Hello all,
I've been running an EVGA 980 SC for the last few months and am having some troubles identifying and solving an ongoing crash issue. I've tried to eliminate as many possibilities as I can, but haven't yet been able to rule out an issue with the card itself.
 
To start, my specs:
i7 2600K
Asus P8Z68-VPro
8GB RAM
EVGA 980 SC
Corsair HX 1000W
Win 10, clean upgrade install
 
Nothing is overclocked (aside from presumably the factory clocking on the 980 SC). As you can tell it's mostly an older system, aside from the 980 SC which I bought as the first part of an ongoing upgrade, but it's well-maintained and otherwise reliable.
 
The crash I've been getting is primarily manifesting when playing GTA V; although it's seems for all intents and purposes random and I'm unable to replicate it when trying to diagnose. Sometimes the game will crash within an hour, other times I'll go for days of heavy play at a time - thinking my most recent tweak may have worked - before it crops up again. I don't want to suggest it's game-specific, but because of the sheer unpredictability of the crash I'm unable to say whether it would occur in many others, and also because almost all of my game time has been spent on GTA of late anyway. The Admin Events list in the event viewer typically lists the following:
 
Faulting application name: GTA5.exe, version: 1.0.505.2, time stamp: 0x561d10fa
Faulting module name: GTA5.exe, version: 1.0.505.2, time stamp: 0x561d10fa
Exception code: 0xc0000005
Fault offset: 0x0000000001183e34
Faulting process ID: 0x25d8
Faulting application start time: 0x01d12bb3e21ec2cd
Faulting application path: C:\Program Files (x86)\Steam\steamapps\common\Grand Theft Auto V\GTA5.exe
Faulting module path: C:\Program Files (x86)\Steam\steamapps\common\Grand Theft Auto V\GTA5.exe
Report ID: 4347b90a-970a-436c-93ca-91e80f3611d7
Faulting package full name:
Faulting package-relative application ID:
 
I've seen references that the 0xc0000005 error relates to memory and drives, so many of my attempted fixes have looked at those, but to no avail. There's not usually any other linked error message related to the crash notification - no Warnings flagged as a display driver error in the event viewer or such, although that has happened twice in the past that I've noticed. The most recent one for example, pasted above, had no associated warning or error and that's the most common way.
 
What I've tried so far:
- Underclocking core and memory by 100mhz in Precision/Afterburner (I'm unsure if this lessens the instances of this crash, but it's my default setting for now)
- Testing aspects of the card with OC Scanner: (no artifacts detected in any test, possibly due to the unpredictability of the crash?)
- Memory testing using Win 10's scanner
- Running chkdsk to find and repair any disk errors
- Various Nvidia driver versions (always clean - no GFE, just driver and PhysX. Currently using 358.50 as that gave me week crash-free not long back), using CCleaner each time
- Reinstalling C++ and NET framework, plus Direct X (although I couldn't find a way to reinstall DX12's full pack. I'm aware the game's not using DX12 - I'm running it in DX10 mode as part of the troubleshooting)
- Reinstalling/verifying the game, altering settings to lower levels, deleting config files. I'm running the game and Steam as admin.
- I don't currently have another PC to test on, but this error didn't occur on my previous 580. It has occured on both Win7 and Win10 since moving to the 980 however.
 
What I haven't tried yet is changing the voltage on the card using Precision X, primarily because that's beyond my experience and I'm aware that it's possible to damage it doing so. I have raised the Power Target/Limit using Precision/Afterburner in conjunction with lowering the clocks but it doesn't appear to have any effect. So, in desperation to try and find a way to avoid the hassle of RMAing, especially if unnecessary, would anyone be so kind to offer advice on how to safely raise the voltage using Precision? I'm unfamiliar with the measurements needed in such a task and it's beyond my experience to tell how much voltage would be dangerous to my setup. Given that the crash can often go for a day or two before appearing I'm hoping it's fixable by tweaks rather than an RMA - assuming, of course, it is a card issue.
 
Alternatively, if anyone else has any potential solutions they'll be gratefully received! Perhaps there's some glaring issue I've missed? Thank you for reading this far, and once again in advance for any advice offered.

Thanks Sajin. It was using a clean install option rather than just the desktop process right over the top of Win 7 - although me having typed it like that probably isn't very clear or helpful; I'll amend it.
 
Thanks for the info regarding the voltage adjustments using the tools - so the main risks are increased heat, power consumption and presumably reduced card lifespan rather than an innate catastrophic risk of failure, then - using those tools, at least. Is there a generally-considered good starting point, or increment to increase by each time when changing the voltage or, perhaps, one of the preset such as Kboost and Overvoltage in Precision are a better bet than manually changing it - or do these operate independently of the voltage slider?
 
Thanks very much for your help so far!

Kboost operates independently and isn't tied to the voltage slider. Overvoltaged is tied to the voltage slider and will need be to turned on before you can adjust the voltage slider.

No problem Mntns,
 
Helping other forum members is what we like to do - good luck

Sajin
0xc0000005 usually points to bad drives/ram. Have you tried any new drives/ram?

Sadly not - I don't have any spares to try with right now. I'd seen that the error's often related to RAM, but all the memory/drive testing hasn't thrown up any errors. 
 

I also found this for ya... 
(OP was having same issue as you on GTA5. Problem was resolved by RMA'ing card.)

Ah! That's not so hopeful, then. Still, I'll give the adjustments previously mentioned a try - since it's so intermittent I'm hoping it might still be fixable. Cheers again!

Hello folks, I thought I'd update after having enough time to test solutions in the hope it helps someone else; my apologies for the delay Sajin's last post but since shortly afterwards in December and then into January I've had barely any time to game at all and the comparatively brief sessions I'd had didn't leave me confident in reporting any changes since I hadn't had any lengthy play time behind them to back up my findings.
 
Essentially, having now had enough time since mid/late-January to play more fully and test properly, I think that upping the voltage in PrecisionX seems to have had the desired effect in effectively eliminating the problem. I've been able to play GTAV - and bunch of other demanding games such as Just Cause 3 and Dying Light - without the previously reported crashes; no other fix I've tried has had the same effect for so long, so for my part I'm thinking this is the fix I've been looking for. At least, it's saved me the hassle of replacing the hardware and I've not noticed any ill effects as a result.
 
Thanks again to both Sajin and Cool for their help, and hopefully this at least helps anyone else trying to diagnose a similar problem.

Thank you again! I'm happy to call this a win given the system's recent performance; all the advice is very much appreciated. Of course, Murphy's Law meant that right after the advice was given I ended up having far less time to test the fix than I normally would...!
 
For the specifics of what I changed, I've found that setting PrecisionX with an Overvoltage of 18mv solved this particular instance with my current setup; that configuration hasn't changed since my original post. Whether this would change per setup, or if it's a unique card-specific solution, I couldn't say - but no other fix has endured for this long without the crash listed in my original post. 

No problem at all, thank you too for all your help with it!