Windows 7 Need help: BSOD caused by AMD Radeon R9 290 series

[Askin]

My specs:

• CPU: QuardCore AMD Phenom II X4 Black Edition 965, 3415 MHz
• Mainboard: M5A78LLE ASUS
• RAM: 8 GB RAM
• GPU: Sapphire Tri-X R9 290
• Power adapter: SF-550P14HE 550W 230V only
• System: Windows 7 Professional

So, whenever I use my Sapphire Tri-X R9 290 my computer crashes absolutly randomly (playing games, using windows, whatever, even if I do nothing):
Link Removed

So, here are the important parts:
I previously had the exact same error with a Powercolor R9 290, what I tried:
- removing all drivers that could cause the problem
- updating my mainboard BIOS
- trying multiple drivers and AMD catalysts
- testing my RAM and CPU
- Windows reinstallation
- many other things I don't remember anymore

This problem does not occur with my Sapphire Radeon HD 6850 and the weird thing is, I can play games on ultra (with the new Sapphire Tri-X R290) and it doesn't crash. Then, I go do something in Windows and it crashes, it is absolutly random... sometimes it crashes 1 minutes after boot, and the next time it may crash 20 minutes after boot.
When I sent my Powercolor R9 290 back, they told me that the GPU was absolutly fine and worked 100% when they tested it.

I really don't know what the hell is going on... I tried to find a solution in the internet, I tried everything. The next thing I'll do is replacing everything (Mainboard, CPU etc.). But it will take some time until I can do that, so fixing the problem now would be really good.

 

[kemical]

Hi Askin,
to properly ascertain what the bsod could be caused by we really need the actual dump files. Please make sure your settings are set thus:

Go to Start and type in sysdm.cpl and press Enter
Click on the Advanced tab
Click on the Startup and Recovery Settings button
Ensure that Automatically restart is unchecked
Under the Write Debugging Information header select Small memory dump (256 kB) in the dropdown box
Ensure that the Small Dump Directory is listed as %systemroot%\Minidump << where your .dmp files can be
found later.
Click OK twice to exit the dialogs, then reboot for the changes to take effect.

If you already have some dump files please read the thread found here and post the info needed:
Link Removed

I see from the above that the bsod is 124 bugcheck which means hardware error, looking at the specs you posted above in relation to the Sapphire R290 I see you are running a 550W power supply with only 16-18 Amps on the 12v rail:
Link Removed
With high end gpu's it's the Amps one has to be careful of as well as the voltage. The system requirement page for the Sapphire card has it requiring a 750W PSU and googling led me to discover it also needs 48 Amps:
Link Removed
The reason your last card worked fine is because it only needs a 500W PSU:
Link Removed
The lack of power will cause the type of issues you have been seeing as the PSU struggles to keep up. You really cannot scrimp on a PSU and one should always buy a branded model which has been researched first.

 

[Askin]

Link Removed

But, if it's just the power supply, why does the card work under full load and not crash immidietly? I'll try to create a dump file.

Uhm, where do I find the dumpfile? It's not in C:...


Nevermind. Is the Sea Sonic M12II-520 Bronze EVO Edition 520W ATX 2.3 (SS-520GM2) good enough?

 

[kemical]

This is the system requirements for the R290:

• [*]PCI Express® based PC is required with one X16 lane graphics slot available on the motherboard.
  [*]1X75 Watt 6-pin PCI Express power connector is required.
  [*]4096MBMinimum of system memory.
  [*]Installation software requires CD-ROM drive.
  [*]DVD playback requires DVD drive.
  [*]Blu-ray™ / HD DVD playback requires Blu-ray / HD DVD drive.
  [*]750 Watt Power Supply is required.
  [*]1X150 Watt 8-pin PCI Express power connector is required.
  [*]64-bit operating system highly recommended
  [*]A display with digital input (HDMI, DisplayPort or DVI) is required

Ref:
Link Removed

Dumpfiles can be found in the Minidump folder which is in the main Windows folder.
Rarely is a system at 100% load for both CPU and GPU unless specifically tasked to do so. This is probably why it happens at odd moments. It's still possible that something else may be in the mix but until I see your dump files I can't really say further.

I did some further searching on the psu needed and Guru3D has the psu down to a 550W-600W whilst other sites mention 600-650W. You probably have enough volts or close to enough it's just the Amps your lacking..

 

[Askin]

I did this:

1. Go to Start and type in sysdm.cpl and press Enter
2. Click on the Advanced tab
3. Click on the Startup and Recovery Settings button
4. Ensure that Automatically restart is unchecked
5. Under the Write Debugging Information header select Small memory dump (256 kB) in the dropdown box
6. Ensure that the Small Dump Directory is listed as %systemroot%\Minidump << where your .dmp files can be found later
7. Click OK twice to exit the dialogs, then reboot for the changes to take effect.

But there are no dumpfiles in my C drive...

 

[kemical]

You need to wait until the bsod occurs again then post the dump file.

 

[Askin]

Here are the dumps: (I was to stupid to find them^^)

 

[Askin]

I ordered the Sea Sonic G-Series G-650 650W ATX 2.3 (SSR-650RM), I hope it'll fix it.

 

[kemical]

I ordered the Sea Sonic G-Series G-650 650W ATX 2.3 (SSR-650RM), I hope it'll fix it.

Very nice.. 54 Amps on the 12v rail... Perfect! Good review here:
Link Removed

Dump File:

*******************************************************************************
*                                                                             *
*                        Bugcheck Analysis                                    *
*                                                                             *
*******************************************************************************

Use !analyze -v to get detailed debugging information.

BugCheck 124, {0, fffffa8007d33038, 0, 0}

Probably caused by : AuthenticAMD

Followup: MachineOwner

Most of the dump files sent were like the above apart from one which was a F4 Bugcheck which basically means a critical process was unexpectedly terminated. Now whilst the above looks like it's saying it's the AMD processor it could in fact be several things which lead to the processor crashing. Looking through the dump file further I note the type of crash was a machine check exception. Causes of this type of crash can be:

• overclocking (which normally increases heat-output)

• poorly-fitted heatsink/computer fans (the same problem can happen with excessive dust in the CPU fan)

• an overloaded internal or external power-supply (fixable by upgrading)

Ref:
http://en.wikipedia.org/wiki/Machine-check_exception

Unfortunately the dump files didn't carry your unloaded driver list but as you said you've recently updated everything it's not a huge worry. I noted your running Norton AV. This anti-virus app can often be at the root of many an issue and I'd recommend you uninstalling it using the Norton removal tool:
https://support.norton.com/sp/en/uk/home/current/solutions/kb20080710133834EN_EndUserProfile_en_us

MSE is extremely stable and good at what it does:
Link Removed

Post any new dump files.

 

[Askin]

I uninstalled Norton and installed MSE. I installed my old card again, to fix the problem for now.

Thank you for your help, I'll wait until the new PSU comes (which is probably tomorrow) and tell you if it fixed the problem.

 

[kemical]

Your welcome and best of luck with the upgrade. (fingers crossed.. )

 

[Askin]

It didn't help. Same problem... I have connected the card on one rail, is that right?

Tried 2 rails, didn't help.

 

[Askin]

Last 2 times it didn't really crash, the screen just freezed. I had to restart the computer myself. I don't know if this is important, but when the PC started, for a short amount of time in the beginning it was running on like 1fps (the mouse was "jumping"). And, when I scroll down web pages, I see little white and gry squares, as it is still loading the graphics...

 

[Askin]

Third time it happend. Again, low mouse fps when windows started and again the PC just freezed. I have no new dumpfiles because technicly the PC does not crash, it just freezes...

And before I forget, it may be coincidence, but it crashed twice when I scrolled down a web page.

New dumpfile, it crashed properly.

 

[kemical]

The card should sit in slot number 1 which is usually the uppermost PCIE slot. Make sure all the extra cables (I think it's two 1x6pin and 1x8pin) are connected to the card.
If the system is not actually blue screening it could be down to some corruption. Try running the system file checker. Find command prompt in the start menu, right click on it and choose run as admin. Type:
sfc /scannow
Press enter and await results.

Also try removing the old gpu driver using this application (display driver remover):
Link Removed
Install the latest version:
Link Removed

 

[Askin]

Scan said everything is fine. Removing old GPU Driver did not help, installing latest driver, beta driver and older drivers did not change anything. Removing CCC did not help...

Sometimes the PC freezes, sometimes it bluescreens...

 

[Askin]

New dump... if it helps. Tried to downclock the card to 1000mhz. Tried to downclock RAM to 800mhz, didn't work.

 

[kemical]

*******************************************************************************
*                                                                             *
*                        Bugcheck Analysis                                    *
*                                                                             *
*******************************************************************************

Use !analyze -v to get detailed debugging information.

BugCheck 124, {0, fffffa80081b4038, 0, 0}

Probably caused by : AuthenticAMD

Followup: MachineOwner

Same as before... Damn. The errors you describe however remind me of RAM or HDD failure when the machine will do strange things and crash at any given point.
Let's do some basic testing to see if we can ascertain what the likely cause is.
HDD
Please run a chkdsk on the C drive:
Link Removed
CPU
Please run the Torture test found in Prime95 and test the CPU temperatures:
http://www.mersenne.org/download/
Monitor temperatures:
http://www.hwinfo.com/

It may also be worth taking the heatsink off and reseating the CPU just to make sure no dust is in with the pins or some other unlikely occurrence
.
RAM
Download the latest version of Memtest86 here:
http://www.memtest.org/
This guide although a little old explains how best to run Memtest86:
http://www.techspot.com/community/topics/tutorial-how-to-use-memtest.62524/

Looking through your drivers I noticed your also running Daemon Tools. This app is well known for causing bsod issues please uninstall to test.

You've also got some pretty old Broadcom drivers on there. Try and update if you can:
http://www.broadcom.com/support/ethernet_nic/downloaddrivers.php

 

[Askin]

I'll do everything you said, but I doubt it'll work. I did memtest for 20 mins when I had the Powercolor... HDD issues are unlikely as I formated the drive when I reinstalled windows (the error did not disappear). Also, my PC works perfectly fine even when the CPU is at it's limits (with my old radeon hd 6800...). Even when the cpu had a cenimeter thick dust layer and overheated all the time the pc did not crash. The PC only crashes when I use the R9 290 series (powercolor and sapphire) and I am not the only one who has the problem (r9 290 causes this on many systems appearently). But I even tried the solutions offered by the others, like downclocking gpu and ram... using older drivers etc. nothing worked. Some say it's a problem with the voltage, and I even tried force voltage with msi afterburner.... did not help. I have no idea what to do, I know it's not the card... but it is definitly caused by the gpu. Btw don't know if worth mentioning, but I only use one HDD with one partition.

 

[kemical]

It would be better if you could run the tests as that at least rules those areas tested out (if they pass of course). As for memtest86 and 20mins it's simply not long enough and it's best left overnight as some errors won't appear until the 12th hour.
It might be that the 290 is stressing your machine in a slightly different way and exposing a weakness. We have to find that weakness and unfortunately this really is the only way to go about it..