Question Help with BIOS settings for Intel Core i5-12600K and MSI PRO Z690-A

[SpaceMouse]

Hello,
I had bought both the Intel Core i5-12600K 3.7 GHz 10-Core Processor and MSI PRO Z690-A DDR4 ATX LGA1700 Motherboard off of the advice of one of the more recent "Top PC builds for gaming" articles. However, I have been plagued with a weird problem.
Unpredictably, the graphics on both screens I have completely freeze. If there is music in a game playing, it will continue, but something like iTunes completely stops. The computer remains like this until I manually restart it by power cycling. The windows logs don't register a problem happening; as far as windows is concerned, there is no freeze. These freezes can happen at any time of day, at any CPU or GPU loads; I have had stress tests last over an hour with no problems, only to have a freeze when just web browsing. These freezes can happen with days of fine computer use, at any and all memory or CPU/GPU loads. I have tried the following:

• Swapping out the RAM sticks individually (the crash has happened on all four, individually)
• Running on intel onboard graphics (freeze happened)
• Reseating CPU and applying new thermal paste with the cooler
• RMA the motherboard
• RMA the CPU
• Enhancing cooling in the case by having the side off and a fan blowing in
• Upgrading the power supply
• Upgrading my UPS
• Complete fresh reinstall of windows 10 on a new solid state drive
• All chipset drivers installed
• Motherboard BIOS updated to latest on MSI's site
• Antivirus / antimalware scans (multiple different ones, all clean, including rootkit scans)
• SFC /SCANNOW (no problems)
• Changing the PCI slot the GPU is on
• Memcheck86+ on all RAM sticks, no problems
• I have even installed the drivers for my monitors

The RAM and GPU both work fine with my older CPU and Motherboard; this problem didn't start appearing until I swapped to the new i5 and Z690. I have tried asking Intel and MSI both about bios settings, and both avoided the question and instead encouraged a RMA (which I did, for both, and the problem continues). I am led to believe at this point it is some kind of bios setting issue, as I experimented turning Intel Thunderbolt on, and had a freeze nearly immediately. The thing I cannot understand is why the sounds continue in the background of games running; if it was a true freeze, shouldn't everything stop? The sounds continuing makes me think there is something weirder happening, specifically in the interaction between the CPU and MB. I've tried monitoring the temperature of the CPU and GPU, and the temperatures don't rise above 65C for either; the case seems to be cooling well.
Is there some "gotcha" setting I should be checking for in the bios settings? I have nearly everything set to Auto, and I'm not trying to do any overclocking. The only similar problem I've ever found someone else have they fixed by replacing the memory, so I've ordered some new RAM, but the thunderbolt setting makes me think this is some kind of bios issue. PCpartspicker doesn't warn me about any problems with my parts, other than noting I need the bios at a version for compatibility with the i5 that the bios is long past in version updates. If anyone has any other idea of what could possibly be going on, I'm all ears, but
Other parts, if you're interested:
CPU Cooler: Noctua NH-U12S redux 70.75 CFM CPU Cooler
Memory: Corsair Vengeance LPX 32 GB (4 x 8 GB) DDR4-2133 CL15 Memory
Storage: Samsung 980 Pro 500 GB M.2-2280 NVME Solid State Drive
Video Card: EVGA GeForce GTX 1070 8 GB SC GAMING ACX 3.0 Video Card
Power Supply: Corsair RMx (2018) 750 W 80+ Gold Certified Fully Modular ATX Power Supply

 

[Nine Layer Nige]

I have the same mobo and cpu combo as you but am stable on 2 sticks, but I dont have 4 sticks of ram.
I recommend that you try on only 2 sticks of ram in the primary slots for now and see if it improves stability.

 

[SpaceMouse]

I have at least had a partial theory there is some kind of issue with the RAM. Are you running with the RAM speed overclocking on or not? Also, have you changed the Try It! Settings for the RAM?

 

[Nine Layer Nige]

I have XMP switched on for auto overclocking and it is at 3600mhz.
My memory sticks are a matched pair. If your are not a matched set of 4, that "could" be the problem which is where trying just 2 sticks of matchs pairs should be stable.
Also, I flashed the BIOS to the latest which you might consider as its possible that an updated BIOS could bring better memory compatibility.

I never used the TryIt! feature. Just used the XMP ON setting.

 

[SpaceMouse]

Replaced the RAM with Patriot Viper Steel 32 GB (2 x 16 GB) DDR4-3600 CL18. XMP on, the other settings disabled or set to auto. freeze after about an hour.

I am again believing it is some problem with the BIOS settings for the CPU. I can confirm I am on the latest (7D25v14) version of the BIOS.

 

[helper800]

Replaced the RAM with Patriot Viper Steel 32 GB (2 x 16 GB) DDR4-3600 CL18. XMP on, the other settings disabled or set to auto. freeze after about an hour.

I am again believing it is some problem with the BIOS settings for the CPU. I can confirm I am on the latest (7D25v14) version of the BIOS.

Stock BIOS settings with XMP will not give you issues with that motherboard. I have 2 friends on the motherboard with similar RAM and CPU selections without issues.

 

[SpaceMouse]

Stock BIOS settings with XMP will not give you issues with that motherboard. I have 2 friends on the motherboard with similar RAM and CPU selections without issues.

I am on stock BIOS settings with XMP on with this motherboard, and I am having issues. If you have other advice about what the problem could possibly be, I would love to hear it.

 

[helper800]

I am on stock BIOS settings with XMP on with this motherboard, and I am having issues. If you have other advice about what the problem could possibly be, I would love to hear it.

I was pointing out that you may be attributing the problem you are having to the motherboard when in all likelihood it is something else. If the motherboard was physically fine but you had some gotcha setting enabled that causes these issues then this forums would be inundated with people on that motherboard and potentially similar MSI motherboards with odd issues. That motherboard is a very popular recommend here on Tom's.

I have personally seen in relatively rare circumstances the CPU cause weird intermittent issues of this kind. Have you really combed over the contact pads on the bottom of the CPU? No TIM on any of the pads? No deep scratches on any of the pads? Motherboard socket looks clean of debris? Motherboard socket has no obvious bent pins?

At the point of diagnosing you are at is where you should either bring the PC into a shop that has multiples of known good parts for everything they swap in to help determine the particular issue or start by getting another motherboard and CPU to swap in for a sanity check.

 

[SpaceMouse]

I was pointing out that you may be attributing the problem you are having to the motherboard when in all likelihood it is something else. If the motherboard was physically fine but you had some gotcha setting enabled that causes these issues then this forums would be inundated with people on that motherboard and potentially similar MSI motherboards with odd issues. That motherboard is a very popular recommend here on Tom's.

I have personally seen in relatively rare circumstances the CPU cause weird intermittent issues of this kind. Have you really combed over the contact pads on the bottom of the CPU? No TIM on any of the pads? No deep scratches on any of the pads? Motherboard socket looks clean of debris? Motherboard socket has no obvious bent pins?

At the point of diagnosing you are at is where you should either bring the PC into a shop that has multiples of known good parts for everything they swap in to help determine the particular issue or start by getting another motherboard and CPU to swap in for a sanity check.

The CPU and Motherboard both have been RMA'd, seperately. The same problem has happened with the combination of:

• first cpu, first mb
• first cpu, replacement mb
• replacement cpu, replacement mb

Given the same problem has happened across them, I can't imagine I've somehow gotten the same defect across two RMAs. Investigating the socket, both in the first and the replacement mb, showed no signs of anything wrong. I've even used compressed air to make sure there wasn't any debris or such in the socket. If there was some actual problem here, I'd figure a stress test that uses every single CPU core should be able to replicate it, yes? The problem is, I can't get the freeze to happen with any stress tests. The CPU and socket are both flawless, physically, as near as I can tell.

The motherboard bios settings is admittedly one of the last longshots I have for ideas. I've never previously had issues where motherboards didn't default setting work with CPUs, and I figured if anywhere would know what could be causing the problem, this place would be the answer. The only even similar case I could find google searching of a complete graphics freeze with games still playing music in the background, the person in question found their issue was with having XMP on. Hence my theory of some possible BIOS setting that needs to be switched from default. I've done absolutely everything else I could think could possibly be the issue. I'd have tried the motherboard swap myself previously, but I don't have a motherboard that has compatibility with the processor.

I don't know of what kind of shop would have that kind of service. I live in a more rural part of Canada, and the nearest Canada Computers is over two hours away. Given how erratic the freeze is, it's a bit difficult to just swap a part and immediately know if it's fixed or not, and even then, people HAVE working pairs of this cpu and mb.

I have noticed the BIOS tends to want to set my CPU Cooler Tuning by default to Liquid Cooler, where my actual cooler is a tower cooler. It's literally the only setting from the default I have to change. I've wondered if the BIOS wanting to do this liquid cooler default means some other setting that should only be on for liquid cooled setups is on by default, and I'd need to change it, but the Noctua was the one recommended specifically in the build article for this cpu/mb combo. But again, I'm not having temperatures out of expected ranges, according to NZXT at least.

 

[helper800]

The CPU and Motherboard both have been RMA'd, seperately. The same problem has happened with the combination of:

• first cpu, first mb
• first cpu, replacement mb
• replacement cpu, replacement mb

Given the same problem has happened across them, I can't imagine I've somehow gotten the same defect across two RMAs. Investigating the socket, both in the first and the replacement mb, showed no signs of anything wrong. I've even used compressed air to make sure there wasn't any debris or such in the socket. If there was some actual problem here, I'd figure a stress test that uses every single CPU core should be able to replicate it, yes? The problem is, I can't get the freeze to happen with any stress tests. The CPU and socket are both flawless, physically, as near as I can tell.

The motherboard bios settings is admittedly one of the last longshots I have for ideas. I've never previously had issues where motherboards didn't default setting work with CPUs, and I figured if anywhere would know what could be causing the problem, this place would be the answer. The only even similar case I could find google searching of a complete graphics freeze with games still playing music in the background, the person in question found their issue was with having XMP on. Hence my theory of some possible BIOS setting that needs to be switched from default. I've done absolutely everything else I could think could possibly be the issue. I'd have tried the motherboard swap myself previously, but I don't have a motherboard that has compatibility with the processor.

I don't know of what kind of shop would have that kind of service. I live in a more rural part of Canada, and the nearest Canada Computers is over two hours away. Given how erratic the freeze is, it's a bit difficult to just swap a part and immediately know if it's fixed or not, and even then, people HAVE working pairs of this cpu and mb.

I have noticed the BIOS tends to want to set my CPU Cooler Tuning by default to Liquid Cooler, where my actual cooler is a tower cooler. It's literally the only setting from the default I have to change. I've wondered if the BIOS wanting to do this liquid cooler default means some other setting that should only be on for liquid cooled setups is on by default, and I'd need to change it, but the Noctua was the one recommended specifically in the build article for this cpu/mb combo. But again, I'm not having temperatures out of expected ranges, according to NZXT at least.

When you get your freeze next time do a test for me as it is frozen. Do you have a keyboard that has a light that goes on/off if you press caps lock or numlock? If so when the computer is still frozen press these buttons on the keyboard. If the light turns on and off on the keyboard that likely means that the computer is still running fine in the background but the video has been cut off for some reason. If the lights corresponding to the keyboard button does not turn on and off as you press it, it likely means that the computer is completely frozen even though you may hear music playing. The music in this case may still be playing because the sound chip on the motherboard is unaffected and somehow continues to play the queue music (I don't know I am not an electrical engineer or a computer scientist). In any case when the computer "freezes" try the Windows Key + Control + Shift + B key command. This resets the graphics driver on Windows 10 and may give you back video display to the computer when it freezes.

Since the freeze seems related to graphics I would assume that even though you can stress test the GPU and not have issues their might be some sort of math rounding error or memory retrieval error with the GPU that is intermittent and can cause issues like this. Do you have a different graphics card to test? Though it is curious that it still happens when on the iGPU of the CPU. This can suggest the same such errors on the CPU but with system memory and the CPU's ALUs or some other such subsystem on the CPU.

I also noticed that you have a relatively odd set of ram as it is 2133mhz cl15 ram with a 4 stick kit. When did you get this RAM? is that particular kit of RAM on the motherboards QVLC list of RAM?

 

[SpaceMouse]

When you get your freeze next time do a test for me as it is frozen. Do you have a keyboard that has a light that goes on/off if you press caps lock or numlock? If so when the computer is still frozen press these buttons on the keyboard. If the light turn on and off on the keyboard that likely means that the computer is still running fine in the background but the video has been cut off for some reason. If the lights corresponding to the keyboard button does not turn on and off as you press it, it likely means that the computer is completely frozen even though you may hear music playing. The music in this case may still be playing because the sound chip on the motherboard is unaffected and somehow continues to play the queue music (I don't know I am not an electrical engineer or a computer scientist). In any case when the computer "freezes" try the Windows Key + Control + Shift + B key command. This resets the graphics driver on Windows 10 and may give you back video display to the computer when it freezes.

Since the freeze seems related to graphics I would assume that even though you can stress test the GPU and not have issues their might be some sort of math rounding error or memory retrieval error with the GPU that is intermittent and can cause issues like this. Do you have a different graphics card to test?

I also noticed that you have a relatively odd set of ram as it is 2133mhz cl15 ram with a 4 stick kit. When did you get this RAM? is that particular kit of RAM on the motherboards QVLC list of RAM?

I've tried Windows+L when a freeze happened, and that didn't do anything, but resetting the graphics card seems like a good idea. I'm led to believe the computer is completely frozen and it's not a display issue, as all internet traffic stops (the plug just stays steady green) and streaming software like OBS no longer is broadcasting. iTunes also cuts when a freeze happens, leading me to believe it is a full freeze, but I'll try.

I've had this happen on both my video card, and onboard graphics with the processor. Doesn't seem to affect the rate of a freeze. Graphics card also works in another computer fine with no freezing, doing the same things. If it only had this freeze with the GPU in, I'd suspect it more as the culprit, but the freeze happening with the CPU's onboard graphics has me doubting that, and a new graphics card would be very expensive, even with the market seeming to get better.

As mentioned in the first post you replied to, the RAM is now instead Patriot Viper Steel 32 GB (2 x 16 GB) DDR4-3600 CL18, two 3600 sticks, as the first post advised trying to swap the RAM out. Freeze still happening on the new RAM, which leads me to believe that isn't the issue. Could be some BIOS setting for ram, but the only thing I have on is XMP (which is noted supported by the ram) and previously switching between it being on or off hasn't changed the crash rate. DRAM reference clock is set to auto, CPU IMC: DRAM Clock is set to auto, DRAM frequency set to Auto. Memory try it! is Disabled. DRAM timing mode also set to Auto. Memory fast boot I've tried on both enabled and disabled; did not affect freezes.

 

[helper800]

I've tried Windows+L when a freeze happened, and that didn't do anything, but resetting the graphics card seems like a good idea. I'm led to believe the computer is completely frozen and it's not a display issue, as all internet traffic stops (the plug just stays steady green) and streaming software like OBS no longer is broadcasting. iTunes also cuts when a freeze happens, leading me to believe it is a full freeze, but I'll try.

I've had this happen on both my video card, and onboard graphics with the processor. Doesn't seem to affect the rate of a freeze. Graphics card also works in another computer fine with no freezing, doing the same things. If it only had this freeze with the GPU in, I'd suspect it more as the culprit, but the freeze happening with the CPU's onboard graphics has me doubting that, and a new graphics card would be very expensive, even with the market seeming to get better.

As mentioned in the first post you replied to, the RAM is now instead Patriot Viper Steel 32 GB (2 x 16 GB) DDR4-3600 CL18, two 3600 sticks, as the first post advised trying to swap the RAM out. Freeze still happening on the new RAM, which leads me to believe that isn't the issue. Could be some BIOS setting for ram, but the only thing I have on is XMP (which is noted supported by the ram) and previously switching between it being on or off hasn't changed the crash rate. DRAM reference clock is set to auto, CPU IMC: DRAM Clock is set to auto, DRAM frequency set to Auto. Memory try it! is Disabled. DRAM timing mode also set to Auto. Memory fast boot I've tried on both enabled and disabled; did not affect freezes.

Are you sure that when you RMAed the CPU and the Motherboard that they sent you new ones?

 

[SpaceMouse]

Are you sure that when you RMAed the CPU and the Motherboard that they sent you new ones?

Both were the kind where they sent me a new one in box, and then I mailed them the old one, so yes. And I'm sure the new ones are the ones I'm using now. The CPU is a different batch number, and the serial on the motherboard underside checks out.

 

[helper800]

Both were the kind where they sent me a new one in box, and then I mailed them the old one, so yes. And I'm sure the new ones are the ones I'm using now. The CPU is a different batch number, and the serial on the motherboard underside checks out.

Okay so here is a checklist:
CPU: Sanity checked
CPU COOLER: Foolproof unless its touching the GPU
MOBO: Sanity checked
GPU: Problem occurs on both this and iGPU. I am not sure this completely rules out a video issue. Have you tried changing the PCIe speed of the slot in BIOS to 3.0?
iGPU: Problem occurs on both this and GPU.
PSU: Sanity checked
UPS: Have you tried plugging the computer directly into the wall? Have you tried the computer or the UPS on a different socket/circuit?
RAM: Sanity checked
STORAGE: How many storage devices do you have? Have you tried just the boot drive connected to the computer? Have you tried using a different boot drive?
CASE: What case do you have? Any motherboard standoff screws not used touching the underside of the motherboard? Tried building the PC outside of the case?
MONITOR/S: What are the exact monitors you are using?
CABLES: How is everything connected? Be elaborate. This is for a sanity check.

I really hope that at a minimum I have helped you exclude some potential issues, but you got a doozy of an issue here.

 

[SpaceMouse]

Okay so here is a checklist:
GPU: Problem occurs on both this and iGPU. I am not sure this completely rules out a video issue. Have you tried changing the PCIe speed of the slot in BIOS to 3.0?
iGPU: Problem occurs on both this and GPU.
UPS: Have you tried plugging the computer directly into the wall? Have you tried the computer or the UPS on a different socket/circuit?
STORAGE: How many storage devices do you have? Have you tried just the boot drive connected to the computer? Have you tried using a different boot drive?
CASE: What case do you have? Any motherboard standoff screws not used touching the underside of the motherboard? Tried building the PC outside of the case?
MONITOR/S: What are the exact monitors you are using?
CABLES: How is everything connected? Be elaborate. This is for a sanity check.

I really hope that at a minimum I have helped you exclude some potential issues, but you got a doozy of an issue here.

I haven't messed with the PCIe speed settings at all. I could check to make sure the slot is set to 3.0. I've tried the GPU in two different PCI slots, and it's currently in the recommended one that's reinforced for stability.
EDIT: Checked the PCIe settings. Max Link Speed set to Auto, PCI Latency timer set to 32 clocks, Above 4G memory set to Enabled, Max TOLUD set to Dynamic, Re-Size BAR Support is Disabled, and SR-IOV Support is Disabled. ASPM is all disabled.

For the UPS I've swapped to a completely different, new UPS, as I was worried there was some irregular power issue. I have the UPS software reporting, and it hasn't detected any problems with the power source. I could swap the plug, but my apartment building is a little old (1960s, i think) and there aren't many options for outlets. I'd need some kind of extension cable to get to another one. I'm wary of plugging the computer directly into the wall, as we seem to be in brownout season, and I have a general rule of having every computer on at least a surge protector (even my rbpis).
Storage is the new SSD that has the new windows install, and two WD brand hard drives, from 2016 and 2021, respectively. Both 2TB. One's a backup drive, and the other is used for steam game storage and video recording storage (i stream, and record them, to edit/clip for youtube/etc). The problem was happening on an old windows install that was on another hard drive, hence why I tried a fresh windows install on a new solid state (since the new mb supports solid state). The SSD is in port 2. So the problem has happened between drives.
Monitors are a Dell P2715Q, plugged into displayport, and a Dell U2412M, plugged into DVI. I was previously advised to try installing drivers for both; I did. When plugged into the onboard video, the onboard doesn't have DVI, so I used an HDMI->DVI converter.
CABLES/CASE: The case is a Thermaltake Mid ATX of some kind, gotten years ago, with minimal bells and whistles. I have the power button and status lights plugged into the MB, as well as the front USB ports, both of which work with usb drives, at least. There's a fan installed on the front facing in, a fan installed in the back facing out, and a single fan on the side panel facing in. No idea what the model is; it has a door. From google searching, seems to be one of the Soprano series. The front and back fans are plugged into SYS_FAN ports on the MB, while the side fan is running off a cable plugged into the power supply, one of the cables the hard drives are on. Two power cables running from the PSU to the MB, as well as the big motherboard plug, two seperate power cables running to the graphics card (i was advised to not use the ones that have two plugs on a single cord, but to plug in two cords, so I did). Two cables running from the power supply for hard drive power and the side fan, not on the same cord just because I had the spare cords from the PSU, and I figured it'd make running the cables in the case a bit easier. CPU fan is plugged into the CPU fan port on the MB. I don't think any cable's touching the metal of the case, especially given how many times I've had to take everything out and reassemble. One standoff screw not used, but as near as I can tell, it isn't touching the motherboard, looking with a flashlight. Would that cause some problem if it was?
I have not tried assembling it outside of a case. Not entirely sure what would be a safe surface to do that on. I'd imagine it would need to be something non-conductive and heat resistant... as well as something my cats wouldn't want to get curious about. Last thing I need is a cat deciding the computer was something worth playing with.

 

[helper800]

I have not tried assembling it outside of a case. Not entirely sure what would be a safe surface to do that on. I'd imagine it would need to be something non-conductive and heat resistant... as well as something my cats wouldn't want to get curious about. Last thing I need is a cat deciding the computer was something worth playing with.

You can safely build the PC in the motherboards box. As far as the cats go I don't know what you have available to attempt cat-proofing. The standoff would only be an issue if your motherboard did not have a hole for it but there was a preinstalled standoff in the case touching the bottom of the motherboard PCB.

So as to my suggestions, disconnect any extraneous and not necessary to function devices. This would include all none OS drives, USB devices, and unplug it from the UPS. I suggest keeping it plugged into a surge protector at least. If you can try a different outlet that would be great but if not because you can then it is what it is. Keep only the DisplayPort monitor attached to the PC as well. If it still happens when in this setup then it has to be some part that is causing issues. What does event viewer say when that happens to the PC?

 

[SpaceMouse]

You can safely build the PC in the motherboards box. As far as the cats go I don't know what you have available to attempt cat-proofing. The standoff would only be an issue if your motherboard did not have a hole for it but there was a preinstalled standoff in the case touching the bottom of the motherboard PCB.

So as to my suggestions, disconnect any extraneous and not necessary to function devices. This would include all none OS drives, USB devices, and unplug it from the UPS. I suggest keeping it plugged into a surge protector at least. If you can try a different outlet that would be great but if not because you can then it is what it is. Keep only the DisplayPort monitor attached to the PC as well. If it still happens when in this setup then it has to be some part that is causing issues. What does event viewer say when that happens to the PC?

Event viewer doesn't say anything other than an unexpected shutdown, timestamped for when I hold the power button down. That's part of what has made this so hard to diagnose, beyond the erraticness in causing it. Windows can't tell the computer's crashing. There's literally nothing in the logs led up to the "The system has booted without cleanly shutting down first", hence why I'm so frustrated with this.

Motherboard box is a great idea.

 

[helper800]

Event viewer doesn't say anything other than an unexpected shutdown, timestamped for when I hold the power button down. That's part of what has made this so hard to diagnose, beyond the erraticness in causing it. Windows can't tell the computer's crashing. There's literally nothing in the logs led up to the "The system has booted without cleanly shutting down first", hence why I'm so frustrated with this.

Motherboard box is a great idea.

If there is no event in the event viewer when the display cuts out and music still plays it seems to me that the display is just cutting out and the computer is still running but without display as was said before. Have you tried switching the DisplayPort cable to another one? Air blast the ports on the monitor and GPU sides. When the display cuts out try disconnecting and reconnecting the DisplayPort cable.

 

[SpaceMouse]

If there is no event in the event viewer when the display cuts out and music still plays it seems to me that the display is just cutting out and the computer is still running but without display as was said before. Have you tried switching the DisplayPort cable to another one? Air blast the ports on the monitor and GPU sides. When the display cuts out try disconnecting and reconnecting the DisplayPort cable.

Clarification: only music in games keeps going. Itunes immediately cuts, as I've said. There's no response to keyboard or mouse input in the games while it's in this state; Minecraft, for example, keeps playing the background music, but won't make any active sounds if I click/spacebar/etc. And as I said, programs like OBS immediately stop broadcasting, and all network activity cuts. Youtube also abruptly stops, with a static 'pop' sound, if I had to describe it.

 

[helper800]

Clarification: only music in games keeps going. Itunes immediately cuts, as I've said. There's no response to keyboard or mouse input in the games while it's in this state; Minecraft, for example, keeps playing the background music, but won't make any active sounds if I click/spacebar/etc. And as I said, programs like OBS immediately stop broadcasting, and all network activity cuts. Youtube also abruptly stops, with a static 'pop' sound, if I had to describe it.

There is nothing specific enough for me to go on that wont lead down the path of more circular reasoning. I think I pretty much covered everything I can think of. What you need is more exposure to the issue here and potentially a shop to help you out or more parts to swap for to see if there is an issue only with one of them. Please make sure to post back if you ever figure out the issue, its a truly vexing one.

 

[SpaceMouse]

Well, I haven't had a freeze since tuesday, and I've had three successful multi-hour streams since then, so knock on desk it seems to be solved. For the record, what I needed to do was disable Intel C-States in the BIOS. Setting them to disabled has (so far) prevented any freeze. I'll provide an update if anything changes, but for anyone in the future having this problem, try that?

 

[i512600k z690]

Have You tried to set MaxTolud manually starting from 2.5Gb without dynamic ?
Try and come back with the result.
Otherwise if you would like to check whayt is max tolud there are two separate manuals dedicated for i5 and z690 chips each over 1000 pages trying to lecture from there is like a 10000 pieces puzzle with only minions on the image ... absolute fun ........