Computer hangs while playing games

[lamoo]

Hey!

Not sure where to start or if this is the correct forum, but I'll wing it. While playing games (whether performance intensive or not) my computer will hang with no crash notice, audio pausing and all. My mouse will not freeze and is free to move (while everything else is frozen, even on my second monitor) but when I click that too freezes. I've found the only way to escape this 'hang' is to manually turn off the power by either holding down the power button on the case or pulling power altogether. I've checked my temps via HW Monitor and am ready and able to post my results if ya'll require it. I've also seen some other posts request a Speccy or Event Viewer log but I'm unsure how to do either of those things. Although I suspect it to be a hardware issue due to my rather low quality specs (which I'll toss at the bottom of this post), it's only about a year and a half to two years old so I'd be surprised if any serious degradation has occurred since purchase. If it matters, it was bought pre-built online from CyberPowerPC and I've had several issues with hangs previously but it seems they've fixed themselves over time.

I'll dump my system specs below, let me know if I left anything out and I'll grab it for ya:

CPU : AMD FX-4300
MOTHERBOARD : Gigabyte GA-78LMT-S2
STORAGE : Toshiba 1TB 3.5" 7200RPM
GPU : Radeon R7 240
POWER SUPPLY : Sparkle 350W SFX
OPTICAL DRIVE : LG GH24NSC0
MONITOR : AOC E2429SWHE 23.6"

Thanks, hope to get an informative response soon so I can get back to my EU4 campaign

 

[Cycloid Torus]

I suspect that the system may be a bit older - the GPU came on the market in 2013 and the CPU is 2012 vintage. I've had good success with Sparkle and the 350w should be sufficient - so if it works, continue to use it.

Since your system may be older than 1.5 years, it maybe time for a thorough cleaning and renewal of the thermal paste (TIM) for both CPU and GPU. Your best bet is to watch several YouTube videos before undertaking this. Doing this also addresses the need to re-seat connections which may have loosened up or picked up a bit of corrosion. It would be helpful if you can identify the motherboard and obtain a copy of the manual.

Third, there might be a driver issue. You didn't mention your OS - you might want to look for updates for the video card - but be careful to read the small print. https://www.amd.com/en/support/kb/faq/gpu-driver-autodetect

Should all the above fail to resolve issue, type 'Reliability History' in search and look for errors, click the links, read the footnotes.

 

[lamoo]

I suspect that the system may be a bit older - the GPU came on the market in 2013 and the CPU is 2012 vintage. I've had good success with Sparkle and the 350w should be sufficient - so if it works, continue to use it.

Since your system may be older than 1.5 years, it maybe time for a thorough cleaning and renewal of the thermal paste (TIM) for both CPU and GPU. Your best bet is to watch several YouTube videos before undertaking this. Doing this also addresses the need to re-seat connections which may have loosened up or picked up a bit of corrosion. It would be helpful if you can identify the motherboard and obtain a copy of the manual.

Third, there might be a driver issue. You didn't mention your OS - you might want to look for updates for the video card - but be careful to read the small print. https://www.amd.com/en/support/kb/faq/gpu-driver-autodetect

Should all the above fail to resolve issue, type 'Reliability History' in search and look for errors, click the links, read the footnotes.

Thanks for the response! So yeah, the parts are a tad bit older than I originally thought (that's what I get for buying a pre-built I suppose). I just recently (within the past 6 months, maybe? right around that mark) put in an aftermarket CPU cooler since my temps were dangerously high and, as you suggested, the thermal paste was thoroughly battered. Long story short, the CPU cooler I have in now is great and keeps the temps down and the thermal paste should be brand spanking new - unless the cooler manufacturer stiffed me on it (didn't bother buying aftermarket paste considering it came with the cooler).

I wouldn't know where to begin on checking for corrosion or re-seating connections, any chance you could give a brief check list? I'm comfortable working around the guts of the PC, just want to make sure I'm checking the right things.

I'm currently messing about with drivers to see if it's actually a software issue, so I'll get back to you on that when it's all done. Thanks again for the quick reply!

 

[lamoo]

-snip-

On another note, I've looked through the Reliability Monitor windows application that you suggested and come up with a few critical events. I'll dump some screencaps for you here:
https://prnt.sc/m1bpya (EU4 crashing, not sure if there's anything special with this one but it was accompanied by a crash as described in the original post)
https://prnt.sc/m1bqyu (This one particularly worries me as it points towards a hardware issue, but that's just judging by the name)
https://prnt.sc/m1br4p (To be honest, no clue what this is but it was accompanied with a crash so figured I would include it)

There's tons of others, but it would take a while to post all the screencaps and most of them are just 'Windows was improperly shut down' with no extra information (which was when I pulled the plug on the PC). Oddly enough, no other event accompanies these improper shutdown events except for the ones described and maybe an outlier or two I missed.

 

[MaikuTech]

If you're tempatures were looking dangerously high, you should have rebooted.
Go into the bios and recheck the cpu tempature.
Any amd cpu reaching past 60c is not good, that gigabyte will shut it down to prevent damages.
If you want put money into a dead beat horse, than I won't stop you but its not worth it in the end. ¯\_(ツ)_/¯

http://www.tomshardware.com/answers/id-3412380/overheating-idle-8320.html

https://www.newegg.com/Product/Product.aspx?Item=9SIAJ8X8MZ6432&Description=cpu heatsink fan&cm_re=cpu_heatsink_fan-_-35-186-092-_-Product
+
https://www.newegg.com/Product/Product.aspx?Item=N82E16835186020&Description=artic cooling mx-2&cm_re=artic_cooling_mx-2-_-35-186-020-_-Product

This should "undangerously" fix that heat problem.
Make sure you know how to clean the cpu of old crusty thermal paste slop first.

 

[lamoo]

-snip-

Oh, this was a while ago when my temps were getting a bit hot. I've downloaded HW Monitor and Core Temp and it doesn't seem like overheating is the issue, which narrows it down to some sort of software malfunction (messed up drivers et cetera) or hardware - but that's the issue, I don't have the expertise to narrow it down past this. Just for good measure though, I'll toss up my temps below:
http://prntscr.com/m1cgu4 (These were recorded just now while I was playing Town of Salem. Although it's a rather low intensive game, the temps don't vary too drastically even when I'm playing more intensive games.)

Edit: Now that I look at it, my CPU actually runs quite cool at a mere 15-20 C* max while doing some more taxing tasks.

 

[Cycloid Torus]

Bad module - chkdsk your hdd first
hardware issue? - more likely driver
EU4 - more likely graphics driver

Reseat is just remove gently, examine, re-insert, confirm fully inserted
See if you can find motherboard manual - I do a disassembly/reassembly (but then I have plenty of time and like details). Normal folks might only bother with GPU.

While GPU is out, you might want to change TIM on that heatsink. GPU service is different as there are thermal pads - check several YouTube videos before doing this.(like

)

 

[lamoo]

-snip-

The chkdsk came out clean with no errors found, thankfully. In regards to the 'hardware issue? - more likely driver' comment, are you referring to the graphics driver that you also mention in your comment regarding the EU4 event or a different driver?

I've tried uninstalling my graphics drivers and reinstalling fresh through the AMD auto-detect tool, is there anything else I should do to make sure my drivers are all good? I uninstalled them via device manager, is that also the correct way to deal with drivers or should I download a separate tool to deal with it? (I see several being advertised, just not sure if they're legit or not - or even if it's something I can easily do myself without the use of another tool)

No matter, once I make sure the drivers are all set I'm planning on grabbing some compressed air and cleaning out my computer as well as reseating everything - may as well do everything in one go so there's no need for separate visits to the guts of my PC.

 

[Cycloid Torus]

My guess is the GPU driver might have a minor corruption or need update. Microsoft tries to push stuff in if the dates are more recent, even if it is not the best. I stick to manufacturer's WHQL drivers only. Download new driver, then run DDU to clean out the old... https://www.techspot.com/drivers/driver/file/information/17830/

If this does not cure issue, then next might be overheating GPU, but I would want to see it in HWMonitor before I messed around. You do have a strange RPM on the GPU fan. Can you run a hot game and the Monitor in separate windows? Can you try other sensor reading utility? See:
http://www.tomshardware.com/answers/id-3781478/radeon-6970-fan-spinning-high-rpm.html

Very glad to hear chkdsk was clean - do backup anyway.

After all that, you can try some demanding games. Look for patterns in any failures. Go to Reliability History for more depth on any issues.

 

[lamoo]

-snip-

If this does not cure issue, then next might be overheating GPU, but I would want to see it in HWMonitor before I messed around. You do have a strange RPM on the GPU fan. Can you run a hot game and the Monitor in separate windows? Can you try other sensor reading utility? See:
http://www.tomshardware.com/answers/id-3781478/radeon-6970-fan-spinning-high-rpm.html

-snip-

After all that, you can try some demanding games. Look for patterns in any failures. Go to Reliability History for more depth on any issues.

Alright, so I've just gone through the process of running DDU to clean out all my visual drivers, and using the AMD auto-detect to reinstall everything. Problem being that DDU also cleared out my HD audio drivers and AMD auto-detect doesn't bother checking for your audio drivers (or the lack thereof) in case it needs to reinstall. I believe (keyword: believe) that I was previously running Realtek but I honestly don't remember as audio drivers were installed before I ever got the PC (pre-built). Should I try my luck and re-download Realtek or is there a set of AMD HD audio drivers that I'm simply not seeing on their website?

In regards to overheating, my GPU is the one component that does (occasionally) run a bit hot as I've done absolutely no aftermarket cooling to it. I do agree that my RPM seems insane, and is similar to the link that you posted - though that issue seemed to be solved by reinstalling drivers as it came down to a wacky temp monitor. I'm going to do some troubleshooting with HW Monitor on my second monitor and see if I can get some screencaps of it (under load and idle) to see if that helps.

 

[lamoo]

Alright, so I've just gone through the process of running DDU to clean out all my visual drivers, and using the AMD auto-detect to reinstall everything. Problem being that DDU also cleared out my HD audio drivers and AMD auto-detect doesn't bother checking for your audio drivers (or the lack thereof) in case it needs to reinstall. I believe (keyword: believe) that I was previously running Realtek but I honestly don't remember as audio drivers were installed before I ever got the PC (pre-built). Should I try my luck and re-download Realtek or is there a set of AMD HD audio drivers that I'm simply not seeing on their website?

-gpu info snip-

Alright, leaving this here to help anybody else who runs into this issue. I tried reinstalling several different installations of Realtek from both my motherboard manufacturer's site and, out of desperation, several other random sites that claimed to host all kinds of different drivers (drivereasy.com is an example). NONE OF THEM WORKED.

The correct solution to reinstall your audio drivers, if you cannot find a proper installation of them from your motherboard manufacturer's site, is to go to Device Manager > System Devices > Audio Composite Bus (or something similar, it should have a blue monitor logo WITH a yellow alert triangle next to it, that part's important) > Update drivers. No other website was pointing me in this direction, so I'm really glad I was able to figure it out.


In other news, now that I've figured out my audio driver situation, I'm going to work on getting you those HW Monitor screencaps @Cycloid Torus!

 

[lamoo]

-snip-

Okay, finally got those HW Monitor results for ya!

Idle:
https://prnt.sc/m1me6w
https://prnt.sc/m1menr (As outlined in red, and as previously mentioned, GPU fan RPM is truly odd. Keep in mind, this is after all drivers have been uninstalled and reinstalled so very little chance of software being at fault.)

Under Load (I chose War Thunder since that's relatively demanding for my system but not enough to cause the crash for a round or two):
https://prnt.sc/m1m22v
https://prnt.sc/m1mevl (Like I said earlier, very odd.)

I do have to add that there were some outliers that showed my CPU clocking in at an insane 120+ Celsius (yes, Celsius) for a second or two before dropping back to a normal temp. My CPU, even before I added an aftermarket cooler to it, would never reach this temp so I'm blaming that one on a failing temperature monitor.

 

[Cycloid Torus]

"The correct solution to reinstall your audio drivers, if you cannot find a proper installation of them from your motherboard manufacturer's site" - an alternative is to remove the device from DM and reboot - Windows will then find it and install what it thinks best. Cleaning out 'hidden' in Device Manager can also help.

"120+ Celsius (yes, Celsius) for a second or two before dropping back to a normal temp" - loose heatsink??

What happens with running War Thunder just before the system freezes? Sensor spikes anywhere? System could be shutting down because sensor says it should.

Facing the bigger issue - wild RPM in GPU fan - insane temps in CPU - if drivers cleanup and reinstall, general cleanup, new TIM & reseating (including power connections and RAM) does not cure problem then I begin to feel that we may be looking at general degradation of CPU &/or motherboard

Warranty?

 

[Cycloid Torus]

Afterthought... just noticed Low Temp on FX4300 was Zero?!?! Are you using this in an outbuilding near the arctic? Wide range in temps can result in strange things - including cracked solder joints, condensation, etc.

 

[lamoo]

Afterthought... just noticed Low Temp on FX4300 was Zero?!?! Are you using this in an outbuilding near the arctic? Wide range in temps can result in strange things - including cracked solder joints, condensation, etc.

Haha, that was my thought as well - seemed odd to me. It may very well be because I'm using a CPU cooler made for far more heat-generating CPUs (aka better ones) than mine. I have no information left hanging about regarding the purchase of the PC on hand but I will search about, though considering I made aftermarket modifications (the CPU cooler) it's quite possible they won't take it back either way. I do want to note that opening up the computer, while I'm willing to do it, is a last resort for me so that's why it may seem like I'm avoiding giving you results on that end.

I've gotten away with playing EU4 for about 4 or so hours now (intermittent breaks between) with no system freeze, so I'm gonna try and watch Netflix on my second monitor at 1080p to try and trigger a freeze. Though I now see I have a few problems elsewhere, I think the freezing may have very well been caused by shoddy display drivers. May be getting ahead of myself there though, so give me about an hour to try and conjure up a crash and I'll get back to ya.