Help with Freezing & Restarting PC

[droscoe]

My PC randomly freezes and restarts. The chances of it freezing/restarting seems to increase when I'm putting a load on my system (such as gaming). Only a few minutes of heavy gaming and it restarts.

On the flip side, when I'm doing nothing at all (sitting on desktop) or doing minimal activity (browsing the internet) it will only SOMETIMES freeze/restart.

Allow me to be more specific - It FREEZES for a full second and then restarts. If there's any sound/audio playing at that moment, the sound does that horrible stuttering thing. Obviously Windows recognizes these events as Kernal-Power Error 41 (unexpected restarts).

Steps I've Taken So Far

• I have no visible damage to system components
• MemTest had no errors after passes
• I've tested individual memory modules, no change
• I've monitored temperatures for CPU, Memory, and GPU and nothing was out of the ordinary.
• Drivers, BIOS, Windows 10 all updated
• Windows Event Viewer shows only 3 events related to the crashes.
  • Error - EventLog (ID 6008) - 8:20:55 pm
  • Critical Kernel-Power (ID 41) - 8:20:51 pm
  • Error - EventLog (ID 6008) - 8:20:55 pm
• I find it interesting and odd that the Critical Kernel-Power error happens BEFORE both Error 6008's happen...but for some reason it's sandwiched in between them?

My Research

• The general answer to unexpected shutdowns include PSU problems, over-heating, and memory modules. I don't appear to have any over-heating issues or memory issues. How do I test PSU issues (before buying a new one)?
• One YouTube video did suggest that it was actually a capacitor that was over-heating due to being covered up by his CPU cooler. This is not the case for me, but how do I know if it's the motherboard causing these restarts? Or capacitors or whatever? How do you test without paying for a new one?
• Many people with the same issue say it's due to an older Windows 10 version, but I've updated everything.
• I've monitored my memory clock speed using MSI Afterburner in-game (Doom Eternal). I have stock memory sticks (not overclocked) that are DDR4-2400, but MSI Afterburner is reporting the Memory Clock at over 3500mhz....do I just not understand memory values or something? Or is this a problem? I have done ZERO overclocking. Or Am I confused as to what Afterburner is actually reporting?

Little Back Story

Months ago I was playing CoD Black Ops 3 and my game kept freezing and crashing. I contacted Activision support and they recommend that I update my BIOS version. I updated/flashed my BIOS according to the ASRock website. Only THEN, suddenly, the crashes/restarts started happening. I was very upset. I updated my BIOS and all of a sudden my system can't handle heavy loads and restarts.

I do have to consider that this is just a coincidence though. I find it incredibly odd that a BIOS update to the newest version would be so broken.

There seems to be no definitive answer for these unexpected Kernal-Power Error 41 crashes. Everyone has a different response, but LOTS of people with AMD Ryzen systems have complained about this issue.

Basically, when I put a heavy load on my system it increases the chances of it freezing and restarting. But this isn't always the case, Sometimes I'm just watching Netflix or browsing Amazon and it freezes/restarts.


CPU: AMD Ryzen 5 1600 (Summit Ridge, ZP-B1)
3200 MHz (32.00x100.0) @ 3392 MHz (34.00x99.8)
Motherboard: ASRock AB350 Gaming K4
BIOS: P5.80, 06/19/2019
Chipset: AMD B350 (Promontory)
Memory: 16384 MBytes @ 1064 MHz, 15-15-15-36
- 8192 MB PC17000 DDR4 SDRAM - Team Group TEAMGROUP-UD4-2400
- 8192 MB PC17000 DDR4 SDRAM - Team Group TEAMGROUP-UD4-2400
Graphics: EVGA GTX 970 SSC ACX 2.0+
NVIDIA GeForce GTX 970, 4096 MB GDDR5 SDRAM
Drive: PNY CS1311 120GB SSD, 117.2 GB, Serial ATA 6Gb/s @ 6Gb/s
Drive: ST1000DM003-1ER162, 976.8 GB, Serial ATA 6Gb/s @ 6Gb/s
Drive: Samsung SSD 860 EVO 500GB, 488.4 GB, Serial ATA 6Gb/s @ 6Gb/s
Sound: NVIDIA GM204 - High Definition Audio Controller
Sound: AMD Zen - HD Audio Controller
Network: RealTek Semiconductor RTL8168/8111 PCI-E Gigabit Ethernet NIC
OS: Microsoft Windows 10 Home (x64) Build 18363.720 (1909/November 2019 Update)
Power: Corsair GS600 PSU

 

[Madmatt2006]

Looks like you have 2 graphics cards?

Graphics: EVGA GTX 970 SSC ACX 2.0+
NVIDIA GeForce GTX 970, 4096 MB GDDR5 SDRAM

Have you tried running one to see it if crashes. to isolate graphic cards

 

[droscoe]

Looks like you have 2 graphics cards?

Graphics: EVGA GTX 970 SSC ACX 2.0+
NVIDIA GeForce GTX 970, 4096 MB GDDR5 SDRAM

Have you tried running one to see it if crashes. to isolate graphic cards

No I don't, that's just the poor formatting of HWinfo64 program. I only have one GPU.

 

[Madmatt2006]

I'd pop graphics card out and use the onboard graphics and see if it still freezes

 

[phillpower2]

No onboard graphics unfortunately Matt

Power: Corsair GS600 PSU

Possible culprit, with the additional hardware that is present and the PSU having the potential to be 7+ years old it could be showing its age, only covered by a 3 year warranty when out of the box new so if over 7 years old one couldn`t complain if it is the PSU.

There seems to be no definitive answer for these unexpected Kernal-Power Error 41 crashes.

That is correct as all the message means is that Windows was not shut down correctly, loss of power and overheating causing the system to shut down can both be reported under this error by Windows.

Download then run Speccy (free) and post the resultant url for us, details here, this will provide us with information about your computer hardware + any software that you have installed that may explain the present issue/s.

To publish a Speccy profile to the Web:

In Speccy, click File, and then click Publish Snapshot.

In the Publish Snapshot dialog box, click Yes to enable Speccy to proceed.

Speccy publishes the profile and displays a second Publish Snapshot. You can open the URL in your default browser, copy it to the clipboard, or close the dialog box.

Download Speedfan and install it. Once it's installed, run the program and post here the information it shows. The information I want you to post is the stuff that is circled in the example picture I have attached.

If you are running on a vista machine, please go to where you installed the program and run the program as administrator.

So that we have a comparison to Speedfan, download, run and grab a screenshot of HWMonitor (free).

To capture and post a screenshot;

Click on the ALT key + PRT SCR key..its on the top row..right hand side..now click on start...all programs...accessories...paint....left click in the white area ...press CTRL + V...click on file...click on save...save it to your desktop...name it something related to the screen your capturing... BE SURE TO SAVE IT AS A .JPG ...otherwise it may be to big to upload... after typing in any response you have... click on Upload a File to add the screenshot.

Screenshot instructions are provided to assist those that may read this topic but are not yet aware of the “how to”.

 

[droscoe]

Thank you for your help phillpower2. Looking forward to your input.

Speccy - http://speccy.piriform.com/results/XcS68TIXEvxISBf0xKep0Qs

 

[phillpower2]

Hello droscoe,

Thanks for the additional info

Only thing of note is the high temp of the GPU which is confirmed by each of the three programs, when out of the box new the idle stock temp would have been circa 28°C and when tested under load running Furmark 63°C I did find other figures being quoted such as 35°C and 73°C but these were not backed up by evidence such as what you will find here

Poor GPU cooling and a weak PSU could both cause this so both need to be checked.

First test that is suggested, power up the computer, do some gaming and while the computer is under load (if you are able to get your hand in) put the back of your hand a small distance away from where the air is expelled by the PSUs internal cooling fan to see how warm/hot the air is that is being expelled.

Second, you have two screens hooked up to the GTX 970, after powering down correctly, disconnect the bigger of the two screens, power up, do some gaming, run Speedfan & HWMonitor, make a note of the temps, do the back of the hand test on the PSU, see if you are able to continue gaming, if not, post back and let us know along with the Speedfan/HWMonitor info + what the temp of the air coming from the PSU was like.

 

[droscoe]

Thanks pillpower2.

So, a couple things. The screenshot of my GPU above was after gaming for a while (with several intermittent crash/restarts), so the GPU hadn't cooled off completely.

After starting up my system, my GPU does indeed begin at 29*C. Also, the air coming out of my PSU is cool.

My "second monitor" is actually my TV (I watch Hulu and stuff on it), but its never on. Regardless, I unhooked it from my GPU so I'm running only 1 monitor.

I loaded up Doom Eternal, played about 7-8 minutes, and it froze/restarted. I had MSI Afterburner displaying the GPU temp and it peaked at 71*C, but usually stayed below that.

Also, the air coming from my PSU vent was warm. Not hot, but warm.
---------------------------------
I'm riding the wave of your assistance, so I don't want to take the reigns here, but I want to mention a couple things.

Ryzen systems that freeze or reboot randomly seem to be common. A lot of people always mention a failing PSU as the culprit, but many are saying AMD had memroy issues and motherboards/BIOS's have been unstable.

Recall that I mentioned that absolutely NONE of this happened until literally the day that I updated my BIOS. Several hours and a lot of deep digging in forums show many people saying ASRock boards have had a recent reputation of random freezes and crashes. No one accuses a single piece of hardware as the culprit, but instead just bad hardware intercommunication, at least when it comes to power management within Ryzen systems. Obviously there's a lot of keyboard "experts" on the internet, but the common theme seems to be sudden drops in voltage, power management, poor BIOS and mobo communication, memory, and Ryzen CPU design.

Regardless, it's obviously good sense, and good practice, to not rely on random internet search results and actively troubleshoot my own hardware. So I look forward to your response on how you suggest I proceed.

 

[droscoe]

I even found an ASRock specific post about how someone had the same issues as I'm having. After updating their BIOS, their system suddenly started freezing and restarting. Evidently, someone said certain boards are "more compatible" with certain Ryzen generations of CPUs and that he needed to flash back to a previous BIOS version or upgrade to the newest STABLE version for 3rd gen CPUs.

The guy responded and said it worked like a charm. I'm afraid to upgrade my BIOS because ASRock has a diclaimer "we do not suggest upgrading to 6.3 if you are using a Summit Ridge CPU". (which I have). I'm not sure why....but I'm on 5.8 and I dont want to destroy my PC. And evidently you can't flash back to older BIOS versions either....I have no idea how this guy was able to do any of it.

 

[phillpower2]

First and foremost you can mention whatever you see fit, it is your computer after all.

To the task at hand, if you suspect that this is a problem caused by updating the BIOS and if you are indeed correct then none of us will be in a position to help as BIOS issues tend to be terminal I`m afraid

Ryzen systems that freeze or reboot randomly seem to be common. A lot of people always mention a failing PSU as the culprit, but many are saying AMD had memroy issues and motherboards/BIOS's have been unstable.

Not sure where you have read this but I am around several forums on average 5 to 6 hrs most days (a lot more atm) and the only issue that you mention that I have seen regular is the RAM issues, your computer had been running fine with your hardware and you know how it has been set up from day one so I never questioned on your choice of RAM, I don`t OC, never have and only ever have to try and help folk repair the damage that they have done by a bad OC.

The crux of it is, from the get go we have seen high GPU temps that were confirmed by three software programs, there are additional devices hooked up to the computer that all draw power and the system is being powered by a PSU that is possibly 7+ years old, where would your suspicions tend to be.

Regarding the RAM, I did notice that your RAM is only 2400MHz and found that when I checked here that AMD say 2667MHz (which is 2666MHz to you and me) is recommended for the AMD Ryzen 5 1600 so yes it is possible that I am wrong and your RAM being slow is the cause, in essence I should have asked if you were OCing your RAM to the required 2667MHz.

 

[droscoe]

Not sure where you have read this but I am around several forums on average 5 to 6 hrs most days (a lot more atm) and the only issue that you mention that I have seen regular is the RAM issues, your computer had been running fine with your hardware and you know how it has been set up from day one so I never questioned on your choice of RAM, I don`t OC, never have and only ever have to try and help folk repair the damage that they have done by a bad OC.

The crux of it is, from the get go we have seen high GPU temps that were confirmed by three software programs, there are additional devices hooked up to the computer that all draw power and the system is being powered by a PSU that is possibly 7+ years old, where would your suspicions tend to be.

Regarding the RAM, I did notice that your RAM is only 2400MHz and found that when I checked here that AMD say 2667MHz (which is 2666MHz to you and me) is recommended for the AMD Ryzen 5 1600 so yes it is possible that I am wrong and your RAM being slow is the cause, in essence I should have asked if you were OCing your RAM to the required 2667MHz.

1. I went into my BIOS and set my memory to 2666. I loaded up Doom Eternal and within 45 seconds it froze and restarted.
2. Doesn't my response above confirm that my temps are fine? Starting at 29*C and peaking at 71*?
3. I didn't see many similar posts here. I've seen dozens and dozens around the internet. AMD forums, ASRock boards, Reddit, and other forum sites. So many people asking the public to help diagnose their crashing Ryzen system.

I wonder if this gives me a clue. I can fool around on YouTube, desktop, or even play other games (Stellaris, Sins of Solar Empire for example) and I can play them for much longer.

All those trivial tasks and playing those strategy games tend to be CPU focused.

However, once I play something that is more GPU intensive, or at least draws more power, such as Doom Eternal, it crashes within seconds. (And its not just Doom, its any other graphically intense game).

I don't know if its power management by the motherboard, or memory management, or what. But it seems as though I can reliably increase the chances of my PC freezing and restarting by doing something more demanding/intensive.

Wouldn't a failing GPU or failing PSU be random in when they crash a system?

Just something about that...the fact that the crashes are not really random, but instead happen more consistently when a demanding game is being played, just has me thinking about what that logically suggests.

Also, the fact that its not a HARD restart. It freezes...it plays whatever sound is playing at the exact split second, but its the terrible looped/repeating/stuttering instance of whatever is playing. The CPU is STILL telling the system to play audio, but it literally sounds like it's stuck. Like the CPU got hung up or something.

 

[droscoe]

The other interesting thing here that I think is important...occasionally my system will SOUND like it wants to crash, but it doesn't. I'll hear the sound stutter for half a a second, but it will continue on and be alright. It just happened to me a few minutes ago. The whole system won't freeze or stutter, but the sound will stutter, just like it does when it DOES freeze/restart.

I never heard of PSUs and GPUs thinking about crashing but decide "ahh, guess I'm alright, let's keep going". lol

I'd like to also note that I changed wall sockets and I just finished playing Doom Eternal, without interruption, for a good 20 minutes. I know I'm probably jinxing myself, but I've read that sometimes voltage drops from older sockets can affect your system. (I live in a 100 year old building).

Of course, now that I say that, it'll crash on me soon. I'll report back if it does.

 

[droscoe]

I just finished a 2 hour stint of playing Doom Eternal, which was impossible before. It did just freeze/crash on me though after a couple hours.

Is this just a coincidence? Would power/current from a wall socket in an old building be unreliable and cause a PSU to lose power for a split second, thus locking up the system?

I also remembered that my 3-prong power cable is actually using a 2-prong adapter, since I literally have no 3-prong outlets in this old apartment. Surely that's not a problem?

 

[droscoe]

UPDATE - I'm not trying to bump my own post. I want to give an update on my plans and I will report back in a couple days. I'm still curious what your thoughts are on this problem though!

I believe I need to invest in a decent UPS. I live in an old apartment, my lights frequently dim multiple times a night, I'm constantly replacing light bulbs around my place. It seems to be plausible that my apartment has inconsistent power and that my computer's PSU is experiencing short dips in power from the sockets, causing it to freeze my system and restart. Many results from Google show people mentioning this as the culprit of random system crashes.

I've already had a 2 hour stint of uninterrupted game-play, which was NEVER possible before, from simply switching wall sockets. I will be buying a UPS and testing it out over a couple days. I will report back!

 

[phillpower2]

All the symptoms that you describe are typical of overheating and/or a poor power supply with a weak PSU not only getting hot itself inside but also causing other hardware components to get hot as well as they struggle to get the power that they need, the GPU being the most power hungry tends to get the hottest first and when it does you get lag, stutter and freezing, you will also get this type of behaviour when there is a hardware bottleneck, your RAM for instance only running at 2400MHz wasn`t able to keep up with the performance of the Ryzen CPU.

As if just by coincidence we have just solved a Ryzen/RAM related thread here

Regarding the wall socket, is the socket that you are using now in the same room or if not in the same room is it on the same floor in the building.

A UPS is always a good idea and especially where there is a known mains power problem and/or there are regular brown outs.

 

[droscoe]

Ok, well I'm still having crashes even with the new UPS I bought.

Are there any other considerations we should make before I purchase a new PSU?

Thanks for help so far. Very helpful!

 

[droscoe]

I'm looking at Speccy and I'm confused as to why it says my DRAM Frequency is 1064.1 MHz?

I have in my BIOS loaded up XMP profile for DDR-2400, so shouldn't Speccy say around 2400MHz?

 

[Bruce]

I've got 2400 RAM and Speccy reports 1200 - it's always half (or there-abouts) based on the following excerpt that I've kept for reference cause it explains it better than I can.

DDR memory speeds

“Like the ticking of a clock, each tick represents a single hertz or cycle (the opening and closing of a transistor gate in this case).
A speed of 1Hz, for example, is one cycle per second; 2Hz is two per second; a MHz is 1,000,000 cycles per second; you get the picture.
The problem is when DDR (or double data rate) RAM came on the scene, it changed how data transfers were registered. Instead of only actuating once on the rise of each clock cycle, it could now also process an additional operation on the fall of that same clock cycle, effectively doubling the rate at which the DIMM could process data.
The figure for accurate measurement of data transfer requests then shifted from MHz to MT/s to adjust for this change, despite the fact that memory still operated at the same frequency.
However, marketing apparently didn’t get that memo, because many companies, in a bid to tout it as the next big thing, ignored the MT/s figure, instead referring it as MHz, while modern day memory quoted at 2,400MHz, for instance, only operates at half that frequency.”
APC Mag, Apr 2018, Issue 453

 

[phillpower2]

The serial presence detect speed of your RAM is 2133MHz which is what the BIOS detects and 2400MHz is what the RAM has been tested and proven to be stable by the manufacturer, dual channel means 2 Xs 1064.1 MHz which = 2128.2MHz which is slightly less than what it should be.

Regarding the wall socket, is the socket that you are using now in the same room or if not in the same room is it on the same floor in the building.

A UPS is always a good idea and especially where there is a known mains power problem and/or there are regular brown outs.

Sorry to hear that the UPS has not helped but being honest swapping in a known good a PSU should have come before the UPS, even if it was only a borrowed one for testing purposes, any power coming from the UPS has to go through the present PSU to get to the hardware which sorry to say you would have been advised if you had answered the question that was asked in my reply #15.

 

[phillpower2]

Any update droscoe?