Monitor shuts off randomly (dumps included)

[TomMe]

Hello guys,

First let me give you my system setup:
- MSI K9N SLI Platinum
- AMD Athlon64 3800+
- 2x Corsair 512MB 5400C4
- AOpen GeForce FX5200 PCI
- Creative Sound Blaster Live!
- SeaSonic S12-430

- AOpen FX5200 PCI Driver
- nVidia nForce 9.16 Driver

- Motherboard BIOS v1.5
- Windows XP + SP2 English

Ever since I've started using my GeForce FX card, I've had a problem with it. Sometimes it will cause my monitor to shut off, and after that my HDD too (looks like both go into standby). On top of that, sometimes after this, my system restarts.

This GeForce FX card is actually a temporary replacement as I've had another problem with my GF7600GT (sometimes vertical bars at boot). Currently, I don't own the 7600GT anymore (sent it back to the store), and am looking to buy a new one in the near future. But at this point I'm stuck with the GF FX.

My motherboard has 3 PCI slots, and whenever I put the card into the 3rd one (with an orange color), the problem is almost replicable by installing ForceWare drivers or playing a game for a couple of minutes. It seems to be more stable in PCI 1 and PCI 2, but the problem still occurs..just less frequently (every week instead of every 10 minutes).

Over the months, I have kept track of the problem through personal notes, although they are a bit messy. I've tried several ForceWare versions (new and old) and a clean Windows install, didn't help. The problem even occured without ForceWare drivers installed.

After the monitor shut off again this morning, I did a reset and everything was back to normal. Then I shut the machine down, and after I started up Windows again an error message popped up saying Windows had recovered from a serious error.

Then I found out about Windows dump files and saw that one was made at the time of the crash this morning. I quickly found there were a couple of dump files that seemed to have been made at dates that correspond with those of crashes like this in the past few months (as indicated in my own notes).

Since people use these dump files on this forum for troubleshooting, I thought I would post mine and see if you guys come up with something that can help. I ran them through WinDbg myself and it seems nv4_mini.sys is the cause, but I don't quite get it since it happened with several video drivers (including standard Windows ones).

I have a couple more of these dump files, but the dates of them don't match with any of my notes..so I'm not sure that they were produces after this type of crash. I'll check them in WinDbg nevertheless.

Here are the dumps. The .txt files include my !analyze -v response.

Thanks in advance!

 

[Tmagic650]

Your monitor shutting off is a symtom of your problem and not the cause. As you already know, this is a problem with the video card/drivers, as the minidumps show. Try another video card and driver set

 

[TomMe]

Your monitor shutting off is a symtom of your problem and not the cause.

That's what I said.. Some people get BSODs, I get a monitor shutdown. I must note that at one time, after I had this and did a reset, I also received a BSOD mentioning "nv_4disp".

As you already know, this is a problem with the video card/drivers, as the minidumps show.

It looks that way, yes. But as I've said, I've tried several drivers. Versions include standard Windows, FW 56.72, FW 66.81, FW 84.21 and FW 93.71.

Try another video card and driver set

I don't think you've read my post fully. I don't have another card, and my 7600GT never had this problem.

Ahh, and also..I have tried this card on an older P3 system, no problem there. But I might not have tested it long enough (!). I've also put it in a friend's PC..no apparent problem there either.

 

[Tmagic650]

This must be an incompatibility between the SLI motherboard and your video card... Your ONLY option is to get another Nvidia video card to use with your existing hardware

 

[TomMe]

Hm..so you're saying there exists a problem between SLI chipsets and certain graphics cards? Do you have some more info on that?

Another problem I have with this card is that sometimes my screen stays blank until I enter the Windows logon screen. I also had this on the P3 system I mention above. Might be unrelated though.

 

[Tmagic650]

"Another problem I have with this card is that sometimes my screen stays blank until I enter the Windows logon screen. I also had this on the P3 system I mention above. Might be unrelated though"...


Again I say TRY ANOTHER CARD!

Save your pennies and buy another card. You can do you own research on what is the best card for your system. Spend a little more time on this yourself

 

[TomMe]

My 7600GT didn't have this problem, as I've already said. So I don't need to "try" another card. I'm just trying to determine the cause of the problem and thought dumps might help.

Believe me, I've spent more time on figuring out what card to buy next and troubleshooting problems (including this one) than I care to admit. So it's not that I'm lazy, on the contrary. I'm just stuck with this problem and am trying to get help. Telling me to buy another card is not help, the next card could have another problem.

 

[Tmagic650]

"Another problem I have with this card is that sometimes my screen stays blank until I enter the Windows logon screen. I also had this on the P3 system I mention above. Might be unrelated though"...


Am I missing something here?

"My 7600GT didn't have this problem".. What card works and what card gives you problems? Is this a hard question to answer?

 

[peterdiva]

The two dumps with the newer video drivers have crashed with a stack overflow, this can be a problem with Norton AV. You also have AVG installed; you should only have one AV program.

Update to the latest video card drivers. If you decide to keep Norton then you will have to extend the STACK.

EDIT: Just had another look at the stack overflow dumps. I now think that they are caused by a recursive memory error. Windows gives a 7F bugcheck and throws a double fault to stop it when it runs out of stack space.

BugCheck 1000007F, {8, 80042000, 0, 0} <-- 8 = EXCEPTION_DOUBLE_FAULT
Probably caused by : nv4_mini.sys ( nv4_mini+393cc )

b95f7080 85ae2838 00000000 b2d05dff 00000001 8(.......]......
b95f7090 85a0aa50 85ae2838 b95f70f8 ffffffff P...8(...p_.....
b95f70a0 f6ea2f82 ffffffff f6ebd550 85ae2838 ./......P...8(..
b95f70b0 00000000 b2d05e00 00000001 f6ebd588 .....^..........
b95f70c0 859e8008 f6ebd550 85ae2838 00000003 ....P...8(......
b95f70d0 00000001 859e8008 f6ea39ea 85a0aa50 .........9..P...
b95f70e0 859e8008 85ae2838 85a0aa50 859e8008 ....8(..P.......
b95f70f0 85a0aa50 b95f7124 ffffffff f6ea2f31 P...$q_.....1/..
b95f7100 ffffffff f6ebd550 85ae2838 00000000 ....P...8(......
b95f7110 b2d05dff 00000001 85a0aa50 85ae2838 .]......P...8(..
b95f7120 b95f7180 ffffffff f6ea2f82 ffffffff .q_....../......
b95f7130 f6ebd550 85ae2838 00000000 b2d05e00 P...8(.......^..
b95f7140 00000001 f6ebd588 859e8008 f6ebd550 ............P...
b95f7150 85ae2838 00000003 00000001 859e8008 8(..............
b95f7160 f6ea39ea 85a0aa50 859e8008 85ae2838 .9..P.......8(..
b95f7170 85a0aa50 859e8008 85a0aa50 b95f71ac P.......P....q_.
b95f7180 ffffffff f6ea2f31 ffffffff f6ebd550 ....1/......P...
b95f7190 85ae2838 00000000 b2d05dff 00000001 8(.......]...... <-- looping, same as above.
b95f71a0 85a0aa50 85ae2838 b95f7208 ffffffff P...8(...r_.....
b95f71b0 f6ea2f82 ffffffff f6ebd550 85ae2838 ./......P...8(..
b95f71c0 00000000 b2d05e00 00000001 f6ebd588 .....^..........
b95f71d0 859e8008 f6ebd550 85ae2838 00000003 ....P...8(......
b95f71e0 00000001 859e8008 f6ea39ea 85a0aa50 .........9..P...
b95f71f0 859e8008 85ae2838 85a0aa50 859e8008 ....8(..P.......
b95f7200 85a0aa50 b95f7234 ffffffff f6ea2f31 P...4r_.....1/..
b95f7210 ffffffff f6ebd550 85ae2838 00000000 ....P...8(......
b95f7220 b2d05dff 00000001 85a0aa50 85ae2838 .]......P...8(..
b95f7230 b95f7290 ffffffff f6ea2f82 ffffffff .r_....../......
b95f7240 f6ebd550 85ae2838 00000000 b2d05e00 P...8(.......^..
b95f7250 00000001 f6ebd588 859e8008 f6ebd550 ............P...
b95f7260 85ae2838 00000003 00000001 859e8008 8(..............
b95f7270 f6ea39ea 85a0aa50 859e8008 85ae2838 .9..P.......8(..
b95f7280 85a0aa50 859e8008 85a0aa50 b95f72bc P.......P....r_.
b95f7290 ffffffff f6ea2f31 ffffffff f6ebd550 ....1/......P...
b95f72a0 85ae2838 00000000 b2d05dff 00000001 8(.......]...... <-- looping starts again.
b95f72b0 85a0aa50 85ae2838 b95f7318 ffffffff P...8(...s_.....

 

[TomMe]

Am I missing something here?

Yes, the fact that I've been talking about the FX5200 the whole time..

peterdiva, it's amazing you can see all that just by looking at those codes! Though I'm not quite following you. Does the recursive memory error have something to do with Norton or not?

I've uninstalled Norton quite a while ago and have been using AVG's latest free scanner. This was the case the last couple of times when I had the crash (like with Mini040907-01.dmp I think). As for driver updating..this card type is relatively old. If it's a driver problem and they haven't fixed it already by version 93.71 (the card comes with 66.81), then they most likely never will.

I'm not sure which dumps correspond to which driver versions..but I'm pretty sure all 4 of them were made after the crash I've described. And there isn't always a dump after a crash either. What do the other 2 dumps tell you? And do you have an idea of what might be causing this (hardware, incompatibility, software,..)?

 

[peterdiva]

I've just noticed that three of those dumps are from November last year. The one from the 9th of April is most likely caused by the old video card driver. You'll need to post some new dump files with version 93.71 installed.

Mini040907-01.dmp:
BugCheck 100000EA, {8507a478, 86c69120, f79cbcb4, 1}
Probably caused by : nv4_mini.sys ( nv4_mini+488e0 )
FAILURE_BUCKET_ID: 0xEA_IMAGE_nv4_mini.sys_DATE_2004_10_01

 

[kirock]

I have the same mobo as your's but I've not tried a PCI video card in it. I have the GeF7900GT and had the 6600Ct before that. What I have seen though with this mobo is OCing the thing at all usually fk's up the PCI bus, whether I do it manually or by MCI software. By this I mean I'd loose my on-board ethernet which runs on the PCI bus. Make sure your BIOS is set to "fail proof" settings and that you haven't set anything to above normal in the performance menu in the BIOS.

In my older AMD ThunderBird 1.4GHZ PC I had the FX5200 AGP card and it was a problem until i updated my chipset drivers and found the right nvidia driver version (I think it was the 66.xx or even the 64.xx). So you might want to roll back your chipset drivers for now and play around with a few video driver versions (older ones) to see if you hit the right combo.

Overall I think the MCI SLI Premium is just not liking the PCI video card or at least the PCI bus is slightly over clocked and is unstable. You might even try turning down the master clock to run the PCI bus a little slower see if this helps (of course some performance lost will be seen, but might be better then BSODs and black screens).

Cheers.

 

[TomMe]

Yes, I'm currently running on FW 66.81 as it is the driver package AOpen provided. I will install 93.71 as soon as I've typed this post. It could take a few days before I get another crash.

Earlier today I had another crash. After the monitor went blank, I waited for about 15 seconds..then the HDD shut down too. After I reset, Windows notified me that a log was created. And indeed there was another dump. I've put it on my webspace in a separate folder.

It seems that a dump is created only when I wait long enough for my HDD to shut down too, and not if I push reset right after the monitor goes blank.

As for the PCI. I haven't had any problems with my PCI soundcard or PCI ethernet card (I used both of them for months at a time). I haven't done any overclocking and I never will either. I'll take a look in the BIOS and post my settings.

Thanks for your help guys, really appreciate it!

edit: 91.73 drivers are installed. Now it's just "sit back and wait".

 

[TomMe]

Okay, after I installed 93.71 yesterday the monitor went out 3 times, plus 2 times this morning. Each time I waited more than a minute for the HDD to shut down too, but it didn't. Windows didn't make any dumps either.

I've noticed a couple of things:
- It looks like the problem occurs more frequently with 93.71 than with 66.81.
- With 66.81, upon Windows startup, my monitor always goes blank for a second, then returns to normal. This doesn't happen with 93.71. Very peculiar.
- The problem often (but not always) happens when I click on something, usually a hyperlink in Firefox.
- Yesterday it happened 2 times while I was playing an MP3. The music stalled about a second after the monitor went out and continued to loop 1 second worth of music for a while, after that the loop became shorter and it sounded like *BRRRRRRRRRRR...*.

I've uninstalled 93.71 because I can't use my computer properly with it. Any other suggestion? For now I'll install 84.21.

Here is a list of my most relevant BIOS settings:

Advanced BIOS Features
- IOAPIC Function [Enabled]
- MPS Table Version [1.4]

Advanced Chipset Features
- MCP55 (SB) to AM2 (CPU) Freq Auto [Enabled]
- MCP55 (SB) to AM2 (CPU) LinkWidth [16 down 16 up]
- MCP55 ACPI HPET TABLE [Enabled]

Power Management Features
- ACPI Function [Enabled]
- ACPI Standby State [Enabled]
- Wakeup Event Setup (all Disabled)

PNP/PCI Configurations
- PCI Latency Timer [64]
- PCI Slot1/2/3 IRQ [Auto]
- IRQ/DMA Resource Setup (all Available)

Cell Menu
- Cool 'n' Quiet [Disabled]
- CPU Frequency (MHz) [200.0]
- CPU Dynamic Overclocking [Disabled]
- Adjust CPU Ratio [Auto]
- CPU Voltage [Auto]
- Memclock Mode [Auto]
- Adjust DDR Voltage (V) [2.00]
- MCT Timing Mode [Auto]
- Bank Interleaving [Enabled]
- Enable Clock to All DIMMs [Disabled]
- CMD-ADDR Timing Mode [2T]
- SoftWare Memory Hole [Enabled]
- PCIE Frequency (MHz) [100]
- Spread Spectrum Configuration (all Disabled)

edit: Just had another crash with 84.21. The HDD shut down and a dump was made. I've uploaded it on my webpage (don't look at the dates there because they're wrong).
edit 2: And another crash. Again HDD shut down and dump added to webpage. It seems that with 84.21 my PC automatically reboots after it crashes. I'm back on 66.81 now.