Figured I'd chime in and say I had to RMA my EVGA RTX 3090 FTW3 Ultra Gaming card.  I see people identifying different versions of this card, so I'll clarify that my card is made in China and has the "red lips".  I bought it at the end of November and I have never seen one in stock since...  I really hope I'm not in for multiple RMAs here because as a Canadian shipping to EVGA's RMA department in California is expensive...  That frustrates me a lot and I would be VERY upset to have to do that more than once if this is a known issue...
 
So far this is probably one of the better threads I have seen on this topic, though these threads seemed to be peppered around the Reddit and the EVGA forums.  I've already done a thousand and one DDU, multiple driver combos, settings, software uninstalls, all lead me nowhere.
 
I saw the dedicated "black screen" thread in the forums here but that was rather unproductive and appears to not be addressing this actual issue.  In fact even in this thread I see people chiming in with a variation which doesn't line up because they are having system shutdowns/restarts indicating more likely an OCP issue with their PSU.  The issue being described in this thread from my point of view is NOT that, so admittedly this is a confusing issue to get to the bottom of...  I have a new Corsair HX850 PSU set to single +12v rail operation mode with 3x dedicated PCI-E cables run to the card.  I also had the same issues on a Corsair RM750 with 2x cables, one daisy chained because that is all this PSU has.  I had the same exact issue with both PSUs, never a full system shutdown or restart though.
 
My symptoms pretty much match the initially described post:
-Running older/less demanding games randomly results in a black screen / no display output and the fans ramping to 100%.  Main games I have played that suffer from this are GTA V and Valheim. Valheim loves to max out ~75%-80% GPU usage I noticed.
-Or the game continues to run and the fans ramp to 100%, EVGA Precision X1 reports fan 3 100% and fan 1&2 0% on the main screen even though I can see all fans spinning, the monitoring tab shows fan 1&2 at ~51% and fan 3 at 98%, eventually the fans ramp down.  Temperatures before this happen are normal, then go super low (like down to 61 Celsius) since all the fans are at 100%
 
Most recently this actually happened for the first time while just browsing in Google Chrome.  I have not been able to get any video output from the card since after re-seating everything multiple times etc.  Card is dead (my RTX 2080 FE card is working perfectly when I swapped that in), though I don't have the much talked about red lights on the PCI-E connectors, so not sure.  When the crashes happen the system is still working (sound continues, windows shuts down normally with the shutdown button etc.).  I actually found that while the card wasn't able to output (including BIOS post screens, so not a windows issue), the system does load into Windows fine (I can see things like my GoXLR mini and Corsair keyboard lighting react to their software being loaded and changing LEDs, HDD activity light etc), but still has no display output.
 
I haven't been able to observe any voltages because I didn't read this thread while I was having recoverable issues...  I haven't overclocked this card at all and I literally played > 110 hours of Cyberpunk 2077, few hours of Shadow of the Tomb Raider, COD Cold war / warzone, with no issue, further emphasizing that these crashes happen with low or fluctuating GPU loads and not max load situations...  Really disappointing for any product, but especially one of the most expensive "gaming" GPUs ever.

Flyingj99
Figured I'd chime in and say I had to RMA my EVGA RTX 3090 FTW3 Ultra Gaming card.  I see people identifying different versions of this card, so I'll clarify that my card is made in China and has the "red lips".  I bought it at the end of November and I have never seen one in stock since...  I really hope I'm not in for multiple RMAs here because as a Canadian shipping to EVGA's RMA department in California is expensive...  That frustrates me a lot and I would be VERY upset to have to do that more than once if this is a known issue...
 
So far this is probably one of the better threads I have seen on this topic, though these threads seemed to be peppered around the Reddit and the EVGA forums.  I've already done a thousand and one DDU, multiple driver combos, settings, software uninstalls, all lead me nowhere.
 
I saw the dedicated "black screen" thread in the forums here but that was rather unproductive and appears to not be addressing this actual issue.  In fact even in this thread I see people chiming in with a variation which doesn't line up because they are having system shutdowns/restarts indicating more likely an OCP issue with their PSU.  The issue being described in this thread from my point of view is NOT that, so admittedly this is a confusing issue to get to the bottom of...  I have a new Corsair HX850 PSU set to single +12v rail operation mode with 3x dedicated PCI-E cables run to the card.  I also had the same issues on a Corsair RM750 with 2x cables, one daisy chained because that is all this PSU has.  I had the same exact issue with both PSUs, never a full system shutdown or restart though.
 
My symptoms pretty much match the initially described post:
-Running older/less demanding games randomly results in a black screen / no display output and the fans ramping to 100%.  Main games I have played that suffer from this are GTA V and Valheim. Valheim loves to max out ~75%-80% GPU usage I noticed.
-Or the game continues to run and the fans ramp to 100%, EVGA Precision X1 reports fan 3 100% and fan 1&2 0% on the main screen even though I can see all fans spinning, the monitoring tab shows fan 1&2 at ~51% and fan 3 at 98%, eventually the fans ramp down.  Temperatures before this happen are normal, then go super low (like down to 61 Celsius) since all the fans are at 100%
 
Most recently this actually happened for the first time while just browsing in Google Chrome.  I have not been able to get any video output from the card since after re-seating everything multiple times etc.  Card is dead (my RTX 2080 FE card is working perfectly when I swapped that in), though I don't have the much talked about red lights on the PCI-E connectors, so not sure.  When the crashes happen the system is still working (sound continues, windows shuts down normally with the shutdown button etc.).  I actually found that while the card wasn't able to output (including BIOS post screens, so not a windows issue), the system does load into Windows fine (I can see things like my GoXLR mini and Corsair keyboard lighting react to their software being loaded and changing LEDs, HDD activity light etc), but still has no display output.
 
I haven't been able to observe any voltages because I didn't read this thread while I was having recoverable issues...  I haven't overclocked this card at all and I literally played > 110 hours of Cyberpunk 2077, few hours of Shadow of the Tomb Raider, COD Cold war / warzone, with no issue, further emphasizing that these crashes happen with low or fluctuating GPU loads and not max load situations...  Really disappointing for any product, but especially one of the most expensive "gaming" GPUs ever.

This is the exact same issue I was having and countless others I've heard here. Only experienced the crash in older dx11 games. After five crashes, the card was dead. Red lips, made in Taiwan for me. How much did you pay to ship from Canada? I'm in Toronto and shipping mine out tomorrow.

It's extremely discouraging that EVGA doesn't even acknowledge what the problem is and therefore can't guarantee that the replacement card won't have the same fatal flaw, all while making us foot the bill for shipping an expensive brick across the continent for them.

Flyingj99

Arrowhead2010
This is the exact same issue I was having and countless others I've heard here. Only experienced the crash in older dx11 games. After five crashes, the card was dead. Red lips, made in Taiwan for me. How much did you pay to ship from Canada? I'm in Toronto and shipping mine out tomorrow.

It's extremely discouraging that EVGA doesn't even acknowledge what the problem is and therefore can't guarantee that the replacement card won't have the same fatal flaw, all while making us foot the bill for shipping an expensive brick across the continent for them.

Cost ~ $145 to ship Via their preferred shipper, UPS.  They did pick it up from my house though which was good for me. 

Damn, that's ridiculous. I was hoping it would be like 50 bucks. lol...

Rewire92
Another update.
 
No further crashes after limiting the card to 1.05 V.

However, I am experiencing large performance losses in high usage games.  Anywhere in the rage from 10-20 FPS loss in games like Cyberpunk, Red Dead Online, Modern Warfare, and Battlefield 4.

I am debating removing the undervolting for higher clock states, and just seeing if this card dies as well.  Very disparaging. 

You must do this every day and every hour. It doenst make sens to have degrade your performance to keep the card alive. You will only postpone death. Sad to see this on 2000$ card with no answer of EVGA until today.

Flyingj99

Arrowhead2010
This is the exact same issue I was having and countless others I've heard here. Only experienced the crash in older dx11 games. After five crashes, the card was dead. Red lips, made in Taiwan for me. How much did you pay to ship from Canada? I'm in Toronto and shipping mine out tomorrow.

It's extremely discouraging that EVGA doesn't even acknowledge what the problem is and therefore can't guarantee that the replacement card won't have the same fatal flaw, all while making us foot the bill for shipping an expensive brick across the continent for them.

Cost ~ $145 to ship Via their preferred shipper, UPS.  They did pick it up from my house though which was good for me. 

Did they offer enough for insurance?  I sent my 1080Ti Hybrid to EVGA and it "only" cost me $85, using Canada Post.  The max I could insure the card for was $1000, which during this stupid time I did because I would need at least that much to replace it if it got lost.  It took 3 business days to get to EVGA, and my replacement should be on my doorstep tomorrow.  

UPS was the cheapest option that could insure the full value of my 3090 as well, for my RMA... 132 beaver pelts to ship it from Alberta to Brea.

Flyingj99
Figured I'd chime in and say I had to RMA my EVGA RTX 3090 FTW3 Ultra Gaming card.  I see people identifying different versions of this card, so I'll clarify that my card is made in China and has the "red lips".  I bought it at the end of November and I have never seen one in stock since...  I really hope I'm not in for multiple RMAs here because as a Canadian shipping to EVGA's RMA department in California is expensive...  That frustrates me a lot and I would be VERY upset to have to do that more than once if this is a known issue...
 
So far this is probably one of the better threads I have seen on this topic, though these threads seemed to be peppered around the Reddit and the EVGA forums.  I've already done a thousand and one DDU, multiple driver combos, settings, software uninstalls, all lead me nowhere.
 
I saw the dedicated "black screen" thread in the forums here but that was rather unproductive and appears to not be addressing this actual issue.  In fact even in this thread I see people chiming in with a variation which doesn't line up because they are having system shutdowns/restarts indicating more likely an OCP issue with their PSU.  The issue being described in this thread from my point of view is NOT that, so admittedly this is a confusing issue to get to the bottom of...  I have a new Corsair HX850 PSU set to single +12v rail operation mode with 3x dedicated PCI-E cables run to the card.  I also had the same issues on a Corsair RM750 with 2x cables, one daisy chained because that is all this PSU has.  I had the same exact issue with both PSUs, never a full system shutdown or restart though.
 
My symptoms pretty much match the initially described post:
-Running older/less demanding games randomly results in a black screen / no display output and the fans ramping to 100%.  Main games I have played that suffer from this are GTA V and Valheim. Valheim loves to max out ~75%-80% GPU usage I noticed.
-Or the game continues to run and the fans ramp to 100%, EVGA Precision X1 reports fan 3 100% and fan 1&2 0% on the main screen even though I can see all fans spinning, the monitoring tab shows fan 1&2 at ~51% and fan 3 at 98%, eventually the fans ramp down.  Temperatures before this happen are normal, then go super low (like down to 61 Celsius) since all the fans are at 100%
 
Most recently this actually happened for the first time while just browsing in Google Chrome.  I have not been able to get any video output from the card since after re-seating everything multiple times etc.  Card is dead (my RTX 2080 FE card is working perfectly when I swapped that in), though I don't have the much talked about red lights on the PCI-E connectors, so not sure.  When the crashes happen the system is still working (sound continues, windows shuts down normally with the shutdown button etc.).  I actually found that while the card wasn't able to output (including BIOS post screens, so not a windows issue), the system does load into Windows fine (I can see things like my GoXLR mini and Corsair keyboard lighting react to their software being loaded and changing LEDs, HDD activity light etc), but still has no display output.
 
I haven't been able to observe any voltages because I didn't read this thread while I was having recoverable issues...  I haven't overclocked this card at all and I literally played > 110 hours of Cyberpunk 2077, few hours of Shadow of the Tomb Raider, COD Cold war / warzone, with no issue, further emphasizing that these crashes happen with low or fluctuating GPU loads and not max load situations...  Really disappointing for any product, but especially one of the most expensive "gaming" GPUs ever.

Do your Valheim crashes happen inside in the crypt type areas or just generally? I can run around in the outside world for hours but as soon as I enter a crypt I know a crash is coming at some point.

Flyingj99
 

B1gbeard


Do your Valheim crashes happen inside in the crypt type areas or just generally? I can run around in the outside world for hours but as soon as I enter a crypt I know a crash is coming at some point.

 
There was no pattern. I had this happen in the menu screen, while exploring, while walking around my base.  Also after playing for a while and  when just starting (ie: everything is cold).  Same thing goes for when I was having the same issue playing GTA V, no real pattern that I could see.
 
To be clear, we aren't talking about a typical driver crash here in this thread.  That is something I wouldn't be surprised to see in an early access  game like this.  This appears to be the hardware going into some kind of protect mode when the fans ramp to 100% and the display output stops.  This seems to be backed up by the eventual dead card some folks have reported here (including myself).  I haven't had any crashing with my RTX 2080 Ti, it has been 100% solid since I swapped it last week.
 
 

That is what is happening to my Gigabyte yes fans ramp to 100% and screen goes blank and only a power down clears it. Happens in older games primarily as reported here randomly but in Valheim it only happens in crypts. Exactly the same crash.

m6monster
So, I was having a similar issue, back in December when I got my card. Under high usage gaming, the card was totally stable no issues whatsoever even with a heavy overclock. However, on the desktop under little to no-load, or in non-demanding games, I would get black screen crashes, sometimes the card would recover, and sometimes it would totally lock up the pc. I looked around and couldn't find a solution. The issue made no sense to me, the only reason I could think of is that there is something in the bios of the card or the driver for the card that causes the GPU to dip to unsustainable voltages when the card was downclocking for power savings under a light load. So, I changed the global power management mode in the Nvidia control panel from "Normal" to "Prefer maximum performance" and I haven't had a single black screen or crash since, and it's been almost three months now. It's not an ideal fix, as the GPU doesn't underclock under low load for power savings anymore, but it sure beats the hell out of the card dying all the time on me. 
 
I don't know if anyone else has suggested this for those of you who are still having black screen issues in light gaming, and I apologize if someone has, but I figured I would mention my experience in the hopes that it might help one of you until Nvidia or Evga issues a proper fix for the issues. 

Does this increase your idle temps substantially?

Fyi. Just open a business account with ups or fedex. After a few ships, youll start getting the discount rate.

Heyo,

I thought people might want a final update to this thread.

I ended up doing a 3rd RMA and have finally secured a card that runs with no issues.  Nothing else about my computer has changed.
 
Serial number 2114, Black lips, manufactured in Taiwan.  No issues, overclocks great, no crashing ever experienced in any games.  Had the card for 3 weeks now.

Cheers all.

So, what's the best advice for now? Do I need to undervolt? If so, at what voltage? 

Because it's a common point of consistency across a wide swath of hardware configurations.

It doesn't matter if people are using 850w gold units with 95w CPUs or 1600w Platinum units running OCed 10900ks and individual 8 pins on each. The FTW3s are failing more frequently than the other SKUs are and they're doing so in games where GPU utilization is so low the cards don't even boost fully.

We also know the FTW3s are hitting voltages they shouldn't be at relatively low clocks and we know they ping-pong from low to high voltage in low load games. In GTA V or MCC, we shouldn't be seeing cycling from .725mv to 1.08v at a clock that should only ever use .725mv.

My 3080 FTW3 does this. My cheaper Gigabyte 3080 OC plugged into the same system doesn't.

EVGA went with a suboptimal voltage controller that isn't found outside of budget oriented boards and it was never intended to be used in a 3x8 pin configuration. They also went with a whack load of fuses for easy repairability. If the problem is reproducable on a wide range of hardware, that points to a design flaw. If undervolting and locking it "fixes" the problem, it again points to a design flaw.

Sorry to barge into this thread, But I have a quick question. 
 
I'm probably worrying for nothing, but I'm also trying to be as vigilant as I can about my investment in this 3080 so I keep an eye on the forums, and to me this is interesting about these things happening in older games, games which should not even be any sort of issue.
 
I decided to run the Precision X1 HWM to see what my voltages are, keep in mind I don't overclock past the factory settings of my FTW3 Ultra, I really haven't had the need to at all as this card is BLAZING fast for me.
 
 This is what I got running P3D v5 (dx12) for about 2 hours, NOTE that my voltage while running the game was low, but while the game was paused is when the voltage spiked up to 1.081 along with the memory clock mhz? Then when I un-paused the game the voltage and memory mhz went back down again.
 
 Is THIS unusual behavior or just normal operating procedure? You would think it would be just the opposite but I am no expert? OR am I not even looking at this correctly?
 
Anyways, should I worry or just go about my business? So far while gaming for hours in Cyberpunk 2077 I haven't broke 76C, and no black screens or high running fans, but I'm almost afraid to pause or alt/tab out of any of my games.
 

 thanks

Anyone in Europe, check this. By EU law you aren’t required to pay anything for warranty repair/replacement.

Articla 3:
”2. In the case of a lack of conformity, the consumer shall be entitled to have the goods brought into conformity free of charge by repair or replacement, in accordance with paragraph 3, or to have an appropriate reduction made in the price or the contract rescinded with regard to those goods, in accordance with paragraphs 5 and 6”

https: //eur-lex.europa.eu/legal-content/EN/TXT/HTML/?uri=CELEX:31999L0044&from=FI

Take out the spaces in the URL as I’m not sure who are allowed to post links here.

Edit: And here is the main site. You can choose your own language for the law text.

https: //eur-lex.europa.eu/legal-content/FI/TXT/?uri=celex%3A31999L0044