Windows Clock drift - Time way off

[roxy99]

DFA,
Don't bother changes the cmos battery. I did so and even tested the old CMOS battery and felt all proud seeing it was 2volts insrtead of 3. New battery was put in and no dif.

The constant is the nforce2 mobo and certain applications. Its obviously a windows and Nforce2 and SageTV issue. Some combination of that. Which nmeans it could be anything,

Not really a fix but you should try underclocking to FSB of 166. Just enter your bios and choose advanced settings and you should find it.

If that fixes the problem then let me know because so far that is th only thing that helps,

[AboveUnrefined]

More recommendations for time issues lie here -- http://forums.sagetv.com/forums/showthread.php?t=15524

clockmon sounded good from what I read but I never had time issues so I didn't really need to test it out.

[DFA]

Roxy99:

Thanks. However, my primary bus speed is already 166MHz. Also, note that my runaway clock problem is sporadic and grossly fast by as much as 25%. Can be as little as 5 sec per minute fast. When it occurs, it is not always consistant in the magnitude of error.

I have been dubious about the battery thing myself but as I Google around on this subject, I keep finding things about CMOS batts. Here is one quote from an OC's Website:

" The A7N8X has some bios issues what with not saving properly, acting stupid, not doing what you tell it to do. Fixing this problem is very easy! If you replace the battery on the motherboard with a new battery (such as an Energizer) of the same value and model these bios problems are no more! The problem, or so we overclockers think, is that the standard battery only gives out 2.95v although it is said to be 3v. If you put in an Energizer, you will get 3.3v! This seems to fix the problems."

That is not to say a battery is my issue but got to try something even if it is wrong. Battery is cheap and easy (in my case not that easy).

[roxy99]

I spoke to soon. I have the same prob with bus 166. Also, the app. L'il atomic clock mentioned earlier here does not seem to activate and reset time.

Its as though Sagetv blocks the time update. Please note- when I DON'T have Sagetv running, the clock keeps perfect sync - does'nt even need l'l atomic clock.

Sagetv may be a great application but its useless to me if I can't get the clock thing straigtened out. I gain about 5min every half hour that Sage is running.

I may try a Sagetv re-install. Failing that, I may have no choice other than a dreaded windows xp clean install and start from scratch. There must be other besides us and there has to be a solution.

[roxy99]

Originally Posted by AboveUnrefined

More recommendations for time issues lie here -- http://forums.sagetv.com/forums/showthread.php?t=15524

clockmon sounded good from what I read but I never had time issues so I didn't really need to test it out.

I'll give Dimension4 a try.

[DFA]

roxy99:

That's about 10 sec/min. fast which is along the order of magnitude that I am seeing.

What I have found that has vastly reduced, but not eliminated, the occurrance of this problem for me is to clear and keep clear the Java temporary download cache of applets that have been put there by various web sites. That can be found by opening the Java applet from "Control Panel".

I do not think this is a direct Sage problem. Perhaps something to do with the Java platform that Sage is built on but not necessarily Sage itself. This is not a wide spread issue for Sage or Java in general or there would be tremendous traffic on this. But it obviously is not entirely isolated events either but rare it seems. Not on anybodys radar which puts us on our own.

I thought I was alone on this and had never bothered posting about it other than a mention in another thread in regard to a different problem someone else was having. I now see that is not entirely the case.

EDIT: What sercvice pack are running on XP. I am at SP1 but thinking it high time to move on to SP2.

EDIT EDIT: It does not make a lot of sense that deleting Java applets would help since these should not be doing anything just by being there and may just have been coincidental for me but I'll take any relief real or imagined. I go long periods of time w/o issue and than run into the problem again. I've been living with for over a year on and off.

Do you recall the battery type/model that the A7N8X uses so that I can have one on hand BEFORE opening and removing the battery only to drive around town and not find one (my luck sometimes).

DFA

[roxy99]

Originally Posted by DFA

roxy99:

That's about 10 sec/min. fast which is along the order of magnitude that I am seeing.

What I have found that has vastly reduced, but not eliminated, the occurrance of this problem for me is to clear and keep clear the Java temporary download cache of applets that have been put there by various web sites. That can be found by opening the Java applet from "Control Panel".

I do not think this is a direct Sage problem. Perhaps something to do with the Java platform that Sage is built on but not necessarily Sage itself. This is not a wide spread issue for Sage or Java in general or there would be tremendous traffic on this. But it obviously is not entirely isolated events either but rare it seems. Not on anybodys radar which puts us on our own.

I thought I was alone on this and had never bothered posting about it other than a mention in another thread in regard to a different problem someone else was having. I now see that is not entirely the case.

EDIT: What sercvice pack are running on XP. I am at SP1 but thinking it high time to move on to SP2.

EDIT EDIT: It does not make a lot of sense that deleting Java applets would help since these should not be doing anything just by being there and may just have been coincidental for me but I'll take any relief real or imagined. I go long periods of time w/o issue and than run into the problem again. I've been living with for over a year on and off.

Do you recall the battery type/model that the A7N8X uses so that I can have one on hand BEFORE opening and removing the battery only to drive around town and not find one (my luck sometimes).

DFA

Hey Dimension 4 really works! Thanks now I at least have a workaround.

First of all, as I said don't bother changing the battery. I did so and it was a waste of time. The battery is probably a CR2032 3.0v lithium battery, about the size of 3 quarters stacked.

I am running SP2. Don't be so kind to Sage, of course its the programs fault. Its not Java's fault nor Windows XP because the only time my clock goes into 'steroid mode' is when SageTv is running. My bank uses Java technology for online users so I have much more faith in Java than Sage.

As far as being isolated, you can bet that every Nforce2 user is having the same problem. There is no flood of complaints simply because most users probably just shrugged their shoulder and started using Dimension 4 to autosync every 5 min. My board is not an Asus A7N8X, but rather a DFI Nforce2 board.

Now about what you did to clean the Java cache, can you be more specific? I wonder why that would cause the problem..

I hope Sage Tech support reads these forums and comes up with a fix.

[DFA]

Roxy99:

The Java applet interface has varied a bit over revisions so what I provide here may or not be accurate. What I am stating is in reference to 1.5_06.

Open Control Panel and find the Java icon and open the Java applet. On the "General" tab, look to the bottom under "Temporary Internet Files" and click on the "Settings" button. In the dialog box that now has opened, click on the "View Applets" button. I delete anything I find there (empty). If you have never done so, you will probably see quite a few applets.

DFA

[roxy99]

Ok I figured it out. I first selected the button delete temp files. Then went to the applets tab & everything was pretty much empty. Maybe deleting the temp cache files took care of the applet files.

I've been testing this dimensions 4 time sync app and works great. What's amazing is that even syncing every 5min, I get a good 20sec clock drift.

Pretty pathetic if no fix can be found, but I am happy to find a workaround. I just hope synching every 5min does't cause other problems like skipped frames.

[BobPhoenix]

Originally Posted by roxy99

As far as being isolated, you can bet that every Nforce2 user is having the same problem. There is no flood of complaints simply because most users probably just shrugged their shoulder and started using Dimension 4 to autosync every 5 min. My board is not an Asus A7N8X, but rather a DFI Nforce2 board.

My DFI NFII Ultra B NForce2 MB (AthlonXP 3200+) does not have this problem. I'm running MCE2005 with all the updates, SageTV 4.1.7 and Java 1.4.2_06 and Sync with SageTV Server and a PVR-250MCE. The time matches my main Intel based SageTV Server.

BobP.

[roxy99]

Originally Posted by BobPhoenix

My DFI NFII Ultra B NForce2 MB (AthlonXP 3200+) does not have this problem. I'm running MCE2005 with all the updates, SageTV 4.1.7 and Java 1.4.2_06 and Sync with SageTV Server and a PVR-250MCE. The time matches my main Intel based SageTV Server.

BobP.

Oh well so much for my theory of Nforce2. So what is causing it then?? If related to Java, then is there something I can cleanup in the registry and then reinstall Java to fix things. I am using Java 1.5.06 so maybe if I can uninstall and use an older Java version then that will work.

I checked the control panel applet of Java and there are no applets installed with Java. I uninstalled/reinstalled Java. System restore is disabled on my system so backtracking is not possible. The only fix for me may be a windows re-install.

[BobPhoenix]

Your idea about it being Nforce2 may be partially correct. Mine might be drifting some (at most 1 minute a day I would estimate - because when I checked - the times matched to the minute and it was < 10 hours to the next EPG update). I think the amount of drift maybe due more to the manufacturer than the Chipset. It is also possible that a newer version of SageTV might cause a bigger drift or possibly the newer Java. It is also posible that MCE2005 might not have a problem that XP does but since it is suppose to be XP without domain stuff I didn't think that was as likely as the other options.

I just wanted to provide a base line of something that appears to me to work.

Also now that I've posted this mine will start to have problems. Always happens when I post "Mine works" .

BobP.

[roxy99]

Originally Posted by BobPhoenix

Also now that I've posted this mine will start to have problems. Always happens when I post "Mine works" .

BobP.

Well look on the bright side- the more people who share the problem, the more brains behind the problem, therefor the sooner a solution. Its much worse being alone with the problem.

I dounbt there is anything wrong with Nforce2 per se since my system is rock solid everywhere else. It is a very new Windows XP installation with decent hardware. Definately it is a software problem.

I'll try ClockMon to provide a hardware based check of system time. I don't like the idea of repolling the internet every 5min- its bound to cause hiccups down the road.

[Steerpike]

Just a thought, after reading through all these recent posts ... what about an interaction between the NForce and the video card? Are you running the client and server on the same system? I'm just thinking that somehow, when Sage is running, it's interacting with the video card for some clocking info. This could somehow impact the system clock in some obscure combinations. Just a wild thought!

[roxy99]

Originally Posted by Steerpike

Just a thought, after reading through all these recent posts ... what about an interaction between the NForce and the video card? Are you running the client and server on the same system? I'm just thinking that somehow, when Sage is running, it's interacting with the video card for some clocking info. This could somehow impact the system clock in some obscure combinations. Just a wild thought!

Anything is possible. However before I start swapping all kinds of hardware I'd like to base it on a theory- otherwise I could end up making a bigger mess.

I installed Clockmon witch rather than syncing to internet, simply polls the hardware clock in the cmos and resets to equal cmos clock if different by more than x ms. I set that that to poll every min. Basically the 5-6 sec clock drift per min is kept in check now every min. Then I also run Dimension 4, an aromic clock synching utilitly and changed it from polling every 5min to poll every 3 hours. With clockmon I may be able to remove Dimension 4. I'll see how it goes. Clockmon does not seem to eatup any resources even a 1 minute intervals. It's probably a lot less of a burden than an antivirus program running in background.

[Opus4]

Originally Posted by roxy99

I hope Sage Tech support reads these forums and comes up with a fix.

As stated many times: no, tech support does not monitor these forums. Have you contacted SageTV?

If you are using a version from several months ago, try using the latest beta.

If you are already using the latest beta, then submit a bug report about it so SageTV can look into it.

- Andy

[gplasky]

Do a search on Google for nForce2 and clock problems. You'll find threads all over and from Linux as well as Windows gamers. This isn't a Sage problem. This is a problem on some of the nForce 2 boards. I didn't even bother searching for any other motherboards. Just wanted to let you know where the root cause probably resides-it's not with Sage. Maybe this might help some users:

A possible cause of this behavior is a clock synchronization issue that
can arise on some nForce2 systems when interrupts are routed through the
IOAPIC, and Spread Spectrum (SS) clocking is enabled. Under certain
conditions, this can cause the IOAPIC to issue multiple interrupts to
the CPU when it should have issued only one.

If you are experiencing clock inaccuracy on nForce2 hardware, take the
following steps to determine if this issue may be the cause:

1. Determine if the hardware you are using may be affected by this
problem. The problem is limited to MCP2 and MCP2-T hardware; it does
not affect MCP2-S or any nForce3 hardware. MCP2 and MCP2-T hardware
may be identified by the PCI device ID of the ISA bridge, which is
0x0060 for these devices.

To read the bridge device ID, use 'lspci -n -s 0:1.0' . The output
should be of the form '0000:00:01.0 Class 0601: 10de:0060 (rev a3)'.
The device ID of the bridge in this example is "0060" following the
NVIDIA PCI vendor ID "10de".

If your ISA bridge device ID is not 0x0060, then this issue is not
the cause of any clock inaccuracy you are experiencing.

2. Otherwise, examine the output from 'cat /proc/interrupts'. If IRQ0
(the timer) is shown to be in PIC mode rather than IOAPIC mode, then
this issue is not the cause of any clock inaccuracy you are
experiencing.

3. Otherwise, reboot your system and enter BIOS SETUP. Check if your
BIOS has a Front Side Bus (FSB) Spread Spectrum (SS) clocking option.
On many systems, this option is located in the "Advanced Chipset
Features" menu. If the option is present and enabled, disable it.
Boot Linux and observe the system clock over several hours to verify
if this has improved its accuracy.

This was from a Linux thread but may be helpful-or least helps explain the problem.

Gerry

[roxy99]

Originally Posted by Opus4

As stated many times: no, tech support does not monitor these forums. Have you contacted SageTV?

If you are using a version from several months ago, try using the latest beta.

If you are already using the latest beta, then submit a bug report about it so SageTV can look into it.

- Andy

Would'nt Sage find a wealth of info from the user forums? I can't see why they would'nt and if they are not , then they should. I think I'll get better help from the user base at large then Sage.

[gplasky]

Not for the beta releases. We can find the bugs but what user is going to have a fix for a newly released beta? If you find a bug in a beta you need to fill out a bug report. Posting in the forums will just give you an idea who else might have it or what you're doing wrong to generate that error. (end user error-not bug.)

Gerry

[Opus4]

Originally Posted by roxy99

Would'nt Sage find a wealth of info from the user forums? I can't see why they would'nt and if they are not , then they should. I think I'll get better help from the user base at large then Sage.

If this is your reasoning for not contacting SageTV or filing a bug report if you've tried it on the latest beta, then don't complain if it turns out to be something that SageTV could have handled, but never got fixed because no one cared to follow up with them about it.

- Andy