SOLVED bsods almost every day for a long long time

[zambas]

This error pops up out of nowhere and most of the times i am not doing anything intensive. Sometimes i just open an internet page, other times i am watching xbmc or watching ytube.
My hard drive is ssd and i dont have any fan pointed to it. Maybe i will put also a fan there.
The latest error i've changed resolution of the screen to 2560 x 1080 to check some games if they perform better.
I've downloaded the logviewer, looks very nice.
With the blue scren viewer and its limited capabilities i see that its always the same error 'atimdag.sys +2781e. That 2781e is the address of the memory that the error took place? If its always the same is it safe to assume its the ram of the gpu that has the problem?

 

[Shintaro]

Maybe it's time to run Windows Performance Recorder.
Microsoft Web page and download HERE.

 

[zambas]

Ok i am installing now.
While installing i had a weird black screen and then an error popped out saying display driver stopped responding and recovered. This is a first.
After i install should i enable something?
Here are the events if it tells you something

 

[Shintaro]

Are you able to drop back to an earlier Video driver?

 

[zambas]

I dont think its the driver Shintaro. I am using just the driver now without ccc and this driver was working for a long time without problems.
Something else is causing all this. It must be temperature. I left the case open now and no problems for 3 days. Also i've moved the hard drive to a different location. What was the temp? 56? In theory ssd drivers shouldnt be able to withstand much more heat? I dont think its the hard drive also but i did it anyway.

 

[kw123]

I dont think its the driver Shintaro. I am using just the driver now without ccc and this driver was working for a long time without problems.
Something else is causing all this. It must be temperature. I left the case open now and no problems for 3 days. Also i've moved the hard drive to a different location. What was the temp? 56? In theory ssd drivers shouldnt be able to withstand much more heat? I dont think its the hard drive also but i did it anyway.

In my experience SSD's aren't as reliable as you would imagine they would be, so perhaps 56 is too high for your SSD to cope with?

Edit: Or is it a HDD? I can't remember what your main drive is... If it is an SSD, it isn't one of those terrible OCZ ones, is it? Seen a fair few of them fail...

 

[Shintaro]

Mate,

See if there are any crash dump files in c:\Windows\Live KernelReports\WATCHDOG

I had a similar problem, that the display would "recover". I dropped back to an earlier driver and it was fixed.

The temperature is from your data that you provided, so unless the sensor is faulty......
But heat and electronics is not a good mix, well unless you intend on cooking eggs on it. . You can buy cheap hard disk fans for $5.

 

[zambas]

Shintaro there are crash reports there. I dont know how to read them i will zip some for u to see. thanks
Kw123 the drive is corsair and reviews put it on top of the list. Either way ssd is what? pcb and smt components which logically would stand over 100 degrees of punishment very easily.
One thing for sure.. one of these days when i decide to take the hammer and send it to its maker i will send you both the video. It will make a ytube hit for sure

 

[kw123]

From everything I've read so far it seems like it's some sort of graphical issue for sure. You're not missing a PCI-e power plug into your GPU by any chance are you?

Have you tried completely uninstalling your GPU drivers + software using device manager and reinstalling from scratch?

Also, have you checked the settings in your BIOS to check that it isn't automatically trying to overclock your GPU or memory? Some motherboards have the option to OC the GPU too you see...

Also, if you have any software like MSI Afterburner or similar I would recommend uninstalling just for now.

Also, it seems like you have rolled back quite a few Windows updates - I would run Windows Update and get it back up-to-date if I was you, seeing as it didn't stop the BSOD's anyway.

 

[zambas]

Everything is double and triple checked. I even changed the modular cable coming from the psu just to be sure. I even changed gpu to a different pci-e slot
Its always from scratch. when i do a driver change i am using display driver unistaller.
I will uninstall msi just to be safe, good call.
The thing about windows updates is that i am sure that its not updates per se that was the problem but maybe some versions were installed where some others needed. But removing the updates did solved the bsods.
My problem now is that the pc shuts down and restarts itself at random times. Like removing power cord this kind of.
And i also had ,3 days ago, the weird driver restored problem which didnt happened again.

 

[Shintaro]

Mate,
They all still point to the video driver. I still maintain trying an older driver.

Debug session time: Sun Jul 20 06:42:48.802 2014 (UTC + 10:00)

Built by: 7601.17514.amd64fre.win7sp1_rtm.101119-1850

System Uptime: 0 days 11:29:17.754

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

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

Probably caused by : atikmpag.sys ( atikmpag+bd7c )

BugCheck 117, {fffffa800ac6e3a0, fffff88004a0bd7c, 0, 0}

BugCheck Info: VIDEO_TDR_TIMEOUT_DETECTED (117)

Bugcheck code 00000117

Arguments:

Arg1: fffffa800ac6e3a0, Optional pointer to internal TDR recovery context (TDR_RECOVERY_CONTEXT).

Arg2: fffff88004a0bd7c, The pointer into responsible device driver module (e.g owner tag).

Arg3: 0000000000000000, The secondary driver specific bucketing key.

Arg4: 0000000000000000, Optional internal context dependent data.

BUGCHECK_STR: 0x117

DEFAULT_BUCKET_ID: GRAPHICS_DRIVER_TDR_TIMEOUT

PROCESS_NAME: System

FAILURE_BUCKET_ID: X64_0x117_IMAGE_atikmpag.sys

¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨``

Debug session time: Sun Jul 20 06:42:41.720 2014 (UTC + 10:00

Built by: 7601.17514.amd64fre.win7sp1_rtm.101119-1850

System Uptime: 0 days 11:29:10.673

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

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

Probably caused by : atikmpag.sys ( atikmpag+bd7c )

BugCheck 117, {fffffa800aad3390, fffff88004a0bd7c, 0, 0}

BugCheck Info: VIDEO_TDR_TIMEOUT_DETECTED (117)

Bugcheck code 00000117

Arguments:

Arg1: fffffa800aad3390, Optional pointer to internal TDR recovery context (TDR_RECOVERY_CONTEXT).

Arg2: fffff88004a0bd7c, The pointer into responsible device driver module (e.g owner tag).

Arg3: 0000000000000000, The secondary driver specific bucketing key.

Arg4: 0000000000000000, Optional internal context dependent data.

BUGCHECK_STR: 0x117

DEFAULT_BUCKET_ID: GRAPHICS_DRIVER_TDR_TIMEOUT

PROCESS_NAME: System

FAILURE_BUCKET_ID: X64_0x117_IMAGE_atikmpag.sys

¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨``

Debug session time: Sat Jul 19 21:35:56.404 2014 (UTC + 10:00)

Built by: 7601.17514.amd64fre.win7sp1_rtm.101119-1850

System Uptime: 0 days 2:22:25.356

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

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

Probably caused by : atikmpag.sys ( atikmpag+bd7c )

BugCheck 117, {fffffa8008b4b390, fffff88004a0bd7c, 0, 0}

BugCheck Info: VIDEO_TDR_TIMEOUT_DETECTED (117)

Bugcheck code 00000117

Arguments:

Arg1: fffffa8008b4b390, Optional pointer to internal TDR recovery context (TDR_RECOVERY_CONTEXT).

Arg2: fffff88004a0bd7c, The pointer into responsible device driver module (e.g owner tag).

Arg3: 0000000000000000, The secondary driver specific bucketing key.

Arg4: 0000000000000000, Optional internal context dependent data.

BUGCHECK_STR: 0x117

DEFAULT_BUCKET_ID: GRAPHICS_DRIVER_TDR_TIMEOUT

PROCESS_NAME: System

FAILURE_BUCKET_ID: X64_0x117_IMAGE_atikmpag.sys

¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨``

Debug session time: Sat Jul 19 21:35:51.448 2014 (UTC + 10:00)

Built by: 7601.17514.amd64fre.win7sp1_rtm.101119-1850

System Uptime: 0 days 2:22:20.400

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

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

Probably caused by : atikmpag.sys ( atikmpag+bd7c )

BugCheck 117, {fffffa8008b4b390, fffff88004a0bd7c, 0, 0}

BugCheck Info: VIDEO_TDR_TIMEOUT_DETECTED (117)

Bugcheck code 00000117

Arguments:

Arg1: fffffa8008b4b390, Optional pointer to internal TDR recovery context (TDR_RECOVERY_CONTEXT).

Arg2: fffff88004a0bd7c, The pointer into responsible device driver module (e.g owner tag).

Arg3: 0000000000000000, The secondary driver specific bucketing key.

Arg4: 0000000000000000, Optional internal context dependent data.

BUGCHECK_STR: 0x117

DEFAULT_BUCKET_ID: GRAPHICS_DRIVER_TDR_TIMEOUT

PROCESS_NAME: System

FAILURE_BUCKET_ID: X64_0x117_IMAGE_atikmpag.sys

¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨``

Debug session time: Sun Jun 8 21:24:48.594 2014 (UTC + 10:00)

Built by: 7601.18409.amd64fre.win7sp1_gdr.140303-2144

System Uptime: 0 days 0:08:49.873

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

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

Probably caused by : atikmpag.sys ( atikmpag+bd7c )

BugCheck 117, {fffffa800a19b010, fffff88005a8cd7c, 0, 0}

BugCheck Info: VIDEO_TDR_TIMEOUT_DETECTED (117)

Bugcheck code 00000117

Arguments:

Arg1: fffffa800a19b010, Optional pointer to internal TDR recovery context (TDR_RECOVERY_CONTEXT).

Arg2: fffff88005a8cd7c, The pointer into responsible device driver module (e.g owner tag).

Arg3: 0000000000000000, The secondary driver specific bucketing key.

Arg4: 0000000000000000, Optional internal context dependent data.

BUGCHECK_STR: 0x117

DEFAULT_BUCKET_ID: GRAPHICS_DRIVER_TDR_TIMEOUT

PROCESS_NAME: System

FAILURE_BUCKET_ID: X64_0x117_IMAGE_atikmpag.sys

¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨``

 

[Shintaro]

I just found THIS article on the AMD website. Might be of use.

 

[kw123]

Everything is double and triple checked. I even changed the modular cable coming from the psu just to be sure. I even changed gpu to a different pci-e slot
Its always from scratch. when i do a driver change i am using display driver unistaller.
I will uninstall msi just to be safe, good call.
The thing about windows updates is that i am sure that its not updates per se that was the problem but maybe some versions were installed where some others needed. But removing the updates did solved the bsods.
My problem now is that the pc shuts down and restarts itself at random times. Like removing power cord this kind of.
And i also had ,3 days ago, the weird driver restored problem which didnt happened again.

When I say PCI-e power I mean the 6+2 pin connectors that come off your PSU into the end of your GPU, your card needs an 8 pin + a 6 pin one connected to work properly I think.

The article Shintaro posted above has lots of things to try that may help, good find!

 

[zambas]

I will leave the pc case open for a few more days and i will see if everything is ok. If not i will change my pc case to a better one.
If i come to a problem again i will update windows again, maybe this time will be ok.
Shintaro this driver i am using is i think the first that supported my gpu, the one that released with my gpu. If anything i will try and update the driver to a newer one and not older.
What do you suggest? Wait and see what i said above or change the drivers?

 

[zambas]

Yes i know kw123. It needs a 6+2 pins and a 6 pin.
Those kind of articles usually points to everything. Its very vague. And even so all the steps there i've done it already 2 times by now

 

[Shintaro]

Mate,
Is it possible to roll back the GPU firmware? So that the GPU firmware matches the approximate date of the release of the Driver software?

That article was purely suggested to make sure that there wasn't anything that was missed.

 

[zambas]

How to roll back? I can flash the firmware with whatever fw suitable. Like asus r9 fw into gigabyte r9 fw but the date? How?
I dont follow..

 

[Shintaro]

GPU

1/ What was the initial version number of the released Firmware on the GPU? (What version of firmware was on the GPU when you got it?)

Do you know what the release date of that firmware was?​

2/ What is the current version of Firmware on the GPU?

Do you know what the release date of that firmware was?​

Driver

Driver Version 13.1 was release on 21st of January 2013

So basically is there a released version of firmware that has a release date around that time?

​

 

[zambas]

Gpu fw doesnt automatically updates. You have to update it yourself and this is not something a reqular consumer will do it or has the knowledge to do it.
Are you telling me to find a suitable fw that was released prior my own? Before or close to 21/1/2013 so that driver and fw will be 'compatible'? But my card was the first r9 290 that came out with the reference cooler and i a m pretty sure the fw written on it is one of the first.
Hm.. I can find lots of fw but as for the date i dont know how i will tell..

 

[Shintaro]

I think you said that you flashed it from R9/290 to R9/290X
Is it possible to go back to R9/290?

So unless your already on R9/290, I am wondering if the R9/290X is causing instability?