Windows 7 crash randomly during gaming

[ifailedhard]

Ever since I played the Division, my comp gave me two BSODs, one ingame and the other just browsing on my comp after playing the game which seem to result from one of my .dll files being corrupted. I did a clean install after that and it solved my BSOD issues, but now my PC would do one of these two things when the HDD light stays on during gaming:

1. Windows 7 would crash on my SSD, but I can still play games and alt tab to other programs that are opened on my internal hard drive. So far, this happened to me like 4 times already the past week.
2. My entire comp would freeze and make a loud buzzing noise with a scrambled pixelated screen after a while. This happened only once yesterday.

I know for certain that it isn't a temp issue as both my GPU and CPU are under 60 degrees celsius. I've ran memtest, chkdsk, clean installed Win7 and updated my drivers and it still happens. I've also check other threads with similar issues which seems to be pointing to motherboard or PSU failure, though I'm unsure if its just my HDD dying or software issues.

Any ideas?

Specs:
AMD FX 8350 8 core processor 3.5 Ghz
Sapphire Dual-X HD 7970 3GB
16 GB ram
1.1k watt PSU (Forgot brand name)
ASUS M5A97 le R2.0 motherboard
120 GB Samsung SSD
1.4 GB Seagate HDD

Apologies if I can't give specifics on certain specs.

 

[Barty1884]

What are you using to confirm your CPU & GPU temps? With an AMD chip, you should be using AMD Overdrive...

Without knowing the brand of your PSU, it's difficult to say. You shouldnt be able to tab out of a crash if the PSU is the issue, but that may be a seperate issue.

Given all the steps you've taken, it definitely sounds hardware related, with the PSU being a likely culprit.....impossible to tell without brand & model.

 

[ifailedhard]

What are you using to confirm your CPU & GPU temps? With an AMD chip, you should be using AMD Overdrive...

Without knowing the brand of your PSU, it's difficult to say. You shouldnt be able to tab out of a crash if the PSU is the issue, but that may be a seperate issue.

Given all the steps you've taken, it definitely sounds hardware related, with the PSU being a likely culprit.....impossible to tell without brand & model.

I'm using MSI Afterburner's On-screen Display to check my temps.

As for PSU, I managed to find the brand:
KENTEK 1000 Watt 1000W 140mm Fan ATX Power Supply 12V 2.3 EPS12V 2.92 Quad SLI-ready PCI-Express SATA 20/24 PIN Intel AMD

If its definitely the PSU that is causing my problem, should I go after a 750W for my build?

 

[Barty1884]

Check your temps via AMD Overdrive first though - you might simply be overheating, at which point it's a (relatively) easy fix, more often that not. And would be cheaper than replacing your PSU.

Remember, AMD OD doesn't measure temps the same way. It's essentially a countdown to zero. If the Thermal Margin hits 10'C or less at load, then you're likely experiencing throttling & then overheating.

Regarding the PSU, I'd never heard of Kentek but a quick google search doesn't really result in good findings - they seem to range from "they exploded & took all my components with it" to "it just doesn't work". Either way, I'd recommend you replace the PSU with a quality unit, Tier 1 or 2 from here:
http://www.tomshardware.com/forum/id-2547993/psu-tier-list.html

 

[ifailedhard]

So far, OD shows Thermal Margin at 27'C at load.
Also, thanks for the recommendation link. I'll see if replacing my PSU will solve my issues.

 

[Barty1884]

How long were you stress-testing it? I'm assuming you're using the stock cooler, so 27'C isn't bad actually. Certainly not low enough it would be throttling. You didn't have any freeze-up issues while testing it?

 

[ifailedhard]

Not using a stock cooler. I'm using ZALMAN CNPS9500A-LED 92mm 2 Ball.
Just did another stability test. Within 3 mins, my Thermal Margin went down to 13'C under heavy load. After stopping the test, Windows disabled Aero. No freeze up issues either.

 

[Barty1884]

27'C with an aftermarket cooler is borderline where I'd like to see it. 13'C may well be getting into throttling stage.

Disabling Aero usually results from utilizing too much RAM....

 

[ifailedhard]

So does that mean its a CPU issue?
Also, I dig around the Event viewer and notice this:

Atapi Event ID 11: The driver detected a controller error on \Device\Ide\IdePort1.

Is that related to my system freezing randomly during gaming? I'm seeing this error popup every 3-4 hours or so.

 

[Barty1884]

Unlikely to be CPU related. Memory leak from RAM - possibly.

Your error though, what is in Port 1? If it's your main drive, that's quite possibly the root of all this.

 

[ifailedhard]

Its my SSD assuming this represents port 1: \Device\Ide\IdeDeviceP1T0L0-1

I had checked to see if it was loose, but it wasn't. I also moved it to a different connector away from GPU since it was very close to it.

 

[Barty1884]

You're getting a controller error - it's unlikely to stem from a cable being loose.

First off, are you sure Port #1 is your SSD? Remember, SATA ports are generally numbers 0,1,2,3 (opposed to 1,2,3,4).

The error could be a result of a failing HDD or SSD, but even worse can be a the actual controller on/in the motherboard is failing.

Update all your drivers, hopefully that fixes the issue.

If it doesn't, run a health check on the drive in question (whichever is currently in Port #1). If it passes, that really narrows you down to the controller on/in the motherboard.

If that's the case, hopefully you're still within warranty on the board - if that's the case, RMA it and get a new one.

If you're not, you have the option(s) of buying a replacement motherboard (and likely a replacement OS) or you can add a PCIe SATA controller card and forget about the onboard SATA ports:
http://www.newegg.com/Product/Product.aspx?Item=N82E16815158365&nm_mc=AFC-C8Junction&cm_mmc=AFC-C8Junction-VigLink2-_-na-_-na-_-na&cm_sp=&AID=10446076&PID=3821802&SID=ily7ebak5n000a1700053

Of course, that card would then take up a PCIe slot.....but if you're not using them.....it would be cheaper than a new board (even moreso if you have OEM Windows as you'd have to buy that again too).

 

[ifailedhard]

I don't have port 0 labeled on my motherboard, but my HDD is on port 1. As for health check, how do I do that?

 

[Barty1884]

Not all motherboards are labled starting at zero. Some do start at 1.

Western Digital & Seagate offer their own software to health check HDDs.

WD: http://support.wdc.com/downloads.aspx Data Lifeguard

Seagate: http://www.seagate.com/ca/en/support/downloads/seatools/ SeaTools for Windows

 

[ifailedhard]

Did a extended health test on my HDD and it resulted it being stuck at a certain sector for almost 2 hours and it was unable to repair the bad sectors. So it seems like my HDD might be the root cause of the crashes. Though I'm unsure how I was sometimes able to still play the game and alt tab when Win7 crashed.
I also ran a quick test and extended test on my SSD and it both passed the test.

Also, I think I might have mistaken with my IDE port. It seems that IDE port 2 might be the HDD since I dig a bit deeper into Event log and I had these errors around the same time as I got these crashes. Though that means my motherboard might also be failing since it reported IDE0 and 1 errors.

Thanks for your help. Seems like I'm gonna have to replace my motherboard, HDD and PSU.

Edit:
I noticed that AMD released a driver hotfix yesterday and I saw a comment on reddit mentioning it fixes the crash/hanging. So I downloaded it and test it on the games that have been causing problems for the past few days. I didn't experience any crashing/freezing within the expected time frame and I manage to play my games for 2 hours without issues. I guess the root cause of my issues was the GPU driver.