ASRock Acknowledges Ryzen CPU Failures Linked to Motherboard BIOS, Offers Warranty Replacement

[Melvis]

But there are still reports of CPU's failing after this update......Steve says

 

[Rightness_1]

From what I've learned AMD specifies a range for those currents and ASRock opted to use the upper limit of those ranges. So you could also blame AMD for specifying ranges that their CPUs may not be able to sustain indefinitely.

There is more to this than being able to go too high with PBO settings. I think their tricking the CPU to take more voltage than it is designed to handle.

 

[JimBob69]

You just need to update the bios. It has nothing to do with the board itself.

And you know this how? How have you confirmed this statement to be factual?

 

[Redwoodz]

Oh the horror.... to think you could actually burn a CPU by overclocking it to far

 

[DudeBeFishing]

These last few years have had a lot burning incidents. CPUs, GPUs, PSUs, Cases. I feel like I'm missing another component.

 

[_roman_]

You just need to update the bios. It has nothing to do with the board itself.

When you believe their step 3 from the 7step or 8D-Report formular.

I highly doubt this is really the root cause, = step 3. Therefore step 4, 5, 6, 7 and 8 are non valid. prevention, doing something it will never happens again and such

i want to see the asrock FMEA for that issue.

AMD. Good luck with that boys!

4 times different AM4 or AM5 mainboard clearly states your warranty is void when you even change your dram settings. Ryzen Master in windows does the same, but i do not use windows and i do not overclock via windows.

those settings are not default. even EXPO voids your warranty on my past msi am4 mainboard, when you really read those warnings in the uefi menu.

It seems someone was faster with that expo thing. = XMP

XMP profile was considered overclocking and you could kiss you warranty bye bye.

how x3D chips

I think 5000 X3d died. I am not sure about the 7000 X3d ones.

 

[Panther_Seraphin]

7xxx x3ds had some horrendous issues early on due to bad Current limits. Asus were really bad at current limits early on.

I didnt hear tooo many 5xxx x3ds dying but I was aware they were really conservative in clocks etc.

 

[AGlezB]

Who makes the best motherboards these days, it used to be ASUS, but is it MSi now? What's your opinion?

The question you need to ask is "who makes the best motherboard at X price point?". For instance, at USD 800+ you can take a blindfold and cover you eyes and pick at random and you'll still get the best product. The brand doesn't matter unless you're looking for a specific feature. Pretty much the same happens below USD 100, were all are equally cheaply built.

For anything in between the answer is: No one. Every OEM will put out good products and bad products so I usually wait for review from sources I trust before buying. The problem is there are a million "reviewers" out there, most of which will base their recommendations on the specs/marketing materials from the OEM because that's the best they can do. And the best reviewers some times get passed over because they tend to not be very likeable or being too technical or too long winded or not making clickbait videos or whatever. Plus it's easier to take advice from sources you like.

That said, I've been using Gigabyte (Aorus) boards for decades and I've never had any issue that's specific to Gigabyte. There have been issues like AM4 bad USB implementation that required like 5 BIOS updates until it was finally fixed but that was AMD's and affected everyone, not just Gigabyte. Just be aware I don't use RGB of any kind so I don't know how good they are at that and I understand some brands have better RBG support than others.

So in the end, the best motherboard is the one you can afford and has all the features you need, as long as it doesn't burn your CPU.

 

[Rightness_1]

Oh the horror.... to think you could actually burn a CPU by overclocking it to far

Except that is not the case here. People running stock have had this problem too

 

[chrcoluk]

Board vendors still not learning from recent history, its just not worth trying to eek an extra 1-2% out of chips at risk of funding RMA's and losing reputation.

 

[b1k3rdude]

So they are replacing the boards, what about the dead CPU's...?

But on a related note, AsRock unlike all the other AIBs only offers 1yr warranties, that should speak volumes to anyone considering any of thier products.

 

[Redwoodz]

Except that is not the case here. People running stock have had this problem too

PBO is overclocking.

 

[NDRE28]

I'm glad that I did not have those issues with my X670E Taichi + R9 9950X combo, though I'm using PBO OC settings.

 

[NoLoihi]

So they are replacing the boards, what about the dead CPU's...?

As mentioned before on this very thread, AMD (allegedly) will. From what it seems (TechYESCity, whom I don’t usually watch, had a video on it), AMD has given AsRock some numbers, AsRock kept to those and now they’re getting all the blame, even though the numbers (it’s said) have come from AMD themselves.
This would also make for a reason as to why AsRock seemingly had a hard time to investigate and assuming blame themselves—they’ve actually stood witihn the lines they’ve been given, their numbers matched their docs. It kinda mirrors the degradation issue with Intel, where A. The hardware was failing to request the voltages it should, but only in brief, specific situations (so, the set values were fine and you needed to know what to look for) and B. mainboard makers across the board (pretty much, AFAIR?) had been skirting Intels operational recommendations in the name of a performance edge, so they actually were doing something wrong, even though the ultimate fault has been found with Intel, not with mainboard makers.

 

[Rightness_1]

As mentioned before on this very thread, AMD (allegedly) will. From what it seems (TechYESCity, whom I don’t usually watch, had a video on it), AMD has given AsRock some numbers, AsRock kept to those and now they’re getting all the blame, even though the numbers (it’s said) have come from AMD themselves.
This would also make for a reason as to why AsRock seemingly had a hard time to investigate and assuming blame themselves—they’ve actually stood witihn the lines they’ve been given, their numbers matched their docs. It kinda mirrors the degradation issue with Intel, where A. The hardware was failing to request the voltages it should, but only in brief, specific situations (so, the set values were fine and you needed to know what to look for) and B. mainboard makers across the board (pretty much, AFAIR?) had been skirting Intels operational recommendations in the name of a performance edge, so they actually were doing something wrong, even though the ultimate fault has been found with Intel, not with mainboard makers.

ASRock told it's customers to take up CPU defects with the retailer or AMD themselves. The guy who runs the motherboard division clearly stated that in the video with Steve, so no confusion, even though it is clearly ASRock's responsibility. I hope AMD honors that warranty, but I would not blame them if they didn't.

The blame lies clearly on ASRock, nobody else's boards are burning up CPU's.

 

[blinnbanir]

Exactly how many are we talking about that have been reported?

 

[R0H1T]

PBO is overclocking.

Well actually turbo/boost clocks is OCing ~ only base clocks are guaranteed by both AMD/Intel & it's been like that since the beginning!

 

[neatfeatguy]

The blame lies clearly on ASRock, nobody else's boards are burning up CPU's.

Actually, that's not true. There have been reports of the same issues from other manufacturer motherboards such as Gigabyte, MSI and ASUS. However, ASRock has had the most that's been reported. With that said, we cannot confirm/deny any of these failures to be solely on the manufacturer (MB or CPU) or the end user due to not knowing all the specifics in each situation.

The only thing we do know is that ASRock says a BIOS update should resolve any possible failures on their boards, now we just wait and see if that's holds true. But we all know how well the microcode update from Intel did on "fixing" failing CPUs.

 

[destruya]

I just bought a X870E Tomahawk to replace my Taichi Lite today. Newegg has them for $329 and I had my local Micro Center price match them.