Hardware incopatibility

[Chasing]

Anything else happened Shian?

BSOD? etc

Or is it running stable now (referring to 1 client being ran)?

[Shian]

New BSOD (takes forever to get one to show up as the system usually restarts and doesn't display it.)
Stop: EA, 882F6020, 8AE0F1B8, B84F7CBC, 1
nv4-disp

-_-

[Chasing]

Could you email me the last 2 mini dumps pls?

You will need to zip or rar it before attaching to an email.

I'll PM you my email in a min.

c:\windows\minidump\ is the directory location.

The filename should be date stamped and be *.dmp

[Guaran]

nv4-disp is the nVidia display driver. That driver crashing could be caused by a bad driver installation (probably not the case), a bad card (seem to remember you had determined the card was ok), or a weak power supply. The video card mainly runs off the 12v rails.

[Chasing]

Okay....

This is what I have found so far looking at the 2 dump files Shian has sent me.

I've found 2 errors, 1 is specific and 1 is generic. There could be a possibility that the generic one is related to the specific one.

I'll focus on the specific one for now.

Here is the debug output (Spam alert)

Microsoft (R) Windows Debugger Version 6.12.0002.633 X86
Copyright (c) Microsoft Corporation. All rights reserved.


Loading Dump File [C:\Documents and Settings\xxxxxxxxx\Desktop\Mini011411-01.dmp]
Mini Kernel Dump File: Only registers and stack trace are available

Symbol search path is: SRV*c:\symbols*http://msdl.microsoft.com/download/symbols
Executable search path is:
Windows XP Kernel Version 2600 (Service Pack 3) MP (4 procs) Free x86 compatible
Product: WinNt, suite: TerminalServer SingleUserTS
Built by: 2600.xpsp_sp3_gdr.100427-1636
Machine Name:
Kernel base = 0x804d7000 PsLoadedModuleList = 0x8055d720
Debug session time: Fri Jan 14 18:15:05.515 2011 (UTC + 0:00)
System Uptime: 1 days 11:53:25.693
Loading Kernel Symbols
.................................................. .............
.................................................. ..............
................................
Loading User Symbols
Loading unloaded module list
................................
Unable to load image nv4_mini.sys, Win32 error 0n2
*** WARNING: Unable to verify timestamp for nv4_mini.sys
*** ERROR: Module load completed but symbols could not be loaded for nv4_mini.sys
************************************************** *****************************
* *
* Bugcheck Analysis *
* *
************************************************** *****************************

Use !analyze -v to get detailed debugging information.

BugCheck 100000EA, {882f6020, 8ae0f1b8, b84f7cbc, 1}

Probably caused by : nv4_mini.sys ( nv4_mini+585f9 )

Followup: MachineOwner
---------

0: kd> !analyze -v
************************************************** *****************************
* *
* Bugcheck Analysis *
* *
************************************************** *****************************

THREAD_STUCK_IN_DEVICE_DRIVER_M (100000ea)
The device driver is spinning in an infinite loop, most likely waiting for
hardware to become idle. This usually indicates problem with the hardware
itself or with the device driver programming the hardware incorrectly.
If the kernel debugger is connected and running when watchdog detects a
timeout condition then DbgBreakPoint() will be called instead of KeBugCheckEx()
and detailed message including bugcheck arguments will be printed to the
debugger. This way we can identify an offending thread, set breakpoints in it,
and hit go to return to the spinning code to debug it further. Because
KeBugCheckEx() is not called the .bugcheck directive will not return bugcheck
information in this case. The arguments are already printed out to the kernel
debugger. You can also retrieve them from a global variable via
"dd watchdog!g_WdBugCheckData l5" (use dq on NT64).
On MP machines it is possible to hit a timeout when the spinning thread is
interrupted by hardware interrupt and ISR or DPC routine is running at the time
of the bugcheck (this is because the timeout's work item can be delivered and
handled on the second CPU and the same time). If this is the case you will have
to look deeper at the offending thread's stack (e.g. using dds) to determine
spinning code which caused the timeout to occur.
Arguments:
Arg1: 882f6020, Pointer to a stuck thread object. Do .thread then kb on it to find
the hung location.
Arg2: 8ae0f1b8, Pointer to a DEFERRED_WATCHDOG object.
Arg3: b84f7cbc, Pointer to offending driver name.
Arg4: 00000001, Number of times "intercepted" bugcheck 0xEA was hit (see notes).

Debugging Details:
------------------


FAULTING_THREAD: 882f6020

DEFAULT_BUCKET_ID: GRAPHICS_DRIVER_FAULT

CUSTOMER_CRASH_COUNT: 1

BUGCHECK_STR: 0xEA

PROCESS_NAME: explorer.exe

LAST_CONTROL_TRANSFER: from b648f965 to b5fd65f9

STACK_TEXT:
WARNING: Stack unwind information not available. Following frames may be wrong.
a163626c b648f965 00000001 8aa47008 b648fb5a nv4_mini+0x585f9
a1636270 00000000 8aa47008 b648fb5a 00000001 nv4_mini+0x511965


STACK_COMMAND: .thread 0xffffffff882f6020 ; kb

FOLLOWUP_IP:
nv4_mini+585f9
b5fd65f9 f390 pause

SYMBOL_STACK_INDEX: 0

SYMBOL_NAME: nv4_mini+585f9

FOLLOWUP_NAME: MachineOwner

MODULE_NAME: nv4_mini

IMAGE_NAME: nv4_mini.sys

DEBUG_FLR_IMAGE_TIMESTAMP: 4cb9d7dd

FAILURE_BUCKET_ID: 0xEA_IMAGE_nv4_mini.sys_DATE_2010_10_16

BUCKET_ID: 0xEA_IMAGE_nv4_mini.sys_DATE_2010_10_16

Followup: MachineOwner

Translation

What all that means that the blue screen was caused by the device driver nv4_mini.sys which as Guaran pointed out is a Nvidia driver (Graphics Card) and is also included as a generic driver in a Win XP build.

The error itself was that graphics card/device driver got pushed into an infinite loop on a process and the system killed the process which resulted in the BSOD.

So whats the possible causes of this error?

1. The device driver (nv4_mini.sys) maybe corrupt
2. The device driver (nv4_mini.sys) may have compatibility issues
3. The graphics card(s) may have a hardware fault

Driver Corruption
I think this is probably the most likely cause. Corruption can be the file itself, damage to entries in the registry, duplicates sitting on the system somewhere etc.

Driver Compatibility Issues
The NVIDIA GeForce GTX 460 graphics card is the latest generation card from NVIDIA. Looking at the NVIDIA official forum there are masses and masses of people that have had problems with this type of card and drivers and MMO's. Also interesting to note is that some of the problems seem to to be related to the actual manufacturer of the card and the motherboard. I can't rule this out as a potential cause, however it might be worth asking Dell support if there is a known issue with the motherboard make and GTX 400 (probably quicker than trying to get an answer from Nvidia).

Hardware Fault
This I see as unlikely as both graphics cards were recently swapped out. However could be a fault on one or both of the cards. I would check if both cards are from the same manufacturer as well.

What can be attempted to resolve the issue?

Couple of questions first of all

1. When you took the computer to 'Microcenter' did they format and rebuild the system from the recovery partition? (All Dell systems have a hidden recovery partition that holds a 'factory reset' version of Windows.
2. When the swapped the graphics cards, did they swap like for like? or was it different models that were taken out?
3. Did 'Microcentre' look at the minidump files?

If you have already tried this then probably no point doing it again.

1. Properly uninstall and reinstall the Nvidia drivers.

Using control panel and add and remove won't physically take a driver off the system in certain cases (most uninstallers leave some kind of rubbish behind). You need to manually remove the affected files.

Instructions are here
http://forums.nvidia.com/index.php?showtopic=169555&hl=

See if this makes any difference, if not then onto 2.

2. Rebuild Windows back to factory reset using Dell partition

Sometimes its easier just to wipe a system and start from scratch. So if this hasn't already been tried then wipe everything (after backing up what you need to keep). Please note I am not referring to using the Windows recovery tool on the setup CD. The Dell recovery partition tools will wipe the hard disc 1st.

Once that has been completed, upgrade your hardware drivers and windows updates. Then before you load anything else (gadgets/software etc) reinstall Istaria and see what happens.

Further possible actions

3. Email Dell CS about GTX 460 (Include Manufacturer Info) and your specific computer build including the motherboard. (as mentioned earlier)

4. Post a bug report on the Nvidia GTX 460 Forums and/or a Nvidia support ticket.

Hopefully this helps a bit Shian. If anyone else can think of something I havent covered about this problem then please advise.

[Spirit]

It's funny that you have a Geforce 460 as I do too (an EVGA Geforce 460 GTX 1024mb), but I've not had any problems with games at all. Initially I installed the very latest Nvidia drivers (260.something..?) and had nothing but problems ... relating to a corrupt display on an hdtv and a projector.

There seem to be problems with the latest driver release ... my solution was to revert to nvidia driver release 258.96. I'm using a Dell XPS 420, with 4gb ram, on Win7 64-bit. I too found a rash of forum posts about problems with the 460, although all the ones I was reading were relating to corrupt displays on versions higher than 258.96. It seems Nvidia changed something.

I'd use the nvidia tool to remove *all* nvidia drivers. My dad's pc has an Nforce chipset, and it worked fine with generics until I installed drivers from nvidia's site. The only gottcha I found was that the ethernet controller wasn't found, so I had to use another pc to download the driver to a usb drive.

Chasing's post is awesome ... it does sound driver related, especially with what happens on a second Hz's instance.

So I'd download the 258.96 drivers (http://www.nvidia.co.uk/object/winxp...driver-uk.html ... link is for XP 32-bit drivers) for your graphics card, along with your nforce drivers.
- Check this is Ok to do for XP, but if it is use the tool to remove All nvidia software and drivers (on my add/remove programs list as "Nvidia Drivers"). Reboot.
- Install the drivers you downloaded earlier.

It's a shame you can't give Windows 7 a try. Yes it's arranged differently, but give it a chance. I found it different, but once you get used to it it's actually very nice. I'd almost bet your problems would vanish.

Incidentally did you try disabling your antivirus, or *temporarily* removing it?

Good luck! I do like the GF 460 ... in SLI it can only be more awesome.

Rakku

Edit: Oh and for the record, Dell CS are bl**dy hopeless. I discovered something very interesting while looking at PSU requirements for the 460, for which at least a 450W supply is recommended. My old graphics card (as supplied with an XPS 420) was an ATI 3870, and that also *required* a 450W psu, and yet Dell fit a 375W psu. Voltages are good, and system is solid as a rock. Will it last I wonder?

[Shian]

Thank you so much Chasing I'll give this a try.

[Chasing]

Interesting Info Rakku

Shian in your first post you said 'since October 09' did you mean 'October 10'?

If it is October 10, look at the date of the driver file in the picture below. (NB Its DD/MM/YYYY for me as I'm in the UK)



Certainly adds more evidence to Rakku's theory.

If the first step I suggested doesn't resolve (manual wipe and latest driver reinstall), then I would redo it using the older driver as Rakku suggested (make sure Windows update is off).

Note, don't read too much into the other 3 errors the picture shows, they are generic and require explaining.

[Shian]

Okay, so older driver just made me go back to crashing during web browsing and other mundane tasks. Later I'll go back to the updated driver again.

Looks like next step is to get another motherboard.

Thankfully Dell let me extend my warranty so I'm not going to be paying $400 for a new motherboard. Now comes the problem of getting their stupid "Dell only" motherboard. After calling Dell CS a few times, they basically have to special order such an "old" motherboard. (One phonecall even had someone trying to sell me a different computer for $400. Yeah, right )

[Spirit]

Okay, so older driver just made me go back to crashing during web browsing and other mundane tasks. Later I'll go back to the updated driver again.

Looks like next step is to get another motherboard.

Thankfully Dell let me extend my warranty so I'm not going to be paying $400 for a new motherboard. Now comes the problem of getting their stupid "Dell only" motherboard. After calling Dell CS a few times, they basically have to special order such an "old" motherboard. (One phonecall even had someone trying to sell me a different computer for $400. Yeah, right )

I imagine if your dell is like mine, the motherboard is mounted on the opposing side of the case when compared to a more normal pc. If you look at mine from the front, the motherboard is mounted on the left side. The expansion slots are flipped as well, so there's no way to fit a traditional off-the-shelf motherboard. The result of this is that I doubt I'll buy another.

Sorry to hear the drivers didn't improve the situation. Good luck with the new board, and glad to hear Dell are doing it under warranty.

Rakku

[Ettanin]

It's certainly your mainboard which does not support your graphics card correctly. Have you tried other games, especially shader intense ones?

I doubt that it's a defective, but a completely incompatible one. Replacing it with another one of the same model will not solve this problem.

That is another reason why i purchase my computers totally in parts and not from a full seller as you may get a working system which however is not compatible to future parts.

[Shian]

Well, here's the bad thing. Istaria is the only game it does this on with the updated driver.

[Ettanin]

Strange. I'm using a GeForce GTX470 (it features the same architecture as your GTX460) with driver version 266.58 (used 260.99 b4) and actually never had problems, be it windowed, fullscreen or 3d vision mode.

If the graphics card were defective, garbage would render right at the desktop (unless it's a shader fault which is only visible in D3D or OpenGL apps and not in Istaria as Istaria does not utilize any shaders).

Another point may be that Istaria (which in fact seems to be most logical to me) does not support SLI. Deactivate one of the graphic cards before starting Istaria.

[Ettanin]

Strange. I'm using a GeForce GTX470 (it features the same architecture as your GTX460) with driver version 266.58 (used 260.99 b4) and actually never had problems, be it windowed, fullscreen or 3d vision mode.

If the graphics card were defective, garbage would render right at the desktop (unless it's a shader fault which is only visible in D3D or OpenGL apps and not in Istaria as Istaria does not utilize any shaders).

Another point may be that Istaria (which in fact seems to be most logical to me) does not support SLI. Deactivate one of the graphic cards before starting Istaria.

To add: I observe a similar error when WinKey+L while running Dungeon Siege. I'm sure Istaria doesn't like SLI and that mechanism killed the engine's rendering scope.