Hello, I built a PC a month ago. But it freezes at idle. Please, help. I would appreciate any fresh ideas.
Here are the components:
Main:
Storges:
Cooling:
Base:
The PC freezes tightly in the following cases:
From the moment I started trying to fix the PC, it began to freeze less often. 3-4 dead lags per day turned into 1-2 per day. (I think disabling DOCP and PBO2 Tuner affected stability). At some point, the PC worked for almost 5 days without freezing. (Active load in games, stress tests, and downtime at night). PC never freezes under gaming load (0 crashes during gaming/stress testing in 150+ hours). There are two types of critical error in the Event Viewer (critical_error_1.txt, critical_error_2.txt
After restarting the PC, it works like nothing happened, only reports in the Event viewer
I have tried the following methods to correct errors:
At the moment, I replaced the RTX 4080 FE with an RTX 3060 Gigabyte EAGLE, and installed the 4080 in an old PC.
Components of the old PC:
Main:
Storges:
Cooling:
Base:
For two days, both PCs have not yet lagged. I actively use only the build on the ASUS ROG STRIX B550-I, and the second one is just idle. But in the build with 4080, two errors appeared in the Event Viewer (error_1.txt, error_2.txt)
Temperatures:
Here is my link to the google drive folder with error descriptions:
https://drive.google.com/drive/folders/1HONt4A5yT19fpWi62mP6FenyL3x8aZXk?usp=share_link
Solved! Go to Solution.
Finish!
What a joke! I spent 6 months looking for a faulty component in my build. I finally fixed it by replacing the Strix board with an Aorus Pro AX. In my use case scenario, crashes only appeared once a week, and my longest streak without any issues was 40 days. This made my repair process much longer. And to top it off, I only found this thread today.
Updated the motherboard bios ?
Can also be the PSU not providing enough power, since you have Corsair SF750 750 W
Hi Macbang,
1)Yes, the last available one is now (2803). I even tried downgrading to a previous version (no result)
2)I checked the required PSU level on the manufacturer's website - exactly 750 W. 5800x3D and 4080 FE at the peak consume 60 + 320 W together.
Ya-Ya, I will work with you but I start by recommending that you do a fresh install of Windows including formatting the C: disk. After installation install only AMD drivers and Ryzen Master. There seems to multiple conflicts between values set in BIOS and those set by Ryzen Master (RM) and users start having problems doing both. In general I advise users to install ONLY BIOS from the MB vendor. Please post screenshots of RM both Basic and Advanced views running Cinebench R23. Images of RM only and if you use the Camera icon to post the images they are much sharper in the forum.
Please try this and after we corner the problem do what ever you want. Thanks and enjoy, John.
Hi John,
1) I did a clean install of Windows, formatting all disks (a single m2 module)
2) Installed google chrome, AMD Ryzen master, amd_chipset_software_4.11.15.342. I updated windows in the System -> "Update & Security" tab (Without installing optional drivers)
3)Attached file RM_CB23.bmp, RM2.bmp or google drive link https://drive.google.com/drive/folders/1HONt4A5yT19fpWi62mP6FenyL3x8aZXk?usp=share_link (Same file name)
All other RM tabs are similar to the one on RM2.jpg, I didn't change anything there
The last error (happened today) was on a system with bios by default(2803, latest on the asus site), never installed Ryzen Master, never installed PBO tuner. DOCP was also disabled in the BIOS.
Before a clean reinstall, Windows passed the GPU, PSU, CPU test in OCCT for 10 minutes. During CB23, all other stress tests and gaming loads (200+ hours) I never experienced any problems. In addition, crashes occurred when this processor was under water cooling.
Also in the google disk folder you can find the DxDiag.txt file, which I made in the DXDIAG.exe program about an hour after the last computer freeze.
Thanks, Ya-Ya. All looks fine to me except for the temperature and that is why it is throttling. What cooler are you using? I suspect that is your problem. Enjoy, John.
I use ALPENFOHN Black Ridge
Run DXDIAG.exe and Save the file to your computer. Open the saved DXDIAG File and go to the very last category.
There is will list all the files including Windows files that are causing problems in your PC.
The Event Viewer code 41 is typical when the PC shuts down by itself unexpectedly.
Try running the freeware OCCT Stress test for your CPU, GPU, & PSU and see if any of the test fails or you see anything abnormal in Temperatures or PSU outputs.
NOTE: When you installed Windows again was it a Clean install or a In-Place Windows Repair type install?
If it was a Clean Windows install and before adding any 3rd party apps did you have the freezes? If you did then it is a hardware issue.
I doubt the freezes are a 3rd party issue but to eliminate that part boot your computer into a Windows Clean Desktop and see if the freezes continue or not.
It's very easy to do and undo. Here is how to do it: How to perform a clean boot in Windows
EDIT: I would also open a AMD SUPPORT - WARRANTY ticket and upload your thread if they believe your CPU should be RMAed to be checked or not: https://www.amd.com/en/support/contact-email-form
Thanks for a lot of useful information.
I contacted AMD warranty. I will try to look for a problem in the software part of the PC for some more time.
But what to do in case of failure? I don't understand what component is causing the problem, because under load everything looks fine. Should I try to send each component for testing under warranty? Or contact a service center? (I do not understand how they will fix the problem, which manifests itself once every 5 days). I would be grateful for any thoughts.
Looking at your Ryzen Master as noted your CPU is overheating and reaching it maximum operating temperature of 90c.
So I would try to get that fixed first before doing anything else. Overheating can cause freezing issues at times.
I saw you posted your German Low Profile CPU Cooler. According to PCPARTPICKER it is compatible with the Ryzen 58003D but looking at the CPU Cooler's Support site showing the specs that CPU Cooler is not strong enough for your Ryzen. It is too weak: https://www.alpenfoehn.de/produkte/cpu-kuehler/black-ridge
The Ryzen 58003D has a TDP Rating of 105 watts but according to your CPU Cooler specs it is made for a CPU rated at a TDP of 95 watts:
You will need to upgrade to a CPU Cooler rated at a minimum TDP of 150 Watts or higher to keep your processor from overheating.
Noctua has some low profile strong CPU Cooler but they are not cheap in case your have a Small PC case.
So I would wait first before starting any Warranty service until you get a proper CPU Cooler.
The best way to check the processor and eliminate the Motherboard is by installing the CPU in another compatible motherboard and see if the same thing occurs or not.
NOTE: A User recently turned in a 58003D under Warranty and AMD is going to send him an upgrade 5900X processor because they didn't have any in stock at that time.
But to answer you question when you are not sure if a CPU or Motherboard is defective I would return both for replacement only to save time.
But if you have time you can return the hardware you believe it most likely defective and wait for a replacement to see if it fixes your problem.
EDIT: There is one point I would like to make, If the PC freezes at idle then it is most likely it wasn't overheating at that time so that can be eliminated unless it was at 90c at idle.
Hello, I recently noticed a strange behavior of the computer during idle time. The problem was the WinSAT process, which used 100% CPU as soon as I stopped controlling the computer (even if a game was loaded or a video was played). The scenario of WinSAT fully coincides with all cases of the problem.
Due to the rarity and spontaneity of the problem, I am forced to wait 30 days with the current configuration to be sure the problem is resolved.
Thank you all for your attention, I will be back in a month to confirm the success of the repair or to continue the repair with you
I tested the system with ARCTIC Liquid Freezer II 240 (200+ TDP) cooler. Exactly the same problem. The processor in stress tests heats up to 75 degrees maximum.
So I will begin the process of identifying the defective component.
to eliminate a 3rd party startup app or driver boot into a Clean Windows Desktop which is similar to running Windows in Safe Mode except after Windows loads.
Here is how to do it: How to perform a clean boot in Windows
Also try changing your Windows Power Plan to "Balanced" with:
Processor State Minimum: 5%
Processor State Maximum: 100%
see if that makes a difference to your CPU frequency and Temperature.
But personally your CPU Cooler should be able to prevent your CPU from going over 90c. Since it is reaching only up to 75c that is well within your CPU maximum operating temperature and show that your new CPU Cooler is doing its job.
EDIT: after exhausting all troubleshooting methods open a AMD SUPPORT - Warranty ticket to see if the CPU needs to be RMAed to be checked or replaced by AMD: https://www.amd.com/en/support/contact-email-form
solved? amd just suck. i had this problem few years ago with 3600 ( every zen cpu has this problem). amd just too bad, because when idle windows put it in low power consumption state, amd cpu dropped power and it just freezes ( rest of the world still running with no issue) . there is no specific fix for this, just undervolt your cpu and don't use amd's power plans
Hi!
I also own a 5800x3d
Asus x570E
2x Corsair 3600cl18 16gb
Asus TUF 4090.
I experience this exact issue for a few months now. Its driving me nuts!
No useful info in event viewer, tried basically everything you listed too. Clean install everything, nothing solves this.
I play 'assetto corsa' often and use 'content manager'. For some reason it happens a lot when using this. (no load in contentmanager, its just a menu basically).
Everything seems to work normal like u say at all times and then randomly locks up. I keep hearing everyone in discord and they still hear me, but about everything else locks up. Its so weird and annoying. Did you find anything yet? Or did amd reply to you?
Many of my friends run corsair memory too and report issues too. My friends pc turns off with no error or event log whatsoever and someone else has bluescreen but instant reboot. All am4 with nvme C drive too.
Hello, the problem still exists.
I decided to change the GPU, CPU and motherboard under warranty. (At the moment I sent the components, and I'm waiting for a replacement)
Nvidia replied that they have a couple of similar cases with the B550 Asus rog strix.
Also Nvidia's comments about the crash dump:
"Going through the crash dump file, a TDR was reported.
The issue happens due to TDR short for Time Out Detection Recovery which is a feature of Windows, which detects response from GPU and recovers to a functional desktop by resetting the card. While the system seems to be "frozen", in reality it is processing a command, which might have lead to Overloading the GPU.
Although TDR is a safety mechanism, it may be caused due to underlying issues such as Faulty power supply, corrupt drivers and operating system, issues with the game file, overclocking, low power received on the GPU or faulty hardware."
I suspect this may be a component compatibility issue. If the replacement of components does not help, then I will try to change the manufacturers of components.
Hi,
I'm also running 5800x3D on the b550i strix board, but with a 4090. Did you find out which component causes the issue? I'm still in the return period so could potentially save me a lot of trouble...
Hi,
I just replaced GPU, CPU and motherbord. Still testing it.
Most likely it's CPU, because each PC crash with the first processor looks one to one, like a PC crash when setting the PBO curve too aggressively.
By the way, my already confirmed 5800x3d replacement was heavily delayed until I asked the support center what the problem was. They decided not to notify me that they didn't have a replacement in stock and just waited. After my letter with a question, they immediately offered me a replacement for the 5950x
Just after replacing the GPU, CPU and motherboard, the problem reappeared.
Very strange, considering that the problem did not occur when I used all the other components with a different base for a month (i5-10400 + RTX 3060 + Z590)
I see two possible problems:
1) some random incompatibility
2) tightened bolts on the board
Hello! So you find solution?
Or just change your motherboard to the B550I AORUS PRO AX
So far my PC has crashed twice in 40 days, which I will consider a success.
I wasn't able to fix the problem completely, but I was able to reduce the number of freezes by removing all extensions from Chrome.
I found a forum full of users with the same problem (RTX 40 graphics card with different components). It seems that the problem is triggered during some internal processes, but in some scenarios it occurs especially often.
From now on, I stop looking for solutions in the hardware and software area, and will just wait for driver and bios updates that should increase the stability of the system.
Finish!
What a joke! I spent 6 months looking for a faulty component in my build. I finally fixed it by replacing the Strix board with an Aorus Pro AX. In my use case scenario, crashes only appeared once a week, and my longest streak without any issues was 40 days. This made my repair process much longer. And to top it off, I only found this thread today.