• Welcome to TechPowerUp Forums, Guest! Please check out our forum guidelines for info related to our community.

RMAd every single component in my PC, and it's still crashing

Status
Not open for further replies.
C

chopinanopolis

Joined
Feb 14, 2024
Messages
44 (0.09/day)
I've been having problems since I built my PC in early February. It just randomly reboots, most of the time it happens in Apex Legends, but not exclusively. Event viewer shows a Machine Check Exception, Cache Hierarchy Error. I just today got my Mobo and CPU back from the RMA I sent it in for, and again, just like with every other component, they've said they couldn't find an issue/not replicate said issue. I'm completely overwhelmed, I don't know what to do anymore. Everyone kept saying it's a hardware issue, now I've sent every single component in for RMA, and not one single component could they find an issue with, but the same damn error still comes up, and the PC still reboots. I've tried everything I've seen online, and nothing has helped. I have no clue what to do anymore. I'm about to give up on PCs, as much as I don't want to.

Specs;
GPU: AMD Radeon 6650XT
CPU: AMD Ryzen 5 5600
Mobo: ASRock B450M Pro4 R2.0
Ram: 16GB Kingston Fury Beast DDR4-3200
SSD: 1TB Lexar NM620 M.2
PSU: 700 Watt be quiet! System Power 9
OS: Windows 10 Pro
 
Last edited:
No specs at all?

I'm guessing AM4 in which case Cache Hierarchy is often fixable only with a new CPU. Did you not RMA the CPU with AMD? If they approve RMA, they're sending you a new one, in which case it's highly unlikely a late production Vermeer CPU will have Cache Hierarchy as a result of production defect.
 
tabascosauz said:
No specs at all?

I'm guessing AM4 in which case Cache Hierarchy is often fixable only with a new CPU. Did you not RMA the CPU with AMD? If they approve RMA, they're sending you a new one, in which case it's highly unlikely a late production Vermeer CPU will have Cache Hierarchy as a result of production defect.
Click to expand...
Sorry, I forgot to add my specs, added them now. No I RMAd with Mindfactory, the online shop I bought everything from.

Shrek said:
Power supply?

[SOLVED] - [SOLVED] Ryzen 5 3500x - WHEA-Logger "Cache Hierarchy Error" Need Help Troubleshooting | Tom's Hardware Forum (tomshardware.com)
Click to expand...
Already tried two different PSUs, and sent the current one I have in for RMA as well, still crashing though, so I don't think it's the PSU
 
chopinanopolis said:
Sorry, I forgot to add my specs, added them now. No I RMAd with Mindfactory, the online shop I bought everything from.


Already tried two different PSUs, and sent the current one I have in for RMA as well, still crashing though, so I don't think it's the PSU
Click to expand...

Sounds like they played you and returned your old CPU then, check against any photos you might've taken of the heatspreader

There is no reason to go to anyone except AMD for an AMD CPU RMA. Fast and efficient as long as you talk your way past the canned response at the start. Sometimes they even cross ship depending on where you are.

For Cache Hierarchy you can really only try and compensate with positive Curve Optimizer values (overvolt) and hope for the best. Regardless, you should be RMAing the CPU if it's outside return window.
 
tabascosauz said:
Sounds like they played you and returned your old CPU then, check against any photos you might've taken of the heatspreader

There is no reason to go to anyone except AMD for an AMD CPU RMA. Fast and efficient as long as you talk your way past the canned response at the start. Sometimes they even cross ship depending on where you are.

For Cache Hierarchy you can really only try and compensate with positive Curve Optimizer values (overvolt) and hope for the best. Regardless, you should be RMAing the CPU if it's outside return window.
Click to expand...
alright, I'll have to send it to AMD then. And yeah, it's definitely the same CPU, I thought when you RMA something you send it in tandem they try to repair it? For every component I sent in they said they couldn't replicate any error and sent the same component back. I bought the CPU early February so I think it'd be out of the return window by now
 
chopinanopolis said:
alright, I'll have to send it to AMD then. And yeah, it's definitely the same CPU, I thought when you RMA something you send it in tandem they try to repair it? For every component I sent in they said they couldn't replicate any error and sent the same component back. I bought the CPU early February so I think it'd be out of the return window by now
Click to expand...

CPUs aren't really something the manufacturer will repair. It's usually just replaced. Sounds like Mindfactory employees are just assholes, I can RMA things thru my retailer (not that I would for CPUs and GPUs, just cut out the middleman) for the first year free of charge, but all they're doing is just handling the RMA with the manufacturer on my behalf. They shouldn't be taking it upon themselves to judge whether an RMA is necessary, unless there really is no grounds for it. Cache Hierarchy speaks for itself, it's not like Bus/Interconnect.
 
you are using some kind of shit software or bios configured poorly.
 
Machine Check Exception, Cache Hierarchy Error
Click to expand...
IIRC, this is the error that popped up when my 1st 5600 'lost a core'. Setting Core Leveling to 5 cores, immediately 'fixed' the problem.

Initiated RMA w/ AMD, they didn't think I actually bought the CPU. I provided 3x proof of my order, delivery, and ownership (including timestamp'd ID pic).
RMA process took about 2 weeks or so.

On the plus side, they only ask for the CPU.
In my case (and probably others') they'll send you back a full retail box - packed CPU ('free' cooler and 'free' case badge)
 
LabRat 891 said:
IIRC, this is the error that popped up when my 1st 5600 'lost a core'. Setting Core Leveling to 5 cores, immediately 'fixed' the problem.
Click to expand...

An excellent diagnostic procedure.
 
Last edited:
tabascosauz said:
I'm guessing AM4 in which case Cache Hierarchy is often fixable only with a new CPU. Did you not RMA the CPU with AMD?
Click to expand...
Unstable VRAM (A.K.A. MCLK) can also cause that machine check exception error on especially Ryzen.

Also, there appear to be some batches of AM4 Ryzens that are unusually intolerant of heat and cause that error to be thrown randomly.

Also, AM4 feels buggy to me, compared to my other platforms. The problem may be the B450 motherboard, additionally.

It seemed to mostly happen with my Ryzen 7 3700X on my MSI B450 Tomahawk, which that was a while before the motherboard stopped powering on. It was more likely to happen when the room temp was 72 F or around there. Seemed like it was less likely to occur at <70 F.
 
Last edited:
You guys need to run B-Die, AM4 is the smoothest platform I have run. No problems.
 
freeagent said:
You guys need to run B-Die, AM4 is the smoothest platform I have run. No problems.
Click to expand...
I have the same CPU and with the same motherboard, it doesn't feel buggy at all, but Windows 11 23H2, even when TM5 passed, (with anta777 extreme) was misbehaving yesterday!

I had to reinstall the latest cumulative update, just to verify it installed correctly and when I did that, the Windows 11 update screen was stuck at 0 percent for an alarmingly long time!
It felt like I was updating Windows 10 on a potato! I usually would see 10's WU take a long time at 0 percent or around there on pre-2018 systems, more so on pre-2017 systems. Never on a Ryzen that I worked on!

When Windows Update isn't acting up, it runs like a dream!

It looks like a Microsoft mea culpa.
 
Last edited:
RJARRRPCGP said:
Unstable VRAM (A.K.A. MCLK) can also cause that machine check exception error on especially Ryzen.
Click to expand...
That's a real kick in the nuts to an AMD system for VRAM to do that.

RJARRRPCGP said:
It seemed to mostly happen with my Ryzen 7 3700X on my MSI B450 Tomahawk, which that was a while before the motherboard stopped powering on. It was more likely to happen when the room temp was 72 F or around there. Seemed like it was less likely to occur at <70 F.
Click to expand...
I still have an original B450 Tomahawk which I got during the time it had glowing reviews. How long were you able to use your B450 Tomahawk before it crapped out?
 
A Computer Guy said:
I still have an original B450 Tomahawk which I got during the time it had glowing reviews. How long were you able to use your B450 Tomahawk before it crapped out?
Click to expand...
Only 2 years, but it looked like on YouTube, that it wasn't necessarily a MOSFET failure. It could very well have been just a faulty SMD connector, which appears to be located by the RAM slots.
I didn't smell anything fried, but literally wouldn't turn on, even with another PSU! IIRC, I had Ubuntu Jammy on it in August, 2022 and I expected it to be on and idling, when I found it off!

But I pulled the MOSFET heatsinks off and found oozing thermal pads, so I chucked it!

I got it on January 3, 2020.

I never even PBO'ed the Ryzen 7 3700X! Nor once raised the frequencies higher than default! On January 3, 2020, I got a Ryzen 3 3200G for it, then in June, 2020, got the Ryzen 7 3700X.
 
Last edited:
Low quality post by Frick
What settings do you run your RAM at? Don't use XMP, try some lower speed and see what happens.
freeagent said:
You guys need to run B-Die, AM4 is the smoothest platform I have run. No problems.
Click to expand...

That is not advice or even helping. No one knows or should have to know what a B-die is to have a stable computer.
 
Last edited:
Status
Not open for further replies.
Back
Top