AMD responds to Ryzen 7 9800X3D CPU failures, blames memory compatibility issues

midian182

Posts: 10,658   +142
Staff member
What just happened? AMD has responded to reports of 108 Ryzen 7 9800X3D CPUs unexpectedly dying in users' machines, with most of the incidents happening on Asrock motherboards. Team Red blames memory compatibility issues for the processors "failing to complete POST," even though the affected chips reportedly did pass POST and worked for anything from half an hour to several months before dying, with many cases showing physical damage.

This week saw a Redditor compile a list of 108 Reddit posts involving a Ryzen 7 9800X3D that died, along with 12 other Ryzen 9000-series chips that suffered the same fate. 98 of the failures, or 82%, occurred on Asrock boards. It's stated that the processors passed POST and worked for varying amounts of time before dying with no signs of failure.

That many instances doesn't represent a huge number compared to the thousands of Ryzen 7 9800X3D CPUs sold each week – the processor is incredibly popular with gamers, and for good reason. But it's not an insignificant figure, either, so AMD gave an official response to some outlets.

Also read: AMD hits record CPU share as new top GPU and OS emerge on Steam survey

"We are aware of a limited number of user reports involving Asrock AM5 motherboards failing to complete POST. Following a joint investigation, AMD and Asrock identified a memory compatibility issue present in earlier BIOS versions, which has been rectified in the latest BIOS," the company wrote, repeating previous statements made by Asrock.

AMD added that failure to POST can be caused by several factors and does not necessarily indicate a non-functional CPU. It recommends that anyone experiencing problems start by updating their BIOS to the latest version for their motherboard. If issues persist, the next step is to contact AMD's customer support team.

Asrock responded to the reports last week. The company said that it had inspected a motherboard from a system where the CPU showed burn damage. It claimed there was no burn damage around the mobo or VRM area. After "cleaning and removing debris" from the CPU socket, the motherboard booted up successfully with the original BIOS and passed long-term stress tests.

AMD is saying that a memory compatibility issue prevented the CPUs from POSTing, stopping them from working entirely – as one would expect. But the Reddit post states that these problematic chips are passing POST and working for a while before failing, sometimes with electrical or burn damage, which doesn't sound like a symptom of the issue AMD is blaming.

One can only conclude that AMD's statement is referring only to the cases where the 9800X3Ds didn't pass POST, and not those that suffered permanent physical damage. Nevertheless, a lot of user response to the explanation has been negative. It's possible that the company will follow-up with another statement at a later date.

Permalink to story:

 
Running my Ryzen 9-9900x at 100% for 2 months, I am pushing it Maximum, running Dual Fan Liquid Cooling, 5200 DDR5 all maxed out, ZERO crashes, ZERO problems
I have 6 Fans on my case, airflow is around 40L a min, machine sits 5f from my window which has cold air blowing in all day. Cpu Temp 78% max
 
AMD made no claim that a failed POST is considered as dead. Any confusing of the RAM compatibility issue with dead CPUs is wholly with the users.
 
This is especially annoying as I’ve been eyeing an Asrock mobo & 9800x3D bundle but I’m unsure if this recent dustup is a real issue or an internet conspiracy.
 
Running my Ryzen 9-9900x at 100% for 2 months, I am pushing it Maximum, running Dual Fan Liquid Cooling, 5200 DDR5 all maxed out, ZERO crashes, ZERO problems
I have 6 Fans on my case, airflow is around 40L a min, machine sits 5f from my window which has cold air blowing in all day. Cpu Temp 78% max
Well, your RAM isn't even up to the speed of the controller on the CPU which is 5600 so your CPU is sitting there like MJ on the court with a high school student.
 
"After "cleaning and removing debris" from the CPU socket, the motherboard booted up successfully with the original BIOS and passed long-term stress tests."
Does that mean that ASRock believes it's just accidental dirty sockets?
 
"After "cleaning and removing debris" from the CPU socket, the motherboard booted up successfully with the original BIOS and passed long-term stress tests."
Does that mean that ASRock believes it's just accidental dirty sockets?

Firstly, Rob Thubron, Techspot, shouldn't have conflated that failure with the RAM compatibility fix.

My understanding is Asrock made no judgement on how the failure happened. They cleaned and tested the board and it worked. To me, the dirt came from the burnt CPU. And I think that's how Asrock viewed it too.

The CPU was fried somehow. AMD is probably looking into it further. It seems to be exclusive the 9000 X3D parts only. These are the ones using the newest assembly methods at TSMC.

EDIT: Reading some of the Redit posts ... it looks like all 9000 model CPUs have died at some point. So can't blame the new X3D assembly arrangement. That probably leaves voltage mismanagement.
 
Last edited:
AM5 was a BIOS nightmare for me until a BIOS was reasonably stable, and it boots up with the speed of an injured walrus.

I miss the era of CPUs and mobod that just worked. It's also really shady to market all of these Intel and AMD CPUs as supporting DDR5-9999 but then not supporting them at EXPO settings.
 
This is especially annoying as I’ve been eyeing an Asrock mobo & 9800x3D bundle but I’m unsure if this recent dustup is a real issue or an internet conspiracy.
MSI might be a safer option.

There are other issues with ASRock mbs. No guarantee you'll get EXPO memory timings to work, or it'll work but with really long boot times.
 
I RMAed my strix x670e I motherboard because of some USB voltage problem hard hold at initial boot launch screen. I ended up just getting a cheaper x870e MSI tomahawk for a midtower build. Just going to sell the itx motherboard. The MSI tomahawk was able to successfully boot my cl30 gdd5 32 gig ram kits to cl 28 with 1 click oc that was prepoppulated. The 9800x3d is stable at 5.425 ghz pbo -15 offset.
 
Is this currently impacting just 9800x3Ds? or impacting X870(E) motherboards as well? I have the 7800x3D on a x870 tomahawk with latest bios but still a little concerned with x3D chips again.
 
Since I looked at the specs of the zen 4 parts, I found it interesting to note that they officially support "max memory speed" of DDR5-5600 using 2 sticks, while only DDR5-3600 when using 4 modules.

e.g. the specs for the 9900X.

But I guess based on motherboard support, and depending on how many modules are used, you can achieve higher speeds using EXPO, but be wary of the number of modules and whether single-rank or dual-rank.
e.g. on the MSI X870e Tomahawk, using 4 dual-rank modules, it can support up to DDR5-4800
 
Running my Ryzen 9-9900x at 100% for 2 months, I am pushing it Maximum, running Dual Fan Liquid Cooling, 5200 DDR5 all maxed out, ZERO crashes, ZERO problems
I have 6 Fans on my case, airflow is around 40L a min, machine sits 5f from my window which has cold air blowing in all day. Cpu Temp 78% max
Why are you running 6 fans when you have liquid cooling?

What is "pushing it maximum"? Constant video encoding? 3D modeling? Most things -including gaming- don't actually hit full tilt on all cores.
 
If I am to read between the lines here, I think what they mean is that ASRock set memory controller voltage too high and that can indeed physically harm the CPU. You'll want to provide nice high voltage to the memory controller if you want stable RAM, so that was probably their goal. The issue ends up being this :D

 
Just be glad you're not running Threadripper... Installing/upgrading RAM on TRX50 boards is not for the faint of heart...

Failure to post? Bah... Do some swearing, flush CMOS, reseat RAM, pray for a bit and try again... and again... and again...

But... it DOES eventually work... Speaking from someone who now has 256GB of V-Color Ram @ 6000 - and a 7980X
 
Is this currently impacting just 9800x3Ds? or impacting X870(E) motherboards as well? I have the 7800x3D on a x870 tomahawk with latest bios but still a little concerned with x3D chips again.
Seems to be all 9000 parts. 9800X3D is naturally overrepresented in the stats because it sold in such high quantities.
 
"Following a joint investigation, AMD and Asrock identified a memory compatibility issue present in earlier BIOS versions, which has been rectified in the latest BIOS"

I'm running a MSI board myself, and I'm seeing similar stuff mentioned in the patch notes.
https://www.msi.com/Motherboard/MAG-X870-TOMAHAWK-WIFI/support
Description:
- AGESA PI-1.2.0.3a Patch A released.
- Fix CPU information is incorrect.
- Fix DRAM message on Post page is incorrect.
- Optimized the memory compatibility of single rank (1R) under 2DPC configuration.
- Optimized the memory compatibility of dual rank (2R) under 1DPC configuration.

MSI Center keep informing me there is absolutely no reason for updating my BIOS if I'm not experiencing any problems. I'm suddenly thinking that perhaps I should update BEFORE I'm starting seeing problems...
 
Why are you running 6 fans when you have liquid cooling?

What is "pushing it maximum"? Constant video encoding? 3D modeling? Most things -including gaming- don't actually hit full tilt on all cores.
running XMrig is at 100% cores to see how long it lasts doing this.If it survives 3 months I will be happy and stop it. 6 Fans include 2 water cooling fans, most people don't realise you need cool air pumping in because your GPU generates a lot of heat, and does it pass out if its not pushed out via the Watercooling fans, not a good idea
 
"After "cleaning and removing debris" from the CPU socket, the motherboard booted up successfully with the original BIOS and passed long-term stress tests."
Does that mean that ASRock believes it's just accidental dirty sockets?

Getting Chernobyl vibes...

"You didn't see it because it's not there"
 
running XMrig is at 100% cores to see how long it lasts doing this.If it survives 3 months I will be happy and stop it. 6 Fans include 2 water cooling fans, most people don't realise you need cool air pumping in because your GPU generates a lot of heat, and does it pass out if its not pushed out via the Watercooling fans, not a good idea
You are stress testing for 3 months? Why? No normal use case even approaches that.

Most people realize that 3-4 fans is plenty if you case isn't crap. I've got 2 watercooling fans bringing in air and another exhaust fan in a Meshify 2 case with no problems gaming, running complex statistical models, or even mining (back when that was profitable). I think you may be overly concerned about a few degrees cooler.
 
Running my Ryzen 9-9900x at 100% for 2 months, I am pushing it Maximum, running Dual Fan Liquid Cooling, 5200 DDR5 all maxed out, ZERO crashes, ZERO problems
I have 6 Fans on my case, airflow is around 40L a min, machine sits 5f from my window which has cold air blowing in all day. Cpu Temp 78% max
Can I ask how exactly does this help? You are here listing out specs that has absolutely NOTHING to do with the reported issue...cringe dude. massive cringe
 
Back