Hello everyone,
I have a problem with my new build PC.
It crashes after benchmarks and in games. If I play games like Farming Simulator 22 or do benchmarks like OCCT, Windows shows BSOD and the PC restarts. Sometimes I got the massage "CPU overheating alert! Please check CPU cooler is firmly attached for working properly.". The benchmarks themselves run without any problems, but just after stopping the benchmark the PC restarts.
I'm using the AMD boxed cooler and didn't overclock my CPU.
If I set the PPT to 45W the PC runs without restarts, but if I go 60+ Watts the PC crashes.
In FS22 the PC crashes at 65-70°C and in benchmarks it runs on every temperature up to 95°C but only crashes after stopping the benchmark.
What I've tested:
reinstalling Windows
reinstalling CPU cooler with new thermal paste
updated BIOS
tested XMP on/off
newest AMD chipset drivers
changed PSU from Thermaltake Smart BX3 650W to Be Quiet Pure Power 12M 850W
My system:
AMD Ryzen 5 5600X with boxed cooler
MSI B550 A-Pro
AMD Radeon RX 7700XT 12G (Gigabyte OC)
Kingston Fury Beast DDR4 16GB dual kit 3600MHz
Kingston NV2 NVMe PCIe 4.0 SSD 1TB
Be Quiet Pure Power 12M 850 W
AeroCool Designer V1
Windows 11 Pro
I'm sorry for imperfect english, it isn't my mother tongue.
Thank you for any help.
Is the AMD stock cooler really so bad, that it can't handle the heat of of my CPU?
Or can I change any settings to solve my problem?
The 5600X comes with the Wraith Stealth, the lowest end version of AMD box cooler. However it should still be adequate to keep the CPU under 95°C and the default PBO power and thermal limits take care of the rest.
The message you report, "CPU overheating alert! Please check CPU cooler is firmly attached for working properly." is not an AMD generated alert. This is likely being generated by some other software, possibly a utility that came from your motherboard manufacturer?
If you are running software from your motherboard manufacturer or any other third party utility that monitors system temperatures check to see if there is a high temperature warning setting and either disable it or set it higher than your max temp of 95°C.
Or purchase a better cooler.
I am with you, FunkZ! I would go further and advise the OP to uninstall any utilities especially those from the MB vendor. Install the latest Ryzen Master (RM) from here. Post a screenshot of RM running Cinebench R24 multicore. Do a Clear CMOS before running the test but remove all poor quality utilities. John!
Hi, I've completely reinstalled Windows and I've omitted every unnecessary utilities.
I've posted some screenshots of my testruns.
It looks like if I enable PBO manually in the BIOS the system goes so hot that the cooler can't handle the temperatures and after stopping the benchmark my PC crashed.
But if I have PBO on Auto it runs fine in this benchmark.
@LP-700 your initial post stated you "didn't overclock my CPU"
Your first Ryzen Master screenshot shows when your PPT is limited to the stock 76W the processor is stable. It limits power which keeps temperature under control.
The second Ryzen Master screenshot shows you have raised the PPT to 1000W which effectively removes the power limit, allowing the CPU to boost up to 95°C which is not stable. This is overclocking your CPU!
Either leave the BIOS settings stock and enjoy a crash-free stable system, or purchase a better cooler if you want to try overclocking. To reduce temperature you can also try setting Curve Optimizer to a negative value to reduce voltage.
This isn't 100% correct. Firstly, I didn't know that switching the PBO via one button click from Auto to Enabled counts as overclocking. But seccondly I've posted another reply which shows a picture of another benchmark called OCCT and there you can see, that the temperatures are rising to around 80°C and after stopping the benchrun my PC crashed but it was limited to 76W. So it isn't crash free
Disable the boost in the bios otherwise this cpu will melt itself and the mainboard to death. Those high temps can cause instability under certain circuumstances. Amd decided to make the cpu boost to and sometimes above the max temp - and the performance gain vs no boost is like 15% at best. Its not worth it. Also 90+ C will significantly shorten the life of the cpu itself and the mainboard because all the condensators around the cpu will be under higher temperatures resulting in faster drying up which means then failing. Under good temperatures a mainboard can last up to 10+ years - but when you run a cpu like this with boost the mainboard will probably last 3-4 years at best - and if it has a little quality issues due to some older condensator used during manufacturing you look to below 2 years - specialy asus mainboards are known to fail early. But asrock mainboards (which belongs to asus) never failed me and ran strong for many many years! (Probably due to the higher sales count - similar to buying vegetables in a cheaper store results often in fresher ones.
gulabon2, do you speak for AMD? How do you know what processor temperature will shorten the life? I suspect you do not know and suggest you refrain from making absolute statements. Enjoy, John.
I've also runned another benchmark which stresses nearly all components of a PC. The temperatures slowly rise to 80°C and seem to stabilise at this temperature. After I stopped the benchmark the PC crashed. Are 80°C really that dangerous for the CPU?
LP-700, I doubt it is the temperature but cannot guess with no evidence. Please look in the Event Viewer and post screenshots of a few of the Critical Error Details tab. Please post no more images of Cinebench. We've seen a zillion of them, simply post the score. Your processor is throttling due to PPT and EDC but also temperature in one case. Have you modified the limits of these values? Do a Clear CMOS and stay out of BIOS or any OCing. Some of this may be a remainder of the crummy utilities you have been running. John.
There are many of these first Errors the others there are only one or two.
A Clear CEMOS means to remove the battery from the mainboard - right?
Do I have to update the BIOS after this and do I need to reinstall Windows?
Thanks, LP-700. I need to translate to English to make sense of it. The first term (fehler) translates to Mistake. I think this is Error not Critical.
Is the same as Critical? Please use Critical in the search. And post a screenshot of the Details tab at the bottom. Clear CMOS should just delete the changes to the BIOS and you should not need to install a fresh copy of BIOS or Windows. The Clear CMOS procedure should be in the MB manual. My board has a button on the back to do a Clear CMOS, yours may not. I never know about these system builders and what they do. John.
Oh sorry that was my fault, I have attached the pictures below with Critical.
I have a flash button on the back of my MB, could this be the same?
The Critical Error in the first picture, I've got about 20-30 times, the other one only once.
Thanks, LP-700. OK, we are making progress. You are getting "4502 WinREAgent". Which is probably taking your system down. I will do some more looking at these errors but please ask your builder about this.
Are there any restrictions on what you can do like installing Windows or clearing BIOS? Please recreate the error and filter the output for the Critical errors and post a few, including the Details tab. Let's see what we can learn. Thanks, John.
My builder is me and he has no glue about this anymore
Clearing BIOS and installing Windows I've done before but it haven't helped me. If I dont't missunderstand you.
I have already posted the critical Errors, they are all saying the same besides this one critical i've posted.
Hey I've done a clear CEMOS but it hasn't solved the problem.
In the first testrun, my PC haven't crashed (this run was with case open, temps around 70°C max 74°C).
In the seccond one my PC crashed again (this time with case closed, temps around 80°C max 82°C)
How can it shut down at 80°C, Windows showed "WHEA_UNCORRECTABLE_ERROR"?
LP-700, I am here but real tied up. I need to see some of the Event Viewer BASIC tab, and a shot of RM when at about 80C and up, I will be back to thinking soon. John.
Hello I've also struggled with this problem for a year now and my only solution was to limit my power usage I believe it is the CPU being faulty and not you're system itself because I've changed literally every part but my CPU and this problem still occurs I've read through multiple reddit post watched videos and amd support pages none of them have helped if anyone actually has found a fix for this I'd gladly appreciate if you'd tell everyone because this has been a problem since last year the reset thing only started happening after a couple of months of having my computer till this day I still limit my power usage just so it doesn't randomly restart while I'm using my computer
joshuaaa, with no real information I cannot even guess. If you would like some help, post all your parts and symptoms as well as screenshots of the Details tab of Event viewer Critical errors into to a newly opened thread. John.
B550M PRO-VDH
Ryzen 5 5600x
RX 6650 XT
700W Thermaltake Smart 80+
32GBS DDR4 Corsair 3600mhz
about 2-3 hours in it restarted again just gonna set it back to low power mode for now
read through some more forums and some mentioned changing the power options in bios to "Low current idle" if you have an old psu supposedly that could be the culprit so far hasn't restarted will update after a couple of days of just leaving my computer on
Hey, I've tried "Low current idle" but it has only made it worse, now after about 1-2 minutes in the benchrun my screen turns black, the rgb of my mouse and keyboard turn off and the fans of my graphics card stop spinning. The fans in my case and my CPU cooler are spinning. The CPU cooler continues spinning at full RPM so I think the benchrun continues. The only thing I can do is to shut down my PC via a long push on the power button.
I had this problem once before about 2-3 weeks ago but it has "solved" itselfs.
I have resetted the setting to which it was before "Auto" but the problem continues.
What can I do now? Can the motherboard be the problem?
I would stop messing around with the settings at this point and upgrade the cooling solution. Spend $100 USD and get yourself a decent AIO cooler (two or three fan, which ever fits in your case) and you'll see your CPU temperature drop significantly. You have a decent PSU already.
Yes I understand, but my CPU temperature limit is at 95°C so it shoudn't shut himselfs down at 80°C. You often see vidoes where CPU's run at 100°C+ (which isn't good I know) but they should throttle themselfs and they don't shut down.
yea I can't help either I'm stuck in the same situation
Which situation, has this setting made the problem also worse or hasn't the setting made any impact on your problem?