News Some Ryzen 7 9800X3D CPUs are allegedly dying prematurely — over 100 cases documented based on user feedback

[TerryLaze]

While article is worrying at 1st, the issue is ballooned way too big than it actually is.

No ,it's not.
People have to be made aware of issues, even if it user error, especially then, you want as many people to know about it so that DIYers can avoid blowing up their CPUs in the future at least.
Keeping it on the hush hush will just lead to more incidents until it does become a bigger issue.

 

[stuff and nonesense]

The hyperbole regarding a limited number of problems, 120 reported is inappropriate but will always happen. That so many were reported against one motherboard vendor suggests that the vendor has done something wrong, perhaps pushing limits to the point where the probability of failure is increased but nominally still within specifications.

While people should be made aware of potential weaknesses the scare stories should be negated. “Matter of fact” is the better way of reporting. Put the failures in context, look for commonalities. Fix the commonalities in the motherboard firmware and prevent the problems from occurring.

You have a choice when you buy a prebuilt PC, a mostly locked down dell/hp etc. or a system integrator using ASUS, MSI etc. that will give you access to the hardware and allow you to play. There are many YouTube videos that say .. do this, do that, there you go free performance. What they don’t say is don’t do this or that if you have a motherboard form Y.
Home assembled PCs are no different from system integrator PCs. You have potentially more choice regarding the hardware but they are no different. They use commodity parts.

If you run your kit at stock you have a more than reasonable expectation that it will be long lived and stable. Failures under warranty will be covered with minimal fuss.
Overclock, you may still get warranty coverage in the event of failure or you might have to tell a few fibs.

Both AMD and Intel are, while not responsible, complicit in the problem. They offer overclocking features in their chips/firmware. Users want this feature, gimme free stuff! Arguably this is better for them and users in so far as if they didn’t ham fisted “enthusiasts” would begin hardware modding motherboards as they did in the past.

Do you have any actual numbers? For both sides.

All we have is reported numbers, the AMD numbers might increase… but small, as of the article, 120. For the Intel problem, all of them 13 and 14 series 65W and up, stock or overclocked. 100% of them IF the bios isn’t updated to last summer’s release.

It isn’t close to being the same type of issue.

 

[TerryLaze]

All we have is reported numbers, the AMD numbers might increase… but small, as of the article, 120. For the Intel problem, all of them 13 and 14 series 65W and up, stock or overclocked. 100% of them IF the bios isn’t updated to last summer’s release.

It isn’t close to being the same type of issue.

It's not comparable though, you are absolutely right that It isn’t close to being the same type of issue, we have zero number of intel CPUs that stopped working due to the issue.
We don't even have any reliable accord on any number of intel CPUs dropping performance below the rated numbers.
We just know that there is some degradation on some CPUs if they were worked at close to their limits, because the bios fault would add to the max settings pushing them beyond the limit.
Tl;dr
Degradation=bad but stopped working=much worse.

 

[Aeacus]

People have to be made aware of issues, even if it user error, especially then, you want as many people to know about it so that DIYers can avoid blowing up their CPUs in the future at least.
Keeping it on the hush hush will just lead to more incidents until it does become a bigger issue.

Even when people are made aware, there are many who doesn't care and do things as they used to.

Prime example would be traffic accidents. Given how many accidents there are out there, whereby 90% are user error, are people really not aware of accidents in traffic? I'm sure that vast majority are, yet, we still get accidents.

Same with PC hardware. User error can not be completely ruled out. Hardware will continue to die in the hands of users. Has been so in the past - will be so in the future. Regardless how aware you make people about it.

If there is a fundamental flaw of something, a recall is issued. Until then - it's business as usual.

 

[stuff and nonesense]

It's not comparable though, you are absolutely right that It isn’t close to being the same type of issue, we have zero number of intel CPUs that stopped working due to the issue.
We don't even have any reliable accord on any number of intel CPUs dropping performance below the rated numbers.
We just know that there is some degradation on some CPUs if they were worked at close to their limits, because the bios fault would add to the max settings pushing them beyond the limit.
Tl;dr
Degradation=bad but stopped working=much worse.

We had, eventually, an acknowledgment from Intel that they had screwed up big time. We had many copium bios updates till they hit upon the bandaid. We had intel blaming everyone but themselves for 6 to 12 months before acknowledging where the fault laid. We got as a way of placating the masses an extension to the warranty period to cover any subsequent faults. Degradation and stopping are equally bad. I want to know my kit is going to be reliable.

I’d suggest you look at level 1 techs videos from the time. Look at Buildzoid’s videos analysing the problems even look at GN for his take.

Try to not conflate the current AMD problem with the intel debacle. Wait for the fixes to come out and then form an opinion.

 

[rambo919]

I don't get the obsession with conflating this with the intel degradation drama.... this is seemingly obviously a memory problem NOT a processor problem..... it's good old fashioned overclocking instability only this time it's NOT the processor being overclocked.... at worst the memory is dying not the processor..... point the fingers at least in the right direction.

In the case of Asrock.... it might not even be AMD's fault at all but it might be.... at least gets the facts validated before you panic and rage. This points to a INDUSTRY wide problem of motherboard manufacturers pushing out models before they properly did all the required R&D because everyone is in a massive rush to get their bling out before the competition..... how are you going to forever prevent randomly pushing garbage under that kind of self imposed pressure?

I wonder if half the cause for all the drama is people not wanting to admit that perhaps it's not really such a good idea to overclock memory in the first place? I am always surprised when I read of someone's RAM, GPU or CPU dying because in 20+ years I have never had either come even close to dying on me.... but then again I never overclock anything nor do I buy the absolute fastest available.

 

[TerryLaze]

for 6 to 12 months

I’d suggest you look at level 1 techs videos from the time. Look at Buildzoid’s videos analysing the problems even look at GN for his take.

Yeah, I'm not going to search through a year worth of the most active tech channels to look for something that might just be your imagination.

 

[stuff and nonesense]

4 element search, 13900k degradation Buildzoid YouTube returns this as the top result..

View: https://www.youtube.com/watch?v=eUzbNNhECp4


Like I said earlier, please don’t conflate 2 completely unrelated issues.

 

[Gururu]

4 element search, 13900k degradation Buildzoid YouTube returns this as the top result..

View: https://www.youtube.com/watch?v=eUzbNNhECp4


Like I said earlier, please don’t conflate 2 completely unrelated issues.

I am not sure anyone is disputing the issue with the 13th/14th Intels, or that Intel took forever to come to terms with it. I think all we can hope for is that the 9800x3d is fixed promptly, especially since it is far worse in severity.

 

[TerryLaze]

4 element search, 13900k degradation Buildzoid YouTube returns this as the top result..

So what does he say that proves that intel blamed anybody!?
He shows a lot of sources for what the issue could be and how to fix it but no sign of intel blaming anybody.

Like I said earlier, please don’t conflate 2 completely unrelated issues.

I would never in my life conflate losing your system with losing 5-10% performance or have the occasional crash if you don't let up on the max overclock.

 

[TerryLaze]

4 element search, 13900k degradation Buildzoid YouTube returns this as the top result..

You know why you think that intel blamed mobo makers and that all CPUs are affected?
Because you are watching entertainment channels on youtube while thinking that they are tech channels...

Look at this, Steve is saying that intel expanded the blame to mobo makers, while the highlighted part of the title he shows says "At this point one could almost believe that intel has simply left the whole thing to the motherboard manufacturers" So first of all it's just an opinion "one could believe" is not stating fact it's expressing an opinion, and second leaving the whole thing to the mobo makers is not blaming them, even if that happen which they don't show any proof of.

And then he says that the issue affects every CPU because the article says that every CPU got higher voltages, while the article also clearly says after that that not every CPU that got higher voltages will be impacted by the elevated voltages issue.

Thanks' Steve, in deed.
Anything for a click.

 

[stuff and nonesense]

Because you are watching entertainment channels on youtube while thinking that they are tech channels...

I take my choices, I look at a wide number of sources. I don’t judge you or your opinion and I will not justify my opinion to someone on the other side of the piece of glass I am currently viewing. Your opinion differs from mine but at no point have I inferred that you are an idiot. From your quote I am inferring that you believe I am and I will not stoop to your level.

 

[TerryLaze]

I take my choices, I look at a wide number of sources. I don’t judge you or your opinion and I will not justify my opinion to someone on the other side of the piece of glass I am currently viewing. Your opinion differs from mine but at no point have I inferred that you are an idiot. From your quote I am inferring that you believe I am and I will not stoop to your level.

The video is online and anybody can go watch it and see that Steve is lying through his teeth.
You believed him blindly, probably because he said exactly what you wanted to hear, just accept it and move on, there is no reason to pretend that you are still right.

 

[stuff and nonesense]

The video is online and anybody can go watch it and see that Steve is lying through his teeth.
You believed him blindly, probably because he said exactly what you wanted to hear, just accept it and move on, there is no reason to pretend that you are still right.

You know nothing about me or my background. You do not know what I read or watch. You are entitled to your opinion. Mine differs. At no point have I called you wrong, at no point have I said you are correct either. End this, do not reply to me.

 

[rambo919]

I wonder how much of this "Steve lies" I see now and then is based on love for Linus.....