Periodic bluescreens and spontaneous restarts

[Zeusface]

Hi,

I’m not too tech savvy and recently I got a pc that someone else built and the guy that built it did a fresh install of windows 10 home before giving it to me. Since then it has encountered random bluescreens after booting up. At first I would get tons of random bluescreens after booting, almost every one being a different stop code. After multiple fresh installs of windows 10 home, updating bios, windows, every driver that I could find, and enabling xmp, most of the bluescreening has stopped. A few days ago I got 3 bluescreens but was only able to catch one stop code: “system_service_exception.” (not 100% sure about this).

I’ve tried scanning/repairing my hard drive (sfc scannow?) and using the windows verifier tool for drivers (but I’m not really even sure what it did other than increase my cpu usage). I’ve run memtest a few times which all came up 100% clear. I checked all my devices in device manager and none of the devices show any notification alerts except one that only shows up when I show hidden devices. lBtUSB_Filter_00 which has no hardware ID and says that it’s currently not connected to my computer.

I also tried uninstalling and reinstalling my graphics drivers and since then my computer has been randomly restarting a few times. Most notably it would boot to desktop then immediately restart and after reaching the desktop it would be fine. It has restarted while idling a couple of times also. Through the event viewer, I saw that there were multiple critical errors “event 41, kernel-power.” I checked all my windows power and startup settings and any automatic restart option that I could find was disabled and I looked and it seems like my psu connections and mobo are fine.

Someone has suggested that my ram is incompatible, which is true when I checked with pcpartpicker, but then someone else pointed out that there shouldn’t be any compatibility with my gskill ram per the gskill compatibility tool online which does say that my ram is compatible. Not really sure what to do with this, so I did order some ram to test which hasn’t arrived yet.

On a separate but perhaps related note, I am unable to play warzone without encountering multiple fatal errors. The only time I was able to play a game was when running it in safe mode.

Appreciate any help I can get! Please let me know if there’s any more information I can provide.

specs
cpu: ryzen 5 3600
gpu: radeon rx 5700xt
ram: gskill ripjaw v series 16gb (8x2) 3200
mobo: asrock x570 phantom gaming itx/tb3
hard drive: wd blue sn550
psu: corsair cx 550 watt 80 plus bronze certified modular psu (cp-9020102-na)
case: nzxt h200

[phillpower2]

Hello Zeusface,

Was going to ask for any crash dmps but no point atm as the first thing that you need to do is get an appropriate PSU.

gpu: radeon rx 5700xt

This requires a minimum of a 600W PSU, Gold efficiency rated PSU and from a proven brand such as Corsair, EVGA or Seasonic, Corsair btw do recommend that their CX range are only used for basic PCs, what they say below;

CX Series Modular power supply units are an excellent choice for basic system builds and desktop PC computer upgrades, offering high reliability, low noise, and the flexibility of modular cabling.

[Zeusface]

I appreciate the reply! I will certainly try swapping out to a different psu. I don’t mean to sound contradictory but I’m curious because this pc was running fine for a few months with this exact build prior to these recent issues after the fresh windows install.

I will be out of town for a few days but will update when I can

[phillpower2]

A weak PSU kills hardware slowly but surely and an add on GPU being the most power hungry device tend to suffer first and the most, your add on GPU may already be damaged so I suggest that if you can you see if you are able to borrow an appropriate PSU to swap in for testing purposes.

As said previously no point in looking at any crash dmps while there is a substandard PSU present, we will provide the steps for uploading any dmps as and when is appropriate.

[Zeusface]

Ok that makes sense. I’ll make a swap as soon as I can

[phillpower2]

(y)

[Zeusface]

Hi,

I was able to swap in a xfx 750w psu but continued to see multiple spontaneous restarts. event viewer still showed event 41, kernel-power

Afterwards I swapped out my two sticks of ram for one stick of 16g ram (crucial CT16G4DFD832A) but then also experienced another bluescreen (system_service_exception)

[phillpower2]

Hello Zeusface,

You will always get event 41 kernel-power errors when a computer powers down while Windows is still running, this regardless of the cause of the unexpected shut down.

You may have two problems going on, one hardware and one software related, you have swapped in a 750W PSU which goes some way to ruling a PSU issue out, two other things that come to mind that can cause random restarts are bad caps and a system short, check for any bad caps on the MB first, see info and example pics here to help with checking.

Will come back to checking the short possibility should we need to.

For checking the software etc, see the below;

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.

1. Copy any dmp files from C:\Windows\Minidump onto the desktop.
2. Select all of them, right-click on one, and click on Send To> New Compressed (zipped) Folder.
3. Upload the zip folder using the Attach button, bottom left of the dialogue input box

[Zeusface]

Ok got it! I suppose I’ve been saying spontaneous restarts with event 41 kernel-power to try to distinguish it from bluescreens. When I experienced spontaneous restarts, the computer just goes black and restarts without any stop code. Didn’t know how to properly reference that.

I’ve kept the computer on since the last successful boot and haven’t had it restart or bluescreen yet. I was also able to successfully run warzone, but now (and I don’t know if this is related or not) google chrome randomly crashes or closes itself periodically. It’s not frequent but has happened a few times in the last several hours, most consistently after resuming from a long time of idling or when chrome has been minimized for an extended period of time, but has also crashed while I was using it as well.

Here is the speccy link: http://speccy.piriform.com/results/U...1QXXP3k2dQDbDX

And I’ve attached the minidump files. All dump files before the the 23rd are before the psu and ram swap.

Thanks and hope you’re having a good Christmas Eve!

[Zeusface]

I may have spoken a bit too soon. After posting, I tried restarting and my pc restarted several times on its own and froze a few times too.

[phillpower2]

Your crashes don`t appear to be driver related as no drivers are named, there is a Realtek device listed in the bug check but this could be because another device tripped it up, possibly the Logitech G430 gaming headset as having audio properties it also uses its own audio drivers.

Mentioned above “You may have two problems going on, one hardware and one software related,”

After looking at the crash dmps (only the last 4 as the others are too old) this does look to be the case, Speccy shows a couple of possible causes.

Windows itself looks to be having issues, it could be corruption or settings that you have and Speccy again may be giving us some clues to which we will go through below.

Power Profile
Active power scheme: High performance

Change the Windows Power Plan to Balanced, Ultra and High Performance are a form of overclocking that is known to cause stability and overheating issues and the setting should only be used for gaming type notebooks that have a discrete GPU that needs the extra power.

Hibernation: Enabled This is a useless setting that is known to cause issues, disable it, this should also disable fast boot which likewise can cause problems but as you are booting from an SSD you don`t need fast boot in any event.

To disable Hibernation:

1. The first step is to run the command prompt as administrator. In Windows 10, you can do this by right clicking on the start menu and clicking “Command Prompt (Admin)”
2. Type in “powercfg.exe /h off” without the quotes and press enter. If you typed it in correctly, the cursor will simply start at a new line asking for new input
3. Now just exit out of command prompt

Turn Off Hard Disk after: (On AC Power): 20 min Not sure why this has been set to 20 minutes as it should be set to Never

Not Installed
12/14/2020 2020-12 Cumulative Update for Windows 10 Version 20H2 for x64-based Systems (KB4592438)
Installation Status: In Progress

The Windows issues previously mentioned are that you look to be having problems installing Windows updates, the one above shown to be still in progress and SearchIndexer.exe was flagged up in your crash dmps.

Make the the three changes suggested above, restart and see how things go for a while then post back with an update for us, we will then take a look at Windows for you.

Last but not least is your RAM, the below from your OP;

ram: gskill ripjaw v series 16gb (8x2) 3200

Not sure where you got this information from but it is incorrect, the RAM is not even the same brand, you have a single 16GB stick of Crucial RAM and because of this the RAM can only work at circa 1600MHz which is 50% of what is required for your AMD Ryzen™ 5 3600 to be able to perform at its best and be stable, that is one heck of a bottleneck.

[Zeusface]

Hello again and Merry Christmas!

I’ve made the three changes you suggested and will continue to monitor the performance.

With regards to my ram, don’t know if you caught it, but after swapping out the psu, I also swapped out the ram as well (from the 2 8gb gskill ripjaws to the 1 16g crucial CT16G4DFD832A). The crucial ram is listed as 3200 mt/s. I thought that the switch would be ok, especially to test out new ram. Here is what I read about ram that might explain the “50% decrease.” Please advise if I misunderstood and if I should swap back to the previous gskill ram modules.

“Modern DRAM modules support double data rate (DDR) transfers, meaning that although they function at a given clock speed, they transfer data during both the rise and fall of a CPU clock cycle, allowing for double the data transfers their native speed, or frequency, would suggest. Some diagnostic or system information software does not express this clearly, only reflecting the modules’ frequency in megahertz (MHz), making such feedback suggest that your parts are only running at half their possible speed. For example, DDR3-1600 supports 1600 megatransfers per second (MT/s) due to its nature as DDR memory, but runs at 800 MHz.”

Thanks again!

[phillpower2]

Hope you enjoyed your Christmas Day (y)

Changes made acknowledged but did you check the caps on the MB;

check for any bad caps on the MB first, see info and example pics here to help with checking.

I confess that I did miss the RAM info in your reply #7 tbh had it been noticed it would have been suggested that you put the original two sticks back in, would suggest we leave as is for now and see how the computer behaves after these three changes.

Explanation about the RAM, DDR and Dual Channel are two entirely different things, the info that you read and post above concerns DDR and not Dual Channel for which you need two sticks of RAM to achieve, your MB has channel A and B and to achieve 3200MHz you need a stick of 3200MHz capable RAM in each, placing a single stick of 3200MHz capable RAM in either slot will as your Speccy report only give you 1600MHz, this because there is no RAM in the other channel, only 2 Xs 1600MHz sticks in channel A and B produces 3200MHz, not a single stick of 3200MHz capable RAM.

Memory
Type: Unknown
Size: 16384 MBytes
Channels #: Single
DRAM Frequency: 1599.6 MHz With two sticks of matching RAM you would be seeing Dual as opposed to Single channel and 3200MHz.

Physical Memory
Memory Usage: 25 % The RAM only working in single channel also results in higher memory usage, in dual channel this would most likely only be about 15%.

[Zeusface]

Changes made acknowledged but did you check the caps on the MB;

Yes sorry I forgot to mention, but yes I did look and everything looked to be fine.

Thanks for the explanation and I’ll continue to monitor the changes. Sorry I’m still trying to learn a lot of this stuff so this would help a lot if I ever run into these things again, but after running “wmic memorychip get speed” in command prompt and the equivalent in powershell, the speeds reported back to me are 3200. The same is also the case when I boot into bios; my bios also says that it’s running at 3200. What do these mean if speccy is saying that it’s at 1600? Because with the other 2 3200 ram sticks, the bios would default them down to and reads them as 2133 and I’d have to enable xmp to have them go at 3200, but this single stick defaults and reads as 3200.

[phillpower2]

Good news that the caps look ok (y)

Rather than me waffle and get tongue tied in the process take a look at the single and dual channel explanations here

Couple of short pointers regarding RAM etc

3200MHz is the maximum bandwidth that your RAM is capable of running at but being in single channel it will report as 1600MHz.

The stock speed of most DDR4 RAM is presently 2133MHz and to achieve anything more XMP or on earlier AMD boards DOCP most often needs to be enabled, all this does basically is increase the RAM voltage from 1.2 to 1.35V.

Your MB memory support list here shows what configurations of RAM are needed to achieve any specific speed, notice that only two or four slots are supported and that not all RAM can be overclocked, this does not mean that using a single stick will not work in the board just that you need two or four sticks to be able to achieve the maximum bandwidth of the RAM.