Asrock confirms Ryzen 9000 failures caused by its BIOS settings, offers to fix motherboards

[midian182]

What just happened? Asrock has gone on record to confirm that it is solely responsible for the more than 108 Ryzen 9000 chips that have failed in the company's motherboards. An executive confirmed that the problem relates to Asrock's BIOS settings, and that it isn't an AMD issue.

Earlier this week, YouTube channel Tech Yes City visited the Asrock booth at Computex to ask about the Ryzen 9000 failure problem on its boards. A rep said it was due to the Electric Design Current (EDC) and Thermal Design Current (TDC). Essentially, it was an amperage problem in the Precision Boost Overdrive (PBO) settings. Asrock said these had been set too high for early CPU samples.

In a later interview with Gamers Nexus host Stephen Burke, ASRock's VP of motherboards, Chris Lee, said the company has rolled out BIOS version 3.25, which alters the PBO settings in the hope of addressing the failing Ryzen 9000 problem. He added that Asrock had not seen any of its motherboards damaged as a result of the original settings.

Asrock said last month 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.

Lee added that Asrock would cover shipping costs both ways if users RMA their motherboards, should they believe they have been damaged.

Lee recommends that affected users send the damaged CPUs back to AMD or wherever they bought them and say they are defective. Anyone who sends a motherboard and CPU to Asrock will have the processor sent back to them so they can carry out the CPU return themselves. It sounds as if this arrangement is unlikely to please customers.

Burke emphasizes that buyers of new Asrock motherboards should not assume BIOS 3.25 is installed and should verify it themselves.

In April, Asrock blamed 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. Asrock released BIOS 3.20 that supposedly fixed this problem, but new cases continued to appear.

It's certainly not a satisfying explanation and still leaves plenty of questions, and one has to wonder why it required interviews with GN and Tech Yes City for Asrock to explain the situation.

The worst part of all this is that there reports of Ryzen 7 9800X3D chips dying in Asrock motherboards with the new BIOS 3.25 update installed. There have also been cases of CPUs failing even when PBO was not enabled.

Permalink to story:

Asrock confirms Ryzen 9000 failures caused by its BIOS settings, offers to fix motherboards

 

[VaRmeNsI]

This could be avoided in future if AMD provided mobo vendors with hard limits.

 

[Loadedaxe]

I guess the takeaway here is....avoid ASRock motherboards. With so many other quality options out there at similar prices, and without the same issues, it just isn’t worth the hassle.

They can blame memory compatibility all they want, but the reality is, no one else seems to be having this problem.

 

[BbN48]

ASRock should be avoided by everyone from now on.

It took them 7 months to acknowledge their fault - and they only did it because GamersNexus forced the issue. Now ASRock are trying to claim it's purely a BIOS issue, yet cpus are still dying on their latest bios.

First, bios 3.20 was supposed to fix the issue, while claiming a different reason for the defect. Now, 3.25 is supposed to fix the issue. Still doesn't.

 

[ShadowDeath]

I guess the takeaway here is....avoid ASRock motherboards. With so many other quality options out there at similar prices, and without the same issues, it just isn’t worth the hassle.

They can blame memory compatibility all they want, but the reality is, no one else seems to be having this problem.

While we're at it throw ASUS at the top of that list of board manufacturers to avoid. Their solution to their version of this was uploading the bios version with the fix as a beta bios which flashing to your board voided the warranty. They only covered it under warranty after everyone called them out on it.

They can blame memory compatibility all they want, but the reality is, no one else seems to be having this problem.

You're focusing on one specific part of their statement while excluding the rest. They blamed more than compatibility. They blamed, CPU defects, user error, ram compatibility, and their own mistakes involving PBO (which I suspect there is more here than what they've spoken about.) A part of me wonders if PBO wasn't being affected even with it turned off which is why they think fixing PBO fixed it for everyone else.

 

[Loadedaxe]

While we're at it throw ASUS at the top of that list of board manufacturers to avoid. Their solution to their version of this was uploading the bios version with the fix as a beta bios which flashing to your board voided the warranty. They only covered it under warranty after everyone called them out on it.

Didnt Asrock spin off of ex Asus employees or something like that?

I can't remember but yeah, explains the continuity between those 2 board manufacturers.

 

[mbk34]

ASRock should be avoided by everyone from now on.

For admitting to a problem and trying to fix it??? That seems a bit harsh.

 

[BbN48]

For admitting to a problem and trying to fix it??? That seems a bit harsh.

Don't be so dishonest

 

[wiyosaya]

For admitting to a problem and trying to fix it??? That seems a bit harsh.

As a software engineer, some problems are not easy to fix. It sounds like they investigated on avenue, thought they might have fixed it, and then found out that they had not so they had to look elsewhere. Some problems are just not easy to fix; however, it does sound like they took problem reports seriously and at least attempted to find a fix.

 

[Loadedaxe]

For admitting to a problem and trying to fix it??? That seems a bit harsh.

Fix it? Well, that depends on how you define "fix."

Did ASRock agree to replace affected motherboards? Technically, yes .... but users had to cover shipping costs both ways.

They patched things up after the Ryzen 7000 series issues, only to repeat the same bios related mistakes with the Ryzen 9000 series. So, was it really fixed? Hard to say, especially when the same problems keep coming back.

With ASRock, it’s tough to know when something is truly fixed, because they have a bad habit of making the same mistakes over and over again.

Now you could say all board manufacturers do the same, and maybe your right, but that does not mean Asrock "fixes" there mistakes, more that they like to repeat them.

lets just hope they keep on "fixing" them

 

[mbk34]

Don't be so dishonest

and how do you think I'm being dishonest?

As a software engineer, some problems are not easy to fix.

Totally agree and I'd imagine it's even harder when you're trying to debug hardware issues. Every user will have different hardware, different RAM settings, different BIOS settings and a different mix of applications running.

Does the manufacturer release the motherboard if it works with the first dozen hardware configurations, the first 100? the first 1000? if a few users say they have issues is it because of your hardware or might it be how they put their system together? Or perhaps it's just something simple like bad cooling?

Sure, it's better if there aren't any problems but, if there are, then I'd like a company to admit there's an issue and come up with a solution. If people stop buying a companies products because they admitted to an issue then no company will ever admit to an issue.

 

[Athlonite69]

Lee recommends that affected users send the damaged CPUs back to AMD or wherever they bought them and say they are defective. Anyone who sends a motherboard and CPU to Asrock will have the processor sent back to them so they can carry out the CPU return themselves. It sounds as if this arrangement is unlikely to please customers.

How's about nope Asrock your mobo caused the damage it's upto you to make it right not AMD as it was not their fault but yours your mobo and crappy BIOS caused the problem. Talk about trying to pass the buck