Arock board with a 9800x3d??

sastusbulbas said:
Not seen anyone on YouTube stating 1000 verified CPU failures with Asrock AM5 yet. No doubt Gamer Nexus is still working on a three hour epic trilogy on the matter.

Gamer Nexus and PC Builder showed far less than 1000 in their last video's and not sure why we would not have the YouTubers happily posting evidence of 1000 verified to have been killed by Asrock failures. They love click bait and attention after all.

The reality is that AMD's own CPU failure rate is probably more likely than the motherboard issue.

And some may be 12 years old account wise. But there is also a fair share of amateurs messing things up and still running 3.15.

Personally I find it a little bemusing that we have went from around 200 a couple of months ago to over 1000 yet no videos or mass articles outside Reddit.

That would suggest 3.15 bios was actually safer than 3.40/3.50 etc despite the claims by Gamer Nexus and PC builder showing a clear decline as more moved to recent bios updates and newer CPU batches.
Click to expand...

A couple of months ago it was up to over 320 reports on the Google form linked to the Reddit megathread, up from ~180 reports 9 months ago, and there has been 1-3 new reports a day in the thread since that publication.

Taking into account the reports on other forums (some of which will overlap with the Reddit thread), and unsubstantiated reports from cybercafes and the likes i.e. https://videocardz.com/newz/cafe-owner-reports-15-dead-ryzen-7-9800x3d-cpus-out-of-150 and retailers claiming they've got into the 100s of CPUs returned consistent with this failure it is fairly likely the actual failures are in the region of 1000 now.
 
Rroff said:
A couple of months ago it was up to over 320 reports on the Google form linked to the Reddit megathread, up from ~180 reports 9 months ago, and there has been 1-3 new reports a day in the thread since that publication.

Taking into account the reports on other forums (some of which will overlap with the Reddit thread), and unsubstantiated reports from cybercafes and the likes i.e. https://videocardz.com/newz/cafe-owner-reports-15-dead-ryzen-7-9800x3d-cpus-out-of-150 and retailers claiming they've got into the 100s of CPUs returned consistent with this failure it is fairly likely the actual failures are in the region of 1000 now.
Click to expand...

All on Asrock motherboards or AMD in general? And only 9800X3D?

One content creator showed that a single 9900x failed in an Asrock motherboard.

The link above was with 9800X3D in Asus motherboards.

Not sure what to believe these days.

I'm still running the Nova 870x but certainly feel a lack of confidence in dropping in the 9950X3D2 later in the year.

With so many complaints about 9000 series I'm wondering if it's safer just to get some last gen CPU or simply avoid AMD in total.
 
Last edited:
sastusbulbas said:
All on Asrock motherboards or AMD in general? And only 9800X3D?

One content creator showed that a single 9900x failed in an Asrock motherboard.

The link above was with 9800X3D in Asus motherboards.

Not sure what to believe these days.

I'm still running the Nova 870x but certainly feel a lack of confidence in dropping in the 9950X3D2 later in the year.

With so many complaints about 9000 series I'm wondering if it's safer just to get some last gen CPU or simply avoid AMD in total.
Click to expand...

All AMD CPUs failing with symptoms consistent with this failure, though 9800X3D is by far the most prominent and makes up most of that number.

9800X3D still has one of the lowest failure rates at retail for similar CPUs - I would say you'd have to be unlucky to experience a failure with one but there have been several posters here who've had problems - but staying away from ASRock and early batches seems to massively decrease the chances of an issue. The 99xx CPUs even less likely.
 
Last edited:
sastusbulbas said:
Left right and center seems a little far fetched.

I may be wrong, but at the last count Gamer Nexus did was it not around 7 CPU failures that happened in 2025 on an Asrock Nova X870E

Have you evidence thousands more Nova have set off CPU's like grenades?

I am assuming Asrock sold more than a dozen Nova X,870E.

I get the loss of confidence. But more people have probably seen the Loch Ness monster than seen an actual live exploding CPU in an Asrock motherboard.

If AMD CPU's have a higher failure rate. How many do you think may be blaming Asrock for a failed CPU that was actually an AMD failure?
Click to expand...

How does GN know of every CPU failure?
 
Darujhistan said:
How does GN know of every CPU failure?
Click to expand...

Well as much as I dislike the guy, over at GN he seemed to make an attempt to collate all the data regarding this issue some time back with some facts and figures from manufacturers and the forum claims instead of wildly claiming they are dying left right and center after reading some Reddit users blaming Asrock for every problem they have with AMD.

Regardless, the fact is Asrock are not blowing up CPU's left right center. The number seems to be anything between 200 to 1000. Unconfirmed and it seems people are assuming its all Asrock and their fault because forums users say so, as we see, one reply here already lists one dealers 15 failures out of 150 CPU's used by one company, yet that was on Asus motherboards before the recent Asus issue was highlighted. 15 out of 150 could simply have been a bad batch from AMD, but 10% of your CPU purchases failing in an occasional batch, who do you blame? AMD or Asus, no wait add it to the Asrock phenomenon!
It seems simply anything goes wrong then it's best to assume the Asrock motherboard even if its a different brand.

AMD 9000 series CPU's seem to have had a problem or two, we see that in the claimed batch failure rates for 2024/2025. Add to that the fact that the AMD CPU failure rate is actually higher than the failure rate of an AMD CPU due to Asrock. And at one point AMD were keen to try and claim any form of overclocking CPU and memory such as PBO or EXPO would invalidate the warranties too before redacting the point, with that in mind it would seem the CPU's being pushed to the limit by AMD and not quite as consistent and robust as we would like.

If Asrock only sold two milllion motherboards and had around 200 failures that was 0.01%, even if sales has stopped dead at 2 million, with a claimed 1000 failures that would be 0.05% failure rate. Asrock have sold far more than 2 million AM5 motherboards at this point.

To quote the Goggle regarding AMD CPU failure percentages, data from major retailers suggests a low overall RMA rate of 0.56% to 1.1%, indicating, for most users, the processor is reliable.
Well if 0.56% to 1.1% is reliable, then the Asrock failure rate is a Godlike reliability and best in class at 0.05%, even if we have 1000 failures on Asrock.
It's like stating your at least 10x more likely to have an AMD CPU in general fail overall regardless of motherboard, than experience an ASrock motherboard fail the CPU. If anything it could be that faulty AMD CPU's are the issue possibly damaging motherboards. Why not, you cannot claim Asrock are to blame for everything if their CPU kill rate is 0.05% and AMD's return a dead CPU rate is 0.5% to 1.1%.
What is the most common motherboard failure when a CPU starts failing. VRM's. Maybe this issue is AMD CPU faults making the Asrock instable.

So if the failure rate of a 9800X3D is at best 0.56%, the simple math points out that the majority of failed CPU's in Asrock motherboards are due to AMD CPU's having issues and not the motherboard. Like nine out of ten due to the AMD failure rate and one out of ten due to the Asrock failure rate. Simplified, but isn't everything.


To put the average 0.56% figure into context.
If 0.56% of two million 9800X3D's failed. That is 11200 dead CPU's due to the CPU manufacturer.
If 0.05% of two million 9800X3D were fried by a motherboard. That is 1000 dead CPU's due to the MB manufacturer.
So with those rough around the edges figures, for every 1000 CPU's Asrock kills, AMD kills 11200 CPU's.

What are the odds that based on such figures, you may experience an AMD failure in an Asrock motherboard due to AMD's failure rate, but assume it's Asrock to blame because people on Reddit think 1000 is a problem, 11200 is not.

So if you would not touch Asrock's 0.05% with a bargepole, why would you touch AMD with their own 0.56% to 1.1% figure?

Looking forward to the Day Asrock and AMD actually find out what was going on and tel lthe consumers it's resolved.
 
It would be curious to find out what was going on.

Personally I'm of the opinion that there is a manufacturing tolerance on the CPU which is skating fine margins, which isn't a problem for most CPUs but pushes the odd one out of tolerance compounded by ASRock boards stretching that margin even thinner.

It is interesting that certain batches seem to be far more affected than others but you can't entirely take away any other factor and the problem is solved.

EDIT: I think this has taken on a bit more prominence than it otherwise would due to enthusiasts jumping on the ASRock boards, and early CPU batches, which seem most affected this time around and so more likely to post about their issues online when it happens vs if it was happening in a more general fashion where the effect was diluted over the average consumer instead.
 
Last edited:
sastusbulbas said:
Well as much as I dislike the guy, over at GN he seemed to make an attempt to collate all the data regarding this issue some time back with some facts and figures from manufacturers and the forum claims instead of wildly claiming they are dying left right and center after reading some Reddit users blaming Asrock for every problem they have with AMD.

Regardless, the fact is Asrock are not blowing up CPU's left right center. The number seems to be anything between 200 to 1000. Unconfirmed and it seems people are assuming its all Asrock and their fault because forums users say so, as we see, one reply here already lists one dealers 15 failures out of 150 CPU's used by one company, yet that was on Asus motherboards before the recent Asus issue was highlighted. 15 out of 150 could simply have been a bad batch from AMD, but 10% of your CPU purchases failing in an occasional batch, who do you blame? AMD or Asus, no wait add it to the Asrock phenomenon!
It seems simply anything goes wrong then it's best to assume the Asrock motherboard even if its a different brand.

AMD 9000 series CPU's seem to have had a problem or two, we see that in the claimed batch failure rates for 2024/2025. Add to that the fact that the AMD CPU failure rate is actually higher than the failure rate of an AMD CPU due to Asrock. And at one point AMD were keen to try and claim any form of overclocking CPU and memory such as PBO or EXPO would invalidate the warranties too before redacting the point, with that in mind it would seem the CPU's being pushed to the limit by AMD and not quite as consistent and robust as we would like.

If Asrock only sold two milllion motherboards and had around 200 failures that was 0.01%, even if sales has stopped dead at 2 million, with a claimed 1000 failures that would be 0.05% failure rate. Asrock have sold far more than 2 million AM5 motherboards at this point.

To quote the Goggle regarding AMD CPU failure percentages, data from major retailers suggests a low overall RMA rate of 0.56% to 1.1%, indicating, for most users, the processor is reliable.
Well if 0.56% to 1.1% is reliable, then the Asrock failure rate is a Godlike reliability and best in class at 0.05%, even if we have 1000 failures on Asrock.
It's like stating your at least 10x more likely to have an AMD CPU in general fail overall regardless of motherboard, than experience an ASrock motherboard fail the CPU. If anything it could be that faulty AMD CPU's are the issue possibly damaging motherboards. Why not, you cannot claim Asrock are to blame for everything if their CPU kill rate is 0.05% and AMD's return a dead CPU rate is 0.5% to 1.1%.
What is the most common motherboard failure when a CPU starts failing. VRM's. Maybe this issue is AMD CPU faults making the Asrock instable.

So if the failure rate of a 9800X3D is at best 0.56%, the simple math points out that the majority of failed CPU's in Asrock motherboards are due to AMD CPU's having issues and not the motherboard. Like nine out of ten due to the AMD failure rate and one out of ten due to the Asrock failure rate. Simplified, but isn't everything.


To put the average 0.56% figure into context.
If 0.56% of two million 9800X3D's failed. That is 11200 dead CPU's due to the CPU manufacturer.
If 0.05% of two million 9800X3D were fried by a motherboard. That is 1000 dead CPU's due to the MB manufacturer.
So with those rough around the edges figures, for every 1000 CPU's Asrock kills, AMD kills 11200 CPU's.

What are the odds that based on such figures, you may experience an AMD failure in an Asrock motherboard due to AMD's failure rate, but assume it's Asrock to blame because people on Reddit think 1000 is a problem, 11200 is not.

So if you would not touch Asrock's 0.05% with a bargepole, why would you touch AMD with their own 0.56% to 1.1% figure?

Looking forward to the Day Asrock and AMD actually find out what was going on and tel lthe consumers it's resolved.
Click to expand...


No need to get so defensive about a motherboard, I hope yours doesn't blow up your CPU, but I still wouldn't touch an ASROCK with a bargepole. It's clearly been an issue much more so than any other MOBO brand though.
 
I bought my AsRock murderboard (X870 Pro RS) in June 2025 and updated the BIOS to 3.25 before popping in my 9800X3D. I've been running BIOS 3.40 since it released in August 2025 and so far everything's running fine (everything running default voltages with XMP on).
BIOS 4.07.AS01[Beta] is the current one for my board but I don't want to be a tester. :cry: BIOS 4.10 is already out for a few AsRock AM5 boards so I'll likely update to it once it's available for my board.

EDIT: As of 15th Feb, I'm running BIOS 4.10 and everything is running nicely as before. Hopefully nothing craps the bed!
 
Last edited:
It is with a heavy heart that I announce the death of my 9800X3D, on 27th February 2026, which was housed in an ASRock X870 Pro RS murderboard. It was only 8-months old. :(

It happened while on the pause screen of Tainted Grail: The Fall of Avalon. At first I simply thought the system had crashed as there were no debug lights lit but I began to worry when I had to switch the system off at the PSU as holding down the power button did nothing.
After powering the system back on, the dreaded CPU and DRAM debug lights lit up and the system fans went to 100%. The system would not POST. Clearing the CMOS resulted in the same. It was at this point I came to the conclusion that the murderboard had claimed a victim.
I retrieved the corpse of my 9800X3D from the murderboard's silicone stained mitts but there was no sign of any damage, neither to the CPU pads or the socket pins. I tried to resuscitate the CPU in an Asus Tuf Gaming B650-E motherboard to no avail, it simply would not POST. The murderboard, however, managed to POST just fine with a 9950X but, worry not, I saved it from the killers clutches.

I contacted OcUK to get a replacement CPU which they were happy to do but they also offered to refund the murderboard. I was glad to send the foul villain to the proper authorities and the poor CPU to its final resting place though sending them together was one final insult (OcUK paid for a DPD collection). OcUK have shipped a new 9800X3D but it's still in transit so I'm using my faithful old 5800X system until it arrives.

I've bought a ROG Strix B650E-E as a replacement motherboard, hopefully that isn't a homicidal maniac too...
 
Aztec2Step said:
It is with a heavy heart that I announce the death of my 9800X3D, on 27th February 2026, which was housed in an ASRock X870 Pro RS murderboard. It was only 8-months old. :(

It happened while on the pause screen of Tainted Grail: The Fall of Avalon. At first I simply thought the system had crashed as there were no debug lights lit but I began to worry when I had to switch the system off at the PSU as holding down the power button did nothing.
After powering the system back on, the dreaded CPU and DRAM debug lights lit up and the system fans went to 100%. The system would not POST. Clearing the CMOS resulted in the same. It was at this point I came to the conclusion that the murderboard had claimed a victim.
I retrieved the corpse of my 9800X3D from the murderboard's silicone stained mitts but there was no sign of any damage, neither to the CPU pads or the socket pins. I tried to resuscitate the CPU in an Asus Tuf Gaming B650-E motherboard to no avail, it simply would not POST. The murderboard, however, managed to POST just fine with a 9950X but, worry not, I saved it from the killers clutches.

I contacted OcUK to get a replacement CPU which they were happy to do but they also offered to refund the murderboard. I was glad to send the foul villain to the proper authorities and the poor CPU to its final resting place though sending them together was one final insult (OcUK paid for a DPD collection). OcUK have shipped a new 9800X3D but it's still in transit so I'm using my faithful old 5800X system until it arrives.

I've bought a ROG Strix B650E-E as a replacement motherboard, hopefully that isn't a homicidal maniac too...
Click to expand...
Rip. Same crap happened to me
 
Aztec2Step said:
It is with a heavy heart that I announce the death of my 9800X3D, on 27th February 2026, which was housed in an ASRock X870 Pro RS murderboard. It was only 8-months old. :(

It happened while on the pause screen of Tainted Grail: The Fall of Avalon. At first I simply thought the system had crashed as there were no debug lights lit but I began to worry when I had to switch the system off at the PSU as holding down the power button did nothing.
After powering the system back on, the dreaded CPU and DRAM debug lights lit up and the system fans went to 100%. The system would not POST. Clearing the CMOS resulted in the same. It was at this point I came to the conclusion that the murderboard had claimed a victim.
I retrieved the corpse of my 9800X3D from the murderboard's silicone stained mitts but there was no sign of any damage, neither to the CPU pads or the socket pins. I tried to resuscitate the CPU in an Asus Tuf Gaming B650-E motherboard to no avail, it simply would not POST. The murderboard, however, managed to POST just fine with a 9950X but, worry not, I saved it from the killers clutches.

I contacted OcUK to get a replacement CPU which they were happy to do but they also offered to refund the murderboard. I was glad to send the foul villain to the proper authorities and the poor CPU to its final resting place though sending them together was one final insult (OcUK paid for a DPD collection). OcUK have shipped a new 9800X3D but it's still in transit so I'm using my faithful old 5800X system until it arrives.

I've bought a ROG Strix B650E-E as a replacement motherboard, hopefully that isn't a homicidal maniac too...
Click to expand...
I’m sorry for your loss at this difficult time.
 
LtMatt said:
I’m sorry for your loss at this difficult time.
Click to expand...
Thank you. At least I have my trusty 5800X's shoulder on which to weep.

And weep I shall as DPD didn't tell me they'd try delivering today, at my folks place for lunch and I get an email through saying they've missed me! :cry:
 
Aztec2Step said:
Thank you. At least I have my trusty 5800X's shoulder on which to weep.

And weep I shall as DPD didn't tell me they'd try delivering today, at my folks place for lunch and I get an email through saying they've missed me! :cry:
Click to expand...
I told you to get an Asus Strix B650E board after my good experiences with it, and knowing your luck, brother! But noooooooooo, you had to be different.

I knew there was something off with that Asrock board, the cold, dead LEDs that smiles never reached...
 
Kirby Wurm said:
I told you to get an Asus Strix B650E board after my good experiences with it, and knowing your luck, brother! But noooooooooo, you had to be different.

I knew there was something off with that Asrock board, the cold, dead LEDs that smiles never reached...
Click to expand...

Yes, telling me to buy a specific motherboard after I had already purchased one was a really good idea...
 
You must log in or register to reply here.
Back
Top Bottom