AMD Isolates Ryzen 9000 Series Boot Issues to a BIOS Update, ASRock Offers a Fix

[AleksandarK]

A small number of AMD Ryzen 7 9800X3D CPUs have had trouble starting up on some ASRock AM5 motherboards. The issue was first noticed in February 2025, and both AMD and ASRock immediately looked into the problem. The investigation found that the older BIOS versions on specific ASRock boards had memory compatibility problems. When a Ryzen 7 9800X3D was used, the computer could not complete its normal boot process, known as POST. ASRock quickly released a beta BIOS update to fix the memory issue, and after testing, a final version of the update was made available as of now. Users who update their BIOS should now see the problem disappear. In one case, a report suggested that a Ryzen 7 9800X3D had been damaged. However, further checks showed no burn marks on the motherboard. The issue was simply due to some debris, and after careful cleaning, the motherboard started normally and passed further tests.

AMD pointed out that many reasons can cause a computer to not start up properly, and a POST failure does not always mean the CPU is broken. They advise users to update their BIOS to the newest version available for their motherboard model. If the problem continues even after updating, users should reach out to customer support for help. In a few cases, if the problem cannot be fixed on the spot, the CPU might need to be replaced. This quick fix shows that AMD and ASRock work well together to solve problems. Even though only a few systems were affected, the issue was taken seriously and fixed fast. With the latest BIOS update, users can now expect their systems to boot without trouble, ensuring that the Ryzen 7 9800X3D offers good performance for everyday tasks and gaming.



View at TechPowerUp Main Site | Source

 

[Dragokar]

I like what ASRock did and AMD said, but still there are nealry daily reports of dead X3Ds on ASRock boards....

Listing, Overview and newest afaik.

 

[Tek-Check]

I like what ASRock did and AMD said, but still there are nealry daily reports of dead X3Ds on ASRock boards....

Listing, Overview and newest afaik.

Not all users report that BIOS was updated or what exactly happened. Also, there's a normal failure rate that may seem blown out of proportion in news because news outlets do not put individually reported cases in proportion to entire industry. Someone needs to measure normal CPU failure rate. Even if we have 1,000 individual news, if AMD sold 1 million CPUs at the same time, that's only 0.1% of failure rate, which is very small. When wccftech says CPU "exploded", it's attention-grabbing and sensationalist headline. Distance, please.

 

[Dragokar]

Not all users report that BIOS was updated or what exactly happened. Also, there's a normal failure rate that may seem blown out of proportion in news because news outlets do not put individually reported cases in proportion to entire industry. Someone needs to measure normal CPU failure rate. Even if we have 1,000 individual news, if AMD sold 1 million CPUs at the same time, that's only 0.1% of failure rate, which is very small. When wccftech says CPU "exploded", it's attention-grabbing and sensationalist headline. Distance, please.

Yeah, if.....but in that case you should see failure rates all across the vendors, but ASRock right now is leading them. That is, at least for me, concerning. This does not say they are at fault for anything, but remember the first round, it took a while to figure it out. I do run several ASRocks at the moment in family and clients and all are fine, but none of them use a X3D. All X3Ds are in MSI boards.

 

[Deleted member 234478]

this is why I never buy new gen/gear of ANYTHING. I dont want to beta test anything. wait a year or 2, get a better price, a polished product, added tweaks. win win win in every way.

 

[soupinator]

When I saw the article title I was hoping they'd actually fixed something but there are tons of reports of actually failed CPUs, some of which are exhibiting suspicious bulges in the exact same location as the spectacular 7800x3D failures 2 years ago.

 

[Mr. Perfect]

So this fix is still referencing the v3.20 BIOS from March 4th?

 

[Dragokar]

So this fix is still referencing the v3.20 BIOS from March 4th?

Yes.

 

[Alan Smithee]

Aren't people reporting that their working systems stop working? RAM compatibility would prevent first boot, not suddenly make a working system stop working.

 

[maxli86]

Oh... I just encountered a rig with AMD Ryzen 7 7800X3D not able to boot properly.
Change motherboard, change different RAM in the end the CPU have issues.
The weird thing is sometimes it can boot, sometime it doesn't.
It take about 30 minutes or even longer than it can post.

If under such circumstances that means motherboard onboard diagnostic isn't accurate
to detect a faulty CPU.
AMD must release a CPU testing software like what Intel has for their CPU.

Than news about AMD Ryzen 7 9800X3D having similar issues is puzzling?

 

[Tek-Check]

Yeah, if.....but in that case you should see failure rates all across the vendors, but ASRock right now is leading them. That is, at least for me, concerning. This does not say they are at fault for anything, but remember the first round, it took a while to figure it out. I do run several ASRocks at the moment in family and clients and all are fine, but none of them use a X3D. All X3Ds are in MSI boards.

I also have two Asrock boards. We will need to wait and see. I am sure they have several of those CPUs and boards in labs for further testing.
We cannot say anything here due to lack of deeper and wider insights. It's all speculation.

 

[_roman_]

Not all users report that BIOS was updated or what exactly happened. Also, there's a normal failure rate that may seem blown out of proportion in news because news outlets do not put individually reported cases in proportion to entire industry

Bios update will not solve that issue.

Backstory:
I bought or got from a refurbish company in the last 18 months four lenovo laptops. #1 died. #2 was replacement. #2 sold because of battery. #3 I returned a few days ago because of bad speakers. #4 i ordered again because #3 was junk. #4 i am very happy. #4 was in 4 years lenovo warranty plan which ended in december 2024. nvme which i just checked in windows smart values - average the device was powered on 1-2 times a day for average 1-2 hours a day for those 4 years according to the samsung smart values for total hours and such.

Main Point: I doubt many people even know about bios or will even update that.
All 4 refurbish lenovo laptops had the first or second bios installed. (to explain for those who do not know: you can check lenovo for uefi updates. it starts with the first bios bios and the number goes up to the last one - usually there are 15-20 different bios version available for download. every few weeks / months you get a new bios version) I assume or I state - that was the bios for the mainboard from the factory.
I doubt many people even know about bios or will even update that.

--

AMD pointed out that many reasons can cause a computer to not start up properly, and a POST failure does not always mean the CPU is broken.

Not every overclocker DRAM will run out of the box. Not every DRAM will run out of the box.
6000 / 6200 MT/s is already overclocking DRAM for AM5.

It's the early adopter tax + user who do not know sometimes what they are doing + user who buys overclocking - not supported DRAM which may work or may not work = the hint about POST = power on self test routine every computer does during the power on phase.

 

[Tek-Check]

I doubt many people even know about bios or will even update that.

If they want to run newly released CPUs, they have to update BIOS on boards released before those CPUs.
But I do agree that some owners do not touch BIOS. Asrock had some issue with memory training for EXPO. We don't know on how many boards.

 

[EatingDirt]

Yeah, if.....but in that case you should see failure rates all across the vendors, but ASRock right now is leading them. That is, at least for me, concerning. This does not say they are at fault for anything, but remember the first round, it took a while to figure it out. I do run several ASRocks at the moment in family and clients and all are fine, but none of them use a X3D. All X3Ds are in MSI boards.

Asrock is 'leading' per... reddit posts. It's anecdotal evidence at best, and you can't underestimate peoples desire to get easy useless digital upvotes.

As someone that just helped a friend with motherboard selection, Asrock generally has the best X870/B850 AM5 lineup. All but the absolute cheapest boards have good VRM, have good rear I/O(Livemixer series stands out), are aesthetically pleasing(Steel Legend & Pro RS both suit white builds) & storage configurations, are aesthetically pleasing and come in at a good $150-$220 mark, which makes them generally good value.

I have no evidence that asrock boards are more popular, but that is what I would expect after looking into the all the X870/B850 lineups.

 

[evernessince]

Asrock is 'leading' per... reddit posts. It's anecdotal evidence at best, and you can't underestimate peoples desire to get easy useless digital upvotes.

As someone that just helped a friend with motherboard selection, Asrock generally has the best X870/B850 AM5 lineup. All but the absolute cheapest boards have good VRM, have good rear I/O(Livemixer series stands out), are aesthetically pleasing(Steel Legend & Pro RS both suit white builds) & storage configurations, are aesthetically pleasing and come in at a good $150-$220 mark, which makes them generally good value.

I have no evidence that asrock boards are more popular, but that is what I would expect after looking into the all the X870/B850 lineups.

People's desire to get upvotes wouldn't concentrate around a single brand. It would be a more random distribution. Your argument hinges on the idea that ASRock sells the most motherboards but the number of cases I've seen on ASRock boards far exceeds that of other brands.

You'd essentially have to be arguing that ASRock is selling nearly all the AMD motherboards in order to account for the fact that they represent the vast majority of reported issues.

I've purchased X370, X470, X570, and X670 ASRock boards and have been exclusively buying ASRock for my main rig since Z68. I don't see the point in deflecting criticism away, that doesn't fix the issue and only hurts ASRock. Heck even the industry as a whole gets worse when customers are willing to play defense for corporations.

It doesn't matter whether ASRock is at fault or not, I want to see them reaching out more to people having these issues to find the problem. I'm not seeing that frankly, they are not active on the ASRock subreddit following up on these posts. Those are your customers having issues on your products, any decent company should want to help them and especially to verify the voracity of the claims being made. Having the motherboards sent in for analysis, heck maybe even paying customers to have the whole CPU, motherboard, RAM setup sent in, makes sense. It's a small upfront cost compared to letting it drag on.

Oh... I just encountered a rig with AMD Ryzen 7 7800X3D not able to boot properly.
Change motherboard, change different RAM in the end the CPU have issues.
The weird thing is sometimes it can boot, sometime it doesn't.
It take about 30 minutes or even longer than it can post.

If under such circumstances that means motherboard onboard diagnostic isn't accurate
to detect a faulty CPU.
AMD must release a CPU testing software like what Intel has for their CPU.

Than news about AMD Ryzen 7 9800X3D having similar issues is puzzling?

Did you try swapping out the power supply? Sometimes older power supplies that don't support lower power states can cause boot issues.

Intel's PDT will only detect the most obvious of faulty CPUs. It's better for validating if the CPU in genuine. Y-Cruncher is far more effective at sussing out issues.

 

[maxli86]

People's desire to get upvotes wouldn't concentrate around a single brand. It would be a more random distribution. Your argument hinges on the idea that ASRock sells the most motherboards but the number of cases I've seen on ASRock boards far exceeds that of other brands.

You'd essentially have to be arguing that ASRock is selling nearly all the AMD motherboards in order to account for the fact that they represent the vast majority of reported issues.

I've purchased X370, X470, X570, and X670 ASRock boards and have been exclusively buying ASRock for my main rig since Z68. I don't see the point in deflecting criticism away, that doesn't fix the issue which only hurts ASRock. Heck even the industry as a whole gets worse when customers are willing to play defense for corporations.

It doesn't matter whether ASRock is at fault or not, I want to see them reaching out more to people having these issues to find the problem. I'm not seeing that frankly, they are not active on the ASRock subreddit follow up on these posts. Those are your customers having issues on your products, any decent company should want to help them and especially to verify the voracity of the claims being made. Having the motherboards sent in for analysis, heck maybe even paying customers to have the whole CPU, motherboard, RAM setup sent in, makes sense. It's a small upfront cost compared to letting it drag on.



Did you try swapping out the power supply? Sometimes older power supplies that don't support lower power states can cause boot issues.

Intel's PDT will only detect the most obvious of faulty CPUs. It's better for validating if the CPU in genuine. Y-Cruncher is far more effective at sussing out issues.

Is a FSP Hydro G Pro 1000 not a old model.

 

[Lost_Wanderer]

I'm waiting on another replacement for my 9800X3D (this will be my third one since January), and second AsRock X870E Taichi Lite. First one was toasted on a Asus ROG Strix B850-E. Second 9800X3D stopped working after about two weeks of use, and it was running BIOS 3.2. No overclocking apart from EXPO. RAM is good, and I've swapped between power supplies (I have two FSP HydroTi's). Fortunately my fallback is an AM4 system with a 5800X3D.

 

[NoLoihi]

It doesn't matter whether ASRock is at fault or not, …

I’ll let that entitlement slip because AsRock really ought to investigate as a matter of course. How else would they know they’re at fault or they’re not?

… I want to see them reaching out more to people having these issues to find the problem. I'm not seeing that frankly, they are not active on the ASRock subreddit follow up on these posts.

Now, there, be careful what you wish for! Following up on posts is done by some Social Media Team or somesuch. The actual diagnostics will be done by engineers, I believe you call them in English (I’m not sure it’s common to call them Ingenieure in German), hence progress might come from less talk and more deed. They might already be investigating with all the power one might expect. Maybe they’re waiting on AMD for the clarification of some specs or whatnot? Who knows? For those with broken machines and no viable backup it’ll suck—I mean, even for those where the backup is okay but they’d rather use their new device—though I don’t see much (from my cursory look on this) they could do other than to pile on, provide data, and maybe even hardware, where requested, and … wait for it to be resolved. No sense increasing pressure much, AsRock probably has figured out all on their own people are waiting.

 

[JustBenching]

Aren't people reporting that their working systems stop working? RAM compatibility would prevent first boot, not suddenly make a working system stop working.

Not necessarily. Imcs can degrade and also mobo trains the ram every boot so...

 

[EatingDirt]

You'd essentially have to be arguing that ASRock is selling nearly all the AMD motherboards in order to account for the fact that they represent the vast majority of reported issues.

And I made the case for Asrock being much more popular on AM5 in my last reply because they have a good balance of boards at a reasonable price, and a few very aesthetically pleasing value boards(B850 Steel Legend, X870 Pro RS & Steel Legend). I would not be surprised if they have a good 10%+ more market presence on new AM5 builds than Asus, Gigabyte & MSI.

Doesn't doesn't matter whether ASRock is at fault or not, I want to see them reaching out more to people having these issues to find the problem. I'm not seeing that frankly, they are not active on the ASRock subreddit follow up on these posts. Those are your customers having issues on your products, any decent company should want to help them and especially to verify the voracity of the claims being made. Having the motherboards sent in for analysis, heck maybe even paying customers to have the whole CPU, motherboard, RAM setup sent in, makes sense. It's a small upfront cost compared to letting it drag on.

From all appearances they are investigating the issue. It's been a weekly news article about the 'issue' and them responding to it. Not having an active subreddit presence(not an official subreddit run by Asrock) is irrelevant. If people have an issue with their setups, and they believe it's the motherboard at fault, they should be contacting Asrock support on their actual website.

 

[geniekid]

I think it's very unlikely ASRock sells so many more boards that their failure rate is proportional to the number of user reported failures, but without actual sales numbers from the manufacturers we're all just guessing.

ASRock's statement claims they've actively reached out to users for investigation and have fixed some issues, but it very much feels like the investigation is ongoing. I'm personally waiting for them to say more before I buy an ASRock motherboard.

 

[blinnbanir]

I have a X870E Taichi. I updated the BIOS when I got my 9900X3D. That was weeks ago and when I read this article and went to look I was already on this BIOS.

 

[evernessince]

Now, there, be careful what you wish for! Following up on posts is done by some Social Media Team or somesuch. The actual diagnostics will be done by engineers, I believe you call them in English (I’m not sure it’s common to call them Ingenieure in German), hence progress might come from less talk and more deed. They might already be investigating with all the power one might expect. Maybe they’re waiting on AMD for the clarification of some specs or whatnot? Who knows? For those with broken machines and no viable backup it’ll suck—I mean, even for those where the backup is okay but they’d rather use their new device—though I don’t see much (from my cursory look on this) they could do other than to pile on, provide data, and maybe even hardware, where requested, and … wait for it to be resolved. No sense increasing pressure much, AsRock probably has figured out all on their own people are waiting.

ASRock has made no statement that they are investigating the physically damaged CPUs. In the meantime their brand is taking a hit and some are even avoiding them. If they are investigating behind the scenes without notifying the public, then that's just a failure on their part to communicate that. It really should not be hard to say "We are looking into this".

And I made the case for Asrock being much more popular on AM5 in my last reply because they have a good balance of boards at a reasonable price, and a few very aesthetically pleasing value boards(B850 Steel Legend, X870 Pro RS & Steel Legend). I would not be surprised if they have a good 10%+ more market presence on new AM5 builds than Asus, Gigabyte & MSI.

10% more market doesn't align with them accounting for 90%+ of the dying CPUs.

From all appearances they are investigating the issue. It's been a weekly news article about the 'issue' and them responding to it. Not having an active subreddit presence(not an official subreddit run by Asrock) is irrelevant. If people have an issue with their setups, and they believe it's the motherboard at fault, they should be contacting Asrock support on their actual website.

Weekly? First off no, there's been 2 articles on TPU about the same ASRock statement. The issue has been occurring for much longer than 2 weeks.

Second, the boot issue mentioned in the TPU articles and physically damaged CPUs are two entirely different things. To date there has been zero progress on the physically damaged CPUs.

 

[EatingDirt]

ASRock has made no statement that they are investigating the physically damaged CPUs. In the meantime their brand is taking a hit and some are even avoiding them. If they are investigating behind the scenes without notifying the public, then that's just a failure on their part to communicate that. It really should not be hard to say "We are looking into this".

Why would you think they're not investigating damaged CPU's? What statement do you want? "We're looking into the issue, but that shit is random, so it will take time, and it's also happening to other vendors motherboards."?

At least 20% of failures are on other vendors motherboards, and that comes from a thread specifically on the asrock subreddit.

 

[Aerpoweron]

I got a 9700X recently. And it has boot issues on an AsRock B650m HDV/m.2. A 7600X3d works fine on it. But the 9700X needs a bios reset to boot again. Memory and CPU stay lit up after a restart with the 9700X. I have the latest bios 3.20 installed.