cancel
Showing results for 
Search instead for 
Did you mean: 

Processors

Kilbot192
Adept I

Fatal hardware error has occurred - Ryzen 9 5950X

 

image.png

 

I keep having this error after crashing randomly. No messages or blue screen. My pc just turns off and boots up again. I've tried resetting the whole pc and reinstalling windows and resetting my BIOS to optimized defaults but I still have no luck solving this problem. Everything should be set to default in Adrenalin and Ryzen Master. I'd greatly appreciate some help with this.

0 Likes
40 Replies
misterj
Big Boss

Kilbot192, please post the Details tab of your screenshot. Thanks and enjoy, John.

0 Likes

Kilbot192_0-1715129495736.png

Kilbot192_1-1715129523079.png

0 Likes

Thanks, Kilbot192. Please look in C:\Windows\Minidump and if it is not empty, please compress it and post it where I ca see it. Thanks and enjoy, John.

0 Likes

There was no such folder there at all.

0 Likes
432hz
Adept III

Verify your RAM isn't faulty with MemTest86 (can take a few hours to run depending on how much RAM you have).

 

RX 6800 • Intel 12600k • G.Skill 6000 32GB • MSI B760M Mortar
0 Likes

I don't think I can run this because I don't have a usb drive or anything like that. But some time ago I was messing around in the BIOS and I tried overclocking the RAM. I just increased the clock speed of the RAM to the value that is in the name of the product. The speed was reading something like 2,000 Mhz so I set it to 3600 as it's advertised. I think this is when I started having problems. But I've reset the BIOS from the BIOS menu and also held down the little button on the motherboard itself. I tried to undo what I had done but something is still wrong.

0 Likes

You found your culprit    ; - )


Either: Dial down memory speed a bit, loosen up the timmings or if you feel bold, increase vSoC or DRAM voltage just a tad bit. But if you are uncomfortable doing the latter, don't do it as you can damage something in the process.

Try using the XMP/DOCP profile instead of just increasing speed.


Good Luck

The Englishman
0 Likes

I selected XMP profile 1. Did another reset as well but I disabled the MSI driver installer that auto starts. I just had another crash but the event viewer just said that it was unexpected and didn't shut down properly but nothing about the processor core this time. 

0 Likes

Kilbot192, I suggest you remove all MB vender software, do a Clear CMOS and get all your AMD drivers here. Drivers from MB venders are often down level and sometime modified. Thanks and enjoy, John.

0 Likes

Kilbot192_0-1715178744938.png

crashed again. Here's what it says.

0 Likes

If its crashing with XMP, make sure the memory kit is present on your boards QVL. If its not, then it wasn't tested and its not 100% sure it will work at the rated speeds. Try 3200mhz instead and stress test.

To be clear, if you run the memory at JEDEC speeds, no crashes occur, right?

I'm following your words.


@Kilbot192 wrote:

 

The speed was reading something like 2,000 Mhz so I set it to 3600 as it's advertised. I think this is when I started having problems.






The Englishman
0 Likes

Well it wasn't crashing until I tried changing things for the RAM in the BIOS. Now it seems to crash no matter what. I've turned the timing configuration from overclock to auto. Now the RAM frequency is reading what it was originally (2166 mhz or something like that). I can't find anything about QVL in the BIOS. I think everything is default now. I don't know what else I can change or anything else that has been changed. I've installed the AMD drivers through Adrenalin, and I installed the chipset drivers through MSI center, I tried reseating my RAM and also switched off the :"ECO" switch on the back of the power supply. Still crashing.

0 Likes

QVL is in the boards vendor site, not on the BIOS.

What is QVL

 

Qualified Vendors List, the board manufacturer will test and certify certain memory sticks at their rated speeds. If its there, then its supported. If your memory sticks aren't on that list, they may be unsupported and will likely only work at JEDEC/Default speeds.

What to do next

If its now crashing with default speeds, its time to do the following: Do a complete power cycle, unplug the PSU from the grid using the PSU power button on the back, keep it off for 5 minutes or so, longer doesn't hurt either. Take the CMOS battery off too.
After all this, put the battery back in, plug and reset BIOS to defaults. Let it boot and test it out.

Which slots to use

Make sure you follow your board user manual on how to insert your memory. If you don't you can get all kinds of errors/crashes.

johnnyenglish_0-1715268909845.png

 

What else can I do

 

Try with one stick only, then try the other, this will troubleshoot each stick for problems, if both work individually, could be the board slot going bad.

Do you have something more for me

Sure, you have a 5950X and I'm not sure what board you have and PSU. Since the 5950X can be a heavy weight, I wouldn't be using a PSU under 650 preferably 750. A good board with good VRMs is also the best pair to that once flagship beast of a CPU.

Enabling ECO mode or disabling core boost can also be a valid troublehooter for power issues.

What your error tells me

Normally, memory crashes won't make the PC reboot, it will just stay on but crashed, with a BSOD or nothing at all but a black/gray whatever screen. CPU issues can reboot the system without warning so it could be the CPU. PSU issues, like insufficient power can also make the system trip into a reboot or a complete power off, in worst case scenarios it will dish out damage to the components if the PSU quality is somewhat sketchy.

However, everything needs to be troubleshooted, even the ram.

I would also advise you to post your complete system specs.


I wish you good luck, keep us posted

 

The Englishman
0 Likes

Kilbot192_0-1715277379634.png

here's that, but I'm not sure my board has a CMOS battery. If it does I have no clue where it is.

0 Likes

I reset the BIOS one more time then unplugged it for a couple of hours. I booted it back up and launched a game. Played for a few hours. Now I'm seeing if it can stay on while streaming a movie in my browser because that's usually when the crashes happened. So far so good but we'll see. Thanks for everyone's help so far. I'll post further updates as a few days go by. 

0 Likes

Well everything seemed fine for several hours. Just had another crash though. Same error. Well that sucks. I can't afford to replace anything so that's that I guess.

0 Likes

I've just browsed your board and in my humble opinion think you've made a poor choice. This particular choice has a very low end VRM configuration for such a power house of a CPU.

Try to enable ECO Mode or disable core boost to see if it helps. 

What is your PSU?

And if this is really your board, here it is, the small coin sized battery:

johnnyenglish_0-1715372731738.png

 




@Kilbot192 wrote:

I can't afford to replace anything so that's that I guess.



Maybe a friend can help if its rocking an AM4 system. Or...

..Do you have Amazon in your country, try ordering either a new Board or AM4 CPU, then return within the 30 day period if still fails, but eventually you'll know where the problem comes from and RMA if inside warranty.

But I think you should have a better board, something along these lines:

X570 TUF PLUS with 12+2 VRM or a B550 RoG Strix E with 14+2 VRM, these are much more suited towards that CPU.

Good Luck

The Englishman

Well I've noticed something interesting. I'm still not 100% about this but it seems to me that I only crash with this error when I'm using Opera GX browser.... I've played games and watched movies with media player, but when I stream in Opera or really do anything like simply typing on this forum, it crashes. I've avoided using that browser all day today and no crashes. What the hell do you think that's about? It's weird lol.

edit: Never mind. It did end up crashing eventually I tried leaving Chrome open to see if that would make it crash and it did while I was asleep.

0 Likes

If you suspect that only happens with opera browser open. Try disabling hardware acceleration and report back.

The Englishman
0 Likes

I always turn that off because it causes a weird flicker when watching youtube or any video in the browser at full screen. So far, other than Opera, it has crashed with Chrome open as well. I haven't seen the issue happen yet while using Edge. So yes, with both of those browsers it still crashes with hardware acceleration turned off.

0 Likes

That is a good input.

 

So, if you game only, no browsers open, it never crashes? Right?

 

 

The Englishman
0 Likes

Correct. As long as those browsers are Chrome or Opera GX. I can leave Edge open and that doesn't seem to cause a crash. It hasn't yet. I left Edge up last night because I had a movie paused on a website, and it's still running fine. So other than Chrome and Opera GX I have not found another way to cause this crash.

0 Likes

So its a software issue. Its been great narrowing down the issue.

What doesn't make sense is, they all run Chromium, even Edge.

Lets try a new thing, run Chrome and do your usual stuff with Incognito mode. I've seen crashes with Chrome that got strangely fixed by this.

Again, sorry for insisting. If you run a game only, it never crashes and if you stress test like hell, using cinebench, OCCT or other benchmark tool, it never crashes too.

Because we may be getting there.

The Englishman
0 Likes

I've been running Chrome in incognito for a few hours now. I'll test it longer so I'll probably post an update tomorrow. 

0 Likes

Okay. No crashes while using Chrome in incognito mode. I left it open all day and all night.

We've found the offender. Its now time to search around chrome.

Start with the most common, reset Chrome to defaults.
If not uninstall it completely all chromium based browsers (not Edge), use a good uninstall application to remove all traces of it.

Reinstall the most up to date Chrome to test it again.


Good Luck


The Englishman
0 Likes

Well I've been running Chrome without incognito all day today and no crashes. I just reset the browser settings. I think if it was going to crash the PC it would have done it by now. I usually wouldn't even be able to type this much because it seemed to love crashing when I'm leaving a comment or something. I think I should now try resetting Opera GX and seeing how that goes now.

Okay. I just got back from hospital. I was admitted the night of the 14th and I left my PC on with Opera GX open. Today, I came back to see my computer was back at profile login as if the PC restarted at some point. I have the same error with "Processor APIC ID: 6" as before logged in event viewer at 2:40 PM, so about 3 hours ago. So at least it was able to go two days without a crash, but it still happened. 

0 Likes

Kilbot192, I suggest you stay out of BIOS and do a Clear CMOS. Make sure you have the latest BIOS. Did you build this machine and what parts are in it? When this settles you may be able to use a memory profile to run your memory at rated speed. I use XMP profile to run my memory at rated speed. Thanks and enjoy, John.

0 Likes

Kilbot192, you should use a profile to change memory speed. Just changing speed will yield problems. Enjoy, John.

0 Likes

Yes, I've tried the XMP profile 1 and I have not manually changed the speed. Just set everything to auto as far as I can tell. 

0 Likes

Kilbot192, do a Clear CMOS. Enjoy, John.

 

0 Likes

My motherboard doesn't have the 3 pins but I have held the little button on the IO side until it stopped flashing red. I've done that a few times but I've done it once more and just booted back up. I'll see what happens this time.

0 Likes

Thanks, Kilbot192. Enjoy, John.

0 Likes

I used the Windows memory diagnostic tool and it came back with no problems. It seems to be able to stay on until I start doing anything like playing a streaming video or just typing into this page. However it did crash again so I don't know what I can try next.

0 Likes
ed6698
Adept III

I would suggest setting the RAM with the profile and add just a touch more to the voltage for RAM. Example: if the profile sets the Ram to 1.35 volts, give it a little more voltage like 1.38 volts, I would not go higher then 1.4 volts

Asus ROG Crosshair VIII Dark Hero - Ryzen 5950X - Noctua NH D15 Chromax Black - 4x8GB 32Gb GSkill DDR4 3800 - Asus RTX 3070 TUF OC V2 - Samsung 980 Pro, 970 Evo Plus, Corsair Force GT - Super Flower 1000W Leadex SE Platinum
0 Likes
_nl_ben
Adept II

Hey Kilbot. Did a quick search for the event ID, and there's three possible explanations.

The first is something AMD would need to solve, namely an error in the (chipset) instructions. (Technically it could also be the motherboard manufacturer who's made a mistake there, but that's normally an area AMD provides with the base template for the UEFI and such)

Second, and that's something you can check for yourself, is the voltage. If the manufacturer of the motherboard (e.g. Asus is a bit more aggressive when it comes to that) has the standard volting profiles set above the recommended maximums, this could incur errors in the chip and then the sequencing after that, incurring the error. Not a certainty, but the higher it goes above the recommended limit the more chance of an error. For the 5950X the recommended maximum safe voltage is 1.35V. If your CPU voltage is above that, I recommend for now to turn it down to 1.3V (because manufacturers who play fast and loose with the voltages tend to also play fast and loose with the noted settings, and 1.35V in such a case might become 1.39V instead).

Third, less likely but still an possibility, is a hardware defect. In that case you'll need to RMA.

0 Likes
byss
Journeyman III

What size power supply are you using? This seems like your not getting enough power for your CPU.

0 Likes

Leadex V 130 pro platinum

0 Likes