At wits end with BSOD's

Csukar · 14 Aug 2014

Hi everyone, i have been plagued with constant bsod's for the last few months, i'll start from the beggining...
I purchased a second hand graphics card from ebay and after a few weeks of using it i started getting the bsod's, my temp's were fine all my drivers were up to date, i ran chkdisk and mem test etc.
In the end i bought a new gpu, but the bsod's still persist, i recently ran memtest again and it came up with two errors, so is my ram the likely cause?
I'm short on money at the moment so i don't want to fork out £60 for some new ram if it isn't going to fix the issue, so could one of you guys explain to me how to upload my minidumps so one of you could take a look for me? It would be greatly appreciated.

Csukar · 14 Aug 2014

Is this what you guys will need?

Arc · 14 Aug 2014

Csukar said:

Is this what you guys will need?

We need something more.

Click on the button below ....

It will download the DM log collector. Right click on the application and run as administrator. It will generate a .zip file on your desktop. Upload the .zip. Dont cut any file down from it.
Screenshots and Files - Upload and Post in Seven Forums

Csukar · 14 Aug 2014

Ok done that here you are :)

Arc · 15 Aug 2014

I checked your crash dumps. I am displaying some of it for you.

With CCC 14.6 RC2 installed:

Code:

BugCheck 1000007E, {ffffffffc0000005, fffff880045d36e1, fffff8800649a768, fffff88006499fc0}

Probably caused by : dxgmms1.sys ( dxgmms1!VidSchiSubmitRenderCommand+149 )
-----------------------------------------------------
    Image name: atikmdag.sys
    Timestamp:        Wed Jul 09 21:04:34 2014 (53BD610A)

RCs may fail, so it is worthy to see what is the behavior with a WHQL installed. Here:

Code:

BugCheck 1000007E, {ffffffffc0000005, fffff88004390998, fffff880053a3768, fffff880053a2fc0}

Probably caused by : dxgmms1.sys ( dxgmms1!VIDMM_APERTURE_SEGMENT::EvictResource+d0 )
------------------------------------------------------
    Image name: atikmpag.sys
    Timestamp:        Fri Apr 18 06:37:07 2014 (53507ABB)

Same with 14.4 WHQL installed. So this issue is not being caused by the display driver, then? We have got the answer too.

Code:

BugCheck 50, {fffff120174414d0, 0, fffff880117f439b, 7}

*** WARNING: Unable to verify timestamp for atikmdag.sys
*** ERROR: Module load completed but symbols could not be loaded for atikmdag.sys

Could not read faulting driver name
Probably caused by : hardware ( atikmdag+76139b )

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

So the source of the issue is the display hardware, Your R9 200 series GPU. Lately we have seen a lot of issues with those GPUs. But, at first disable any overclocking that you may have. Then take a stress test.
Video Card - Stress Test with Furmark

Let us see how it takes the stress.

Csukar · 15 Aug 2014

Hmm so I've just been incredibly unlucky and got two bad gpu's in a row? It's fine under stress, infact it crashes less frequently under stress, recently it seems to be occurring when browsing Facebook or when idle on the desktop.
Also could it be a bad pci-e slot? My card wobbles quite a bit when inserted properly, a few days ago i put some bluetac over the metal bracket bit where it's meant to screw into the case so that it doesn't wobble, and since then the crashes have been a lot less frequent.
Is it worth RMAing my card?
And thank you for spending your time helping me :)

Arc · 15 Aug 2014

Hmm so I've just been incredibly unlucky and got two bad gpu's in a row?

So you had problems with another GPU too, and you have replaced it? Can you let us know exactly when the GPU is replaced? Not having this information might be misleading, because the crash dump that shows the display hardware issue is of 31/07/2014. May be it is of the previous GPU?

infact it crashes less frequently under stress,

Frequency is not vital, but ..... did it crash during stress test? Let us know it clearly.

Also could it be a bad pci-e slot?

The question rightly appears when multiple GPUs fail in the same way. You may get this answer. Do you have access to any other computer? Your own second machine, or any computer of the family members or friends? Apply the GPU there. Observe how it works. If everything goes good with the other machine, then the PCIe slot is the victim. Otherwise, if the other machine crashes in the same way too, you know it is the GPU.

Also I will request my Guru essenbe to have a look at your issue, lest I am missing something.

Csukar · 15 Aug 2014

Yes i replaced my old GPU, my R9 is only just over a month old, my previous GPU was a GTX 560. The crash dump from the 31st was when i was using my R9 i am certain of that.
Sadly i do not have access to another computer so i can't test in another machine.
And i'm running the stress test now, i'll let you know the results, how long should i let it run for?

Csukar · 15 Aug 2014

Also forgot to mention incase you ask, my gpu temperatures are fine under strees, never seen them go over 70 degrees.
UPDATE: Ran the 15 minute 1920x1080 preset in furmark, everything was fine and no crashes.

essenbe · 15 Aug 2014

You need to replace the faulty ram. No one can say if it will solve all your problems, but your computer will never run right unless the ram is without fault. Your problem very well may be the faulty ram. If the ram faults, the computer will crash and the debugger in your computer will blame whatever driver is most in use. Again, I can't promise you that good ram will solve all your problems, but until you have good ram none of the dump file readings are reliable. I can tell you that with bad ram, you will always be plagued with BSODs.