Burning Saga Continues, This Time it's an AMD Ryzen 7 9800X3D CPU

[AleksandarK]

A new case of catastrophic CPU failure has emerged involving AMD's Ryzen 7 9800X3D processor, marking the latest in a series of reported incidents within the last few days involving high-performance GPUs and CPUs. The failure occurred during routine use when a system equipped with the 9800X3D and an ASRock Nova X870E motherboard suddenly shut down, resulting in visible thermal damage to both components. The incident is particularly noteworthy as the system operated under stock settings, with only AMD EXPO memory optimization enabled. The affected user, who has two decades of PC building experience, reported that the system had been operational for approximately 20 days before the failure, with no temperature anomalies recorded through HWMonitor during its operation.

The hardware was running the ASRock BIOS version 3.16. This case differs significantly from the previous AMD Ryzen 7 9800X3D failure, where user error during installation was identified as the primary cause, with the user force-installing the CPU in the socket. The timing of the failure—during a low-intensity workload of streaming video content—further complicates the investigation into root causes. While isolated cases of hardware failure are not uncommon in the PC component market, this case may be the first one caused entirely by the CPU/motherboard combination, not user error. The user also faces uncertainty regarding warranty coverage, as the CPU and motherboard were purchased separately from different retailers. We hope the warranty case goes well, and the user gets a replacement!



View at TechPowerUp Main Site | Source

 

[lepudruk]

I've got exact same cpu&motherboard combo in my system. The difference is I updated bios to 3.18 (by flashback) before putting it all together. No problems so far, all system components working fine.

 

[kondamin]

thats a bit of a bother for the users that lost both parts, unless this is happening to multiple people it’s nothing but spreading FUD and I’m fairly certain we have enough of that around us at the moment

 

[Xeanoa]

Am I reading this right?

A singular failed processor, out of how many thousands or even millions? This could very well have been a speck of dust that's been smoldering for three weeks before carbonising enough to become conductive. Is it a slow day or something?

 

[stickleback123]

This is complicated equipment sold relatively cheaply and in large volumes. Some will fail.

 

[O.0]

AMD EXPO is enabled

 

[usiname]

Am I reading this right?

A singular flailed processor, out of how many thousands or even millions? This could very well have been a speck of dust that's been smoldering for three weeks before carbonising enough to become conductive. Is it a slow day or something?

You can't even take the words of the person for granted, the system could burned on the first boot, or under extreme overclock. It could be defect with the CPU or with the motherboard socket. With just single reported problem for few months since the release it doesn't matter at all.

 

[AusWolf]

The CPU burned up at stock, with only EXPO enabled (khm...) 20 days after installation? Something doesn't add up.

 

[dont whant to set it"']

Did the motherboard disable any and or all of the thermal protections?
Even without a heatsink,the CPU will boot at most , depending on the POST time, before thermal throttle kicks in followed by over temperature shutdown if the first mechanism cannot solve the issue.
Le: on second thought, assuming no install/fitting error: Overclocking can be fun and sometimes expensive like that. Buck up.

 

[_roman_]

Were there bend pins in the cpu socket before this mainboard was powered on?

The standard am5 socket is not to my liking. I ditched the standard am5 socket quite fast. Some parts are kept of course. Best 9€ spend.

AM5 Secure Frame RED – Thermalright

www.thermalright.com

4 screws nothing can move. Less likely to get thermalpaste much closer to the pcb of the cpu as with the standard am5 socket

--

Was anyone patient enough to use the picture and check what pins show the burn mark? I would consider only the cpu itself, not the socket.

 

[Chomiq]

AMD EXPLO is enabled

There, fixed it for you.

 

[PixelTech]

My question is if the motherboard was brand new? I'm thinking that maybe the motherboard was refurbished and some of the CPU pins were bent...

 

[bogardandy]

It's so funny that this post got so much traction because was posted in pcmasterrace sub, BUT if you go to AsRock sub , you will find plenty of them , including 1 that that keep count of them (30+). It is more common on AsRock boards.

 

[kondamin]

My question is if the motherboard was brand new? I'm thinking that maybe the motherboard was refurbished and some of the CPU pins were bent...

You go reddit and ask there, shouldn't be on TPU until it's certain there is an actual problem with the product.

 

[wNotyarD]

My question is if the motherboard was brand new? I'm thinking that maybe the motherboard was refurbished and some of the CPU pins were bent...

Were the pins bent, would it reach ~20 days of use?

 

[rv8000]

The CPU burned up at stock, with only EXPO enabled (khm...) 20 days after installation? Something doesn't add up.

Seems fishy to me. Very little details. I find it more likely to be something set in the bios by user/ASRock. Even ASUS has some weird voltages set by default on x870; VDDGs are 850mv by spec/default, if I set expo on my board or leave it at auto when memory is above jedec, its defaulting to 1050mv for VDDGs. It happens with other voltages as well, especially if someone was enabling 10x scalar with PBO enabled, feeding 1.4v +/- vcore.

 

[Robin Seina]

Well, user can send it to Gamers Nexus, they like cases like this for study purposes at least.

 

[Space Lynx]

My AsRock mobo just got this BIOS update, its the first one since 3.16, so hopefully it fixes whatever issue this is... does anyone know what Patch A actually entails?

 

[Redwoodz]

Burned? That looks like bent motherboard pins damage from here. LMAO

 

[Neo_Morpheus]

Am I reading this right?

A singular failed processor, out of how many thousands or even millions? This could very well have been a speck of dust that's been smoldering for three weeks before carbonising enough to become conductive. Is it a slow day or something?

Hey, Ngreedia white knights (on all parts of the spectrum) cannot allow the news to only trash their beloved company.

Check the wording on the Ngreedia titles and then on this one.

I swear that AMD is treated like a convicted felon that was released on a technicality and everyone must make sure to make their lives as much as a living hell as possible.

 

[LittleBro]

Am I reading this right?

A singular failed processor, out of how many thousands or even millions? This could very well have been a speck of dust that's been smoldering for three weeks before carbonising enough to become conductive. Is it a slow day or something?

Yes, right? If 4 reported RTX 5090 problems are not surge, then this is definitely not a saga lol.

I'd like to know SoC voltage.

 

[rv8000]

Yes, right? If 4 reported RTX 5090 problems are not surge, then this is definitely not a saga lol.

I'd like to know SoC voltage.

VSOC cannot exceed 1.3 set; depending on board, bios, llc, and droop, this can be anywhere from 1.29-1.31 and shouldn’t kill/burn the chip.

This seems more like pins may have been damaged, and user let loose on PBO & Scalar settings in bios which could send 1.4v vcore depending on load (more normal operation is around 5.225ghz ~1.1-1.2v max). Curious to know what pins in the socket the burned area account for in terms of voltage delivery.

 

[LittleBro]

The CPU burned up at stock, with only EXPO enabled (khm...) 20 days after installation? Something doesn't add up.

Exactly! Should have been three days sooner.

 

[evernessince]

I believe I've seen reports that this happens mostly with the Nova. Could be a problem with just that motherboard or a combination of factors that need addressing by AMD and ASRock.