Latest driver killing 68xx/69xx ?!

A professional repair company in Germany released this video.

Driver in question is 22.11.2.

My guess it is the MBA versions overheating.. any thoughts?

Pictures of the card in the video and it got a destroyed die:
https://www.facebook.com/krisfixde/...bhZb22gPGUxmoXFgVbeDpha23SZfkPhdRrcR5F6pAS7El

 

AMD Radeon RX 6000 GPUs Mysteriously Start Dying, German Repair Shop Receives 48 Cards With Cracked Chips

 

What a joke

Last driver for RX 6000 series and older are from 1st Dec.... where a thousands of affected users spamming forums, where are vids of burning cards etc.?

 

Bogus

 

RX68/6900 cards were still (mostly) covered by warranty on December, and yet, somehow, for some unknown reason all those "separate buyers", chose not to RMA their dead (never before opened, completely untouched) cards but instead trusted a "repair shop" in Germany...only this specific repair shop...all of them.

Somebody's mining rig (maybe his own...) got fckdup and he tries to stir some "anger/rage/reactions", after the 7900XTX cooler mess.

 

It is a possibility but check out the guys YouTube channel it's not just any repair shop. He seems to really know his stuff.

The video where he reballs at 3090 is pretty awesome.

 

If so, that's really messed up. A driver killing older cards. Even though it is not entirely sure, just to be clear. The trick here is to not be conclusive yet, but wait for more findings and even more reports to force AMD to look at this issue.

If in the end more findings are being made, it will be undeniable.

 

I didn't say he doesn't know what he's doing/saying.

I said that the cards are (were, at the time that this supposedly happened) still under warranty, so there is absolutely no need to go to a repair shop...legit or otherwise. RMA, new card/refund, new warranty period.

There is absolutely no logic in letting a repair shop void your warranty, when you can get a new one.

 

If this were really a thing then you'd see the likes of Jayz2c , GN Steve not to mention der8auer would all be all over this like white on rice but are they no they're not why because it isn't a thing

 

Wonder if some one can check the new driver vs old one. an look inside the code to see if the driver has boost the card or not ?

 

Igor' Lab has an update saying it's not driver related.
Let's wait to se if there are more developments in this matter.

 

No it isn't.
Unless the cooler was already struggling.

edit: Oh...... it seems ~100 degrees was already possible, even on nitro.



but cracked die.... just no way.

 

still, cracked dies are unrelated even if the driver makes the cards even more toasty than before.

 

I would not have expected something like that... I'm curious to see how this turns out in the end.

 

Being toasty in some part while cold in the other could crack the die due to uneven thermal expansion, link.

aggressive clock boosting combine with aggressive power gating could cause some part of the chip heating up while the other remain cold.

 

Ive been running mine at 320W and 2600mhz for maybe 8 months and zero issues, i am on water though and it never goes over 65C on the junction.

 

I dont think it could happen unless for some reason driver thinks you have some GPU that can withstand such temps.
High end GPUs and CPUs run at so high temps and power limit that i am not even surprised.
If you start a game and pc turns into jet engine its time to turn down your settings and limit fps.

 

"repair shop" and cracked chip :V Blame the drivers :V

I still struggle to get my 6700 xt to reach the TDP.

 

Highest I've seen my 6700XT run is 65C.....I thought there might be a problem with it.

 

Mine gest to 85-90 at junction, with chill enabled.