Gaming PC instabiliity

[Sydneyblue]

Im trying to find what is causing it. The first system i built was around November 2024
12700kf hyper 212 cooling
Z790 Tuff DDR latest bios
DDR5600 Fury 16gbs and *6400 renagade 32gbs hardlocked with both in running at xmp profiles
*ROGthor 2 platinum 1000w
*Asus TUFF 4070ti super OC
Samsung 1tb NVME gen 4

I was getting the buzzing and restarting in cyberpunk and BF2042 so i returned the mobo cpu and NVME
I keep the PSU and GPU
so after christmas i grabbed
14700k with Ryujin III 360 latest bios
Z790 E strix DDR5
*6400 renagade ddr5 32gbs
*ROGthor 2 platinum 1000w
*Asus TUFF 4070ti super OC
Samsung evo plus 2 tb

couple weeks i blue screen while optimizing shaders in COD black ops 6
yesterday i keep crashing on the shader screen in hogwarts legacy

I order a exact replacement PSU ROGthor 2 platinum 1000w and hogwarts loading fine but i decided to double check the older thor and hogwarts loaded fine so im like wtf. Also been getting usb issues where i would get the error where the usb didnt reset.

Im completly stumped on this . I looked into the case and didnt notice anything weird like a cap bulging.
Any help appreciated

 

[FreedomEclipse]

SInce youre running at 6400mhz on the ram. kick it back to 6200 or 6000 and see what happens. 6400 could be putting a lot of stress on the memory controller which could be where your instability is coming from.

 

[stahlhart]

Did you stress test either build for stability, and if so, how?

 

[Toothless]

Check your radiator and feel if one side is getting really hot. If it is also check thermals with hwinfo64. The Ryujin coolers require a software update to run properly otherwise you'll hit 100c near instantly.

 

[Sydneyblue]

SInce youre running at 6400mhz on the ram. kick it back to 6200 or 6000 and see what happens. 6400 could be putting a lot of stress on the memory controller which could be where your instability is coming from.

Im going to try that if it crashes with new psu in

Did you stress test either build for stability, and if so, how?

I use occt i run it for about hour

Check your radiator and feel if one side is getting really hot. If it is also check thermals with hwinfo64. The Ryujin coolers require a software update to run properly otherwise you'll hit 100c near instantly.

I know escecially with the 14700k lol

 

[FreedomEclipse]

Im going to try that if it crashes with new psu in

Dont worry about new PSU. You probably didnt need it. New PSU takes time putting in. Downclocking ram can be done in less than a minute. Youre just creating unnecessary work for yourself.

This is also part of the diagnosing process. If it turns out that it was the memory controller that couldnt handle the memory OC. You've just wasted money on an expensive PSU that you didnt need that could have been sent back for a refund. Its not a cheap powersupply either.

 

[Bill_Bright]

Im going to try that if it crashes with new psu in

Since EVERYTHING inside the computer case depends on good, clean, stable power, verifying you are providing it is not a waste of time. In fact, verifying good power as one of the first steps is standard hardware troubleshooting 101.

Computers should run find and stable at their default settings. So I recommend you reset and return all settings to their out-of-the-box, default setting and see how your system runs.

If still having stability issues, then I agree with Toothless and you should check your temps. I use and recommend Core Temp to monitor CPU temps in real time. Under Options > Settings > Notification Area, I have mine set to display "Highest temperature" only.

HWiNFO64 is also very (if not overwhelmingly) informative. When using for the first time, I recommend you select “Sensors only” when prompted. Then scroll down to your motherboard’s section to see the temps.

HWiNFO64 also includes an excellent notification/system “Tray Icon”, though it is a little more challenging to setup than CoreTemp because of the massive amount of information and options provided by the program. Under Sensors Status > Configure Sensors (the gear icon), select the Tray Icon tab, scroll down to and click on “Core Max” to highlight. Then click on “Show in Tray”. Then OK to set. If you want HWiNFO64 to run all the time you will need to select “Auto Start” from the "Main Settings" window.

If your temps are good and still having stability issues, then you need to verify power.

That said, not sure what you meant when you said this,

I order a exact replacement PSU ROGthor 2 platinum 1000w

Did you already order this? Have you received it? Or do you have another PSU to try?

I would not recommend "buying" another PSU at this time. But if you have a spare on hand or can temporarily borrow one from another computer or a friend, then I would do that. Swapping in a spare PSU for testing purposes is easy. If careful, you don't even have to unmount the old or remove it from the case. If the computer runs fine at default settings with the second PSU, then consider buying a new (or returning the first is still under warranty).

 

[DarkDreams]

couple weeks i blue screen while optimizing shaders in COD black ops 6
yesterday i keep crashing on the shader screen in hogwarts legacy

That sounds like typical raptor lake instability. Is your mobo bios up to date? Running intel defaults or any overclocking/undervolting? I would also check ac/dc loadline to make sure the bios isn't doing any "hidden" undervolting.

 

[FreedomEclipse]

That sounds like typical raptor lake instability

12th Gen Alder lake.

 

[DarkDreams]

12th Gen Alder lake.

Was replaced with a 14700K if I understand OP correctly (2nd paragraph)

 

[FreedomEclipse]

Was replaced with a 14700K if I understand OP correctly (2nd paragraph)

I missed that. You were right. I was still working off the basis that he had the KF. If its the 14700k then it could very much be degradation.

 

[Sydneyblue]

Since EVERYTHING inside the computer case depends on good, clean, stable power, verifying you are providing it is not a waste of time. In fact, verifying good power as one of the first steps is standard hardware troubleshooting 101.

Computers should run find and stable at their default settings. So I recommend you reset and return all settings to their out-of-the-box, default setting and see how your system runs.

If still having stability issues, then I agree with Toothless and you should check your temps. I use and recommend Core Temp to monitor CPU temps in real time. Under Options > Settings > Notification Area, I have mine set to display "Highest temperature" only.

HWiNFO64 is also very (if not overwhelmingly) informative. When using for the first time, I recommend you select “Sensors only” when prompted. Then scroll down to your motherboard’s section to see the temps.

HWiNFO64 also includes an excellent notification/system “Tray Icon”, though it is a little more challenging to setup than CoreTemp because of the massive amount of information and options provided by the program. Under Sensors Status > Configure Sensors (the gear icon), select the Tray Icon tab, scroll down to and click on “Core Max” to highlight. Then click on “Show in Tray”. Then OK to set. If you want HWiNFO64 to run all the time you will need to select “Auto Start” from the "Main Settings" window.

If your temps are good and still having stability issues, then you need to verify power.

That said, not sure what you meant when you said this,

Did you already order this? Have you received it? Or do you have another PSU to try?

I would not recommend "buying" another PSU at this time. But if you have a spare on hand or can temporarily borrow one from another computer or a friend, then I would do that. Swapping in a spare PSU for testing purposes is easy. If careful, you don't even have to unmount the old or remove it from the case. If the computer runs fine at default settings with the second PSU, then consider buying a new (or returning the first is still under warranty).

Temps from what i have seen are very good under 80c under heavy stress R23 and OCCT
I did get the new PSU and popped it in. should i send it back ?

 

[ShrimpBrime]

Update the bios and ME firmware. Then update the ME software.

 

[Sydneyblue]

Update the bios and ME firmware. Then update the ME software.

The bios is latest

I guess what i was thinking was since i had the same psu on both systems and both systems frooze then it had to be the PSU and not her ram because i did get the freeze with the 5600 Fury kit and the 6400 kit

 

[phanbuey]

can you see your event log and see what errors are there?

 

[Sydneyblue]

can you see your event log and see what errors are there?

hyperviser failed to migate

 

[Vya Domus]

RAM or CPU instability. Run the RAM at 4800Mhz, if it still happens I'd assume it's the CPU.

 

[Bill_Bright]

I did get the new PSU and popped it in. should i send it back ?

If the problem occurs with both PSUs, then either send this one back or keep it as a spare/test PSU. Or sell it.

 

[ShrimpBrime]

The bios is latest

I guess what i was thinking was since i had the same psu on both systems and both systems frooze then it had to be the PSU and not her ram because i did get the freeze with the 5600 Fury kit and the 6400 kit

OK, is the ME firmware updated? Windows ME software updated also?
Fresh windows install?

Intel® Management Engine Drivers for Windows 10* and Windows 11*

Intel® Management Engine Drivers for Windows® 10 and Windows 11*

www.intel.com

 

[Sydneyblue]

OK, is the ME firmware updated? Windows ME software updated also?
Fresh windows install?

Intel® Management Engine Drivers for Windows 10* and Windows 11*

Intel® Management Engine Drivers for Windows® 10 and Windows 11*

www.intel.com

Thanks for the reply the issue wherei would get a hardfreeze with buzzing happened on 2 different systems with the same PSU GPU so at the moment im testing wit hthe new psu to see if the issue happens

 

[Princess Garnet]

14700k

couple weeks i blue screen while optimizing shaders in COD black ops 6
yesterday i keep crashing on the shader screen in hogwarts legacy

My first suspicion would be there. Crashing during shader compilation or error messages about not enough memory to allocate a rendering resource are usually the telltale circumstances/symptoms of instability/degradation on those Raptor Lake CPUs.

For clarity, it ultimately means the CPU is unstable or the RAM is too high of a frequency to be stable, so this occurrence is by no means exclusive to Raptor Lake systems (I saw someone with a Zen 3 system having crashing during shader compilation in Silent Hill 2, and lowering RAM speed resolved it). But when it happens on a Raptor Lake CPU, it's a Red flag since they have their known issues, and this is usually the first circumstance they expose themselves.

Test bt reverting to JEDEC RAM speeds (disable RAM XMP profile) first.

In other words, do this...

Computers should run find and stable at their default settings. So I recommend you reset and return all settings to their out-of-the-box, default setting and see how your system runs.

If your issues go away with RAM at lower speeds, I would suspect your CPU may have degraded and is no longer stable at higher RAM frequencies because one of the "resolutions" to the CPUs that have degraded was precisely that; to lower RAM speed to around as lo as 4,800 MHz (the other was to limit boost frequency down to the lower 5 GHz range).

That being said, these could also be a cascading result of a broader instability (such as a poor PSU). Random freezes/reboots have an endless list of causes.

 

[eidairaman1]

GPU possibly since thats the only thing common between both and its locking up

 

[Sydneyblue]

GPU possibly since thats the only thing common between both and its locking up

I am Currently testing the PSU with the replacement so that would be the only thiing left

 

[phanbuey]

can you turn off virtulaization and see if the crashes persist.

If so there are some registry settings you can use to stop the crashes.

 

[RJARRRPCGP]

SInce youre running at 6400mhz on the ram. kick it back to 6200 or 6000 and see what happens. 6400 could be putting a lot of stress on the memory controller which could be where your instability is coming from.

Sounds like possibly fatal WHEA for "Cache Hierarchy Error" or other error that points to RAM.