News Defective Vapor Chamber May Be Causing RX 7900 XTX Overheating Issue

Toggle sidebar Toggle sidebar
the melting connector is worse then this, at least with this, you could just replace the cooler on the card, and keep using it, the power connector on the rtx cards melting, means need a replacement card.

thisisaname said:
A great example of pride coming before a fall,
Click to expand...
i wonder how many times nvidia did the same to amd only to have smething go wrong with their own products.
 
  • Like
Reactions: thisisaname, King_V and phenomiix6
Will AMD take as much time as nvidia to come foward and explain whats going on, and how are they going to really deal with this ?

I guess I'm lucky I don't have enough spare money to get any new tech product for now.
 
USAFRet said:
Both problems should never have made it past inhouse testing.
Click to expand...
People failing to plug in their cables all the way isn't really Nvidia's fault. Also, in-house testing is only meant to catch problems that may arise when the product is used as intended and the plug isn't intended to be 2-3mm short of fully inserted. There does not appear to have been any further public reporting on this issue after it got confirmed as user error in just about every case that got looked into back in October.

AMD however appears to have a design flaw that may affect all GPUs using the reference cooler. I'm looking forward to heavily discounted 7000-series GPUs due to this epic fail giving the series a bad reputation.
 
  • Like
Reactions: panathas, RandomWan, daworstplaya and 7 others
InvalidError said:
People failing to plug in their cables all the way isn't really Nvidia's fault...
Click to expand...

This is very much Nvidia's fault since you must take user error into account when you design these things. Besides, the simple fact that most of us have been dealing with half-plugged cables (USB, SATA, IDE, power, you name it) our entire life without any of them catching fire should tell you that this wasn't normal.

But then I guess both companies were more too worried about making a quick profit off their vastly overpriced products to worry about quality control...
 
  • Like
Reactions: thisisaname, PEnns, atomicWAR and 4 others
Buildzoid was first to find out why Asus mobos catch fire. GN was first to find out why NVidia connectors melt. And now derBauer is first to find why AMD cards are cooking themselves.

May I ask, where the hell are quality control departments in hardware companies these days? It's great that we have famous youtubers doing their work but come on - it is not supposed to be this way!
 
  • Like
Reactions: Hooda Thunkett, atomicWAR, carvedinside and 3 others
atomicWAR said:
Leave it to AMD to make the over priced 4070ti look like a decent buy through defective(?) Founders edition cards of their own.

They better get a handle on these cards or it will be way worse for them than it was for Nvidia and the power connector.
Click to expand...
I will agree this is arguably worse than the power connector thing not because of the potential fire hazard or cards failing, but because it's way more common and will, more than likely, affect way more people in operation. The PCB and GPU (chip) themselves are fine, but the cooler for REFERENCE (not the same thing as nVidia's; FE is exclusive to nVidia and partners can't build FEs) cards has a problem. If you bought a RX7900 for water cooling from a reference model, then you're fine. This issue won't impact you.

BX4096 said:
This is very much Nvidia's fault since you must take user error into account when you design these things. Besides, the simple fact that most of us have been dealing with half-plugged cables (USB, SATA, IDE, power, you name it) our entire life without any of them catching fire should tell you that this wasn't normal.

But then I guess both companies were more too worried about making a quick profit off their vastly overpriced products to worry about quality control...
Click to expand...
Not really... I don't disagree the design of the 12VHP connector is yanky/questionable, but it does have a way to be connected properly and as long as people was careful connecting it, then they wouldn't have the issue. Yes, 'd like a better design for it, but as it stands, it's not "unsafe" as long as you know what you're doing. I like high power connectors to be a tad more "idiot proof", but oh welp.

AMD's problem comes from the factory and it's affecting more people in a tangible way. It has a, potentially, easy fix, but it's still something a $1K card should not be having. AMD better do early adopters good and recall them. Also, just as I said, the PCB and GPU itself doesn't have a problem, just the cooler has the problem, so if you want to ditch the reference HSF, the problem disappears.

Regards.
 
  • Like
Reactions: RandomWan, digitalgriffin, Coffee Fueled Curmudgeon and 1 other person
DRagor said:
May I ask, where the hell are quality control departments in hardware companies these days?
Click to expand...
That's one of the first departments to get trimmed down when companies get big to the point that moving volume starts to overshadow quality.
Even if publicity on a product is poor, it doesn't really hurt if profits are still being made.

Where's that statement about, "There's no such thing as bad products?"
 
  • Like
Reactions: King_V, carvedinside and DRagor
InvalidError said:
People failing to plug in their cables all the way isn't really Nvidia's fault. Also, in-house testing is only meant to catch problems that may arise when the product is used as intended and the plug isn't intended to be 2-3mm short of fully inserted. There does not appear to have been any further public reporting on this issue after it got confirmed as user error in just about every case that got looked into back in October.

AMD however appears to have a design flaw that may affect all GPUs using the reference cooler. I'm looking forward to heavily discounted 7000-series GPUs due to this epic fail giving the series a bad reputation.
Click to expand...
In the 16 pin connector case, it really was NVIDI's fault. Most likely it's also AMD's fault. Both companies were surely involved in the new custom connector used for this interface. Confronted with a similar situation we always used 2 corner pins on opposite edges of the connector to verify that the connector was seated before allowing the system to exit the reset state. Since they had full control of the connector specification, a single sense pin could have been added to the center of the connector as well. If the sense pin does not make contact due to the connector not being fully inserted, the GPU would not power up.

Are they going to fix the problem with an updated connector, or will they continue to blame customers for their design failure?
 
  • Like
Reactions: Hooda Thunkett
DRagor said:
Buildzoid was first to find out why Asus mobos catch fire. GN was first to find out why NVidia connectors melt. And now derBauer is first to find why AMD cards are cooking themselves.

May I ask, where the hell are quality control departments in hardware companies these days? It's great that we have famous youtubers doing their work but come on - it is not supposed to be this way!
Click to expand...
Charging so much money and yet.. they are cutting costs everywhere... XD
 
Did AMD only have BTX systems in-house to test these on?

On the bright side, if you have one of these cards you can use it for now by simply turning the whole ATX tower case upside-down so gravity can make up for the defective wick structure in the vapor chamber. Without a wick, the hot-side must be on the very bottom which is why case on the side in the video failed to work.
 
BFG-9000 said:
On the bright side, if you have one of these cards you can use it for now by simply turning the whole ATX tower case upside-down so gravity can make up for the defective wick structure in the vapor chamber. Without a wick, the hot-side must be on the very bottom which is why case on the side in the video failed to work.
Click to expand...
In Debauer's video, once the GPU experiences the "wicking stall issue" from being horizontal, switching it to vertical doesn't restore it to expected behavior. If the wicking can fail to recover with an orientation change back to the "known good" orientation on an otherwise "known good" card, then failure in any other orientation should likely be considered probable, albeit with reduced probability.
 
  • Like
Reactions: daworstplaya, Hooda Thunkett and atomicWAR
Did we watch the same video? He had the card completely upside down, which did not work. Then he tried vertical on its side which worked OK until he turned it completely upside down again so the water would get trapped in the cold parts, after which it did not work when vertical again because the liquid was stuck so far away from the heat source. At no point did he try with the heatsink right side up which would drain the liquid to back over the die.

Heatpipes without wicks at all have to have the heat source at the lowest point in order to work. And unfortunately ATX puts the GPU die at the highest point when in a tower case.
 
  • Like
Reactions: Hooda Thunkett and PEnns
BX4096 said:
This is very much Nvidia's fault since you must take user error into account when you design these things. Besides, the simple fact that most of us have been dealing with half-plugged cables (USB, SATA, IDE, power, you name it) our entire life without any of them catching fire should tell you that this wasn't normal.

But then I guess both companies were more too worried about making a quick profit off their vastly overpriced products to worry about quality control...
Click to expand...

Actually no, user error is not the manufacturers problem. Car manufacturers still let people drive even though they mess up all the time.
It is up to the user to make sure they are using a device as directed.
 
  • Like
Reactions: RandomWan, daworstplaya, Why_Me and 1 other person
You must log in or register to reply here.

TRENDING THREADS

Latest posts

Moderators online

Share this page

COMPANY
RESOURCES
FOLLOW
Top Bottom