AMD Radeon RX 7900 XTX May Feature Faulty Coolers, Causing Overheating

[AusWolf]

Igor has published a new in-depth analysis of the supply chain and quality control mechanisms regarding the issues with the vapor chamber.

Unfortunately it is only available in German: https://www.igorslab.de/einzelfaell...x-7900-xtx-kuehler-problem-samt-qm-qa-und-qc/

Here is a short rather long summary:

• The whole card is manufactured by PC Partner, the cooler is made by Cooler Master, the vapor chamber is supplied by AVC (Asia Vital Components).
• As we all know by now the issue is caused by too little liquid in the vapor chamber. As far as Igor knows one charge of vapor chambers from AVC was affected which usually consists of 10000 units.
• Not all vapor chambers from that charge are fully defective, some work ok-ish, some show the 110°C issue.
• This slipped through quality control at Cooler Master and PC Partner which casts a damning light on their quality processes.
• To make matters worse vapor chambers and coolers are not serial number matched. That means the defective charge of 10000 vapor chambers is split among several charges of coolers and was mixed with non-defective charges.
• This means AMD has no way to match the serial numbers of the cards with the defective coolers, they can not issue a list of defective cards and the customer has to find out by himself.
• The amount of cards affected is quite high, AIBs (like Saphhire and Power Color) report ~10% of RMAs of their cards.
• At the beginning AMD support offered replacements within 2 weeks, now they say they can not provide a delivery date as their shelves are empty and rather propose a refund. This might be worsened by the problem that they themselves do not know which cards are affected and do not want to ship out another faulty card to an already unsatisfied customer.
• For us end consumers this is annoying, for the bigger system integrators who build hundreds of systems this is a real problem since they now sit on unfinished PCs they can not sell.

This is probably the most useful post along the 10 pages in the entire thread. It should be pinned to the top.

I wonder where der8auer is now to tell us that it's all crap and AMD hates everyone.

 

[nguyen]

Man AMD is an endless source for meme 2023

 

[OfficerTux]

At least it's just pref hit, no real so even if affected you can keep use it albeit at the cost of noise. When summer come it will only get worse.

That is not totally true. In the cases where the hotspot hits 110°C the DRAM often also is at > 100°C and it is not rated for those temps so it might shorten its lifespan.

(Example: https://www.digikey.com/en/products/detail/micron-technology-inc/MT61K512M32KPA-16-B/10321310, rated up to 95°C)

 

[kapone32]

Igor has published a new in-depth analysis of the supply chain and quality control mechanisms regarding the issues with the vapor chamber.

Unfortunately it is only available in German: https://www.igorslab.de/einzelfaell...x-7900-xtx-kuehler-problem-samt-qm-qa-und-qc/

Here is a short rather long summary:

• The whole card is manufactured by PC Partner, the cooler is made by Cooler Master, the vapor chamber is supplied by AVC (Asia Vital Components).
• As we all know by now the issue is caused by too little liquid in the vapor chamber. As far as Igor knows one charge of vapor chambers from AVC was affected which usually consists of 10000 units.
• Not all vapor chambers from that charge are fully defective, some work ok-ish, some show the 110°C issue.
• This slipped through quality control at Cooler Master and PC Partner which casts a damning light on their quality processes.
• To make matters worse vapor chambers and coolers are not serial number matched. That means the defective charge of 10000 vapor chambers is split among several charges of coolers and was mixed with non-defective charges.
• This means AMD has no way to match the serial numbers of the cards with the defective coolers, they can not issue a list of defective cards and the customer has to find out by himself.
• The amount of cards affected is quite high, AIBs (like Saphhire and Power Color) report ~10% of RMAs of their cards.
• At the beginning AMD support offered replacements within 2 weeks, now they say they can not provide a delivery date as their shelves are empty and rather propose a refund. This might be worsened by the problem that they themselves do not know which cards are affected and do not want to ship out another faulty card to an already unsatisfied customer.
• For us end consumers this is annoying, for the bigger system integrators who build hundreds of systems this is a real problem since they now sit on unfinished PCs they can not sell.

I surmise the first batch of reference cards is usually the only mass produced run. Maybe there now will be another run though. At least they caught it quickly and AMD are working hard to fix it. When you talk about QC there is help needed everywhere. I went to Niagara falls late last year and in my hotel they had 1 person working at the Starbucks downstairs.....in a Vacation town on a Saturday morning? This is not good but at least it is much clearer now. Thanks for posting.

That is not totally true. In the cases where the hotspot hits 110°C the DRAM often also is at > 100°C and it is not rated for those temps so it might shorten its lifespan.

(Example: https://www.digikey.com/en/products/detail/micron-technology-inc/MT61K512M32KPA-16-B/10321310, rated up to 95°C)

My memory sits at 58C for Memory Junction temp at idle in HWinfo64. I had a reset and zero fan was reapplied I have seen that temp as high as 96 C.

 

[OfficerTux]

This is probably the most useful post along the 10 pages in the entire thread. It should be pinned to the top.

I wonder where der8auer is now to tell us that it's all crap and AMD hates everyone.

First of all, thanks for the kudos.

But I feel like many here did not see the full videos from der8auer but just pick the parts that make him look bad. For an example in his follow up video he explicitly says that the issue with the broken card he checks may be related to the vapor chamber or not and that you have to be careful with single cases and that they do not reflect the majority of cards.

I like his videos, but that's just my opinion. I watch them in German though, maybe his English translations are not as good.

 

[AusWolf]

First of all, thanks for the kudos.

But I feel like many here did not see the full videos from der8auer but just pick the parts that make him look bad. For an example in his follow up video he explicitly says that the issue with the broken card he checks may be related to the vapor chamber or not and that you have to be careful with single cases and that they do not reflect the majority of cards.

I like his videos, but that's just my opinion. I watch them in German though, maybe his English translations are not as good.

I watched his (English) videos on the matter and wasn't impressed. His testing was lacking (he talked about throttling, but he didn't show any data related to clock speeds or game performance), and then he made his conclusion entirely about his own opinion that he based on anecdotal evidence which isn't related to his own testing (how widespread he thinks the issue is). Then he went on to create an opinion piece on the AMD statement that he thinks is wrong because "AMD should know the serial number of each and every cooler on each and every card that they sold", not accounting for the partner cards that also come with the same cooler. His vapor chamber analysis was alright, but it didn't turn up anything useful.

Never mind... thanks for the Igor's Lab reference and the summary.

 

[Dirt Chip]

First of all, thanks for the kudos.

But I feel like many here did not see the full videos from der8auer but just pick the parts that make him look bad. For an example in his follow up video he explicitly says that the issue with the broken card he checks may be related to the vapor chamber or not and that you have to be careful with single cases and that they do not reflect the majority of cards.

I like his videos, but that's just my opinion. I watch them in German though, maybe his English translations are not as good.

Yep, you can see correlation of over-criticism toward der8auer and under-criticism toward AMD on that matter.
He`s done proper in-depth HW testing for that matter and reported very transparently with many reservation for his own finding plus the thing you shouldn't conclude incorrectly from his test.

 

[big_glasses]

Thank you @kapone32 very interesting article from Igors

He did post a translated one also (machine translated? was readable at least)

Vapor gate including failure of the entire quality management? Background on the cooler problem of the AMD Radeon RX 7900 XTX (MBA) | igor´sLAB

Before I mention more numbers about the Radeon RX 7900 XTX and the cooler problem, as well as a few words about the supply chain and its failure in quality management, we also have to talk about…

www.igorslab.de

Yea, proper hard to find the scope of the issue when it's like that. And might also be NDA's or similar agreements, that prevents early statements.

 

[AusWolf]

Yep, you can see correlation of over-criticism toward der8auer and under-criticism toward AMD on that matter.
He`s done proper in-depth HW testing for that matter and reported very transparently with many reservation for his own finding plus the thing you shouldn't conclude incorrectly from his test.

I missed quite a few things from his tests (see above - clock speeds, game performance, stability through extended periods, maybe heatsink temperature, infrared pictures, etc.), and his conclusion was just unrelated hot air. And then his opinion video on the statement without knowledge on how the supply chain works is so unprofessional.

But let's agree to disagree.

 

[Dirt Chip]

I missed quite a few things from his tests (see above - clock speeds, game performance, stability through extended periods, maybe heatsink temperature, infrared pictures, etc.), and his conclusion was just unrelated hot air. And then his opinion video on the statement without knowledge on how the supply chain works is so unprofessional.

But let's agree to disagree.

No wrong in criticize him but don`t forget to do the same to AMD as it is their responsibility, after all.
His reporting isn`t perfect, but non of it make AMD situation better or his finding wrong. At least I can see his effort to be accurate and to inform you when it`s his opinion and not hard fact`s. Can you say it about AMD in this case?
According to igors lab write (I just read the whole piece) ,AMD statement sound detached and based on wishful thinking or just an attempt to dwarf the situation, as it is not a small issue. ~10% RMA, when the usual total rate is sub 1%, and no way to track those affected units which makes it even worse is a blander and deserve criticism just as der8auer deserve it, if not more.

And as you suggested, I agree to disagree

 

[AusWolf]

No wrong in criticize him but don`t forget to do the same to AMD as it is their responsibility, after all.
His reporting isn`t perfect, but non of it make AMD situation better or his finding wrong. At least I can see his effort to be accurate and to inform you when it`s his opinion and not hard fact`s. Can you say it about AMD in this case?
According to igors lab write (I just read the whole piece) ,AMD statement sound detached and based on wishful thinking or just an attempt to dwarf the situation, as it is not a small issue. ~10% RMA, when the usual rate is sub 1%, and no way to track those affected units which makes it even worse is a blander and deserve criticism just as der8auer deserve it, if not more.

I don't think der8auer's testing was even half as extensive as it should have been. Giving us only temperature data with a brief mention of some kind of "throttling" that he only deducted based on some reduced power consumption figure that also wasn't shown? It's very weak.

As for AMD, they're stuck between a rock and a hard place. They could recall the entire line, but that would come at tremendous effort and cost if you consider worldwide shipping, storage, testing, redistribution of working units, etc., not to mention PR damage. RDNA 3 would be doomed for good.

The other option is what they chose instead: let the individual owners get in touch. This also comes with costs (support line, testing, replacement units, etc), albeit a lot less, as they only have to move the faulty units and the replacements, not all of them. It also doesn't come with such a massive PR damage. The flip side is that it's more effort for the consumer.

You can hate AMD for not going with the more costly, but more consumer-friendly option, but I personally don't. The end result for you is the same: your card gets replaced. Nobody knows when, as this fiasco wasn't planned. It sucks, I don't deny that. As for AMD, the house is on fire already, the question is how they want to save what can be saved. There is no single good option here, imo, and I can't blame them for picking the one that potentially causes less PR damage, especially since an error in cooler manufacturing wasn't their fault. It's Coolermaster that should pay the price, if anyone.

 

[OfficerTux]

It's Coolermaster that should pay the price, if anyone.

Or AVC, or PC Partner. We will probably never know who pays for this in the end, I think it would be really interesting to know how they handle this internally.

AVC and PC Partner have quoted shares though, so we might see some movement there in the future (in negative direction):

AVC: https://www.google.com/finance/quote/3017:TPE
PC Partner: https://www.google.com/finance/quote/1263:HKG

 

[AusWolf]

Or AVC, or PC Partner. We will probably never know who pays for this in the end, I think it would be really interesting to know how they handle this internally.

Agreed, although I also don't think we'll hear much of it, unfortunately.

 

[Dirt Chip]

You can hate AMD for not going with the more costly, but more consumer-friendly option, but I personally don't. The end result for you is the same: your card gets replaced. Nobody knows when, as this fiasco wasn't planned. It sucks, I don't deny that. As for AMD, the house is on fire already, the question is how they want to save what can be saved. There is no single good option here, imo, and I can't blame them for picking the one that potentially causes less PR damage, especially since an error in cooler manufacturing wasn't their fault. It's Coolermaster that should pay the price, if anyone.

No hate whatsoever as I don`t have an affected card (or any) by them.
I think that if they really can`t trace the bad units than a recall from already sold units is irrelevant as no real danger is present.
It is very good to suggest consumers an immediate refund and mentioning this option it in the video by Scott could have help the PR wise.
But they do need to stop the shipment of MBA XTX and contact vendors and inform them about the situation now that they know of the problem, unless they put a big disclaimer on each unit that the product might behave below expected...
To knowingly keep selling products that 10% of them are affected and to really on the consumer to find for them the bad ones (because they can't do the serial match to bad coolers) is not cool at all. It will also keep this whole situation much longer and AMD will be the one to take the hit.
In the end, I worry that AMD might pave the way to more bad PR and elongation of this 'Vapor gate' saga (as Igor called it) by their own way of handling things. It can cost them more then a recall (PR and $ wise) if things will turn wrong and the ground is shaky atm.

 

[sith'ari]

Igor has published a new in-depth analysis of the supply chain and quality control mechanisms regarding the issues with the vapor chamber.

Unfortunately it is only available in German: https://www.igorslab.de/einzelfaell...x-7900-xtx-kuehler-problem-samt-qm-qa-und-qc/

Here is a short rather long summary:

• The whole card is manufactured by PC Partner, the cooler is made by Cooler Master, the vapor chamber is supplied by AVC (Asia Vital Components).
• As we all know by now the issue is caused by too little liquid in the vapor chamber. As far as Igor knows one charge of vapor chambers from AVC was affected which usually consists of 10000 units.
• Not all vapor chambers from that charge are fully defective, some work ok-ish, some show the 110°C issue.
• This slipped through quality control at Cooler Master and PC Partner which casts a damning light on their quality processes.
• To make matters worse vapor chambers and coolers are not serial number matched. That means the defective charge of 10000 vapor chambers is split among several charges of coolers and was mixed with non-defective charges.
• This means AMD has no way to match the serial numbers of the cards with the defective coolers, they can not issue a list of defective cards and the customer has to find out by himself.
• The amount of cards affected is quite high, AIBs (like Saphhire and Power Color) report ~10% of RMAs of their cards.
• At the beginning AMD support offered replacements within 2 weeks, now they say they can not provide a delivery date as their shelves are empty and rather propose a refund. This might be worsened by the problem that they themselves do not know which cards are affected and do not want to ship out another faulty card to an already unsatisfied customer.
• For us end consumers this is annoying, for the bigger system integrators who build hundreds of systems this is a real problem since they now sit on unfinished PCs they can not sell.

This is probably the most useful post along the 10 pages in the entire thread. It should be pinned to the top.

I wonder where der8auer is now to tell us that it's all crap and AMD hates everyone.

Well if you actually watch Der8auer's video , at 5:15 he is saying that it was Igor who pointed out that those vapor-chambers have a QR code ,thus the defective-batch can be detected.
Now i'm seeing here that Igor says that those vapor chambers don't have a serial ,thus the defective ones can not be detected.
Clearly these are conflicting info and you can't blame Der8auer if what Igor is saying is conflicting. (and i won't believe that Der8auer would ever mention Igor if Igor didn't actually said such a thing ).
--P.S. also , if we look Igor's quoted notes here , we clearly see that AMD's AIBs report around 10% RMAs ,and that they can be up to 10.000 defective vapor chambers ,so clearly NOT a "very small batch" of defective products like Herkelman claimed.

 

[TheoneandonlyMrK]

Seems to me like the most invested in this debate are Nvidia owners who self profess they wouldn't own AMD anyway but clearly like Der8aure, or to regurgitate add nauseum there asinine argument.

As for Der8aure, I'm done with his shit, 3 videos on this bullshit only one of which had any effort invested IE the first.

Beyond that the guy is looking like a click bandit who has been paid to defame AMD.

Like showing an already broken card somehow break more, truly revealing.

My hot take f him I'm unsubbed now.

 

[OfficerTux]

or to regurgitate add nauseum there asinine argument

Damn, that were a lot of words I had to look up, thanks for expanding my vocabulary

 

[AusWolf]

No hate whatsoever as I don`t have an affected card (or any) by them.
I think that if they really can`t trace the bad units than a recall from already sold units is irrelevant as no real danger is present.
It is very good to suggest consumers an immediate refund and mentioning this option it in the video by Scott could have help the PR wise.
But they do need to stop the shipment of MBA XTX and contact vendors and inform them about the situation now that they know of the problem, unless they put a big disclaimer on each unit that the product might behave below expected...
To knowingly keep selling products that 10% of them are affected and to really on the consumer to find for them the bad ones (because they can't do the serial match to bad coolers) is not cool at all. It will also keep this whole situation much longer and AMD will be the one to take the hit.
In the end, I worry that AMD might pave the way to more bad PR and elongation of this 'Vapor gate' saga (as Igor called it) by their own way of handling things. It can cost them more then a recall (PR and $ wise) if things will turn wrong and the ground is shaky atm.

This is certainly the longer way to the end, but I don't think it'll incur more cost and damage than a recall would. AMD decided to go this route, and I assume, not without a reason. They're taking a hit either way, the question is only how big. We'll see anyway.

Well if you actually watch Der8auer's video , at 5:15 he is saying that it was Igor who pointed out that those vapor-chambers have a QR code ,thus the defective-batch can be detected.
Now i'm seeing here that Igor says that those vapor chambers don't have a serial ,thus the defective ones can not be detected.
Clearly these are conflicting info and you can't blame Der8auer if what Igor is saying is conflicting. (and i won't believe that Der8auer would ever mention Igor if Igor didn't actually said such a thing ).
--P.S. also , if we look Igor's quoted notes here , we clearly see that AMD's AIBs report around 10% RMAs ,and that they can be up to 10.000 defective vapor chambers ,so clearly NOT a "very small batch" of defective products like Herkelman claimed.

Well, I do blame der8auer for going ahead guns blazing to make an opinion piece on something he clearly doesn't know. Just because Igor said something, it doesn't mean you need to make a video about it and make a fool of yourself in the process. If Igor says jumping in front of a train solves the issue, would you make a video doing it? One source is no source, as we used to say at uni.

Sure, 10% RMA is not a small number, but what's your point?

In one of his news videos, Steve from Gamer's Nexus said he didn't report on this issue because most of his team was on holiday, and he was too busy with something else. Given the magnitude of the case, I don't think that's entirely true. I think he just didn't want to step in this hot mess, and be another der8auer reporting half news with half-done tests, and I don't blame him.

 

[oliverprescott]

This is certainly the longer way to the end, but I don't think it'll incur more cost and damage than a recall would. AMD decided to go this route, and I assume, not without a reason. They're taking a hit either way, the question is only how big. We'll see anyway.


Well, I do blame der8auer for going ahead guns blazing to make an opinion piece on something he clearly doesn't know. Just because Igor said something, it doesn't mean you need to make a video about it and make a fool of yourself in the process. If Igor says jumping in front of a train solves the issue, would you make a video doing it? One source is no source, as we used to say at uni.

Sure, 10% RMA is not a small number, but what's your point?

In one of his news videos, Steve from Gamer's Nexus said he didn't report on this issue because most of his team was on holiday, and he was too busy with something else. Given the magnitude of the case, I don't think that's entirely true. I think he just didn't want to step in this hot mess, and be another der8auer reporting half news with half-done tests, and I don't blame him.

does Igor say where he gets the numbers from in his article? I can't seem to find any citations for that source? other than saying he has some sources and some info is industry standard I'm curious where he gets the batch number sizes and the RMA percentages. that seems like super super important info and with the reporting of this so far I'm not giving anyone the benefit of doubt. that last video from der8auer really demonstrates he doesn't know what he's talking about and making very definitive sweeping statements without a full understanding of the situation destroys his credibility.

 

[sith'ari]

This is certainly the longer way to the end, but I don't think it'll incur more cost and damage than a recall would. AMD decided to go this route, and I assume, not without a reason. They're taking a hit either way, the question is only how big. We'll see anyway.


Well, I do blame der8auer for going ahead guns blazing to make an opinion piece on something he clearly doesn't know. Just because Igor said something, it doesn't mean you need to make a video about it and make a fool of yourself in the process. If Igor says jumping in front of a train solves the issue, would you make a video doing it? One source is no source, as we used to say at uni.

Sure, 10% RMA is not a small number, but what's your point?

In one of his news videos, Steve from Gamer's Nexus said he didn't report on this issue because most of his team was on holiday, and he was too busy with something else. Given the magnitude of the case, I don't think that's entirely true. I think he just didn't want to step in this hot mess, and be another der8auer reporting half news with half-done tests, and I don't blame him.

My point is that 10% RMA is a very high number ( as you also say ).
And if you accept Igor's quoted statements (you said it yourself that this was the most important post on this thread) ,then this means that the 10% is also accepted.
If so , then why nobody accuses Herkelman who tried to downplay the significance of the matter ,and used the term "very small batch" while many here seem to have a problem with Der8auer's analysis ?
Herkelman's analysis was also incorrect at best , misleading at worse , if Igor's AIB-RMA numbers are accurate.
But if we can use Igor's statements as a tool in order to blame Der8auer , then we have to also use Igor's statement as a tool in order to blame Herkelman , and clearly this is not what happening here.
That's my point.

 

[AusWolf]

My point is that 10% RMA is a very high number ( as you also say ).
And if you accept Igor's quoted statements (you said it yourself that this was the most important post on this thread) ,then this means that the 10% is also accepted.
If so , then why nobody accuses Herkelman who tried to downplay the significance of the matter ,and used the term "very small batch" while many here seem to have a problem with Der8auer's analysis ?
Herkelman's analysis was also incorrect at best , misleading at worse , if Igor's AIB-RMA numbers are accurate.
But if we can use Igor's statements as a tool in order to blame Der8auer , then we have to also use Igor's statement as a tool in order to blame Herkelman , and clearly this is not what happening here.
That's my point.

I think some terms are wrong, or at least misunderstood here.

1. You think 10% RMA is a high number. I agree. Herkelman does not. I still don't see what the point is here, or how it changes anything. Do you think AMD should change their approach to the situation? I personally do not. Recalling the bad 10% along with the 90% of perfectly good products is a huge waste of money and effort that also comes with tremendous PR damage. It's unnecessary, impractical and unreasonable.

2. Igor making a statement on his opinion (or on whatever), and then der8auer parroting it are two different things. If you comment on something that you don't know, you're making a fool of yourself. But if you comment on someone's ignoramus comment, then doubly so. Not to mention if you're a public figure, you should at least do the courtesy to your fan base to come up with something original. Fact checking is also not totally unheard of, you know. Just because Igor says something, it doesn't mean it's fully right, and you should make a video on it. Like I said before: one source is no source. If you go ahead and make an opinion piece on it without checking or trying to obtain first-hand information anyway, then you're the fool. Even the analysis that he did do on his own was weak and lacking data that could have supported his claims a lot better. I equally disapprove of JayzTwoCents for making a video on der8auer's (weak) analysis, commenting totally false things like XT models are also affected, and such. Jay should know better to make his own content instead of parroting someone else's.

TL,DR: I'm not using Igor's statement to blame der8auer. I'm blaming der8auer for using Igor's statement without checking facts or using multiple sources to make a quick clickbait video.

3. What do you mean "Herkelman's analysis"? What did he analyze? As far as I remember, he only issued AMD's statement on the matter. There was no analysis. Whether he thinks that 10% is a high number or not is irrelevant. Owners of faulty cards need to contact AMD support for a replacement or refund. End of.

 

[shovenose]

Take any of these articles and YouTubers with a grain of salt they all parrot and blow this stuff out of proportion because they benefit from that.

yes, this is a true, bad issue but we have no idea how much of that conjecture is true.

 

[sith'ari]

1. You think 10% RMA is a high number. I agree. Herkelman does not. I still don't see what the point is here, or how it changes anything. Do you think AMD should change their approach to the situation? I personally do not. Recalling the bad 10% along with the 90% of perfectly good products is a huge waste of money and effort that also comes with tremendous PR damage. It's unnecessary, impractical and unreasonable.

First of all i don't care if AMD will recall or not those faulty parts. This is something that should mostly concern AMD's GPU customers and i'm not one of them.
(*although i always prefer AMD when it comes to CPUs compared to Intel )
The initial reports were saying that those parts have QR code ,so they can be spotted(* thus they could be recalled as well ).
A later report says that it seems that those parts don't have anything that could lead anyone to spot them(very weird from logical point of view , but anyway) .
So as i said before ,these are conflicting info , thus i can't conclude or think anything before this matter doesn't get completely clear.

But what i can judge is the morality of Herkelman's/AMD's statements :
So ,apparently , what you ,or myself , and i'm sure the vast majority of customers as well , would judge as a high RMA percentage , another person (Herkelman) who also ,coincidentally ,happens to be the representative of the company who sells those faulty products ,speaks about a .... "small batch".
If you combine that ... "evaluation gap" , with the fact that also Herkelman reffers to a "small performance delta" from the throttling ,while Der8auer ,through his tests reports a hit in performance up to 10-20% (in his video he mentioned that many of those faulty coolers can lack about 70-80watt of heat dissipation ,compared to the good ones ) .

so ,with these 2 combined ,it leads me to conclude that either :
A)AMD's representative has a (very)serious issue whenever he tries to evaluate percentages/numbers !!! , or
B)There is something wrong with AMD's approach on the matter , trying to lower its significance to the(their) customers.
Which is more likely , i wonder ?
Anyway ,i don't have the intention to keep this discussion going ,it seems futile ,so , just like you wrote : "End of"

 

[kapone32]

For me this issue is dead AMD have already responded. I have some experience in Manufacturing and Logistics and what Igor describes is so common in the market that it made complete Common sense. Even though the GPUs have a QR code that does not mean that the Cooler and heatsink will be as granular in their information. AMD may not have stock because the Container(s) may be enroute or sitting on a dock as we speak.

There is so much propaganda about both Companies that it is sickening but watching Youtube tech "Influencers" making videos asking the question as if a Driver could actually make a GPU explode but leave the fuses intact. Electronics 101 speaks to major conductive contact on that GPU but I guess that is not in the wheelhouse of some of these individuals. The problem is that Millions of people think that these people know more than engineers about PC and will be quick to jump on the bandwagon.

Even here on TPU I have heard the opinions of some that are worrying. Someone said that Nvidia make better cards in construction but is there a difference between the MSI TRIO 6800XT or 3080 in terms of construction? People will say that AMD is Garbage and the Management is lost without realizing that Freesync, FSR and AMD software were all free and adopted. I sadly saw a user write at least a page of comments about how the 4070TI was a better GPU to get than the 7900XT or 7900XTX because they were not as good at DLSS and Ray Tracing. Yes reviews are part of the problem as (I say this all the time) when Gaming became Rasterization it started us down this rabbit hole of Features for Gaming. The problem is Freesync is that good at what it does that as long as your output is in the range 48-165HZ you will have no issue with what use to plague Games traditionally but people are fixated on FPS. So much so that users will argue with owners of the X3D chips on how it cannot be a better experience than the 5800X because of Clock speed.

 

[AusWolf]

First of all i don't care if AMD will recall or not those faulty parts. This is something that should mostly concern AMD's GPU customers and i'm not one of them.
(*although i always prefer AMD when it comes to CPUs compared to Intel )
The initial reports were saying that those parts have QR code ,so they can be spotted(* thus they could be recalled as well ).
A later report says that it seems that those parts don't have anything that could lead anyone to spot them(very weird from logical point of view , but anyway) .
So as i said before ,these are conflicting info , thus i can't conclude or think anything before this matter doesn't get completely clear.

But what i can judge is the morality of Herkelman's/AMD's statements :
So ,apparently , what you ,or myself , and i'm sure the vast majority of customers as well , would judge as a high RMA percentage , another person (Herkelman) who also ,coincidentally ,happens to be the representative of the company who sells those faulty products ,speaks about a .... "small batch".
If you combine that ... "evaluation gap" , with the fact that also Herkelman reffers to a "small performance delta" from the throttling ,while Der8auer ,through his tests reports a hit in performance up to 10-20% (in his video he mentioned that many of those faulty coolers can lack about 70-80watt of heat dissipation ,compared to the good ones ) .

so ,with these 2 combined ,it leads me to conclude that either :
A)AMD's representative has a (very)serious issue whenever he tries to evaluate percentages/numbers !!! , or
B)There is something wrong with AMD's approach on the matter , trying to lower its significance to the(their) customers.
Which is more likely , i wonder ?
Anyway ,i don't have the intention to keep this discussion going ,it seems futile ,so , just like you wrote : "End of"

Are we arguing semantics now? "Small" and "high" can mean different things for different people. For example, you say 10% performance difference is a lot. For me, it's nothing. I don't even consider looking at a new GPU if it's not at least 50% better than the one I have (unless I'm really curious about the architecture or something). These are all subjective, and in the end: meaningless.