settings (http://www.shrapnelcommunity.com/threads/download.php?Number=509773).

[se5a]

set it an hour forward so when it looses an hour it will be right?

might be an interesting test to see if it looses one hour or two.

[Atrocities]

I am having the same problem fyron. On all my XP machines

[douglas]

Somewhere in the menu brought up by double-clicking on the system clock is a simple checkbox for synchronizing the clock through the internet. Either uncheck this box or change your time zone setting.

[Fyron]

I'll try setting it an hour ahead, but I have my doubts...

douglas said:
Somewhere in the menu brought up by double-clicking on the system clock is a simple checkbox for synchronizing the clock through the internet. Either uncheck this box or change your time zone setting.

As I said in the OP, I have already tried every combination of those settings, and the system can no longer synchronize with Internet time... This is not an issue of how to use the system, it is a bug that has cropped up for which I am trying to find a solution. Also, that isn't really a solution anyways, because it will only perform the sync once a week. Even when the box was still able to do that sync, the clock would still get set back an hour, about an hour later. I wrote a little script a week ago to manually perform the sync (w32tm /resync) once an hour, which worked to fix the clock, but the time service is no longer able to sync even manually on the command prompt.

[PvK]

It sounds like it's the Daylight Savings Time bug.

This year, various places (including most or all of the USA) changed the date when Daylight Savings Time went into effect. Microsoft Windows requires an update, or it will follow the old schedule and get mixed up, etc. It's especially nice since many networked computers rely on a time server, and will thus systematically re-set to the wrong time; and some M$ network software reportedly requires clocks to be in sync in order to authenticate http://forum.shrapnelgames.com/images/smilies/rolleyes.gif . Win 2003 and Win XP SP2 should patch it if you go to the Microsoft Update web page and run their software (especially if you have their insulting "Genuine Advantage Genuinization Tool" or whatever-the-frack they call it). Earlier Windohs sufferers (including Win2K, and XP SP1 it appears) need to manually install a patch, which does exist but isn't the easiest thing to find...

As for why the frick Microsoft Windohs won't even let you set your own system clock manually and have it not be a problem, well, I'd chalk that up to typical M$ over-engineered "we know what you want - here, let us jam it down your throats" design.

PvK

[Captain Kwok]

Yeah, that bug is affecting some of my co-workers. Their e-mails are constantly being sent 1 hour in the past. http://forum.shrapnelgames.com/images/smilies/tongue.gif

[Renegade 13]

I have that exact same problem on my computer as well...I just readjust the clock and then ignore it, since once the "normal" DST time comes around, it'll fix itself I'm sure http://forum.shrapnelgames.com/images/smilies/tongue.gif

[Fyron]

Yeah, this didn't start until after I installed the DST patch. It doesn't matter if I turn off the DST updating, it still sets back an hour. I even uninstalled the DST patch, but no dice.

Setting it 2 hours ahead didn't matter.

I think I'll try living in Arizona for a while... see if that helps. hehe

[Fyron]

Arizona time zone didn't help.

[Baron Munchausen]

PvK said:
It sounds like it's the Daylight Savings Time bug.

Earlier Windohs sufferers (including Win2K, and XP SP1 it appears) need to manually install a patch, which does exist but isn't the easiest thing to find...

As for why the frick Microsoft Windohs won't even let you set your own system clock manually and have it not be a problem, well, I'd chalk that up to typical M$ over-engineered "we know what you want - here, let us jam it down your throats" design.

PvK



Yay! It's finally catching. http://forum.shrapnelgames.com/images/smilies/happy.gif My meme is finally spreading. Spell it WINDOHS to remind people how cruddy it is. Not only does this recall Homer Simpson's "D'ooh!", it carries some association with Playdoh, too. http://forum.shrapnelgames.com/images/smilies/happy.gif

[narf poit chez BOOM]

Playdoh is for people who can't make modelling gunk themselves.

We've got a recipe.

[PvK]

Sounds like the DST patch doesn't work... sigh.

Ya, Baron Munch., the fine products of M$: M$ Windohs, M$ Outhouse, M$ Abscess, M$ Turd...

[Atrocities]

there is a problem and microsoft does NOTHING to fix it.

[Baron Munchausen]

PvK said:
Sounds like the DST patch doesn't work... sigh.

Ya, Baron Munch., the fine products of M$: M$ Windohs, M$ Outhouse, M$ Abscess, M$ Turd...



Don't forget Internet Exploder. http://forum.shrapnelgames.com/images/smilies/laugh.gif

[Baron Munchausen]

Atrocities said:
there is a problem and microsoft does NOTHING to fix it.



Actually, there is a problem because Microsoft tried to fix something. This is typical. The 'fix' for the animated cursor attack is known to break some software packages. They released it anyway because they were under a lot of pressure to 'do something' about this security hole. They are probably frantically scurrying around in their bureaucracy right now, trying to figure out how to fix the DST thing properly.

[Fyron]

Geeze.. you guys will take any opportunity you can to bash MS. It's not appropriate in threads where people are genuinely trying to get help; take it elsewhere.

[Baron Munchausen]

We are just noticing that there is no help to be had -- not entirely off topic. http://forum.shrapnelgames.com/images/smilies/tongue.gif

[Will]

Perhaps it is the BIOS system clock that is off? Maybe your hardware is trying to be too smart and compensating for DST, while the DST patch doesn't know how to deal with it.

Long shot, but if you've already tried everything else...

[Santiago]

If it wasn't for windows, third party software mfg's would have nothing to bash to be able to push their own products and little incentive to make anything better.

As for the time problems, I have had no problems on 4 pc's with the DST update so I'm sorry to be of little help. Plus I don't have the time displayed.

Also I have retreated from keeping track of time per se. I don't wear a watch anymore, eat when I'm hungry not at some arbitrary time set by some clock , ie Noon. I sleep when I'm tired, get up if it's daylight out. My internal biotime has always served me well by waking me a few minutes before the alarm, so I don't use alarm clocks anymore. I don't miss my flights and I'm very relaxed nowadays. And generally feel better for it.

[AngleWyrm]

I found an hour. It was in a tray with the spare change and car keys. It's in fairly good shape -- looks like it was only used a couple times to watch TV. If anyone's clock is still missing an hour, I'll post a link for downloading it (takes about an hour).

[narf poit chez BOOM]

/me hands AngleWyrm a medal.

Cause that post deserves one.

[Atrocities]

AngleWyrm said:
I found an hour. It was in a tray with the spare change and car keys. It's in fairly good shape -- looks like it was only used a couple times to watch TV. If anyone's clock is still missing an hour, I'll post a link for downloading it (takes about an hour).



Please post the link.

[se5a]

aye, did you come up with that yourself?

[AngleWyrm]

heh, ya. Was gonna make a countdown timer page to link to also, but...it'd take me more than an hour http://forum.shrapnelgames.com/images/smilies/tongue.gif

For a more interesting hour, Gravity Probe-B (http://einstein.stanford.edu/) preliminary reports are due next week. Frame dragging and spatial distortions.

[narf poit chez BOOM]

Cool.

One thing I've wondered is, if mass is partially relative to the speed of light, what is the speed of light relative to?

[douglas]

narf poit chez BOOM said:what is the speed of light relative to?


Anything and everything that isn't accelerating, as long as the reference frame you use has the object in question at rest.

[narf poit chez BOOM]

Ah, but at rest relative to what?

[douglas]

Your reference frame. To do meaningful calculations in any situation where relativity has a significant impact you will at some point simply have to arbitrarily define what is at rest. You can choose absolutely anything you want, it will all work out correctly regardless, but you have to pick something.

[Possum]

Hehehe, read this and had to chuckle.

Yes, it's the change in daylight savings time, but that's not why I'm snickering.

I'm snickering because 98SE doesn't have this problem at all. I set time to PDT manually on the new day. On the usual day, I fired up my bawx, and 98SE politely said, "Hey, boss, it's time to set the clock ahead. You want me to do that?"

I replied "no", and that was the end of it http://forum.shrapnelgames.com/images/smilies/happy.gif

Why, why, why are more sophisticated mechanisms, (and there can be no doubt that XP is far more advanced and sophisticated than 98SE), always more subject to quirky little problems like this?

[Possum]

douglas said:
Your reference frame. To do meaningful calculations in any situation where relativity has a significant impact you will at some point simply have to arbitrarily define what is at rest. You can choose absolutely anything you want, it will all work out correctly regardless, but you have to pick something.



Somewhere, the shade of Herr Doktor Werner Karl Heisenberg is laughing heartily http://forum.shrapnelgames.com/images/smilies/happy.gif

[AngleWyrm]

How about a squirrel busily gathering nuts? I'm at rest relative to that.

Energy of squirrel = Mass of nuts times Chewing velocity. Squared.

Cause everyone should have a square meal.


narf poit chez BOOM said:
One thing I've wondered is, if mass is partially relative to the speed of light, what is the speed of light relative to?


That'll probably come up again with lots of jargon in it, if/when we break the light-speed barrier.

"Just cause it got from here to there and back before light did, doesn't mean it was actually travelling faster than light. Relative to itself, the photons experienced phase shifting dopplar blahzidy blah effects, so really it's not going faster. Einstein be praised!"

[se5a]

that question always confused me too narf.

[AngleWyrm]

If mass and energy are the same thing, what happens when I move my coffee cup from one side of the table to the other?

[narf poit chez BOOM]

Ok, so then, what's the difference between the speed of light here and at jupiter?

Remember, the planet Earth and the planet Jupiter are moving at different speeds in different directions.

Now think about that some.

[douglas]

Zero. The speed of light relative to Earth in a reference frame where Earth is at rest is exactly the same as the speed of light relative to Jupiter in a reference frame where Jupiter is at rest. This simple fact is the basis of most or all of the theory of relativity.

[narf poit chez BOOM]

I might point out that earth and jupiter are not at rest relative to each other.

Nor am I contradicting you.

It just seems to have some interesting implications.

[se5a]

nor is the whole system at rest eaither...

[Fyron]

narf poit chez BOOM said:
I might point out that earth and jupiter are not at rest relative to each other.

Which doesn't matter when you aren't comparing Earth and Jupiter.


Why are we talking about this in a windows issue thread, again?

[narf poit chez BOOM]

...I dunno.

Fyron...If the speed of light is relative to the nearest large body...Think about it.

[Phoenix-D]

The speed of light isn't relative to anything; that's the entire point of relativity. Through vacuum it always travels at C. (which is what causes all the other weird side effects of relativity, like time dilation)

[narf poit chez BOOM]

Phoenix-D said:
The speed of light isn't relative to anything; that's the entire point of relativity. Through vacuum it always travels at C. (which is what causes all the other weird side effects of relativity, like time dilation)



douglas said:
Zero. The speed of light relative to Earth in a reference frame where Earth is at rest is exactly the same as the speed of light relative to Jupiter in a reference frame where Jupiter is at rest. This simple fact is the basis of most or all of the theory of relativity.


...Hey, I'm just working off what other people say.

[Phoenix-D]

Which doesn't contradict me at all, he just didn't explain it the same way. http://forum.shrapnelgames.com/images/smilies/happy.gif

[narf poit chez BOOM]

Hmm, yes, then he didn't answer my question, because I specifically asked the difference between the speed of light here and at jupiter *While both are moving*. Not at rest.

[Fyron]

The speed of light is the same here as it is near Jupiter. It is not relative to the nearest large body. That is the point; from all reference frames, the speed of light is the same.

This contradicts what you would expect, given the behavior of matter. eg: If I throw a ball, its speed of travel is just the speed at which I threw it, when we are discussing the reference frame of sitting here an Earth. However, the ball is really moving a lot faster than that. Its absolute speed is the speed at which I threw it, plus the speed of the Earth's rotation, plus the speed of the Earth's orbit around the sun, plus the speed of the suns movement in the galaxy, plus the movement of the galaxy... (though we really should be talking about velocities, not speeds.)

But with light, none of that is relevant; it travels at the same speed regardless of reference frame. Shine a flashlight from the surface of the earth, the light moves at c. Shine a light from (a stable, geosynchronous) orbit, the light moves at c. Shine a light from orbiting Jupiter, where you are moving at a different speed than orbiting Earth, the light still moves at c.

[narf poit chez BOOM]

Ah, well, thanks. That answers my question and brings up a different interesting thought - That, using the speed of light, it should be possible to calculate an absolute frame of reference.

Anybody done that yet?

[AngleWyrm]

Might be able to do something like this with red/blue doppler shifts seen in surrounding stars.

For instance, take a given star and measure it's color-band shifts in spring when the earth is travelling in one direction, then again in fall when the earth is travelling in another.

[narf poit chez BOOM]

Why not just measure the speed of the earth relative to the speed of light?

[douglas]

narf poit chez BOOM said:
Ah, well, thanks. That answers my question and brings up a different interesting thought - That, using the speed of light, it should be possible to calculate an absolute frame of reference.

Anybody done that yet?


The whole point of relativity is that an absolute frame of reference simply does not exist. No matter what frame of reference you use, light travels at c in that frame of reference. If you measure the speed of a planet as 300 km/s and then measure the speed of light relative to that planet in the same direction of movement, you will get c - 300 km/s. In the opposite direction you will get c + 300 km/s. If you then change your reference frame by landing on the planet so you are moving along with it and the planet is at rest from your perspective, repeating your measurements would tell you that light is moving at c relative to the planet in all directions.

[se5a]

sooo...
the speed of light IS a frame of reference?

if I have a torch (flashlight) moving at 0.5c and I turn it on, the light from the torch would be moving at c right?
but, ralitive to the torch the beam is moving at 0.5c or 1.5c depending on weither the torch is facing forwards or backwards.

what narf is saying, is if you use said torch, and mesure the amount of time it takes for light to get from the torch to an known distance, you should be able to work out how fast, and in what direction the torch is moving, and therefore be able to say what "still" is.

[douglas]

se5a said:what narf is saying, is if you use said torch, and mesure the amount of time it takes for light to get from the torch to an known distance, you should be able to work out how fast, and in what direction the torch is moving, and therefore be able to say what "still" is.


What I'm saying is that this exercise will just tell you that whatever you had already decided was "still" for purposes of calculation was, in fact, correct. Simply stating that something is moving at speed X inherently requires a definition of what is at rest. The point of relativity is that this definition can be absolutely anything and the speed of light will be the same regardless.

For example:
Suppose you are on Earth trying to carry out that torch experiment. You carefully measure the distance between the torch and a mirror, you light the torch, and you time how long it takes for the reflected light to get back to you. You will calculate from this data that you (actually your measuring device) are at rest.

Next, you move to Mars and repeat the whole experiment. You will again calculate that your measuring device is at rest even though it is clearly moving at a different velocity than it was on Earth. Consideration of this simple fact is how the entirety of the Theory of Relativity was derived.

[se5a]

that contradicts what fyron said.
I think...

[narf poit chez BOOM]

I think it may have something to do with the time slowdown.

[AngleWyrm]

douglas said:
For example:
Suppose you are on Earth trying to carry out that torch experiment. You carefully measure the distance between the torch and a mirror, you light the torch, and you time how long it takes for the reflected light to get back to you. You will calculate from this data that you (actually your measuring device) are at rest.




There was no mirror in the original torch example. Whether or not it has travelled the same distance/taken the same time in both directions is what is at question.

In this illustration, a red bulb sends a signal to a red mirror. Time goes by, and the bulb/mirror experiment moves to position green. The pulse strikes the green mirror and heads back. More time goes by, and the bulb & mirror move to position blue. The light pulse returns to the bulb at position blue.
http://home.comcast.net/~anglewyrm/roundtrip.jpg
If it takes a different amount of time on each leg, then we can say that the experiment moved (as illustrated), and the speed of light was the same throughout. And as Narf said, it implies that the speed of light is relative to something other than the bulb&mirror. If it comes back in the same amount of time that it took to go out, then something is amiss.

Shouldn't the stars change color sleightly as the earth approaches them, and then as the earth pulls away? If earth's motion about the sun is about 29km/sec, or roughly 0.001% speed of light, then there should be a +/-1% red/blue shift in colorbands throughout the year.

[douglas]

AngleWyrm said:If it comes back in the same amount of time that it took to go out, then something is amiss.


According to any device that is moving at the same velocity as the experiment, that is exactly what would happen. Scientists tried to do something like that to determine the speed of the Earth as long ago as 1881 with the most famous "failed" experiment ever (http://en.wikipedia.org/wiki/Michelson-Morley_experiment). No matter what speed you are moving at, your perceptions of time and distance will conspire to give the same result.


AngleWyrm said:Shouldn't the stars change color sleightly as the earth approaches them, and then as the earth pulls away? If earth's motion about the sun is about 29km/sec, or roughly 1% speed of light, then there should be a +/-1% red/blue shift in colorbands throughout the year.


Yes. The speed of the light involved remains the same, however, no matter how much the light wave's frequency may change.

[AngleWyrm]

The Michelson-Morley experiment was trying to detect wind from the earth passing through aether--the fluid that conducts light waves. It failed to notice any, and is considered proof that there is no aether.

your perceptions of time and distance will conspire to give the same result.
I recently read a philosopher's quote, that I'll paraphrase: A theory which claims to be unmeasurable is probably best left unspoken as well.

Walter Ritz proposed Emitter Theory (http://en.wikipedia.org/wiki/Emission_theory) that suggested light speed was added to local motion, but that was not observed in binary stars as they orbited each other. The example given is that light from the approaching orbit would appear to be C+orbital speed, and therefore the light would overtake previous light, and our observation would be...strange. (maybe reversed)

[douglas]

AngleWyrm said:
your perceptions of time and distance will conspire to give the same result.
I recently read a philosopher's quote, that I'll paraphrase: A theory which claims to be unmeasurable is probably best left unspoken as well.


Except it's not unmeasurable. There are all sorts of experimentally measurable implications of the constancy of the speed of light independent of reference frame, and many of them have been tested and found to be correct. For example, take two atomic clocks and synchronize them. Then put one of them on a supersonic jet, have it fly around at high speed for a while, and put them back next to each other. One of the implications of the speed of light being constant regardless of reference frame is that the two clocks will now be out of sync - they will disagree about how much time has passed since they were synchronized. The exact amount of the difference can be calculated using equations derived strictly from conceptual analysis of what would have to be true for the observed fact about the speed of light to be true. This experiment has actually been conducted, and the result was well within experimental error of the calculated value.

[AngleWyrm]

http://home.comcast.net/~anglewyrm/workthatpremise.jpg
(Tricia Helfer, Battlestar Galactica)

[capnq]

narf poit chez BOOM said:
Why not just measure the speed of the earth relative to the speed of light?

Because that results in the inconveniently small number (0.001% of c) that AngleWyrm mentioned further upthread.

[narf poit chez BOOM]

...I'm totally lost. Too many explanations.

[Atrocities]

Ok so I got sick and tired of manually adjusting the clock every few days so I sent in a customer / tech request to Microsoft asking for help.'


Even after I applied the DST update to my computer it keeps resetting the system clock back an hour every few days. What the hell is the problem? I have to manually reset the time on my pc every few days even after applying the most recent update from the DST web site. I have tried to reinstal the patch again and it won't install because it is ALREADY installed. Please fix this bug, as there are many others experiencing the same issues. Thanks




This is the response I got.


Thank you for contacting Microsoft Online Customer Service.

I understand that your computer shows the time an hour behind even after installing DST Update.

For assistance with your issue, I would like you to know that if you are experiencing issues with Daylight Savings Time (DST), please use the automatic patches provided by the DST Help Center through the following link:
http://support.microsoft.com/gp/cp_dst

However, if you have additional questions or issues with the updates, please contact our Support Professional at 1-800-MICROSOFT.

Ray, I hope the above information is helpful and I appreciate your patience in this regard.

Thank you for using Microsoft Products and Services.

Basavaraju
Microsoft Online Customer Representative




*Sigh*


Um, Basavaraju, or whoever reads this, your response to my concern is, as usual, of absolutely no help to me. Did I or didn't I state in my message that I had already been to the DST web site and tried all the "fixes?" I have to be honest, I am not at all impressed with your cookie cutter response. I find it proof positive that Microsoft is exactly what people think it is, the home for low brow software programmers who couldn't get a job at a real software development company. I mean, had you taken the time to actually read my concern you would have known that I did go to the DST web site BEFORE I contacted Microsoft with this problem? I sincerely hope that at some point in time you die a horrible painful slow death. And thank you for your assistance.



What can I say, I am utterly disgusted. /threads/images/Graemlins/Envy.gif

[Suicide Junkie]

capnq said:

narf poit chez BOOM said:
Why not just measure the speed of the earth relative to the speed of light?

Because that results in the inconveniently small number (0.001% of c) that AngleWyrm mentioned further upthread.

Actually, it results in a relative speed of 1.0 c (relative to light)

You could measure your speed relative to the earth, or relative to the sun, or relative to the calculated center of the galaxy, or relative to the middle of the local cluster.

But light itself always travels a 1.0c relative to the observer.

[narf poit chez BOOM]

Gotta be the time dilation.