navi 31 AMD Declines Radeon RX 7900 XTX RMA For Hitting 110C Junction Temps, Says “Temperatures Are Normal” (Updated #2)

[BiG StroOnZ]

Jay talks about the updated information from der8auer about the cooler and what some others have said:

 

 

[CommanderAlex]

6 minutes ago, BiG StroOnZ said:

Jay talks about the updated information from der8auer about the cooler and what some others have said:

 

 

This guy (Jay) literally just rips off others work. I swear I found der8auer's videos on this to be more interesting than the summary video. 

[BiG StroOnZ]

34 minutes ago, CommanderAlex said:

This guy (Jay) literally just rips off others work. I swear I found der8auer's videos on this to be more interesting than the summary video. 

 

He typically makes his own content. Sometimes for his talking head videos however, he may reference other Tech Tubers. Der8auer's videos will definitely be more in depth, but it doesn't hurt to hear other people's opinions on the subject. Jay is an acquired taste though. 

[Quackers101]

9 minutes ago, BiG StroOnZ said:

He typically makes his own content.

Jay is an acquired taste though. 

agree, but lately it might feel like just reposting others or no content in the video. else it might just be rants and goofing around, but a few times might test products they have.

[BiG StroOnZ]

3 minutes ago, Quackers101 said:

agree, but lately it might feel like just reposting others or no content in the video. else it might just be rants and goofing around, but a few times might test products they have.

 

From what I can tell, it might be because he's trying to increase the volume of videos he's putting out weekly. I noticed it has gone up recently, so he might feel obligated to cover whatever he can to simply be able to put out a video. I'm not complaining because it gives me more content to watch from my YT Subscription feed. 

[CommanderAlex]

17 minutes ago, BiG StroOnZ said:

 

He typically makes his own content. Sometimes for his talking head videos however, he may reference other Tech Tubers. Der8auer's videos will definitely be more in depth, but it doesn't hurt to hear other people's opinions on the subject. Jay is an acquired taste though. 

Yeah I get that, but his own content has just gone out the door. 

 

6 minutes ago, Quackers101 said:

agree, but lately it might feel like just reposting others or no content in the video. else it might just be rants and goofing around, but a few times might test products they have.

For sure definitely and the fact that the day his videos tend to be a couple days behind newer content on other channels is just something I don't want to hear. I knew in the first 10 seconds of that video @BiG StroOnZshared here knowing it was going to be on that. 

[BiG StroOnZ]

1 hour ago, CommanderAlex said:

Yeah I get that, but his own content has just gone out the door. 

 

For sure definitely and the fact that the day his videos tend to be a couple days behind newer content on other channels is just something I don't want to hear. I knew in the first 10 seconds of that video @BiG StroOnZshared here knowing it was going to be on that. 

 

I've been enjoying most of his content lately, I think for many Tech Tubers it's hard to remain fresh in this space. So I understand if content starts to become long in the tooth for others. That's why I subscribe to a lot of different content on YouTube, that way from the titles I'm able to differentiate what I'm going to be interested in watching. 

 

Anyway, found this on VideoCardz:

 

Quote

It is reported that the undisclosed OEM suspects at least one faulty batch of Radeon 7900 series might have left the factory. The possible issue being described is insufficient coolant added to the vapor chamber.

 

 

 

https://videocardz.com/newz/amd-partner-suspects-a-faulty-batch-of-reference-radeon-7900-gpus-is-affected-by-thermal-issues

 

This was before der8auer released the video of him opening up the vapor chamber. I've yet to watch it, but it doesn't seem like many conclusions were drawn from that video though.

[CommanderAlex]

21 minutes ago, BiG StroOnZ said:

 

I've been enjoying most of his content lately, I think for many Tech Tubers it's hard to remain fresh in this space. So I understand if content starts to become long in the tooth for others. That's why I subscribe to a lot of different content on YouTube, that way from the titles I'm able to differentiate what I'm going to be interested in watching. 

 

Anyway, found this on VideoCardz:

 

 

https://videocardz.com/newz/amd-partner-suspects-a-faulty-batch-of-reference-radeon-7900-gpus-is-affected-by-thermal-issues

 

This was before der8auer released the video of him opening up the vapor chamber. I've yet to watch it, but it doesn't seem like many conclusions were drawn from that video though.

Hmm interesting to see. Der8auer has stated in the latest video, that possibly is the cause but is unsure as he doesn't completely know the engineering behind vapor chambers, that they might not have enough water inside (I doubt it's water, wouldn't it corrode in copper and turn green like the Statue of Liberty?? Not an expert here but I do know some knowledge on common rxns). 

 

 

Jay is right towards the end of the video though how AMD has not said a word about this but yet they were so giddy to not use 12VHPWR like Nvidia, while AMD is facing an issue with the reference design of their own graphics cards and how, like @WereCat said the same thing in this post just recently. This should raise some eyebrows unless people are so tunnel-visioned that Nvidia is the bad guy here while AMD is just as bad.  

[HenrySalayne]

8 hours ago, CommanderAlex said:

(I doubt it's water, wouldn't it corrode in copper and turn green like the Statue of Liberty?? Not an expert here but I do know some knowledge on common rxns). 

You need more than just water for corrosion. A vapour chamber contains only water (gaseous and liquid) and maybe a inert gas to adjust the partial pressure (probably not).

Copper being exposed to the elements turning green is a more complicated process. It basically involves oxidation and a further reaction with CO₂ to form a carbonate.

[BiG StroOnZ]

Well looks like there's been some thorough developments to this story, will probably update OP ~

 

Thousands of AMD Radeon RX 7900 XTX reference cards could be affected by 110°C GPU hotspot issue. System integrators already returned hundreds of Radeon RX 7900 XTX MBA cards:

 

Quote

The problem with reference XTX cards thermal throttling has now been localized to the vapor chamber. In an email received from AMD by one of the affected customers (system integrator), we learn that an official AMD statement was to be made yesterday:

 

Quote

The problem is already known for more than two weeks and is communicated between AMD and the distributors (internally) .. How exactly the problem will be communicated to the end customer has not yet been finally decided. … AMD’s statement was originally supposed to be made on January 3, 2023 at 6 p.m. CET, but it was postponed.

[..] As assumed, the cause is the evaporator chamber… Several batches are affected. Currently, 4-6 batches and thousands of graphics cards are assumed. Only MBA cards are affected. 

 

— Quote from AMD system integrator email 

 

As mentioned, integrators have already returned hundreds of cards including those which were already assembled in complete systems. Returned cards included all AMD board partners cards (but only MBA cards).

 

It is very important to note that the problem only affects the RX 7900 XTX model and only AMD reference design (Made by AMD/MBA). This issue not impacting any of the custom cards which use non-reference cooling solutions. Furthermore, the reference RX 7900 XT model with cut-down Navi 31 GPU is not affected either.

 

Quote

[…] only the RX 7900 XTX’s functionally limited vapor chamber is such an essential problem that it requires an RMA. This is because it is not only the hotspot temperature, but also the storage tank, which can get as hot as 110 °C and thus operate well above the permissible temperature limit.

— Igor Wallossek

 

AMD has issued a short statement, finally admitting that there is a problem with reference cooling design:

 

Quote

We are working to determine the cause of the unexpected performance limitation of AMD Radeon RX 7900 XTX graphics cards. Based on our observations to date we believe the issue is related to the thermal solution used in AMD’s reference design and is occurring in a limited number of cards sold. We are working to resolve this issue for the affected cards. Customers experiencing this unexpected limitation should contact AMD support.

 

https://videocardz.com/newz/thousands-of-amd-radeon-rx-7900-xtx-reference-cards-could-be-affected-by-110c-gpu-hotspot-issue

[starsmine]

Official response is Small batch of MBA vapor chambers did not get enough water
Also sounds like an unofficial recall with a streamlined RMA process going forward. Sounds like at the warehouses they can swap parts, but I dont know if they pulled any from retail shelves that was not commented on, if the batch is small enough they might just be hopping it just sorts itself out.

 

[Radium_Angel]

Is this the first recall of GPUs? I can't...errr...recall any others happening, but then again I haven't been paying attention to such things in the past.

[leadeater]

56 minutes ago, Radium_Angel said:

Is this the first recall of GPUs? I can't...errr...recall any others happening, but then again I haven't been paying attention to such things in the past.

Not the first, EVGA had a couple. There are others I can't quite remember the specifics enough to mention them properly. There were some in the GTX 10 generation, the whole MLCC capacitor thing for the RTX 20 or 30 series, I forget which.

 

They all tend to be big news for all about 30 seconds.

 

I think GTX 970 technically ended up being a recall.

[Kisai]

1 hour ago, leadeater said:

Not the first, EVGA had a couple. There are others I can't quite remember the specifics enough to mention them properly. There were some in the GTX 10 generation, the whole MLCC capacitor thing for the RTX 20 or 30 series, I forget which.

 

They all tend to be big news for all about 30 seconds.

 

I think GTX 970 technically ended up being a recall.

The MLCC thing was the 30 series, specifically the 3080's MLCC vs POSCAP. 

https://www.gigabyte-data.com/media/news/48255

https://www.hardwaretimes.com/asus-rtx-3080-gpus-use-the-best-caps-followed-by-fe-zotac-has-the-worst/

[tim0901]

10 minutes ago, Kisai said:

The MLCC thing was the 30 series, specifically the 3080's MLCC vs POSCAP. 

https://www.gigabyte-data.com/media/news/48255

https://www.hardwaretimes.com/asus-rtx-3080-gpus-use-the-best-caps-followed-by-fe-zotac-has-the-worst/

Wasn't that all concluded to be more or less complete BS though, with the problem eventually being solved in software? Hence it vanished as a talking point after a few weeks once the new driver rolled out? I certainly don't remember there being actual recalls over it.

[Mihle]

On 1/4/2023 at 12:47 PM, HenrySalayne said:

You need more than just water for corrosion. A vapour chamber contains only water (gaseous and liquid) and maybe a inert gas to adjust the partial pressure (probably not).

Copper being exposed to the elements turning green is a more complicated process. It basically involves oxidation and a further reaction with CO₂ to form a carbonate.

Copper oxide, before it turns green because another reaction, can be different colours depending on what type and form it is in. It can be different kinds of red/brown, but also black. It can also be yellow in some forms.

[porina]

1 hour ago, leadeater said:

I think GTX 970 technically ended up being a recall.

I recall nvidia losing the class action over the 3.5GB VRAM thing, but don't recall a recall, unless it was for something else I don't recall?

[leadeater]

2 hours ago, porina said:

I recall nvidia losing the class action over the 3.5GB VRAM thing, but don't recall a recall, unless it was for something else I don't recall?

Yea, I'm just not sure as a result of that what they technically did was also a manufacturer recall. I think it was a compensation payout only.

 

2 hours ago, tim0901 said:

Wasn't that all concluded to be more or less complete BS though, with the problem eventually being solved in software? Hence it vanished as a talking point after a few weeks once the new driver rolled out? I certainly don't remember there being actual recalls over it.

MSI issued a recall on their RTX 3080 and RTX 3090 cards. Each AIB handled it differently. EVGA had sent pre-production card to reviewers with different capacitor design than to the retail cards so that was resolved by sending reviewers retail cards.

[divito]

On 1/3/2023 at 7:31 PM, CommanderAlex said:

This guy (Jay) literally just rips off others work. I swear I found der8auer's videos on this to be more interesting than the summary video. 

This is naivety. Are you expecting tech Youtuber's to never discuss tech news or the work of others just because they didn't personally do something? 

It would make sense if Jay produced no projects, sponsored videos, etc... but he intersperses "news" type videos when they are active and to what he finds relevant.

[Ripred]

All this is familiar  OOh yeah NAVI #1 

 

In which case there will be an update or GN mod video

[Deadpool2onBlu-Ray]

This whole GPU gen is honestly a giant turd

[CommanderAlex]

1 hour ago, divito said:

This is naivety. Are you expecting tech Youtuber's to never discuss tech news or the work of others just because they didn't personally do something? 

It would make sense if Jay produced no projects, sponsored videos, etc... but he intersperses "news" type videos when they are active and to what he finds relevant.

The only reference's Jay has been making to produce as evidence is other people's work. To me, that's being lazy and click bait. 

[divito]

5 minutes ago, CommanderAlex said:

The only reference's Jay has been making to produce as evidence is other people's work. To me, that's being lazy and click bait. 

So your assertion is that Youtubers, and news channels are lazy and click bait if they report stories without personally replicating? Never mind that, with that logic, any and all essays and research papers are the result of lazy people because they only reference other people's work. 

[starsmine]

31 minutes ago, divito said:

So your assertion is that Youtubers, and news channels are lazy and click bait if they report stories without personally replicating? Never mind that, with that logic, any and all essays and research papers are the result of lazy people because they only reference other people's work. 

Man We wish Jay was up to the standard of a high school research paper.

[CommanderAlex]

48 minutes ago, divito said:

So your assertion is that Youtubers, and news channels are lazy and click bait if they report stories without personally replicating? Never mind that, with that logic, any and all essays and research papers are the result of lazy people because they only reference other people's work. 

You're judgment is pretty shortsighted and clearly do not watch his channel that much, or just trying to pick a fight b/c I called it out.