Solved New GPU causing 'ntoskrnl' BSOD randomly

[jadenred]

I recently purchased a Gigabyte R9 290X which i installed into my custom built computer to replace the R9 270X. The previous card ran fine and i had no problems with. Obviously before installing the new card i ran DDU (Display driver uninstaller) to remove the previous drivers. After installing and powering the PC up, i installed the latest catalyst software, pretty sure its 14.9, but since then it randomly blue screens.

I have tried older drivers, MEMtest, checking for faults in my SSD and HDD.

I also ran FurMark at benchmark for 15 minutes to check for over heating etc. and the card ran fine and didnt crash for the entire 15 minutes, it then crashed sitting on the windows start page.

Any help would be greatly appreciated.

 

[lapetus]

Most of the BSODs are related to this Stop 0x124, so have alook here:

http://www.sevenforums.com/crash-lockup-debug-how/35349-stop-0x124-what-means-what-try.html

While I know how annoying it is to RMA, I think if you can at all you should because graphic cards are temperamental at their best imo.

EDIT
Here are some more ideas.
The first thing I would do is check if the extra power supply is definitely connected to the card and working, and that the temps on card are all normal.

Then I would try using verifier.exe to stress the Ati drivers and see if it's them. But I'm not sure on this because say it is the Ati driver- I think it could happen that stressing a different driver causes the Ati one to trip up. Regardless if verifier.exe doesn't cause any problems I think it is then more likely the card has a fault. If verifier.exe does give you the BSOD, I suppose look for updated BIOS of the card and the latest Catalyst.

There are others with similar problem, multiple suggestions on this page:

AMD Support and Game Forums - Sapphire R9 290 BSOD

Here are 2 from there that are probably worth a punt:
1.
"What worked for me...
1. Uninstalled Catalyst Control Center (CCC).
2. Disabled Hardware Acceleration in Adobe Flash (mainly used for YouTube)."

2.
"I had the same problem as you for the longest time. I tried everything. New ram, fresh windows install, different drivers. Nothing worked. Then I ran Afterburner and went into Settings>General> and checked the box labelled Force Constant Voltage. This fixed it. I havn't had a BSOD in over a month, before they were happening multiple times per day. Apparently the issue is caused by the voltage changing so quickly that the motherboard enacts a failsafe to protect it from damage. Constant voltage may use more power but it works, and its not going to hurt your components."

I know how annoying these BSODs are, best of luck.

 

[lapetus]

Actually the thread below might solve your problem and is Gigabyte specific. Read to the end!!
Start here before you try my well intentioned but ill-informed "advice"

http://www.sevenforums.com/bsod-help-support/347734-bsod-while-watching-videos-online-2.html

 

[jadenred]

Firstly thankyou for your reply and help.

I ran the verifier.exe and came up with no problems, i then went about updating the bios.

It was flashing the bios when my computer bluescreened and now i have no video at all coming form the card- so im think i screwed up majorly here.

i could run the onboard graphics but then i cannot update the bios on my gpu

Stuck for ideas

Update: Managed to update the Bios and get video coming back through the GPU. Problem still persists.
In CCC there is no mention of presets for voltages which many be causing the card to fail. I also have no GIGABYTE software installed, only the CCC so it cannot be that either

 

[lapetus]

I realised I didn't link to how to use verifier.exe - excuse me if you found this and already did this, but here is the link to set it up correctly anyway :

http://www.sevenforums.com/crash-lockup-debug-how/65331-using-driver-verifier-identify-issues-drivers.html

If verifier doesn't crash on the ati drivers or others, maybe there is a conflict with onboard graphics. Maybe try disabling onboard graphics in BIOS.

The fact that it BSODed when you were flashing BIOS makes me want to RMA it. My only other advice is to remove card, reboot, uninstall everything to do with the card and try re installing now that you've flashed it. You could try re installing windows, but I wouldn't want to do that either . One way around this is to partition your drive and install win on the partition and see if card runs then (that way u can keep your current set up).

Can you take the card to a friend (or acceptable enemy) and see if it works on their computer?

 

[jadenred]

Thanks angain,

I used the verifier again to check the drivers and followed the steps listed on that article.

It is still BSOD which leads me to return the card because im ready to destroy the thing.

Im thinking if it could possibly be my PSU? Would a 700w Aerocool GT-R PSU not be able to run it?

 

[lapetus]

Fill out these calculators and see:

eXtreme Power Supply Calculator - The most trusted PSU Wattage Calculator

ASUSTeK Computer Inc. -Support- Recommended Power Supply Wattage Calculator

It's possible. Your best bet is to test it in another working system, if it's failing there definitely return it.

Sometimes I think it would be nice if it choose other colours than blue.. random colour screen of death.. lol

 

[jadenred]

I checked the overall voltage and im well within the range, i have contacted the seller of the card to return it as i think the card is literally broken. Its a shame really, i wanted to upgrade

Thank you for your support, and youre right, id much prefer a green screen of death

 

[lapetus]

No problem. I'm new to this forum but I think you should probably mark as solved, maybe a final post with "I took these steps and had no joy so I returned the card". If you get a replacement and it works maybe just update this thread confirming card was bad. This would be handy for others with similar problem.

Also I think you mean you checked overall wattage

green screen of death would be awesome... although I always thought there should be a pixel art Bill Gates just pointing and laughing at us, with a Microsoft stock counter slowly growing in the corner

 

[jadenred]

Wattage... My bad, Im even studying physics -.-
New revelation!- I entered Easy Boost (Gigabyte utility) and lowered the clock control on the Over clock section to as low as it can go, i also lowered the memory to as low it can go and now system stability is vastly improved, i am even playing Advanced Warfare as we speak!

The current OC settings are : (GPU: 309Mhz) (Memory: 150 MHz)

The question remains, what does me turning these settings down mean?

Im surprised you are not a Microsoft Employee!

 

[lapetus]

What is too funny...I'm doing math/physics..and in the past played the hell out of OGR, GRAW1/2 lol.

Reducing speeds would produce less heat, but you've checked temps and they're ok. I had a Geforce 7800 GTX in the old days that slowly over about 6 months developed artefacts (weird rays all over the screen in games,like GRAW-caused by dying memory). It was also improved on lower performance.

My point being there likely a problem on the card as adjusting settings on it alone are changing BSOD frequency. I think returning it is the way to go.

 

[jadenred]

Well, im following through the return procedure and will get my money back, probably to buy another AMD GPU

Thankyou for your support lapetus, i wish you the best.

Thanks again

 

[dwarfer66]

green screen of death would be awesome... although I always thought there should be a pixel art Bill Gates just pointing and laughing at us, with a Microsoft stock counter slowly growing in the corner [/QUOTE]

Fantastic mate!