5950x WHEA Logger Errors

[greatash]

Just got a ton of these earlier playing cyberpunk, BSOD with whea error every 30 min or so. Tried turning ram back to XMP stock, turned off curve optimised still got whea errors.

Ive fixed it now by just locking the 5800x at 4.6ghz 1.25v so it's definitely something to do with the voltages spiking. Temps dropped by 10c as a bonus, in game now 60-64c max with very low pump/fan speed. Voltages are way too high on these at stock for my liking.

 

[Nicholas Eagle]

I was getting the interconnect WHEA errors and random reboots on my build, appears that setting VDDG to 950 and VDDP to 900 has stopped the WHEA errors at least and appears to be stable now with no reboots so far.

Also I have 2 GigaByte X570 Aurous Masters same Rev 1.2 (ended up with 2 as I RMA'd it and couldnt be without the motherboard) - and I previously had a 3900x in them both, the other x570 motherboard did not have the random reboots, when I moved the 3900x to this other x570 it had the random reboots and was only fixed by changing those VDDG and VDDP above... Since putting the 5900x in it appears that the VDDG and VDDP is still needed but this might be coincidence as I haven't tried the 5900x in the other board.

 

[Shilz]

For what its worth I have the 5900x with Asus Strix E X570 - BIOS 2802 - AM4 AGESA V2 PI 1.1.0.0 Patch B - Team Group 3600Mhz - 6800XT

Default CPU settings - Resize bar on and SAM enabled - Everything else left on auto apart from my ram which is overclocked to the relevant speed and timings adjusted.

Been running completely stable with general everyday use and gaming.

 

[ltron]

Have been having the same WHEA random reboot issues. System is a 5950x coupled with a ASUS Crosshair VIII Impact (3003 bios). Only happens on idle and very random as it could be fine for days and then it happens a couple of times within a few hours.

Noticed ASUS release a 3101 beta bios so tried that but unfortunately it won't boot my machine with DOCP enabled. If I set memory to Auto (running at 2133) the system boots and I'm going to see if that is stable. My memory is on the QVL for the board and was working fine for months with my 3950x before I upgraded (Corsair CMW32GX4M2C3200C16).

This never happens under heavy load, in games or running video encoding, only on idle and based on the other threads might be due to one of cores spinning up from idle and not getting enough voltage which hopefully a BIOS update will fix as very frustrating.

TomDD

If you turn XMP off is it stable?

 

[TomDD]

If you turn XMP off is it stable?

So far but it's only been a couple of days.

TomDD

 

[4K8KW10]

If the WHEA appear when the CPU is idle, then the voltage spikes should be from the memory.
I have seen that my restarts appear when the memory usage changes, for example when I open a browser, when I play a YouTube video, or when I am back on the desktop after a gaming session.

Must be some settings with the DRAM. My voltage is 1.37-1.38 V in BIOS, when it should be 1.35 V.

 

[TomDD]

If the WHEA appear when the CPU is idle, then the voltage spikes should be from the memory.
I have seen that my restarts appear when the memory usage changes, for example when I open a browser, when I play a YouTube video, or when I am back on the desktop after a gaming session.

Must be some settings with the DRAM. My voltage is 1.37-1.38 V in BIOS, when it should be 1.35 V.

I’m seeing something similar with my 1.35v Corsair RAM showing as 1.39v in the BIOS. Wondering if this is why it won’t boot with DOCP. Running at stock seems fine so far.

TomDD

 

[4K8KW10]

I’m seeing something similar with my 1.35v Corsair RAM showing as 1.39v in the BIOS. Wondering if this is why it won’t boot with DOCP. Running at stock seems fine so far.

TomDD

I disabled XMP Profile 1 and now the DDR4 is at 2133 MHz to try to find some more stability and then will gradually increase the frequencies up intil it's again stable.

My CMK32GX4M2B3200C16 Corsair "DDR4-3200" 2x16GB kit is rated by Corsair at 1.35 V tested voltage but the DRAM manufacturer Nanya nowhere states the same.
It's 1.2 V at 3200 MHz DDR4.

 

[TomDD]

I disabled XMP Profile 1 and now the DDR4 is at 2133 MHz to try to find some more stability and then will gradually increase the frequencies up intil it's again stable.

My CMK32GX4M2B3200C16 Corsair "DDR4-3200" 2x16GB kit is rated by Corsair at 1.35 V tested voltage but the DRAM manufacturer Nanya nowhere states the same.
It's 1.2 V at 3200 MHz DDR4.

I've got a similar kit of 2x16GB Corsair CMW32GX4M2C3200C16 DDR4-3200 although it shows the DRAM Manuf. as Samsung. XMP-3200 shows 1.350V but as you say 1.20V at 2133MHz. Its even on the QVL list for the motherboard but guess that was written before 5950x. I had it working with my 3950x with zero issues for over 6months.

Older BIOS's ran at 3200 ok but I got the WHEA errors.

TomDD

 

[4K8KW10]

I've got a similar kit of 2x16GB Corsair CMW32GX4M2C3200C16 DDR4-3200 although it shows the DRAM Manuf. as Samsung. XMP-3200 shows 1.350V but as you say 1.20V at 2133MHz. Its even on the QVL list for the motherboard but guess that was written before 5950x. I had it working with my 3950x with zero issues for over 6months.

Older BIOS's ran at 3200 ok but I got the WHEA errors.

TomDD

You have an Asus board, I am not so lucky

 

[Alexander Burton]

Hi Guys!

I've been having these issues, but nothing when idle. Only when loading into or playing a game. Updated the latest BIOS and enabled XMP and was stable for 9 hours, then got a reboot with "Cache Hierarchy Error" - All of my WHEA are fatal/uncorrectable and cause a reboot.

Could this be a faulty chip, or still BIOS issues? My latest crash today was after 9 hours, when i changed my game from my 1080p monitor to my 4k monitor. Hit apply, all fine, navigated the menu and went to play a game, screen changed then boom, reboot.

I've requested an RMA of the CPU as it may be the chip but still wanting to do more tests as the wait for a new 5900x will suck...

EDIT - It also passed all stress tests (Prime95, Cinebench, OCCT) with flying colours and was playing/downloading games all day, while watching twitch streams and all sorts. Surely if the chip was faulty those things would have shown that, right?

 

[RichL]

I was getting WHEA errors at idle (and the CPU voltage was dropping very low as the CPU clock dropped). I've updated the X570 chipset driver with the latest on AMD.com and all seems fine so far, and no severe vcore drops. It seems quite a few people are having crashes at idle on 5000 series, all the fixes I've seen involve increasing the min CPU voltage one way or another, be in locking the core voltage or modifying the curve optimizer. Chipset driver seems to have worked for me though.

 

[Daylight_Invader]

I am unfortunately still getting a WHEA Logger error generally during boot every day or two. Alternatively I get a page fault error. I am on the latest X570 family of bios from Asus which is now on the 1.2.0.0 AGESA code. Admitdedly these do seem to happen at DOCP (3600), but I have seen them at default before too. Memtest x86 is completely clean, so this does not appear to be a memory related issue.

 

[TomDD]

I am unfortunately still getting a WHEA Logger error generally during boot every day or two. Alternatively I get a page fault error. I am on the latest X570 family of bios from Asus which is now on the 1.2.0.0 AGESA code. Admitdedly these do seem to happen at DOCP (3600), but I have seen them at default before too. Memtest x86 is completely clean, so this does not appear to be a memory related issue.

Have you tried changing the SoV voltage? I've not had one WHEA (touch wood) since manually setting my SoC voltage. Also running on latest 1.2.0.0 AGESA.