The weirdest problem I have ever seen

[TomasVan]

Hello

I have been directed here by Patrick Barker following our e-mail exchange so this si my first post.
I have executed the instructions in https://www.sysnative.com/forums/bs...g-instructions-windows-8-7-windows-vista.html.
There was a problem because the program displayed "Almost finished" and then cycled repeating "Waiting for system info".
This might be due to the fact that my computer is French. Just a guess.
However as the folder with files was created, I attach it as instructed.

The requested answers are :

· OS - Windows 7
· x86 64 bits
· What was original installed OS on system? Vista
· Is the OS an OEM version (came pre-installed on system) or full retail version (YOU purchased it from retailer)? I purchased the W7.
· Age of system (hardware) : 3 years with the exception of the GC. Bought a GTX 690 4 months ago.
· Age of OS installation - have you re-installed the OS? I abanodonned Vista and installed W7 1.5 years ago. Didn't reinstall W7 since.

· CPU : I7 975 EE
· Video Card : GTX 690 (Gigabyte)
· MotherBoard : Asus P6T WS Professional
· Power Supply - brand & wattage : Antec Signature 850 W

· System Manufacturer : A small company assembled the components I selected





Here is the description of my problem :

4 months ago I have been plagued by black screens and computer rebooting when playing games. I spent a month writing on all possible computer boards but to no avail. The problem being apparently TDR and nothing helping I decided to give away my GTX 295 and buy a new GC.
I bought a GTX 690 and for 3 months everything was perfect - the problem seemed to be solved.

But 3 days ago it reappeared again with exactly the same symptoms.
Whenever I play a game (I tried 3 but I suppose that it is ANY game) within as little as 1 minute and as much as 5, the screen goes black, the sound stays 2-3 seconds and then it goes too and either the power stays on and the fans turn despite the black screen (general case) or the computer reboots (rare case).
It happened once that the black screen didn't come inside a game but on the screen after W7 boot where the password is asked.
If I manually cut the power and boot again, W7 boots and works normally with the 1 exception mentionned above.
As I did no modification 3 days ago - neither software nor hardware, the sudden reappearance of the old problem seems to defy the causality laws.

Things I already did :

- verify for virus (ran MSE in complete analyse) . No virus detected.
- memory check with the W7 tool (no error found)
- chkdsk (no error found)
- install the latest Nvidia driver (320.49) instead of the one I had during these last 4 months (311.06)
- GC and CPU températures (normal below 50°C). The crash would anyway appear even after a cold start as soon as I enter a game.
- VDDC (0.987 V - here I don't know if it is normal but it has always been around that value)

Thanks for your help. Despite being an engineer, I have not a beginning of an idea about what could be happening here.

 

[TomasVan]

I would like to add a big compliment for the instructions featured in https://www.sysnative.com/forums/bs...g-instructions-windows-8-7-windows-vista.html. They are absolutely clear, detailed, foolproof. Perfect.
Beats MS system "advices" which are unclear, cryptical and in 90% of cases totally useless every day.
MS should take inspiration of how to write useful advices and instructions for people who are not Windows system analysts with 30 years experience.

 

[Patrick]

Hi Tomas, glad to see you registered. You're in good hands!

In regards to the perfmon, everything looks okay. My French is 'okay' so I just see mention of no antivirus being installed as a notification. That's about all I can see of relevance by attempting to read French :+)

Following the DMP files, they are all of the VIDEO_TDR_FAILURE (116) bugcheck. This indicates that an attempt to reset the display driver and recover from a timeout failed. This is why your screen is going black. When it's going black, your display driver is trying to recover. If and when the display driver fails to recover during the timeout period, you will be greeted with the *116 BSOD. If it can recover however, that's when everything will return and for example if you were AMD, Catalyst Control Center would say 'AMD Display Driver successfully recovered', etc.

Taking a list at the loaded drivers list, I see:

ASACPI.sys - Mon Oct 30 22:09:12 2006

^^ Asus ATK0110 ACPI Utility (a known BSOD maker in Win7 and Win8). Also a part of many Asus utilities. Yours is dated from 2006, the pre-2009 version is very very troublesome. I imagine you either installed this from Windows Update via Optional, or the motherboard CD?

Navigate to - P6T WS Professional - Motherboards - ASUS

and click the + next to Utilities and download ATK0110 driver for WindowsXP/Vista/Win7 32&64-bit (2009.12.07).

Moving on . . .

- VDDC (0.987 V - here I don't know if it is normal but it has always been around that value)

VDDC is the voltage, so is this the default voltage? If not, return everything to default.

- install the latest Nvidia driver (320.49) instead of the one I had during these last 4 months (311.06)

If you are already on the latest video card drivers, uninstall and install a version or a few versions behind the latest to ensure it's not a latest driver only issue. If you have already experimented with the latest video card driver and many previous versions, please give the beta driver for your card a try if such is available.

- memory check with the W7 tool (no error found)

If after updating the ASACPI.sys driver via the software and various different video card drivers, let's run a Memtest for NO LESS than ~8 passes (several hours) to test your RAM:


Memtest86+:

Download Memtest86+ here:

Memtest86+ - Advanced Memory Diagnostic Tool

Which should I download?

You can either download the pre-compiled ISO that you would burn to a CD and then boot from the CD, or you can download the auto-installer for the USB key. What this will do is format your USB drive, make it a bootable device, and then install the necessary files. Both do the same job, it's just up to you which you choose, or which you have available (whether it's CD or USB).

How Memtest works:

Memtest86 writes a series of test patterns to most memory addresses, reads back the data written, and compares it for errors.

The default pass does 9 different tests, varying in access patterns and test data. A tenth test, bit fade, is selectable from the menu. It writes all memory with zeroes, then sleeps for 90 minutes before checking to see if bits have changed (perhaps because of refresh problems). This is repeated with all ones for a total time of 3 hours per pass.

Many chipsets can report RAM speeds and timings via SPD (Serial Presence Detect) or EPP (Enhanced Performance Profiles), and some even support changing the expected memory speed. If the expected memory speed is overclocked, Memtest86 can test that memory performance is error-free with these faster settings.

Some hardware is able to report the "PAT status" (PAT: enabled or PAT: disabled). This is a reference to Intel Performance acceleration technology; there may be BIOS settings which affect this aspect of memory timing.

This information, if available to the program, can be displayed via a menu option.

Any other questions, they can most likely be answered by reading this great guide here:

FAQ : please read before posting

We'll start with this for now.

Regards,

Patrick

 

[TomasVan]

Thanks Patrick

Asus ATK0110 ACPI Utility (a known BSOD maker in Win7 and Win8). Also a part of many Asus utilities. Yours is dated from 2006, the pre-2009 version is very very troublesome. I imagine you either installed this from Windows Update via Optional, or the motherboard CD?

Indeed it must have been the MB CD because I never tinker with drivers or hardware (never oclck) when not knowing what the consequences might be. I followed your instructions and suppose the 2009 version is now installed.

If you are already on the latest video card drivers, uninstall and install a version or a few versions behind the latest to ensure it's not a latest driver only issue. If you have already experimented with the latest video card driver and many previous versions, please give the beta driver for your card a try if such is available.

This is something I have been repeatedly trying back 4 months ago when I had the problem and STILL the old GC (GTX 295).
Back then I must have tried a good dozen of drivers from 2 years old to the last to a beta. It never solved anything.
Here I remind that 3 days ago when the problem reappeared I have been using 311.06 which was not the latest. That's why my second action (just after running MSE virus check ) was to go to the Nvidia site and to install the latest driver (320.49).
The weirdest thing here is that as soon as I changed the GC 4 months ago, everything has been working fine during 3 months with a non updated Nvidia driver (311.06) and even with the hopelessly old Asus driver that you mentioned and then the same problem suddenly reappeared despite the fact that I changed, installed or updated nothing. As a summary - the problem appeared with an old driver 311.06 and continues with the last 320.49.

Download Memtest86+ here: Done. As I have neither an empty CD nor an UCB key, I can't run the test yet.

 

[Patrick]

Understood, thanks for the update. Good work on the ASACPI and in regards to Memtest, just run it when you can. Keep me updated on the behavior of the system since updating ASACPI.sys. Heading to bed now, I'll check the thread as soon as I wake up to see if you've provided any new updates.

Regards,

Patrick

 

[TomasVan]

Update

- I launched World of Tanks and the crash with black screen came after some 5 minutes as usual

- I bought a CD and did the Memtest. After 2 passes (1.5 hours) it reports 0 errors so I stopped it. I doubt that doing more passes would bring any useful information because whatever is causing the crashes, it acts very fast and dependably (it is ironical to see that the only thing one can depend upon is that the crash will occur) . The temperature is also not an issue. The necessary and sufficient condition to systematically reproduce the crash and the black screen is to enter a game and wait between 2 and 5 minutes.

 

[TomasVan]

Update 2

I have also been running the driver verifier.
As there are only few non Microsoft drivers , it was easy.
Sofar nothing special happened, I am letting it run for 24 hours as requested.

 

[satrow]

Hi Tomas,

It's obviously something related to graphics card/drivers/DirectX/etc., uninstall Intellipoint and test some games (mouse drivers must interact with DirectX and games, etc. and I've seen Intellipoint involved in BSODs before).

If that has no effect, borrow a standard (non -Logitech) mouse, uninstall the Logitech mouse drivers (best done in Device Manager, accept the driver deletion when/if prompted) and disconnect it. Connect the standard mouse, test again.

Do the games crash if you reduce the graphics GPU and memory clocks by 5-10%?

I also see you have the 'Evil Update', KB2670838 installed A platform update is available for Windows 7 SP1 and Windows Server 2008 R2 SP1, can you uninstall IE10 and also this update (it's involved in a lot more problems than the MS page admits to, it caused me multiple issues and I don't have any of the listed hardware). It may be relevant to your problem, when was it installed?

 

[TomasVan]

Do the games crash if you reduce the graphics GPU and memory clocks by 5-10%?

I don't know how to do that. I thought to try to set the graphics inside a game to minimum but didn't come to it yet. In all games the settings are generally on maximum because the GTX 690 is much more powerful than what is required. In any case the settings didn't change 3 days ago when the problem reappeared.

If that has no effect, borrow a standard (non -Logitech) mouse, uninstall the Logitech mouse drivers (best done in Device Manager, accept the driver deletion when/if prompted) and disconnect it. Connect the standard mouse, test again.

I used already a non Logitech mouse. But I never uninstalled the drivers. Will do.

I also see you have the 'Evil Update', KB2670838 installed A platform update is available for Windows 7 SP1 and Windows Server 2008 R2 SP1, can you uninstall IE10 and also this update (it's involved in a lot more problems than the MS page admits to, it caused me multiple issues and I don't have any of the listed hardware). It may be relevant to your problem, when was it installed?

No idea. I generally approve all the updates that MS proposes automatically (mostly called security updates).
But it is a pretty good guess that both were installed before this period of 4 months after I changed the GC and the system was running perfectly.
What is not a guess but is sure is that I installed nothing such 3 days ago when the black screens came back again.

P.S
Driver Verifier has been running for 5 hours now. Still no problems.

 

[Patrick]

Hi,

To me, at this point already, if satrow's suggestions don't provide any light, the black screen and such raise flags for video card failure. All the bugchecks are consistent nVidia video driver crashes with the display driver failing to recover, and there are some Direct X Kernel calls in the stack. I would recommend when you get a chance seeing if the same thing happens when running Furmark:

Furmark:

FurMark Video Stress Test - free from here: FurMark: VGA Stress Test, Graphics Card and GPU Stability Test, Burn-in Test, OpenGL Benchmark and GPU Temperature | oZone3D.Net
FurMark Setup:
- If you have more than one GPU, select Multi-GPU during setup
- In the Run mode box, select "Stability Test" and "Log GPU Temperature"
Click "Go" to start the test (Looks like it's "BURN-IN test" now)
- Run the test until the GPU temperature maxes out - or until you start having problems (whichever comes first).
- Click "Quit" to exit

Regards,

Patrick

 

[TomasVan]

I would recommend when you get a chance seeing if the same thing happens when running Furmark:

OK, now I am going to bed but will do it tomorrow.
The Driver Verifier is still running. No problems, no crashes.
I have difficulty to believe in the GC (hardware) failure.
I had this same problem with the previous GC GTX 295.
I change it and buy a GTX 690.
The problem disappears and during 4 months all is perfect. No crashes.
3 days ago it reappears.
How high is the probability that 2 different GC present exactly the same problem and the second one in only 4 months ?

 

[Patrick]

Hardware failure is hardware failure, it happens. Regardless of whether or not a piece of hardware is the problem, we aim to do the diagnostics anyway to form a checklist so we can easily progress and ultimately find the culprit.

Right now, high on the priority list is Memtest and Furmark + satrow's suggestions. If you don't crash within ~24 hours with verifier enabled, please disable it.

Regards,

Patrick

 

[TomasVan]

Right now, high on the priority list is Memtest and Furmark + satrow's suggestions. If you don't crash within ~24 hours with verifier enabled, please disable it.

- Memtest done. No errors found.
- Verifier still running. It has been 15 hours now.
- In 9 hours when I stop the Verifier, I will do the other suggested tests.

 

[Patrick]

Hi,

Good work, and I'll look forward to your next update. You ran Memtest for ~8 passes, correct?

Regards,

Patrick

 

[TomasVan]

Update

- Stopped Driver Verifier after 25 hours. No problem found.

- Ran Furmark. The temperature maxed more or less out at 90°C on GPU1 (GPU2 being around 80). Little time later the system crashed with a black screen. Minidump attached as Furmark.

- Went to a game (World of tanks) to verify températures. The game runs at 58°C and this time I had no crash during 3 hours. However once it crashed at last, it began to crash immediately after the reboot as soon as I arrived on desktop. This is something that I have sofar never seen. The crashes were sofar ONLY happening when in a game. I had to boot in the safe mode in order to be able to do this update. Minidump of the last crash (while being on desktop) is attached as Last. The temperature at this moment was around 40°C.

 

[Patrick]

Hi,

Both dumps still of the *116 TDR bugcheck. This is seems like clear cut and dry video card failure to me. Do you have by any chance have access to integrated graphics? If so, uninstall your video card drivers, shut down, remove video card(s), enable integrated graphics, install latest integrated graphics from motherboard product page, use the system as you regularly would.

Regards,

Patrick

 

[Patrick]

Hi,

Are you still with us, Tomas?

Regards,

Patrick

 

[TomasVan]

Hi,

Are you still with us, Tomas?

Regards,

Patrick

Yes but we apparently ran out of tests. My MB has not integrated graphics so I couldn't do what you suggested.
However something even weirder happened.
The day after I did my last post with the crash dumps, the system began to crash already during booting.
I wanted then to boot in the safe mode and it crashed too (on the screen asking the PW after W7 has booted).
So as apparently the last option left was to send the GC to Gigabyte for exchange, I needed the Serial Number so I unplugged the power and removed the GC to read the SN.
After that mechanically and without thinking I put the GC back again, plugged power in and pressed the power button.

This time I had no crash but the BIOS wanted me to put a boot device what it shouldn't be doing. Entering the set up, I noticed that the BIOS options have been initialised - the list of HD, the booting order, the system clock etc. I don't know why but it looked like an initialisation so I put back the right HD to boot from and the system booted without crash.
Since that time and that is 3 days now, I had no black screens and no crashes in games or otherwise.
So my conclusion was that I didn't know why it was crashing and I don't know why it started working now.
As we did tests that excluded RAM, HD, drivers, temperature and the last (unvoluntary) experiment excludes apparently GC hardware, I don't know what's left as the possible black screen cause.
Hence it can potentially start crashing again in 1 hour or 1 week.

 

[Deek]

Ah, the old take it apart and put it back together fix...I forgot about that one...works like a charm. In your case, a good blow-out of the slot might be in order if the issue returns. It only take a tiny piece of dirt to cause issues like this.

The bios resetting is almost always attributed to you needing a new motherboard battery, they are usually a CR2032 coin cell found just about anywhere for a couple bucks.