Still BSOD

[CrazyAviator 10]

Even after patch 1.03 and numerous changes to the graphics settings in ARMA II I havent been able to make my BSOD go away. It didnt for awhile with 1.02 but its back now with a vengence.

The BSOD I get is your typical "You have encountered a problem, windows needs to shut - if you have installed anything recently please uninstall or contact Admin". The error codes are like 00000FFFF000000D%353 or some shit.

Does anybody have any clue how to fix this problem or has anyone had this same issue?

Specs:

Dual core Athlon X2 6400+ (3.21 ghz)

4 GB Ram

nVidia 8800GT

Windows Vista x64 SP2

p.s. Running with most settings to Normal now, even though FPS isnt an issue. Also windowed mode.

[Bulldogs 10]

Next time it happens write down the exact error code and put it up here.

Also check device manager (control panel->system->advanced system settings->Hardware->device manager) and check if you have any conflicts.

Also try removing any devices you don't need (even sound card, if on-board sound then disable in BIOS) then run Arma 2 and see if you still get the crash.

And just in case you haven't done it yet, install the latest Directx9 Distro.

[CrazyAviator 10]

Here you go, Ive crashed about 4 more times in the last hour. Once was a "your video device has stopped working properlly", the others were BSOD.

Problem signature:

Problem Event Name: BlueScreen

OS Version: 6.0.6002.2.2.0.768.3

Locale ID: 1033

Additional information about the problem:

BCCode: 116

BCP1: FFFFFA8003CF5010

BCP2: FFFFFA6003417F30

BCP3: 0000000000000000

BCP4: 0000000000000002

OS Version: 6_0_6002

Service Pack: 2_0

Product: 768_1

Files that help describe the problem:

C:\Windows\Minidump\Mini081709-03.dmp

C:\Windows\Temp\WER-32401-0.sysdata.xml

C:\Windows\Temp\WER8545.tmp.version.txt

[Bulldogs 10]

BCCode : 116 relates to a video error.

More aptly it means the video drivers crashed and it attempted to reset the display driver but failed causing the crash.

It's more common than not a problem with either corrupt drivers, corrupt windows recovery or a lack of access permission.

The first and last are often caused by insufficient permission when initially installing drivers (not likely by windows permission, more likely Anti-Virus/Anti-spyware).

A good thing to do is to use driver removal software to completely remove you graphics card drivers then make sure Anti-Virus software is disabled (including atni-spyware like Windows Defender) and make sure UAC is disabled then install a fresh video card driver.

Hope that helps.

[cashxx 10]

Sorry but I don't think thats going to work. I am having the same issue and have installed fresh copies of WIndows XP, Vista, and Windows 7 and the exact same problem. So the issue is happening on fresh installs with everything updated with only Arma 2 being installed without other software being added like Antivirus etc.

I don't think its an over heat issue, overclocking, lack of permissions, corrupt drivers, etc. Everyone keeps saying these things and its not the case.

[cashxx 10]

Just to double check I turned off UAC, never done that. Disabled Windows Defender and Antivirus. Uninstalled Nvidia Drivers and used Drive Sweeper and booted into safe mode and removed the drivers. Turned off hyperthreading, AA off, set resolution a little lower to 1600x1200 instead of 1920x1080. Tried various graphic settings in arma. Using riva tuner to set the fan at 100%. GPU doesn't go over 52c. No conflicts in device manager. Like I said before its happening on fresh installs with latest MS updates and only arma installed. Various Nvidia drivers tested.

[CrazyAviator 10]

Interesting update:

I uninstalled the 7/7/2009 nVidia drivers in safe mode and reinstalled (as admin and all antivirus disabled) in safe mode new 7/14/2009 nVidia drivers. To my astonishment the BSOD went away for 2 full days of game play. Then I started getting them again last night, frequently. Wtf. It happend shortly after bumping my view distance up to 5000 - I'm starting to wonder if the new drivers allow the video card to run cooler but then the view distance killed it again.

Can anyone verify if view distance is directly responsible for BSOD? I am going to experiment with this today.

[nyran125 10]

you have to play around with the settings a bit Aviator i went back to 3000 view distance and it stopped crashing maybe even a bit under i also tried turning Post process onto low that helped a bit. Some settings crashed my desktop if they were low and other graphic settings crashed my desktop if they were high,lol. Seriously, if this game didnt have the editor mode i would never have bought it. But that editor mode SEEMS to be completely Bugfree. Its almost like they made this game JUST for the editor to run well for everybody. I love the Editor mode but without it , this game wouldnt be worth its weight in Bronze let alone GOLD. But they have the editor so im content.

[Bulldogs 10]

It really sounds like an overheating issue. Even if it seems that it's not overheating, Arma 2 puts a lot more stress on CPU/GPU than most other games and can spike very fast causing the crash before the sensors pick it up. If you haven't tried it already, open the side of the case and aim a large fan in there to see if that helps (don't get the fan to close to the case to avoid static)

If your video card is spiking in heat fast then it could shutdown for safety then try to recover, but the intense heat hitting it so fast would prevent a recovery causing the crash (if it did manage to recover you'd get a "device freed" error)

That said, it could also be too much pressue put on your graphics card (try underclocking the card)

Many people forget that keeping a card cool isn't the only thing, and in the cases of overclocks (or stock clocked older cards) after running 3d mark for 8 hours, even if your card survives that it doesn't necissarily mean it's entirely stable in all areas.

Oh, and if you haven't yet, download and run 3d mark on loop for 8 hours

[wooly-back-jack 10]

try without overclock and with these settings

PhysX == Off

AF == application controlled

AA == OFF

texture clamp == use hardware

error reporting == ON

extension limit == OFF

force mipmaps == NONE

max pre-rendered == 8

multi display/mixed GPU == single

anisotropic == ON

negative LOD == ALLOW

TF Quality == High performance

TF Trilinear == ON

Threaded optimization == AUTO

Triple buffering == ON

Vsync == OFF

[oktane 0]

As others have said more or less, software doesn't cause bluescreens. The only thing that can cause bluescreens are kernel drivers (like video and audio drivers), faulty hardware, and in really screwed up cases, a hosed OS. (I personally do not think your os is hosed)

Arma does not have kernel drivers. If a crash happens during an Arma game, it is a faulty driver or malfunctioning hardware. There is no maybe's about that, Arma cannot cause bluescreens at all, it's impossible.

Please read this and find out what is crashing, by debugging the dump file that windows makes. Usually it will point you to a driver that malfunctioned, or if it is very vauge, likely something else like memory corruption (use windows memory diagnostic) or a dying power supply.

http://forums.majorgeeks.com/showthread.php?t=35246

[CrazyAviator 10]

As others have said more or less, software doesn't cause bluescreens. The only thing that can cause bluescreens are kernel drivers (like video and audio drivers), faulty hardware, and in really screwed up cases, a hosed OS. (I personally do not think your os is hosed)

Arma does not have kernel drivers. If a crash happens during an Arma game, it is a faulty driver or malfunctioning hardware. There is no maybe's about that, Arma cannot cause bluescreens at all, it's impossible.

Please read this and find out what is crashing, by debugging the dump file that windows makes. Usually it will point you to a driver that malfunctioned, or if it is very vauge, likely something else like memory corruption (use windows memory diagnostic) or a dying power supply.

http://forums.majorgeeks.com/showthread.php?t=35246

Yes Im beginning to agree with you that its a hardware issue, but created BY Arma 2. I downloaded the debugger and followed the instructions - this is what I got when I loaded my most recent minidump (from tonight, just before reinstalling the video drivers again).

Microsoft ® Windows Debugger Version 6.11.0001.404 X86

Copyright © Microsoft Corporation. All rights reserved.

Loading Dump File [C:\Windows\Minidump\Mini082009-01.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 Server 2008/Windows Vista Kernel Version 6002 (Service Pack 2) MP (2 procs) Free x64

Product: WinNt, suite: TerminalServer SingleUserTS Personal

Built by: 6002.18005.amd64fre.lh_sp2rtm.090410-1830

Machine Name:

Kernel base = 0xfffff800`01e0b000 PsLoadedModuleList = 0xfffff800`01fcfdd0

Debug session time: Thu Aug 20 17:22:30.390 2009 (GMT-7)

System Uptime: 0 days 6:25:55.106

Loading Kernel Symbols

...............................................................

................................................................

.............................

Loading User Symbols

Loading unloaded module list

...........

*******************************************************************************

* *

* Bugcheck Analysis *

* *

*******************************************************************************

Use !analyze -v to get detailed debugging information.

BugCheck 116, {fffffa8008c4d010, fffffa60034e5140, 0, 2}

Unable to load image \SystemRoot\system32\DRIVERS\nvlddmkm.sys, Win32 error 0n2

*** WARNING: Unable to verify timestamp for nvlddmkm.sys

*** ERROR: Module load completed but symbols could not be loaded for nvlddmkm.sys

Probably caused by : nvlddmkm.sys ( nvlddmkm+e1140 )

Followup: MachineOwner

---------

now I hit !analyze -v and this is what I got

VIDEO_TDR_FAILURE (116)

Attempt to reset the display driver and recover from timeout failed.

Arguments:

Arg1: fffffa8008c4d010, Optional pointer to internal TDR recovery context (TDR_RECOVERY_CONTEXT).

Arg2: fffffa60034e5140, The pointer into responsible device driver module (e.g. owner tag).

Arg3: 0000000000000000, Optional error code (NTSTATUS) of the last failed operation.

Arg4: 0000000000000002, Optional internal context dependent data.

Debugging Details:

------------------

FAULTING_IP:

nvlddmkm+e1140

fffffa60`034e5140 ?? ???

DEFAULT_BUCKET_ID: GRAPHICS_DRIVER_TDR_FAULT

CUSTOMER_CRASH_COUNT: 1

BUGCHECK_STR: 0x116

PROCESS_NAME: System

CURRENT_IRQL: 0

STACK_TEXT:

fffffa60`0310ba08 fffffa60`03f2d9d8 : 00000000`00000116 fffffa80`08c4d010 fffffa60`034e5140 00000000`00000000 : nt!KeBugCheckEx

fffffa60`0310ba10 fffffa60`03f2d80b : fffffa60`034e5140 fffffa80`08c4d010 fffffa80`08d56540 fffffa80`08c4d010 : dxgkrnl!TdrBugcheckOnTimeout+0xec

fffffa60`0310ba50 fffffa60`03ee73aa : 00000000`00294b1d 00000000`00000000 fffffa80`08d56540 00000000`0016a508 : dxgkrnl!TdrIsRecoveryRequired+0x16f

fffffa60`0310ba90 fffffa60`03fa8347 : 00000000`00000000 00000000`00000002 00000000`00000002 00000000`00000002 : dxgkrnl!VidSchiReportHwHang+0x312

fffffa60`0310bb40 fffffa60`03fa6f49 : fffffa80`08d56540 fffffa80`00000000 00000000`0016a508 00000000`00000000 : dxgkrnl!VidSchiCheckHwProgress+0x7b

fffffa60`0310bb70 fffffa60`03eeb718 : ffffffff`ff676980 00000000`00000000 00000000`00000000 00000000`00000000 : dxgkrnl!VidSchiWaitForSchedulerEvents+0x199

fffffa60`0310bbf0 fffffa60`03fa6d69 : 00000000`00000000 fffffa80`08d5abe0 00000000`00000080 fffffa80`08d56540 : dxgkrnl!VidSchiScheduleCommandToRun+0x398

fffffa60`0310bd10 fffff800`0206ff37 : fffffa80`08d6e060 fffff800`01e5f507 fffffa60`032562b0 00000000`00000001 : dxgkrnl!VidSchiWorkerThread+0x95

fffffa60`0310bd50 fffff800`01ea2616 : fffff800`01f7e680 fffffa80`08d6e060 fffff800`01f83b80 fffff800`01f7ee10 : nt!PspSystemThreadStartup+0x57

fffffa60`0310bd80 00000000`00000000 : 00000000`00000000 00000000`00000000 00000000`00000000 00000000`00000000 : nt!KiStartSystemThread+0x16

STACK_COMMAND: .bugcheck ; kb

FOLLOWUP_IP:

nvlddmkm+e1140

fffffa60`034e5140 ?? ???

SYMBOL_NAME: nvlddmkm+e1140

FOLLOWUP_NAME: MachineOwner

MODULE_NAME: nvlddmkm

IMAGE_NAME: nvlddmkm.sys

DEBUG_FLR_IMAGE_TIMESTAMP: 4a5ce19c

FAILURE_BUCKET_ID: X64_0x116_IMAGE_nvlddmkm.sys

BUCKET_ID: X64_0x116_IMAGE_nvlddmkm.sys

Followup: MachineOwner

---------

0: kd> lmvm nvlddmkm

start end module name

fffffa60`03404000 fffffa60`03ecff00 nvlddmkm T (no symbols)

Loaded symbol image file: nvlddmkm.sys

Image path: \SystemRoot\system32\DRIVERS\nvlddmkm.sys

Image name: nvlddmkm.sys

Timestamp: Tue Jul 14 12:50:52 2009 (4A5CE19C)

CheckSum: 00ADAD70

ImageSize: 00ACBF00

Translations: 0000.04b0 0000.04e4 0409.04b0 0409.04e4

[Dwarden 1125]

tried last public beta build?

[CrazyAviator 10]

What?

[no use for a name 0]

What?

search BETA...

[CrazyAviator 10]

Well on a hunch I decided to download a temperature program (called SpeedFan). Cool little program that appears to work well. It automatically detects temps based on stock OEM sensors and somehow the speedfan can adjust the fans in your case to try to keep the temp as low as possible.

Anyway, when I loaded up speedfan it gave me warnings right away. Apparently my GPU was running at 80 degrees C. Thats almost 180 F :eek: That was after getting a BSOD and sitting in windows desktop! Wow, so I decided to stick a floor fan on the side of the case and im down to almost 60 degrees C now. I'm guessing this is still a little on the hot side but I hope it stays cool ingame. Im going to try again.

[Bulldogs 10]

60 deg isn't much to worry about except it's not great while not under pressure. 80 is high, not extremely high but on an older card that's been used for a long time it can start to wear down on it.

I'd suggest giving the card a good clean out and if it's still running hot then you may want to take off the fan, clean the GPU and put some fresh cooling gel on there (artic silver is good) then replace the fan and monitor the temps slowly. If you put cooling compound on there then just be careful not to put too much (1mm thickness is enough)

[ziiip 1]

I've been having these freaking BSODs since 1.03 and the beta patches dont help. :( And my VGA is made by Sapphire and has a huuge FAN so heating cant be an issue AFAIK.

BCCode : 9c BCP1 : 00000000 BCP2 : BA344050 BCP3 : B66B4000

BCP4 : 00000135 OSVer : 5_1_2600 SP : 3_0 Product : 256_1

Edit: 3DMark06 runs without a hitch and I get normal scores.

Edited August 21, 2009 by ziiip

[Bulldogs 10]

I've been having these freaking BSODs since 1.03 and the beta patches dont help. :( And my VGA is made by Sapphire and has a huuge FAN so heating cant be an issue AFAIK.

BCCode : 9c BCP1 : 00000000 BCP2 : BA344050 BCP3 : B66B4000

BCP4 : 00000135 OSVer : 5_1_2600 SP : 3_0 Product : 256_1

Edit: 3DMark06 runs without a hitch and I get normal scores.

This ones very difficult to locate the problem.

BCCode : 9c, means that your system (more aptly, your CPU) encountered a non-specific problem. Code 9c (Machine Check Exception) is hard to verify because it is a very broad error.

The second part (BCP1) dictates where in the MCE an error occured, but being at 00000000 states that the CPU doesn't know where the error occured or the error occured before the CPU could get any information about it.

9c is also one of the very few BSOD's that can be caused by software (reading and writing software that confuses the CPU)

But more commonly it is caused by an overheating CPU or an overload in the power supply.

[oktane 0]

Yes Im beginning to agree with you that its a hardware issue, but created BY Arma 2.

Sorry, again, not possible.

The dump file you posted says your video card driver is causing the crash. Make sure your fan is operating correctly and if needed use a 3rd party app to crank it to 100 and monitor the temperature.

Arma2 and other games push the video card very hard. You may try something to really test out the card like FurMark and I bet it will crash.

Edit, glad to see you got a temp.. There are some apps like RivaTuner which can display an overlay in game, to tell you your temp too. Also if you have 2 monitors you can just put the monitoring app on the other one while you play, you'll be surprised how fast it goes down when you stop playing. I bet it was higher than that!

Heres a cool app for simple stats http://www.techpowerup.com/gpuz/

---------- Post added at 10:57 PM ---------- Previous post was at 10:48 PM ----------

I've been having these freaking BSODs since 1.03 and the beta patches dont help. :( And my VGA is made by Sapphire and has a huuge FAN so heating cant be an issue AFAIK.

BCCode : 9c BCP1 : 00000000 BCP2 : BA344050 BCP3 : B66B4000

BCP4 : 00000135 OSVer : 5_1_2600 SP : 3_0 Product : 256_1

Edit: 3DMark06 runs without a hitch and I get normal scores.

You need to run the dump file analyzer and post that, this paste doesn't mean much, the analyze will be much more detailed.. FurMark is the only app I use to find faulty video hardware, I'm sure there are others though that really stress the video hardware. But most people want to blame the games or 3d apps. Again, games can't cause BSOD's. If a game sends invalid data to the video card driver, it should handle it, not shit the bed and crash your system. And in most cases it does handle it, except in the case of bugged video drivers or malfunctioning hardware.

GPU's and Ram on video cards goes bad all the time guys. I have a pile of cards as high as a small child at work, all burned out. Symptoms, crashes of applications, corrupted video and artifacting, BSOD's, you name it. It isn't the applications fault. We have RMA'd quite a few 8800 and 200 series cards lately.

If Arma2 truly had a BSOD causing problem, a ton more people would get it. Since only a handful of people get them, its incredibly unlikely. It's hard being a PC game developer when you can't control the end users hardware environment, they like to blame all kinds of things on you.

Also Ziip, at home I have the same card and don't have any problems. I modified my fan curve on my card though, and when that didn't help, I installed an aftermarket cooler.. the 4850's get ridiculously hot, and they get up there or if you try to OC them, they will crash the system. That's my experience at least.

Edit: My temp when running the game and the GPU is at 98% load is 52C max. 44C idle. Sapphire 4850 with aftermarket cooler. Bios was modded to lock memory clocks to the same level so when it clocks down to save power, it doesn't glitch the secondary display. (damn ATI) I noticed something very strange.. If you turn up the view distance, the GPU load goes DOWN. As if theres a CPU bottleneck and can't get the data to the GFX card fast enough. 5000VD will give me 98% load, 10000 will give me 60~

Ignore the fan speed because the aftermarket one doesn't plug into the card.

Edited August 21, 2009 by oktane

[Bulldogs 10]

Sorry Ziiip, I should've been a little less cryptic. I forgot to fully mention that by "Broad problem" I mean that it's a broad problem around the CPU area. The error occurs at the CPU but what's causing it could be overheating of the CPU or power spikes, or confusion on part of the CPU due to read/write software.

It's not GPU related in any way though so don't worry about your graphics card but monitor you CPU temps closely. As Oktane mentioned dual monitor is good for that.

And CrazyAviator, try EVGAPrecision. It's like Fraps but it's overlay can also show you your GPU temps in game as well as FPS.

[ziiip 1]

I've tried Crysis as well and voila...reboot, so you are right. But I get normal scores in 3Dmark06 so I dunno. I'll dl a cpu heat monitor and see what happens.

Here are the files that the machine wanted to send to Microsoft, packed into a rar.

http://dev-heaven.net/attachments/2144/crash_report.rar

When not in-game I dont get any reboots.

Edit: I restored the CPU clocks to the original value, but its still 58-60 celsius degrees warm at idle. Maybe it was a shitty idea to apply the fan by myself (although I've been having this cpu for 4 months and it started going crazy a week ago). The AUX (whatever that is) is 55 C at idle.

Btw my case is open and I dusted the inside yesterday.

Edit2: after playing ArmA2 for like 1.5 min the temp is 66 C.

Edited August 22, 2009 by ziiip

[ViperNL 10]

Hello guys, I had my first 2 BSOD ever in this game but it only showed for a second or 2 before rebooting (I have it set to reboot on bsod) so I dont know the specific message/code.

I do know this though: there is only 1 difference with my pre-BSOD config: I hadnt Installed Teamspeak! After installing TS, I played for a few hours before getting the BSOD. I rebooted, restarted TS and arma and I get another BSOD before im even in the menu!

Rebooted again, didnt start TS and played fine for 5 hours.. (Yes, I played a whole fucking day till 3am) :D

Now this may have been a coincidence of course, but it may also mean there's a conflict between the 2 programs (probably with the sounddriver, since I dont think TS uses much else).

I don't know if this is true or if either program is to blame, but I suggest turning off ALL programs you're running in the background, no matter how small or stable they are. You've probably tried this already, but i thought id suggest it just in case if you havent, because there may be other programs like TS that may be causing a conflict with your hardware for whatever reason.

Hope this helps! GL Officer! ;)

[oktane 0]

I've tried Crysis as well and voila...reboot, so you are right. But I get normal scores in 3Dmark06 so I dunno. I'll dl a cpu heat monitor and see what happens..

Heres a stressing app i just tried out on a system i'm OC'ing for a friend (i7 to 4ghz!)

http://www.ocbase.com/perestroika_en/index.php?Download

Pretty awesome!

[ziiip 1]

Thanks, but apparently I have a heating problem too. Now with stock clock speed, I can maintain a chilly 80 C degrees while gaming, so no BSOD. :D I wonder at what temp the PC reboots itself.