just bought a palit 2080TI..

[xkm1948]

Sorry to hear that you are having troubles.

Meanwhile this EVGA 2080Ti is still going strong here. I bought the extra warranty to extend to 5yra in total just in case.

 

[blued]

this card passed all the testing i gave it with flying colours.. two months down the line its starting to fail..

not a good sign some kind of degrade has taken place.. most of the fuss about failing 2080ti cards has died down.. but if my card is anything to go by the story isnt over yet..

my card has had roughly 200 hours gaming use.. mostly on ashes and farcry 5.. its now f-cked.. he he..

currently its still working at -1000 (the max down-clock) on the memory speeds.. i expect it to give up the ghost completely soon..

this hasnt been a gradual thing its pretty much happened rapidly over the last few days..

trog

Pls also post this at Nvidia forum, get your details added to the growing list there.

https://forums.geforce.com/default/board/227/geforce-rtx-20-series/
https://docs.google.com/spreadsheet...tHgGDGrG-3Vf7_b82KX_lZ4-9M/edit#gid=216338597

 

[eidairaman1]

This card comes with 3+2 year extended warranty, I just hope they honor it.
Unlike XFX that basically told me to F myself with a cactus when by Vega 56 went on fire.
I promptly F my self with a cactus by getting a 2080ti.
XFX / AMD can shove their Radeon VII up where the sun don't shine.

Not amd but xfx fault for it, amd just provides the gpu die, xfx builds the pcb.

 

[Zubasa]

Not amd but xfx fault for it, amd just provides the gpu die, xfx builds the pcb.

When I complained to the Consumer Council, XFX reply was it was against AMD policy etc, and with a screenshot with an e-mail from AMD.
So yeah AMD at the very least played along with XFX on that matter, and it was a reference card.
A card that ignited on its own during normal use is apperently "Constumer Induced Damage".
Complete with a PDF showing warranty terms where damage to pci-e connector by IMPACT DAMAGE is not cover by warrranty.
I showed them pictures of the PCI-E connector being charred from the heat, while the PCB is structurely intact with no dents / nicks.
So 11/10 for AMD / XFX not to read their own PDF. BTW there are no warranty terms etc in the box either.
I guess crashing your car is the same as the car burning up on the highway with you in it.

The PCI-E slot on my X399 board melted as well, thank god Asrock didn't just out right deny warranty on something that might not be their fault.
It also left soot on my Xonar STX II sound card.
XFX also denied themselves as the manufacturer of all things.
It really doesn't matter for the costumer, that who made the thing.
When an HP / Dell machine failed you go to HP / Dell for warranty, not RMA your cpu to Intel / AMD so their policy shouldn't concern you either way.

Spoiler: Image

 

[trog100]

i have posted this in the other 2080ti issues thread..

i now have an rma arranged for my two month old palit gaming pro oc 2080ti card.. having spoken with an Ebuyer UK service engineer they are aware of issues with these cards and how they are failing he seemed apologetic and did say they were getting less palit cards back than others..

i confirmed a pretty standard UK warranty period of two years and this will start again from when i receive the replacement card..

having checked around the internet i cant find any comment regarding 2080ti failures later than mid November.. its just as if the problem has gone away but it quite clearly hasnt..

the problem seems to be a memory issue artifacting and lock ups or crashing to the desktop..

i assume sooner or later whatever is causing the problem will get fixed at source and we will know a little more..

my advice to anyone that has one of these cards is to check your warranty.. you may need it..

a couple of weeks back i would have been one of those people saying "my 2080ti card is working fine" .. sadly this is no longer the case..

trog

 

[EarthDog]

What is happening??

 

[spaceZ]

@trog100 - Did you have success with Palit RMA?

I have Palit RTX2080Ti GamingP OC 11G since early Dec. 2018, just started to go bad on me now after 4 months of no problems. "space invaders" like artifacts and lock-up when card hits 70+ C under load - game or stress test.

Will talk to Scan.co.uk about return on Monday.

 

[Russ64]

There is clip on Youtube that says this is now a known issue and it is due to the Micron GDDR6 memory chips failing. I have EVGA RTX 2080 XC which according to GPU-Z has Samsung GDDR6.

 

[trog100]

@trog100 - Did you have success with Palit RMA?

I have Palit RTX2080Ti GamingP OC 11G since early Dec. 2018, just started to go bad on me now after 4 months of no problems. "space invaders" like artifacts and lock-up when card hits 70+ C under load - game or stress test.

Will talk to Scan.co.uk about return on Monday.

i dont think you will have any problems.. i didnt but mine came from ebuyer not scan..

your warranty should also start over again from when you get your new one..

trog

ps.. the reason for these cards failing is still a mystery.. ones with samsung memory have also failed..

 

[Russ64]

Yes, I have seen some posts that say that. I will not be pushing my memory and maybe that is why Afterburner runs Memory clock at 6800 instead of rated 7000.....

 

[bug]

i dont think you will have any problems.. i didnt but mine came from ebuyer not scan..

your warranty should also start over again from when you get your new one..

trog

ps.. the reason for these cards failing is still a mystery.. ones with samsung memory have also failed..

The cards just can't take the price tag

 

[Raptor_i]

There is clip on Youtube that says this is now a known issue and it is due to the Micron GDDR6 memory chips failing. I have EVGA RTX 2080 XC which according to GPU-Z has Samsung GDDR6.

Is there is a way to know what's the card's memory before buying although I've read that even cards with Samsung memory have failed too , I want to upgrade to 2070 but these artifacts issue holding me off.

 

[bug]

Is there is a way to know what's the card's memory before buying although I've read that even cards with Samsung memory have failed too , I want to upgrade to 2070 but these artifacts issue holding me off.

I thought these problems were confined to 2080Ti.

 

[Raptor_i]

I thought these problems were confined to 2080Ti.

No 2070s are failing too also "space invaders"

https://www.reddit.com/r/buildapc/comments/9u3l4f

 

[eidairaman1]

Skip it rtrt/rtx is broken, grab a 1660, 1080ti R7/VII or V56/64...

 

[EarthDog]

It really wasn't a big issue in the first place. You'll have as good a chance to get a failed non Turing based card as you would Turing really. If the issue was prevalent, we'd still be hearing about it all over the web.

 

[spaceZ]

@trog100 Thanks. I'll try to remember to post on how process goes.

I'm thinking of ordering a cheap video card -GT 1030 maybe - to use for for non-GPU intensive work during replacement interval as my 7940x CPU has no graphics capability. This is also in case of future problems with the replacement 2080Ti...

 

[Vayra86]

No 2070s are failing too also "space invaders"

https://www.reddit.com/r/buildapc/comments/9u3l4f

"I have been hearing" is a good sign you can completely ignore a reddit post.

Do that.

 

[Russ64]

I guess there is a "normal" failure rate for all products and I have seen numbers of 3 - 5% mentioned but of course nVidia will not publish actual numbers. Sales are also down due to the collapse of cryptomining and the high price of RTX. Plus in forums you only read about the failures and not all the good ones.

I just ran OC scanner on mine and got up to 2085 GPU Clock (power limit at 130%, no OC on voltage or Mem clock). Looks stable so hope it says that way, I am Folding on it but at lower clocks (-135 GPU to keep temps at ~66c).

 

[trog100]

@trog100 Thanks. I'll try to remember to post on how process goes.

I'm thinking of ordering a cheap video card -GT 1030 maybe - to use for for non-GPU intensive work during replacement interval as my 7940x CPU has no graphics capability. This is also in case of future problems with the replacement 2080Ti...

mine was pretty quick.. i think it took five days in all..

trog

 

[bug]

It really wasn't a big issue in the first place. You'll have as good a chance to get a failed non Turing based card as you would Turing really. If the issue was prevalent, we'd still be hearing about it all over the web.

Nvidia could have killed this story fast, releasing numbers for these and other models or generations. For some reason, they chose not to

 

[EarthDog]

Nvidia could have killed this story fast, releasing numbers for these and other models or generations. For some reason, they chose not to

I'd imagine the percent may be a bit higher, but, the world was not crumbling around them like some users here thought/wanted it to.

 

[64K]

Nvidia has kind of admitted that there was a problem with the early production 2080 Ti cards. They posted this on their GeForce forum in Nov last year:

"Limited test escapes from early boards caused the issues some customers have experienced with RTX 2080 Ti Founders Edition. We stand ready to help any customers who are experiencing problems. Please visit www.nvidia.com/support to chat live with the NVIDIA tech support team (or to send us an email) and we’ll take care of it."

They switched to Samsung VRAM chips on the next runs and as far as I recall there weren't as many problems as the first runs using Micron chips although Nvidia has denied the switch to Samsung chips had anything to do with the cards artifacting and dying.

https://www.techspot.com/news/77445-nvidia-addresses-failing-geforce-rtx-2080-ti-cards.html

 

[EarthDog]

OFN....all of it. A good reminder though.

Surely someone will still blame the Micron memory even though its been posted here there and everywhere nothing was proven.

 

[bug]

Nvidia has kind of admitted that there was a problem with the early production 2080 Ti cards. They posted this on their GeForce forum in Nov last year:

"Limited test escapes from early boards caused the issues some customers have experienced with RTX 2080 Ti Founders Edition. We stand ready to help any customers who are experiencing problems. Please visit www.nvidia.com/support to chat live with the NVIDIA tech support team (or to send us an email) and we’ll take care of it."

They switched to Samsung VRAM chips on the next runs and as far as I recall there weren't as many problems as the first runs using Micron chips although Nvidia has denied the switch to Samsung chips had anything to do with the cards artifacting and dying.

https://www.techspot.com/news/77445-nvidia-addresses-failing-geforce-rtx-2080-ti-cards.html

Well, every failing card is a problem. My gut feeling is Nvidia's refusal to out numbers on it (or state the nature of the failure, for that matter) only made this one seem worse than it really was.
At the end of the day, it's not like anyone was left with a broken card.