System Restarts or BSOD While Gaming

[Blake1]

Hi All. Huge thank you in advance for anyone taking a look at this and helping out.

Short Version:
When gaming my system will restart (power does not completely cut). Usually I get a frozen screen that goes pixilated, and and then my system restarts. No error messages. OR I'll crash and get the WHEA_UNCORRECTABLE_ERROR message. Both happen while gaming.

What I've Done:

• Updated bios to latest version
• Latest windows 10 updates
• Latest GPU driver udpates
• trying monitored temps (see below)
• Uninstalled MSI Dragon Center
• Formatted system and fresh install of windows 10
• Set bios to default (except I have XMP on and custom fan curve settings)
• Reduced Memory Speed to 3400hz, then 3200hz
• Removed 2 of 4 memory sticks (then swapped 2 of the memory sticks I took out with the ones that were in the system)

System Specs

• AMD Ryzen 5900x (no overclock) - Bought at beginning
• NZXT Kraken X73 Liquid Cooler
• Nvidia RTX 3090 FE - Bought 1 month ago
• MSI X570 Tomahawk - Bought in Nov 2020
• Corsair Dominator Platinum RGB 32GB (4x8GB) DDR4 3600 (PC4-28800) C18 - Bought in Dec 2020
• Samsung 850 Pro 256 SSD - Bought around 2016
• Samsung 850 EVO 500 SSD - Bought around 2016
• Corsair HX1000i Platinum PSU (refurbished) - Warranty Exchange in Oct 2020
• Windows 10 (fresh format and install about 3 weeks ago, I had problems prior too)
• x64
• dmp files & System Diagnosis files here. I'm not sure if I got the system Diagnostic Report files correctly, so let me know if there are problems and I can try again.

Temps
I used hwmonitor to check my temps, but I would do it by alt+tabing over from my game to occasionally check it. So I don’t know the exact temps at the time of crashing/restart.

cpu: most of the time it was 68-72 but I never saw it higher than 75
Gpu: most of the time right at 68, but I didn’t see it higher than 70

Detailed Version:
My system is fairly newly built and I wasn't experiencing any problems for the first few months. Then I started crashing with the WHEA_UNCORRECTABLE_ERROR message. It was always while gaming and at first it was after an hour or longer then it got worse and happened in 30 minutes or less, unless I shut my system off and came back to it later and then I could usually game over an hour without an issue. The change that I had made when this started happening was that i updated my bios to the latest version (not in beta) and that is when I started having the issue. So I rolled back my bios and the issue went away for several weeks.

Then I noticed I started having the same issue again and this time the change I made that I thought may have caused it was updating the MSI Dragon Center. I started reading that a lot people were getting problems with dragon center. So I uninstalled it, and the issue went away for several weeks. Then it started happening again. This time I couldn't figure out why. So I just formatted my system entirely, reinstalled Windows 10 and started fresh. I didn't have problems for a few weeks, but then I started freezing in the middle of games followed by a restart. I didn't have dragon center installed so I knew that wasn't it and I just did a fresh format and install of windows 10. I ignored it for a while because it was fairly infrequent and I guess I just didn't want to deal with it.

Today I got a sudden restart while gaming with the WHEA_UNCORRECTABLE_ERROR message. So the next steps I took were updating my MB bios to the latest version (which was a newer version that the version I had before I rolled back before) and also making sure my GPU drivers and Windows had the latest updates. I started gaming again and after maybe 45 minutes I got the WHEA_UNCORRECTABLE_ERROR message again. I let my system restart and then I jumped back into gaming and I got the message again after about 15 minutes. So it seems to be getting worse. I'm not really sure what is going on at this point.

I read something about checking of the events in event view in windows computer management. I did find a few things that make me think there is a problem with my processor. I'll post them here to get others opinions on it.

Here are the links to my screen shots.
https://imgbox.com/tKjxTbd2#
https://imgbox.com/x2FzHoVf

 

One other thought I had that I wanted to share and get thoughts on. my MB has 2 plugins for power, and 8 pin and an extra 4 pin. From what I read the 4 pin wasn't necessary unless you were overclocking, which i'm not at the moment. So I don't have the 4 pin plugged in, but I'm wondering, could that be the issue?

Edited June 11, 2021 by Blake1
I had my memory listed wrong

[Guest]

I  think plugging in the extra 4 pin may fix the issue.

[Analog]

26 minutes ago, Blake1 said:

Hi All. Huge thank you in advance for anyone taking a look at this and helping out.

Short Version:
When gaming my system will restart (power does not completely cut). Usually I get a frozen screen that goes pixilated, and and then my system restarts. No error messages. OR I'll crash and get the WHEA_UNCORRECTABLE_ERROR message. Both happen while gaming.

 

You have already done quite a bit of troubleshooting so the only thing that I can think of are memory and primary drive issues.

 

You you said your memory is 3200MHz, but you where running it at 3400MHz, but since you had issues you put it back to "stock." This 3200MHz is it XMP or JEDEC? If it is XMP, disable it and run them at the standard JEDEC specified speed for these modules (2133MHz or 2400MHz). See how it the system will be performing then. Most often than not XMP doesn't really work properly, as the motherboard is left with too many values to figure out by itself, and not all motherboards / BIOS revisions do a good job of it. 

 

The other thing to look into is whether the primary drive that you have is working properly. If you are having issues with the primary drive this can also result in WHEA errors. 

[jaslion]

20 minutes ago, Analog said:

 

You have already done quite a bit of troubleshooting so the only thing that I can think of are memory and primary drive issues.

 

You you said your memory is 3200MHz, but you where running it at 3400MHz, but since you had issues you put it back to "stock." This 3200MHz is it XMP or JEDEC? If it is XMP, disable it and run them at the standard JEDEC specified speed for these modules (2133MHz or 2400MHz). See how it the system will be performing then. Most often than not XMP doesn't really work properly, as the motherboard is left with too many values to figure out by itself, and not all motherboards / BIOS revisions do a good job of it. 

 

The other thing to look into is whether the primary drive that you have is working properly. If you are having issues with the primary drive this can also result in WHEA errors. 

I'd say run a memtest86 with the different mem configs to see f the memory is good or if there is a stability issue.

 

Also I see they used a refurbed hx1000i. That thing is ancient by now and since the warranty ended in 2020 it is most likely a 2013 unit and I doubt it is still in a good enough shape to power a system like this especially with the powerspikes.

[Blake1]

9 hours ago, Analog said:

 

You have already done quite a bit of troubleshooting so the only thing that I can think of are memory and primary drive issues.

 

You you said your memory is 3200MHz, but you where running it at 3400MHz, but since you had issues you put it back to "stock." This 3200MHz is it XMP or JEDEC? If it is XMP, disable it and run them at the standard JEDEC specified speed for these modules (2133MHz or 2400MHz). See how it the system will be performing then. Most often than not XMP doesn't really work properly, as the motherboard is left with too many values to figure out by itself, and not all motherboards / BIOS revisions do a good job of it. 

 

The other thing to look into is whether the primary drive that you have is working properly. If you are having issues with the primary drive this can also result in WHEA errors. 

@Analog thank you for the reply! I realized that I listed my memory stats wrong in my original post. My Memories clock speed is 3600MHz (I updated my original post). So when I was running it at 3400MHz then 3200MHz I was clocking down form 3600MHz, but I was still using an XMP profile to do that. Understanding that I'm assuming you'd still make the same suggestion in that I take off the XMP profile and let it run on the standard JEDEC, right?

 

One clarification, I don't know what JEDEC actually is (I'll do some reading), but if I go into my BIOS and just turn off the XMP profile and leave the bios settings at default, is that essentially using the standard JEDEC that you are suggesting?

 

I'll consider my primary SSD as a possible issue to rule to rule out as well. I'll test my memory like you suggested and if I still have the problem do you have any thoughts on how I can test my SSD? Or is it simply just removing it and using a different drive? 

[Blake1]

9 hours ago, jaslion said:

I'd say run a memtest86 with the different mem configs to see f the memory is good or if there is a stability issue.

 

Also I see they used a refurbed hx1000i. That thing is ancient by now and since the warranty ended in 2020 it is most likely a 2013 unit and I doubt it is still in a good enough shape to power a system like this especially with the powerspikes.

@jaslion I'm not familiar with memtest86, but I'll look into it. Thank you for the suggestion and help! Sounds like a great way to narrow down if it is my memory or not.

 

Also, I'm a little suspicious of my PSU as well. My last motherboard would give me a message if it was a power surge related issue. I haven't seen any of those message yet, but I don't think my new motherboard will give me those messages. I'm going to give Corsair a call and see if they have a way I can further diagnose it. Thank you again!

[Blake1]

10 hours ago, ardwivedi16 said:

I  think plugging in the extra 4 pin may fix the issue.

@ardwivedi16 I'm suspicious of this, but a lot of what I'm reading is that the 8pin on its own is plenty of power for my CPU. But I'm not ruling this out. Unfortunately my PSU didn't come with another 4pin for me to use, so I'd need to look at picking up an extra one. I think I'll test out some of the potential issues with memory and my SSD and then dive into my PSU a bit more.

[Blake1]

6 hours ago, Blake1 said:

@Analog thank you for the reply! I realized that I listed my memory stats wrong in my original post. My Memories clock speed is 3600MHz (I updated my original post). So when I was running it at 3400MHz then 3200MHz I was clocking down form 3600MHz, but I was still using an XMP profile to do that. Understanding that I'm assuming you'd still make the same suggestion in that I take off the XMP profile and let it run on the standard JEDEC, right?

 

One clarification, I don't know what JEDEC actually is (I'll do some reading), but if I go into my BIOS and just turn off the XMP profile and leave the bios settings at default, is that essentially using the standard JEDEC that you are suggesting?

 

I'll consider my primary SSD as a possible issue to rule to rule out as well. I'll test my memory like you suggested and if I still have the problem do you have any thoughts on how I can test my SSD? Or is it simply just removing it and using a different drive? 

I just tried turning of XMP profile in my BIOS and I gamed for abut 45 minutes and got BSOD. I also just read someone having a similar situation that used XMP put slow increased voltages because memory wasn't getting enough power. That seemed to stabilize it for them. Is this something I should potentially try? I don't mess with memory voltages much.

 

I'm going to try some testing with memtest86  next. 

[Blake1]

@jaslion I ran the memtest86 and before the test completes my computer just restarts. I tried it twice now and both times it restarted about 5 minutes into the test. I did not get an test complete message or anything, just a flat out restart. Does this indicate it's a memory issue or would it be something else since the test isn't even able to complete? Thank you again in advance.

[jaslion]

10 hours ago, Blake1 said:

@jaslion I ran the memtest86 and before the test completes my computer just restarts. I tried it twice now and both times it restarted about 5 minutes into the test. I did not get an test complete message or anything, just a flat out restart. Does this indicate it's a memory issue or would it be something else since the test isn't even able to complete? Thank you again in advance.

Yeah your memory is pretty much messed up then so I would go and get that in warranty.

[Blake1]

@jaslion so I decided to test each stick individually and found 2 sticks wouldn't complete the test and just cause my system to restart during the test.  The other two passed the memtest86 with no errors. 

 

So I just took out my two sticks that were causing my system to restart and left my other 2 in so I could keep using my system until i got the warranty taken care of. Well, bad news, I gamed for about 30 minutes and my computer just rebooted on its own. Is it possible that I have bad memory and a bad PSU? Kind of surprising if that is the case. Is there anything else I can trouble shoot to help me figure out what is going on?

[Blake1]

So update here. I've been speaking with Corsair's support team and they seemed to think it was either my memory or my PSU. After a lot of memory testing it was not still clear what it could be. They sent me a brand new power supple (Mine was still under warranty) and after hooking it up I'm unfortunately still getting the same problem.

 

Prior to getting the new power supply I ran each individual memory stick on the memtest86. All of them passed except for one that did not pass or fail, my system just restart when testing it. I ran the test a second time on this same stick and it did pass. So it seems hit and miss. 

 

So I'm wondering if the Memory is the issue at this stage? I even question the mother board or CPU. If anyone has any other ideas on how to diagnose, I 'd love to hear them. Thank you in advance!

[Blake1]

Another update here. I've replaced my memory and just recently my MB. I'm still get restarts while gaming. 

 

I got my new MB yesterday and I've played for about 2 or so hours and then my system froze in the middle of the game and restarted. I did not get a blue screen. Previously sometimes I'd get a blue screen other times I would not. So it seems like the same problem.

The game that this occurred in was Gears of War 5. I've had this happen in other games before (and even once or twice just in windows). But this time I played several other games for a hour or so and then gears for about an hour and then it did this. I don't think it's unique to Gears, but I'm mentioning it just incase. I'm going to keep playing some other games to see if I get the same issue.

So would the next culprit to replace be my processor? Maybe one of my SSDs? I'd love anyone's thoughts, at this point I'm completely lost as what it is. I thought for sure it would be one of the parts I've replaced so far. I'm feeling a little like I'm not going to be able to fix this issue. its hard not to get frustrated.

One last note. I did not reinstall windows when I got my new MB. I did format both hard drives and reinstall windows when this first started happing (didn't fix) but should I be doing that again with new hardware? I'd think not, but again. I'm just thinking out loud.

Thank you in advance for the help.