Question PC freezing with buzzing noise while playing games.

[nooblet0]

Ok so this is going to be a long one.
About 3 days ago I experienced a kmode_exception_not_handled bsod while playing apex legends.This occured within 2 minutes of dropping into the match area.After the system rebooted I was able to play 3-4 matches without any issues(1+ hour).
The next day I load up apex and this time 2 minutes after dropping in my pc freezes up(on the last frame) and I hear a buzzing noise from my speakers.I had to restart the pc with the restart button on my cabinet.On rebooting I tested with dmc5 and the game froze up with a buzzing noise on a loading screen(roughly 2 mins into the game)once agiain(kb and mouse stop working but cabinet power light stays on).After restarting again I tested with metro exodus.same outcome(game freezes up with in 2 mins during a cutscene).At this point I'm thinking its probably my gpu.So I ran a furmark stress test for about 5 mins without any issues.I start apex legends again and guess what?same crash within 2 minutes.During this time there were a couple of instances where even after the restart the pc froze up as soon as it got to the desktop screen and the buzzing noise started again.Now my gpu(rx 580)is 1 month old and has been performing as expected so far.I haven't noticed anything wrong with in game temps either(it doesn't go beyond 75 degrees).There are no frame drops or artifacting before the freezes occur.After about 5 or 6 reboots I fire up apex again and suddenly it doesn't crash anymore.I played 2 games without any issues.After that I played the prologue section of dmc5 again without any crashes.Then I played some pes18 for a total of 3+ hours.I figured whatever it was had fixed itself.
So yesterday I booted up my pc.i ran a furmark test for about 20 minutes : no crashes.I load up apex and again the game freezes up within a minute (w/ buzzing noise)I restart and after 3 minutes into the desktop screen my pc gets a bsod error dpc_watchdog_violation.After this I haven't played anything else yet.
After booting up my pc today I ran furmark for about 40 minutes with no crashes.I have yet to play test a game.Any ideas about what's going on..


Link for folder containing the dmp file dxdiag and other stuff:
https://drive.google.com/open?id=1t-LSJO1ns9IRVR3OiF0hygsWG3vIZcwY

I'll try and post a recording of the buzzing sound the next time that happens.

 

[Colif]

atikmdag.sys is part of radeon drivers.

I will get @gardenman to look at dump files and we see what they show apart from that.

buzzing can just be the sound of drivers dying out.

 

[nooblet0]

atikmdag.sys is part of radeon drivers.

I will get @gardenman to look at dump files and we see what they show apart from that.

buzzing can just be the sound of drivers dying out.

Just stress tested the gpu again.If something was wrong with drivers wouldn't it crash during the stress test.Nevertheless I'll do clean install of the drivers using ddu asap.
Link for a recording of the sound: https://drive.google.com/file/d/1tOOm9WuFQo3hSfe07vpBn9tjiWsVoEEU/view?usp=sharing
I use headphones,hence the shit quality.

 

[gardenman]

Hi, I ran the dump file through the debugger and got the following information: https://pste.eu/p/eN8B.html

File information:	041119-35234-01.dmp (Apr 11 2019 - 09:52:08)
Bugcheck:	KMODE_EXCEPTION_NOT_HANDLED (1E)
Probably caused by:	memory_corruption (Process: r5apex.exe)
Uptime:	1 Day(s), 17 Hour(s), 50 Min(s), and 58 Sec(s)

Comment: The overclocking driver "AODDriver2.sys" was found on your system. (AMD Overdrive)

Motherboard: Gigabyte 970A-DS3P
BIOS you currently have installed: Version FB

This information can be used by others to help you. I can't help you with this. Someone else will post with more information. Please wait for additional answers. Good luck.

 

[Colif]

your bios isn't on the list for motherboard, it had FA & FC but not the one you have... that is different. Either way, there is FD now, you appear to have bios that came with board. You have revision 2 of the board (easy to tell, completely different range of bios versions)

https://www.gigabyte.com/au/Motherboard/GA-970A-DS3P-rev-2x#support-dl-bios

Updating bios might help but as its from 2014, it might not change anything.

wish I knew what MEMORY_CORRUPTION_LARGE meant, is it a bug of the debugger which doesn't like how win 10 does memory compression or something else? Its on almost every BSOD I see.

Overdrive is from 2013, see if there is a newer version - it might be okay, I don't see it often. Most pre June 2015 drivers aren't win 10 drivers but its possible AMD have re used a Win 8 driver.

 

[nooblet0]

So it's been a while and I figured it's time for an update since I see a lot of threads on here reporting similar issues and most of them remain unresolved. Sadly I have not been able to fix the issue in it's entirety nor have I been able to conclude exactly which component/driver is faulty.However I have managed to prevent the crashes from occurring for long periods of time(2-3 weeks). I was able to finish RE2 and HitMan2 during this time with only minor crashes. There was one notable crash among these when the pc froze up for a few moments while playing RE2(w/ buzzing noise) and then resumed functioning normally without me doing anything.

Temporary solution:
What seems to fix the issue for me is usually re inserting my single ram stick. If that doesn't work Reseating the gpu/disconnecting the power cable and reconnecting it to the gpu also seems to do the trick and the game will usually not crash after rebooting. Also crashes no longer occur after 30 seconds in game like it used to before. Nowadays it's more random.

I've had days when I've been able to play for 5+ hours without crashes, days when the game has crashed within minutes and days when the game has crashed after an hour or so. The timing of the crash seems to be independent of GPU load as crashes will occur in loading screens/team selection window. When I say "crash" I'm of course referring to the freezing w/ buzz.

As reported before I haven't noticed any graphical bugs/artifacts in games or stress tests yet.

Changes made/parts tested after last post:
Given where I live it is extremely difficult to find another rig w/ specs that will allow me to individually test my components however I was able to test my GPU for about 4 hours in a different rig but no errors were detected in that time. I was also able to test my ram stick for about an hour or so(not paired w/ my gpu) and guess what? no freezes but I can't really conclude anything from that as I've had longer crash free gaming sessions on my own rig. The nature of the issue and the frequency at which it occurs makes it all the more difficult to pinpoint exactly what is causing it.

Thinking that the hdd might be the issue I purchased an ssd(samsung evo 860 latest firmware) and reinstalled windows(legit) on it.
The crashes occur on games installed on the ssd too(most recent crash was 2 days ago while playing control, pc froze up after about 40 minutes in game ). I've also uninstalled and re installed gpu drivers multiple times (driver verson doesn't seem to make a difference).

As colif pointed out my bios version was FB which is not listed(I suspect that it was the backup bios which gets loaded on dual bios mobos on bios corruption). I updated my bios to the latest version and that hasn't fixed the problem either.

Weird Occurrences:
Remember how I said that I haven't had any artifacts; well that wasn't entirely true. About a month ago I noticed red dots/flickering lines on the screen while playing a game(Middle Earth Shadow of War) . The lines were still present after I closed the program . This was followed by the display becoming dimmed(the monitor didn't exactly flicker but it kinda went dark) and then it flickered a few times before everything became normal. I changed the molex power connector and the issue hasn't occurred again.

I've pretty much given up hope on ever fixing this thing but any help is appreciated.

 

[Colif]

I can give you a few tests you can run on PC to narrow it down?

CPU - https://www.mersenne.org/download/
Ram - Try running memtest86 on each of your ram sticks, one stick at a time, up to 4 passes. Only error count you want is 0, any higher could be cause of the BSOD. Remove/replace ram sticks with errors.
GPU = well, you tested it already but if you want more
don't run at same time
https://geeks3d.com/furmark/

https://benchmark.unigine.com/heaven
HDD - https://www.hdsentinel.com/hard_disk_sentinel_trial.php (would be strange for it to be hdd)

PSU
the paper clip method - https://forums.tomshardware.com/threads/what-is-the-paperclip-method-of-testing-a-psu.1336402/
or multimeter,
or in the BIOS to check the +3.3V, +5V, and +12V. - https://www.lifewire.com/power-supply-voltage-tolerances-2624583

Motherboard - there are no tests for motherboarrd. You check everything else and only blame it when everything else has been confirmed as good.

Are you still getting BSOD?

 

[nooblet0]

Hi,
No BSODs after cleaning my componenets after the first time this happened. No application errors show up on event viewer either.

I've also ran furmark and unigine heaven benchmarks multiple times with varied results. PC has never frozen up during furmark tests(I've tested for well over an hour)

PC has frozen up during heaven benchmark twice about five minutes into the test.
Relevant thread : https://forums.tomshardware.com/thr...hmark-but-stable-during-furmark-test.3471751/

I'll do the memtest as soon as I can.


Update: Ok so remember how I mentioned I got random artifacting months ago in a game. Well they happened again about half an hour ago while watching a video , the artifacting lasted for about 30 seconds or so(I was unable to get a screenshot of it). I've been trying in vain to re create them by running some gpu heavy applications but no luck so far.
Will update again when I've checked my ram for errors but this is increasingly looking like a gpu issue(?)

 

[nooblet0]

Update: Ran memtest . 4 passes no errors. Crashing is still erratic. On two different occasions system froze while watching a stream.

Edit: This happens about 2 minutes into prime95.Copying from result.txt . I ran a blend test on 8 threads.

[Thu Oct 03 10:37:16 2019]
FATAL ERROR: Rounding was 8388608.012, expected less than 0.4
Hardware failure detected, consult stress.txt file.
FATAL ERROR: Rounding was 0.4853713801, expected less than 0.4
Hardware failure detected, consult stress.txt file.

 

[Colif]

Thu Oct 03 10:37:16 2019]
FATAL ERROR: Rounding was 8388608.012, expected less than 0.4
Hardware failure detected, consult stress.txt file.

did you look in the stress.txt file? its in the install folder

it seems the advice is to increase voltage - https://forums.tomshardware.com/threads/prime95-cpu-test-error-do-i-change-voltage.1705478/

 

[nooblet0]

Update:
I haven't modified the voltage settings yet. Decided to ran some more tests. As things stand it's looking like a CPU issue. Ran prime95 blend test two more times both occasions resulted in errors in 2 workers. Ran Smallest FFTs(this is supposed to stress test the cpu only?) and PC froze up except there was no buzzing noise(because nothing was playing at the time and although the audio drivers probably crashed, there was no audio loop).
Some other interesting observation:
Whenever the crashes occur almost every peripheral component stops functioning(I assume the USB drivers crash) however num lock light on the kb stays on until I press a key(no idea why this happens). My xb360 controller(upper right quadrant) light also stays on until I disconnect and reconnect it. All cabinet/cpu/psu/gpu fans continue to operate normally. GPU fans gradually slows down as the temperature goes down and eventually stops spinning. The monitor screen never changes at all , it remains frozen on the same frame(I left it like that for over 2 hours) The crashes seem to be occurring more frequently now.