Comp crashes now from BSOD after a Display Driver has stopped responding error that keeps happening!!!!

[Gares]

Ok this is pissing me off now cause I really don't know what is going on.

I haven't changed anything since I started boxing and every once in awhile my comp will just freeze and go black and then flash my screen 5-6 times with the Display Driver has stopped responding....then it drops me to BSOD.

System setup:

Windows Vista Home Ultimate 64 SP 1
4 Gig of RAM
ATI Radeon 4850

I have the latest ATI Catalyst drivers dated August 2008. I also disabled the Catalist AI in it because it was mentioned to do it somewhere.

I checked the temperature of everything on my comp that matters (Vid Card, RAM, HD's) and they are all well below the max.

I honestly don't know what the BSOD of death actually says cause I only see if for like 1-2 seconds before a reset (I really should disable that)

My game/connections on some slaves sometimes are laggy as hell. 4 screens have green conn and 1 not.

It still seems like my clones lag behind for no apparent reason and I don't know why (I get this is solo play as well)

I HAVE NO ADDON's installed!


Anyone have any suggestions that I should do? Maybe turn my graphics down in the game on my main screen and/or slaves?

I honestly really don't know what to do....

I seen a macro somewhere about turning down the graphics on my slave screens. Maybe I should go use that? Would it really make that big of a difference?

Maybe I should get 8 gigs of RAM instead of 4? But that really won't solve my Graphics card crashing problem.

Anyone help? Suggestions?

[keyclone]

well, sounds like you're beyond the basics and you've checked your drivers... so i'll go with hardware...

check your power supply... is it enough?
next, i'd guess the memory is hosed. got any way to check it? how about just swapping it out?
could your motherboard be hosed? i had a 750i SLI FTW that was whacked... changed it out and all was better (i was getting hard system lockups)

let me know if that helps.

Rob

[Gares]

well, sounds like you're beyond the basics and you've checked your drivers... so i'll go with hardware...

check your power supply... is it enough?
next, i'd guess the memory is hosed. got any way to check it? how about just swapping it out?
could your motherboard be hosed? i had a 750i SLI FTW that was whacked... changed it out and all was better (i was getting hard system lockups)

let me know if that helps.

Rob

Ok so my power supply is 700w which is more then enough.

As for the memory I could swap it out, but I don't have any extra laying around (or the money to buy some extra ATM) so swapping it out is probably out of the question. It doesn't really lead me to believe that its that because it only happens (BSOD) so far when I am boxing and in nothing else at all. I haven't been solo playing much (Different WOW folder and HD with addons) but the only problem I have with that is choppy/soft lag gameplay that I get even when I am boxing which is starting to annoy me.

Again I don't think my MB is hose cause I only get this problem with WOW and boxing so I swear its gotta be something mixed with Graphics Pressure against my card or something along those lines. I know I am not using all my RAM up (Around 75%) as my indicator shows my RAM and CPU usage when I start doing stuff...

Very puzzled and annoyed..

[Freddie]

I honestly don't know what the BSOD of death actually says cause I only see if for like 1-2 seconds before a reset (I really should disable that)
This is a good place to start -- with the info that Windows collects to help figure out the cause of crashes. Depending on how your options are set, Windows collects a lot of it. To find it:

Go to Start Search, enter "wercon", then "View Problem History."

Microsoft provides on-line assistance for analyzing crashes. For XP this is at:

http://oca.microsoft.com/en/Welcome.aspx

I'm not sure how it's been rearranged for Vista. It may be all automated and built into Wercon now, I don't know.

Edit: If you can find reports for several instances of the same type of crash, are the memory/register numbers the same each time? If so I think it's probably a software problem, not a hardware problem.

[Gares]

I don't know what I am looking for or should copy for you to look at?

I see alot of errors but I don't know specifically what I should look at/for?

This is the Video Error I get listed 6 times before my comp shuts down..

Product
Windows

Problem
Video hardware error

Date
14/09/2008 9:29 PM

Status
Not Reported

Description
A problem with your video hardware caused Windows to stop working correctly.

Problem signature
Problem Event Name: LiveKernelEvent
OS Version: 6.0.6001.2.1.0.768.3
Locale ID: 4105

Files that help describe the problem
WD-20080914-2128.dmp
sysdata.xml
Version.txt

Extra information about the problem
BCCode: 117
BCP1: FFFFFA80042D00D0
BCP2: FFFFFA6002836390
BCP3: 0000000000000000
BCP4: 0000000000000000
OS Version: 6_0_6001
Service Pack: 1_0
Product: 768_1

Then this is the actual shutdown problem:

Product
Windows

Problem
Shut down unexpectedly

Date
14/09/2008 9:31 PM

Status
Report Sent

Problem signature
Problem Event Name: BlueScreen
OS Version: 6.0.6001.2.1.0.768.3
Locale ID: 4105

Files that help describe the problem (some files may no longer be available)
Mini091408-01.dmp
sysdata.xml
Version.txt

Extra information about the problem
BCCode: 116
BCP1: FFFFFA8006FFF4E0
BCP2: FFFFFA6002836390
BCP3: 0000000000000000
BCP4: 000000000000000D
OS Version: 6_0_6001
Service Pack: 1_0
Product: 768_1
Server information: b405d52d-2a67-4893-b9de-0df582da84c2


I have this error listed for each time my comp has looped with a display driver stopped responding and then shutdown problem.

[bugilt]

Did you take a temp reading while the computer was peak stress?

Pull out the video card and check it out. Blow it off with some compressed air (Making sure to hold the fan still to not over spin the bearing.)

[Yamio]

"Description
A problem with your video hardware caused Windows to stop working correctly."


I wouldn't be trusting that card now. Blow it off, take care of the fan, but I'm thinking that thing is on it's last leg.

[Freddie]

The first thing I'd notice is the BC code ("bug check code") for the initial problem, 117. We can look that up on Microsoft's website:

http://msdn.microsoft.com/en-us/library/aa469203.aspx

That page explains, "The VIDEO_TDR_TIMEOUT_DETECTED bug check has a value of 0x00000117. This indicates that the display driver failed to respond in a timely fashion." In other words, the driver froze. In a case like this, the operating system tries to re-initialize the driver. Which brings us to the second BC code, 116:

http://msdn.microsoft.com/en-us/library/aa469184.aspx

That page explains, "The VIDEO_TDR_ ERROR bug check has a value of 0x00000116. This indicates that an attempt to reset the display driver and recover from a timeout failed."

So the sequence of events seems to be, the driver froze and the OS tried to restart it. The OS couldn't restart it, so it shut down the machine.

Unfortunately we don't know why the video driver froze. (The dump files contain information that could help explain this, but it may be meaningful only to the people who wrote the driver.) One possibility is a bug in the driver, so you could try an earlier driver.

Another possibility, like other folks are saying, is that the video card malfunctioned due to heat or some other reason. In a case like that the driver should handle the problem gracefully and not lock up, but maybe the driver's error handling is incomplete for this particular case.

Another thing you can do is check to see whether Microsoft has any advice for this problem. You probably did that already -- I think you have to push a button on the WER panel. I'm on an XP machine atm so I can't look.

Anothing thing would be to send the WER info to ATI customer support.

If you want to send me the set of files for 117 I'll look at them but probably I won't be able to say anything useful. My email is freddie at hotkeynet.com.

[aNiMaL]

The atikmdag error probably.

Note: most of this info is from ~april but I suppose it's still unchanged.

It is a known error for several cards in any version of Vista64.

Either turn of the secondary screen or turn of the aero theme worked for me.


I got introduced to it when AoC was launched and I installed my new computer with Ultimate Vista 64.
It's not related to anything you're doing, some have reported their system crashed before they could even pass the windows login.
Vista tries to recover from the crashed driver, sometimes manages to do so even,
but most of the time gets into a loop of recovering and crashing again, which gives you your BSOD in the end.

The error has been around since vista 64 launched and till now there hasn't been a decent solution which works for everyone.
Some fixed it by setting the gpu fan to a fixed 50% or more,
some fixed it by adding more memory,
and a lot of other things i've tried before turning off the aero theme..
(After all, it's a "new" os, I wanted it's new look,
so I tried all the other options I could find before turning off the aero theme.)

Not only ati cards have this trouble btw,
many reports have been around of people with nivida cards with the same troubles,
they just have "nvlddmkm" in stead of "atikmdag" in the error.

[Freddie]

The atikmdag error probably.
I'm a little skeptical there's any single "atikmdag error," even though many people describe it that way. I think it's more likely that there are many different bugs and incompatibilities that make atikmdag lock up, causing this same series of error messages.

I just googled "aktikmdag bsod" and noticed that people report many different ways they fixed "the" problem. On some machines they fixed "it" by adjusting their BIOS. Other people fixed it by replacing RAM. Other people fixed it by cleaning out their drivers and installing from scratch. Etc.

Bad RAM ... incompatible BIOS settings ... improperly installed drivers ... it goes on and on. It's a pretty good rule of thumb that when the solutions are different, the bugs are probably also different.

For example:

http://www.repairyourpcnow.com/resources/atikmdag-has-stopped-responding.php

[Gares]

Did you take a temp reading while the computer was peak stress?

Pull out the video card and check it out. Blow it off with some compressed air (Making sure to hold the fan still to not over spin the bearing.)

My system is only 3 weeks old. There is no dust in it.


"Description
A problem with your video hardware caused Windows to stop working correctly."


I wouldn't be trusting that card now. Blow it off, take care of the fan, but I'm thinking that thing is on it's last leg.

As I said above my video card is only 3 weeks old.


The atikmdag error probably.

Note: most of this info is from ~april but I suppose it's still unchanged.

It is a known error for several cards in any version of Vista64.

Either turn of the secondary screen or turn of the aero theme worked for me.


I got introduced to it when AoC was launched and I installed my new computer with Ultimate Vista 64.
It's not related to anything you're doing, some have reported their system crashed before they could even pass the windows login.
Vista tries to recover from the crashed driver, sometimes manages to do so even,
but most of the time gets into a loop of recovering and crashing again, which gives you your BSOD in the end.

The error has been around since vista 64 launched and till now there hasn't been a decent solution which works for everyone.
Some fixed it by setting the gpu fan to a fixed 50% or more,
some fixed it by adding more memory,
and a lot of other things i've tried before turning off the aero theme..
(After all, it's a "new" os, I wanted it's new look,
so I tried all the other options I could find before turning off the aero theme.)

Not only ati cards have this trouble btw,
many reports have been around of people with nivida cards with the same troubles,
they just have "nvlddmkm" in stead of "atikmdag" in the error.

Turn off the Aero theme? How do you do that? Also I have a box setting for using only one monitor which I still had the same problem occuring.

[Ughmahedhurtz]

I have another suggestion: download the dual/quad core version of Prime95 and run a stress test on it. If it reports hardware errors, I'd be calling someone to arrange a warranty service. ;)

[Gares]

I have another suggestion: download the dual/quad core version of Prime95 and run a stress test on it. If it reports hardware errors, I'd be calling someone to arrange a warranty service. ;)

You can't do it with Vista: Windows Vista: WARNING: Vista no longer supports services that interact with the desktop. Quoted straight from the website

[Ughmahedhurtz]

I have another suggestion: download the dual/quad core version of Prime95 and run a stress test on it. If it reports hardware errors, I'd be calling someone to arrange a warranty service. ;)

You can't do it with Vista: Windows Vista: WARNING: Vista no longer supports services that interact with the desktop. Quoted straight from the websitelolwut? Are you saying Prime95 isn't compatible with Vista?

[Gares]

I think that is the case, unless I read it incorrectly. However there is no option to download for Vista users but there was some reference to XP within and something else, so I'm not entirely sure.

Maybe I should set all my slaves and main to low settings? (I have all the slaves at low settings ATM but not the main)

Also would the fact that I may not have Keyclone's Maximizer and in game resolution set properly maybe cause some problems like follower lag and hell even this crash?

I'm not sure if this is correct but I am using 2 monitors (22 inch ant 1680X1050) and (17 inch at 1280X1024)

So I have maximizer set to 2560X960 for my main and 1440X1050 for my slaves in keyclone.
My game resolution is 1680X1050 for my main screen and 800X600 for my slaves.

BTW my slave screen is split into 4 equal parts in each side 640X512 in maximizer and my main is 1680X1050.

I'm not sure if I should really fix this or not cause it doesn't really cause me problems when I PIP swap but maybe this is causing my performance issues?

[brandonbwt]

I have received the Nvidia version of this TDR error....after starting WoW, I typically get a game freeze after about 1 minute (whether I'm playing 5 windows or 1) and then the 'nvlddmkm.sys" video driver has stopped working but has been restarted following the 5-6 screen blinks, then straight to a half second BSOD + reboot.

This problem only happens when I'm playing WoW or some other game that starts to juice the graphics card.

I've also tried several suggested solutions including:

1) booting into safe mode, removing the Nvidia drivers using DriverCleaner and reinstalling via the EXPAND command at the Command Prompt.

2) wholesale uninstalling the Nvidia drivers and reinstalling from within the normal Vista 64 environment

3) manually removing and re-pasting a freshly unzipped nvlddmkm.sys file from an unzip folder

Nothing has worked so far...it has made WoW unplayable on my desktop since the bug popped up a few weeks ago. I've cleaned the case, monitored the temps on GPU, CPU and Northbridge. I've also run memtest86+ on all 4 2GB sticks and have 100% error free over a 18 hr test period. Prime95 reports no 'under load' hardware errors afte a 48 hr stress test.

Just an anecdote, this driver/hardware/whatever issue has generated a 113 page thread on the Nvidia support forums with no official response from the company. It has been present over the last 10 driver releases from Nvidia and I would take a wild guess that ATI/AMD has seen a similar response from the community.

[Zub]

sorry didnt read the whole thread (wouldn't understand half of it anyway >.<)

I used to have BSODs every now and again on my machine (after 2 minutes, or after half an hour, or anywhere in between), while boxing three WoW on the one machine.
Then someone mentionned having separate wow folders to avoid libraries clash etc.
I duplicated my wow folders - never had the BSOD again.
(i have since sym-linked all the data/interface fodlers and it's still running fine.

Try disabling all the wow display options as well, at least to test it.
Maybe that would help.

Good luck
-Zub

[Gares]

I'm not sure if that is the case. I do run seperate folders for main and slave only, but not one for each.

Problem is I have this problem on a normal basis in windows sometimes as well (Not BSOD) but the graphics error, however my comp recovers from this each time and doesn't go to BSOD.

So I'm really thinking its not a wow folder etc issues its a driver/windows issue.

[Vyndree]

Not only ati cards have this trouble btw,
many reports have been around of people with nivida cards with the same troubles,
they just have "nvlddmkm" in stead of "atikmdag" in the error.


Both Suvega and I had the dreaded "nvlddmkm" error on our "main" PCs that we just recently purchased (random reboots, sometimes with a bsod with "nvlddmkm"). It plagued us until just recently -- a nvidia driver update appears to have fixed it for our computers. These two computers are identical except for the video card -- Suvega's running the 9800 GX2 and I'm using the 9800 GTX -- very similar.

A driver update appears to have fixed my issue (I rarely, if ever, run into reboot problems. The reboot issues I do run into is usually on a very hot day, when our cooling in the room hasn't yet turned on. So I attribute the few reboots to heat.)

I believe Suvega hasn't run into too many problems recently either, now that he's updated his drivers. He still will lock up every once in a while -- often enough for it to still be suspicious. Often, Suvega's reboots happen during raids, so it's quite frustrating.


In any case, both machines are Vista x64 -- and if you search for "nvlddmkm" (and, I'd assume the ATI version of "atikmdag") you'll find the problem is notorious on many forums. There's much theorycraft about not overclocking anything, changing RAM, mobo issues.... NVidia refused to take responsibility -- they claim it's a vista problem (even though it's their driver listed in the bluescreen). However, I've updated my drivers and I seem to be totally fine, which makes me think it's a driver issue. Suvega also experienced reduced lockups/reboots after he updated drivers.

[brandonbwt]

@Vyndree:

Quick favor: can you post the full version details of the Nvidia drivers that you are using? (version, date, etc)

For Vista Ultimate 64x, the most recent Nvidia driver package could find was 175.19 [dated 5/16/2008] which I downloaded from the Nvidia website.

I reinstalled the nvlddmkm.sys file again using this version of the driver...previously the Nvidia uninstaller wasn't removing an older version (dated 12/7/2007) of the nvlddmkm.sys driver file from my C:/WINDOWS/System32/drivers folder.

After manually removing that old .sys file and pasting the new 5/16/08 version into my system32/drivers folder, I assumed it would be fixed, but upon reboot and starting up WoW, got the same TDR error after 45 seconds of being logged onto the server.