Crashing during Games..again

[El Diablo]

I'm back again with blue screens and black screens this time.

At the end of November I installed a Gigabyte 670 with a little trouble, cleaned the old drivers out and it worked fine after that for about 2 weeks.
Now here we are with the same kind of issues I had with my old 560ti 8 months ago and again about 2 months ago.

I've also had several black screens where the computer restarts itself afterwards, but mostly just blue screens, no beep codes.
Most of the crashes have been from Battlefield 3, although I did get one from Garrys mod and I know that Arkham City would probably crash too since it seems to be sensitive to these things.

Both rkill and tdss killer have been run in safe mode and a normal boot along with my Eset antivirus, no infections.
I've also run chkdsk several times which did find corrupted files once. I tried cleaning and reinstalling my graphics drivers several times with no results, it just seems to work when it wants to.

On occasion I can play Battlefield just fine, some games haven't had any trouble yet either.
Other fixes I've tried are removing a USB 3.0 PCI bracket that came installed in the computer, downclocking my gpu just in case the card was having trouble with the factory overclock none of which yielded results.

Looking at the crashes in BlueScreenView just points to core windows files like Ntoskrnl, one also suggested Ntfs so I suppose the hard drives could be the issue.
I've never had memtest or seatools tell me there were any problems out of all the times I've run them, but it has been awhile.

So I'm stumped on what to try next, any suggestions?

Thanks for taking the time to read through my post if you did.

Regards

EDIT: Added another dump file from Killing Floor.

 

[El Diablo]

Here's a small update:

I ran Memtest for 17 passes over a 14 hour period with 0 errors,
also the last time I tried to play a game I received something like this;

The instruction at ************ referenced memory at ************.
The required data was not placed into memory because of an I/O error status of *************.

I didn't catch what numbers were listed as explorer crashed soon after requiring me to hard reset.

Before running memtest I stessed my gpu with Furmark, running the 15min burn in test with no problems of overheating or overloading the psu.

I'll run seatools later tonight.

 

[El Diablo]

I haven't got around to running Seatools yet but I think I might have found the problem.

One of the connections between one hard drive and one sata cable is kinda loose,
so I tried swapping out the cables with brand new sata cables.
I played several games successfully without crashing, but there is still a little wiggle in the cable.
I'm positive this is the hard drive itself and not the cable (manufacturing error possibly) as the other drive is just fine, almost no movement.

Here's where it gets suspicious, I wiggled the cable a little bit to see how loose it was with the new one before running the next game,
20 minutes in or so it did end up crashing while trying to load a new area. I restarted and checked again with Battlefield 3 which was the biggest problem before,
and although it plays longer than it used to it still crashes just like before.

It's definately possible I'm completely wrong with my allegation, but the evidence is there.

It doesn't help that windows is giving me this message:
Windows was temporarily unable to read your hard drive. We don't know the exact cause of the problem.
In most cases, this type of condition is momentary and doesn't indicate a serious problem, but sometimes it means that a hard drive is failing.
I don't know if that's just from the blue screens or what's causing the blue screens but I'll guess the latter.

I'm attaching two more dump files from earlier today.

 

[El Diablo]

Ok, so the computer finally blue screened outside of a game.

I left it for a few hours with just Steam, iTunes and Internet Explorer open, when I returned it was sitting at a blue screen. I'll attach the minidump.

I suppose you could call it progress if you wanted to but I just see it as more evidence for a failing hard drive.

I'm still hoping that someone will eventually read through my minidumps just to be extra sure that it's not a driver.

 

[Core]

Ntoskrnl just refers to the kernel and will always be present in crash dumps.

I think you've done a good job isolating the issue and I really think you're on the right track.

Can you pull some SMART figures from your hard drive? They'd be a good indication of whether it's dying or not, although I'd bet that it is.

Oh and since you've got both HDDs in a raid 0 array, I'd back up their data since the failure of one will mean total loss of everything.

I'd also throw in a question: did these crash issues begin around the time you striped your drives in a raid array? I am only asking because I've experienced system instability in a raid 0 configuration before; whether it was a bad raid controller or simply the fact that the drives weren't identical I cannot be sure of. Just something to chew on in case the SMART data doesn't indicate mechanical failure.

 

[El Diablo]

Hello, thanks for relpying to my post.

I'm having trouble getting SMART data as Speedfan doesn't see any hard drives, SMART is turned on in the BIOS too.

The drives have been in raid 0 since the computer was built, I've already contacted warranty about a new drive and will be getting one in a week or two.

My important data is backed up to an external and I've refrained from downloading anything else that I would want or need to back up so all is good there.

If replacing the one drive with the loose sata plug doesn't work, the whole computer is still under warranty for 10 months so I'll just have to send it in if that's the case.
It really is down to the one hard drive (maybe both) or the raid controller I think, a clean windows install might help as well as that is inevitable once I get the replacement.

Thanks again for your input.

Finally got around to running Seatools, both drives passed the long and short tests just fine although I still think that the loose sata port on the drive is the problem.