WHEA Logger Fatal Hardware error

[Mussels]

Apologies, my RAM is running at 3600mhz--the default XMP setting with the click of one button, so I haven't touched anything extra relating to voltages. If I start getting Whea-logger event ID 18 BSODs again, I'll look into bumping up the SoC value.

Here are all the tabs of cpu-z:

zentimings shows a lot more of the RAM settings, but yeah 3600 - you need that SoC voltage raised

CPU-Z doesnt show your current SoC voltage while zentimings will - can be anywhere from 0.95v to 1.05v on auto/stock, 1.15v is generally enough for any setup

I'm running 2x 32GB 3600MHz CL18 (2x2R) on a 5900X/X570S with no adjustment to any voltages, never had any issues like this. Previously ran 2x 16GB 3600MHz CL18, also 2x2R, on a 5800X/X470 - again with no voltage tweaks and no WHEA errors. I think you're a skewed sample size due to your penchant for over-/under-clocking, or maybe your ITX board was just a dud.

Uhhh, I've got four ryzen systems and helped build over 20 more and then dozens of help threads here, other forums, FB, etc.

You're literally running one of the top CPU's with better binning on the second gen x570s chipset, the last released one. Almost like it skews the odds in your favour?
Temp spikes also make the IMC's less stable, which again - you're on a dual CCX CPU, so you get less of those.

It's something that i'm dealing with several times a day, not just on my own systems

 

[n_n]

zentimings shows a lot more of the RAM settings, but yeah 3600 - you need that SoC voltage raised

CPU-Z doesnt show your current SoC voltage while zentimings will - can be anywhere from 0.95v to 1.05v on auto/stock, 1.15v is generally enough for any setup

Thanks, I just DL'd ZenTimings. I'm noob at this kind of info, though. Is "VSOC" the value of interest here?

 

[Mussels]

Yup, that's the one.

SoC is system on chip - the other half of your CPU that controls a LOT of your motherboards hardware - from PCI-E lanes to USB ports, some devices are hardwired in to the CPU, so errors from one part of that (like a high memory overclock or a faulty PCI-E riser cable) can cause the other one to error out - this was the origin of those ryzen "USB dropouts" in the past

Your default is a pretty much normal - tends to be 1.085v

Some boards have better voltage control than others, so it can droop down under certain types of load and cause issues

 

[n_n]

Yup, that's the one.

SoC is system on chip - the other half of your CPU that controls a LOT of your motherboards hardware - from PCI-E lanes to USB ports, some devices are hardwired in to the CPU, so errors from one part of that (like a high memory overclock or a faulty PCI-E riser cable) can cause the other one to error out - this was the origin of those ryzen "USB dropouts" in the past

Your default is a pretty much normal - tends to be 1.085v

Some boards have better voltage control than others, so it can droop down under certain types of load and cause issues

Thanks a lot for the info! I went into my MSI BIOS just now and changed "CPU NB/SoC Voltage" from auto to override with the value of 1.1500V. Hopefully that sounds right?

 

[Mussels]

That's the highest you'd want to set it, yes.

1.20v is perfectly safe for 24/7 use but truly only needed when ram overclocking - so 1.15v is that bit lower while still likely to help, and hopefully helps your issues. If it doesnt and you forget about it, it also cant do any harm.


So much hardware is part of the SoC and inter-connected, you can think of it like having every light in a house on dimming them all slightly, and when one is switched off the others flicker for a second - you're keeping that minmum level up, so even if it droops it should be high enough that nothing craps out


As an example, my VR headset drew enough power to cause issues with a lower SoC voltage while nothing else did - that motherboard even had options to raise the voltages on the USB ports directly which didnt help, but the SoC voltage did

 

[FoxtrotVictor]

Hey everybody! I created this account just to post here. I’m having the same issues with my rig.
My PC specs are:

Ryzen 1600AF
B450M Steel Legend
2x8Gb XPG 3000mhz
RTX 2060 Super
NZXT 650W
1x XPG 256Gb NVME
1x WD Blue 1TB Sata m.2
1x HDD 1TB

I just bought a Ryzen 5600 (w/o X or G) a few days ago and I'm having suddenly reboots on my pc after that. The only thing I did was change the CPU.

I already updated and reset the BIOS, I even downgraded to P4.60 version.
I disconnected my XPG and HDD, created a new partition on my WD Blue SSD and installed a new fresh Windows 11. The frequency I get reboots is very low now, but sometimes the PC just reboots for like, 3 times in a row, and then works fine for HOURS…

I saw in this post that many of you had a problem with WD drives. I’m gonna change that in the future just to see if it’s the problem.

I already tested with all USB disconnected, changed memory slots, XMP on and off, new PBO setting… nothing works for me until now.

The PC works fine on benchmarks and playing games, runs everything perfect, but out of nowhere it reboots. I got WHEA errors 1 hour after installed Windows 11, but not anymore.

I’m gonna get back in the future to tell you guys if something else worked for me… The only thing I have to do now, is turn of my WD and install a new Windows 11 in other drive… let’s see…

 

[Mussels]

Hey everybody! I created this account just to post here. I’m having the same issues with my rig.
My PC specs are:

Ryzen 1600AF
B450M Steel Legend
2x8Gb XPG 3000mhz
RTX 2060 Super
NZXT 650W
1x XPG 256Gb NVME
1x WD Blue 1TB Sata m.2
1x HDD 1TB

I just bought a Ryzen 5600 (w/o X or G) a few days ago and I'm having suddenly reboots on my pc after that. The only thing I did was change the CPU.

I already updated and reset the BIOS, I even downgraded to P4.60 version.
I disconnected my XPG and HDD, created a new partition on my WD Blue SSD and installed a new fresh Windows 11. The frequency I get reboots is very low now, but sometimes the PC just reboots for like, 3 times in a row, and then works fine for HOURS…

I saw in this post that many of you had a problem with WD drives. I’m gonna change that in the future just to see if it’s the problem.

I already tested with all USB disconnected, changed memory slots, XMP on and off, new PBO setting… nothing works for me until now.

The PC works fine on benchmarks and playing games, runs everything perfect, but out of nowhere it reboots. I got WHEA errors 1 hour after installed Windows 11, but not anymore.

I’m gonna get back in the future to tell you guys if something else worked for me… The only thing I have to do now, is turn of my WD and install a new Windows 11 in other drive… let’s see…

The RAM you have sounds like it may be from the era that wasn't ryzen friendly, and you may have got lucky with your prior CPU. Yours sounds like like an infinity fabric issue, rather than the hard drive issue this thread focused on. Entirely fixable, just problematic on automatic settings. Theres a chance enabling XMP will stabilise it, if you haven't done that already.

Post a screenshot of Zentimings and i'll be able to help out there - The 5000 series have more BIOS options for RAM and infinity fabric than the 1600AF (which is a 2000 series Zen+ CPU in disguise)
Oh and check what slots your RAM is installed in - slots 2 and 4 are the best to use.


Personally i'd go into the BIOS and load optimised defaults, enable XMP and set the SoC voltage to 1.10v - odds are that'll stabilise the system completely. If it works save it as a user BIOS profile so you can restore it in the future, such as it resetting due to a flat CMOS battery (also very possible, the boards a few years old now)

 

[FoxtrotVictor]

Thank you for the rapidly reply!

I did a lot of things and just one thing actually looks like make the system runs stable, at least for now. I disabled CPB (Core Performance Boost). Now the Ryzen 5600 runs at 3.5Ghz, and, at least until now, no suddenly system reboots.

The CPU now runs at 0.95V, max 40W of TDP. I have XMP enabled at 3000mhz @1.35v. I'm not a technican at all, but I think it's a problem of a bad CPU silicon or bad motherboard energy management.

I'm gonna use my PC the way it is right now for a while, and see if it's really stable. In the future, I might wanna change the CPU clocks, to like 3.8Ghz, or 4.0Ghz, and see the limits of this silicon/motherboard.

What do you think?

 

[Mussels]

That's crippled your CPU utterly.
It's at 3.5GHz, but it's a 4.4GHz CPU

you can't do all core overclocks like on older CPU generations, your approach there is all or nothing. It's going to be related to the memory you have, as written above.

It's details like WHICH kit of 3000 ram you have that make all the difference - theres a dozen things it could be, and you'll never be able to solve the problem without that information.