• Welcome to TechPowerUp Forums, Guest! Please check out our forum guidelines for info related to our community.
  • The forums have been upgraded with support for dark mode. By default it will follow the setting on your system/browser. You may override it by scrolling to the end of the page and clicking the gears icon.

WHEA LOGGER 18(Cache Hierarchy Error) KERNEL POWER 41. READ DESCRIPTION please help

Wasn't there an issue where this only happened to very-very-early Vermeer batches? (or other WHEA crashes) (pre-'21 batches)
Apparently, multiple people reported the issue. Even my first Vermeer purchase was clear from that period.

So, if later than 20xx, then it's just a fluke, it looks like.

Unlike Raptor Lake, where if crashes start, it's too late and the CPU must be replaced. The BIOS fixes in that case, must be applied before any symptoms. Yes, the Ryzen must be replaced if new symptoms occur, but crashes and/or errors, are not known to be a widespread issue, unlike 13th-gen and 14th-gen Core i.
No, the issue hasn't gone away. Just lower failure rates Zen2 to Zen3.

As we praise AMD for cache errors and blame Intel for similar, if not same exact issues.

As if everyone expects 100% fab without error. Gimme a break. Please don't be that click bait guy. You know better.
 
Ok lets wait and see what the op does next, no need to argue, both intel and amd flub sometimes, lets move on.
 
Ok lets wait and see what the op does next, no need to argue, both intel and amd flub sometimes, lets move on.
This is peaceful conversation.

It's hard to come to terms with it. It's rare, but those are the posts where all diagnostics where exhausted, no extra hardware and at the mercy of warranty.

I only share what would work accurately as a resolve. I don't care what the OP does honestly.
 
This is peaceful conversation.

It's hard to come to terms with it. It's rare, but those are the posts where all diagnostics where exhausted, no extra hardware and at the mercy of warranty.

I only share what would work accurately as a resolve. I don't care what the OP does honestly.
Its sort of like having to explain to someone that certain vehicle suspension parts wear out over time like rubber, some understand, some don't.
 
Its sort of like having to explain to someone that certain vehicle suspension parts wear out over time like rubber, some understand, some don't.
What man... was the click bait comment too much? I didn't mean to harm. I can easily be asked to edit my posts via pm, or while you seem to moderate, perhaps remove it for me quietly...

I have no idea why he can't Speak for himself, nor why you derailed the convo.
 
whea 18 is amd ryzen 5xxx b1 cpu error.... mostly is cause by boost too high and cause the error...disable turbo boost or tweak to lower speed if no more rma.... or curve optimiser potive some golden core....
else rma it
 
I was getting Whea 18 and cache hierarchy errors on my 5700X/B450 Mortar max. My fix was moving to AM5. It stopped then.
 
I would expect to see it only if you're:
- running unstable memory OC
- have applied CO to the CPU and it isn't stable (CPU ID will match the logical core number of the core that crashed)

Update BIOS, run optimized defaults, turn on XMP and adjust DRAM voltage to 1.35V. Make sure you fabric clocks match the required frequency (1800 MHz) to match the XMP profile of your memory kit. Then run MemTest5 and check for memory errors.
I was getting Whea 18 and cache hierarchy errors on my 5700X/B450 Mortar max. My fix was moving to AM5. It stopped then.
So you changed out the entire platform to "solve" the issue. Great!
 
Built a couple b450 pc's in the past (b450 tomahawk both) that also couldnt quite manage 1:1 with 1800 (3600mts ddr4) Maybe try 3433 as well and if ti goes away, well, there's your answer
 
I've built hundreds of Zen2 and Zen3 systems, and I've RMA'd a dozen or so CPUs with cache heirarchy bluescreens.

You might be able to restore stability with a voltage bump and a negative boost override to stop it trying to reach the highest boost clocks, but AMD absolutely do swap these CPUs out because they're not stable at spec/stock speeds and voltages.

As long as you are within the returns window, it's faster to return for a refund and buy something else. If you are outside of the returns window, raise an RMA.
 
I've built hundreds of Zen2 and Zen3 systems, and I've RMA'd a dozen or so CPUs with cache heirarchy bluescreens.
Did the Vermeers have a pre-21xx date code? It seemed that my Ryzen 7 3700X (Matisse) (Zen 2) stopped having those episodes.
 
Back
Top