PC only freezes when playing games for a set amount of time (30mins to 1hour)

[izt_is]

Hi guys, I need your help! PC is acting up, and while I usually know how to fix these problems, this one has really stumped me.

Problem: PC just freezes after a set amount of time when gaming (40 mins to 1 hour). The game just freezes, the mouse and keyboard are unresponsive, and the sound just freezes as well and goes on that loop. I have 2 screens, so I can see the GPU and CPU monitoring gadgets on the other screen, and they are frozen as well. I have to hard reset me computer to do anything else. Thing is, there’s no BSOD or dump file (I use blue screen viewer for this), and I can’t figure out what the problem is. This freezing is not the end of the world, but I kinda sucks when you’re engrossed in the middle of an intense multiplayer match or single player mission.

This only seems to happen when I'm playing games though, because I can leave my PC on for days on end (>2.5 days at the least) doing work, typing, browsing, reading etc. and the computer never once locks up or gives any problems.

I used to run 2x 560Ti in SLI and thought that was the problem, so I eventually gave up on it and tried running each 560Ti in single, but the freezing still happened. Anyway I sold off the 2 graphics cards to fund my 3 week old GTX 770 but this freezing still continues to happen. My PSU is not entirely old either, and I only replaced it 3 months ago to try out SLI. I have freshly installed Windows 7 once in this period when I got my new SSD, and before I started running the GTX 560Ti in single.
I’ve tried over 4 differed freshly installed nvidia drivers at this point, including the latest beta one for BF4, but this does not seem to be the issue. Even tried updating my Realtek audio drivers to the latest ones, but to no avail.
One thing to note, is that I’ve tried switching my RAM to 2 different slots, then the ones they are in now, and my PC will have some trouble booting into the bios, so I’m not sure if those slots are fault or what’s the story. I want to suspect it’s a RAM problem, so I can easily get them replaced and solve all the problems, but this doesn’t explain why the freezing happens while gaming and not if my PC is on for a few days busy with work.

PC Specs:
i7 860 Stock with Xigmatek Cooler (temps never go above 64 degrees Celsius on 100% load)
8gb DDR3 Ram XMP profile available, but running on stock
P55a UD4P motherboard
Samsung 840 Pro 128gb SSD – This is the drive the OS is on and software like browsers and MS Office, no games etc.
4 other WD HDDs, Blu-Ray drive
XFX 800W PSU (Silver)
Gigabyte GTX 770 4GB (temperatures never go above 72 degrees Celsius on 100% load)

What do I do? Thanks in advance.

 

[Ketchup]

Welcome to the forums. Pretty darn good diagnostic work if you ask me. How many sticks of RAM do you have, and have you tried anything else?

Are you overclocking the CPU?

Have you tried gaming without the other WD hard drives plugged in?

The only thing nice about your situation is that you can try different things, and you will know within an hour if that fixed your problem (if you have time for gaming that is).

 

[izt_is]

Welcome to the forums. Pretty darn good diagnostic work if you ask me. How many sticks of RAM do you have, and have you tried anything else?

Are you overclocking the CPU?

Have you tried gaming without the other WD hard drives plugged in?

The only thing nice about your situation is that you can try different things, and you will know within an hour if that fixed your problem (if you have time for gaming that is).

Hi ketchup thanks for your help.

The CPU is at stock and not overclocked.

I have only 2 sticks of RAM 4gb each. Their G.Skill Ripjaws X which are supposed to be decent and not just your run of the mill generic value ram.

Do you think it could be the other HDDs causing the problem?

My games/apps are all installed on a separate HDD and my Windows 7 OS is on an SSD. I thought it might be a bad windows install but then it doesn't explain why this freezing only happens when playing games.

I've updated all drivers/firmware/bios including nvidia, realtek audio, SSD firmware and drivers, motherboard drivers etc.

 

[AnonymouseUser]

Run memtest86+ overnight, and if you have any errors run again but with one stick at a time in different slots.

Alternatively, you could run Windows' built-in memory test, which is not as thorough, but may still detect any problem/s.

Also, my first suspicion is the motherboard. Either it's overheating or you may have bad caps.

 

[Ketchup]

I don't know that it's your hard drive, but there is some suspicion considering that your games are on a hard drive (as opposed to an SSD), and playing those games are the only time the crash occurs. I think the best way to rule that out would be to put one of the games on the SSD and play it until it crashes.

Do you do anything else that puts your computer under heavy load?

As AnonymouseUser mentioned, memtest (or just running the machine one stick at a time, as memtest isn't 100% accurate) is a good idea to see if memory is the culprit.

 

[Steltek]

Did your system exhibit the freezing issue before you replaced the power supply to move to a SLI system? The type of freezing you describe can also be the result of a power supply problem. Just because it is new doesn't mean it is working properly, despite the fact that XFX normally provides very good power supplies.

 

[sm625]

You should try downclockign your gpu. If it only happens when your gpu has been loaded for 30+ minutes then it has to be a gpu load temperature/voltage issue. Maybe the voltage needs to go higher after the temp hits 72C and its not doing that.

 

[frozentundra123456]

Does this happen in all games or just certain ones? I suspect overheating of some component, maybe the motherboard as someone else suggested. If you turn down the game settings to decrease gpu load does that reduce the problem?

I had a similar problem with my old gaming PC, except that instead of locking up, the frame rate would just slow to a crawl after about 30 min of gaming. CPU and gpu temps looked fine. I never did solve the problem, but could minimize it by turning down the graphics quality. It also seemed to only occur on certain games, which was really weird.

 

[izt_is]

Thanks for all the suggestions guys. I really doubt it's the graphics cards since it happened with 3 different cards both old and new.

I'm not running SLI anymore as well now and just am using a single brand new 770 so I really doubt that's the problem. I will be pissed if I have to downclock my card just to get it working normally, seeing as I paid so much for it. But anyway temperatures rarely go up above 75 degrees Celsius on full load so that should be more than fine. Wonder if setting it to maximum GPU usage in nvidia control panel will help maybe?

Either way I'll give mem test a shot. I've been dreading doing it as it takes forever and is the biggest waste of time, but it seems like the only option at this point ):

Any ideas on how to check if it's a PSU problem? I really hope it's not. I wonder if I'm not feeding certain components enough voltage in the bios, maybe I need to up the ram or CPU voltage.

And any ideas on how to check if it's a motherboard problem? It's a good gigabyte board with all that Ultra Durable crap, I thought it should be pretty stable. As well my ambient temperature is around 18 degress or less for the past few years with the exception for a few weeks in summer, so I doubt the MOBO could have been damaged. I also never overclocked since I first got it.

Thanks again guys.

 

[AnonymouseUser]

That motherboard has some pretty large heatsinks on it, and they are attached with typical but very weak plastic clips. If you've ever bumped them when installing components, then you may have knocked the thermal paste loose which could cause overheating of the components underneath. I have a Gigabyte AMD board that the north bridge overheated, with a similar heatsink installed, and eventually I just replaced the heatsink entirely.

I would recommend reapplying thermal paste on those heatsinks if the memtest is clean, and check your caps anyway.