MrBaracu
I wanna see something that show it.

 All you have to do is search and you will get this: google.com/=new+world+gpu+issues

idclip

liud21
People waiting or skipping New World, like it's going to fix the problem with their card. Lol, you are fools to think that way. What happens when you run into a game after your warranty that triggers the problem and kills your card out of warranty? Better to find out now, get a replacement, then to find out later and be crap out of luck.

There's a 3 year warranty on this card, I'm not in any hurry to be inconvenienced by an RMA. When it happens, I'll get it replaced.

 
Exactly.  And some of us extended to 5 years.. so if the car dies 5 years and 3 months later.. it goes in the trash and I just buy what's new that I can afford.

Any news from EVGA on this ? Did they even communicate on the issue (Recently)?

Ravena

SwordaDEV
Any news from EVGA on this ? Did they even communicate on the issue (Recently)?

nope I spoke to several evga people and they said just wait don't put your card at risk if you have tried new world and it crashes but still works stop and wait for an update. evga, nvidia and amazon are most likely working on a solution. RMAing cards for the sake of RMA a card is not good. I don't think its "faulty card" for say I think its an issue with vbios, new world or driver related issue. though I am not saying the design of the card is good or anything, its just that its most likely isn't the card at fault. don't blow up your card cause there most likely a patch coming soon ish

Thanks for the info. I won't try & burn my card for the sake of it.
 
I already crashed twice & now it's analyze time !
 
I dug up in crash logs & found 2 things :
 
Following codes are to be used in Pastebin (Since we cannot put links in posts..)
First crash : umH3Uguq
Second crash : p1e8rYLP
 
On the second crash you can see : [Info] (AZCore): [Rendering] Present failed, error code -2005270523. Attempting to recover...
 
And in the NewWorld forum someone commented this:

Nullsan
 
Aight fam so here’s a rundown of what I think is going on.
The error code -2005270523 (decimal) which is 0x887a0005 in Hexadecimal is an error from the operating system.
It decodes to this error.
DXGI_ERROR_DEVICE_REMOVED
0x887A0005
The GPU device instance has been suspended.
So the video card has “disappeared”. The most common reason for this is that the video card device driver crashed. You might find entries in the Windows Event log recording the crash.
Some video card crashes are also related to memory leak issues (low available RAM causes a crash are a long period of heavy load, but the real bug is a memory leak in the driver).
Video card device driver problems like this are also common. If you Google 0x887A0005, you’ll get 1000s of people reporting this error with various software.
Combining this with reports that other services and applications are killing the GPU with this error due to rendering time elongation means that this is possible here. The reason for this is that In order to prevent any rendering or GPU computation from locking up the system, the Windows operating system kills the GPU driver whenever a rendering takes more than a few seconds. When the driver is killed, the application using it crashes automatically. It is not possible to know how long a rendering task or a computation may take (it depends on the GPU, the drivers, the OS, the mesh size, the texture size, etc.), therefore it is not possible to put a limit on how much the computer should process and avoid the crash from the application level.
What that does to the card after the fact is beyond me, but people have made claims that it’s bricking their card.
All-in-all this means it’s not limited to only one card series, brand, generation, and not limited to just NW. If we see improvements it will most likely be with future driver updates and game patches. If a suggestion someone’s made works for you, GREAT! It won’t always work for everyone else. Be sure to keep sending logs, reading through everyone’s tips and suggestions, and keep posting till we get a fix.
With that, I wish you all the best, and with luck, I’ll see you in Aeternum!!!
NOTE: I AM NOT A TECH EXPERT. THIS IS MERELY MY FINDINGS WHILST SCOURING THE WEB LOOKING FOR OTHER REPORTS AND SOLUTIONS TO THIS ERROR!

 
So it would be tied to a driver issue finally ?
I will continue to dig more & report if I find anything new.

Rock2k
Anyone Fried their 3090 kingpin with this game yet? or is the Kingpin Strong?

Sorry not gonna try to let people know. But I did not hear of any such report.

i hope evga is building a huge file of cards that died to new world to recoup costs in some kind of a class action suit. its clear that the game is the culprit here as it's an anomaly that's happening with all different cards and NOT just evga.

youcanteatthat
i hope evga is building a huge file of cards that died to new world to recoup costs in some kind of a class action suit. its clear that the game is the culprit here as it's an anomaly that's happening with all different cards and NOT just evga.

I laugh at these sorts of posts. The game is using standard DirectX API calls. They are doing nothing wrong. Why are they at fault? If anything, it would be Microsoft that has a bug in their DirectX API so sue Microsoft.

flg2010

idclip

liud21
People waiting or skipping New World, like it's going to fix the problem with their card. Lol, you are fools to think that way. What happens when you run into a game after your warranty that triggers the problem and kills your card out of warranty? Better to find out now, get a replacement, then to find out later and be crap out of luck.

There's a 3 year warranty on this card, I'm not in any hurry to be inconvenienced by an RMA. When it happens, I'll get it replaced.

 
Exactly.  And some of us extended to 5 years.. so if the car dies 5 years and 3 months later.. it goes in the trash and I just buy what's new that I can afford.

Assuming you can pick up a 4090 or what ever Nvidia comes out with 4 years from now... Lol

Ravena
its just that its most likely isn't the card at fault.

Say that to people that say "software cant kill your card".

Anyone knows what exactly is triggering this? It's a bit insane that a userspace application can fry a card.

varunpriolkar
Anyone knows what exactly is triggering this? It's a bit insane that a userspace application can fry a card.

Back in Oct and Nov of 2020, when 3090s we're dying, people were blaming Power supplies. Now, they're blaming the game/software....

varunpriolkar
Anyone knows what exactly is triggering this? It's a bit insane that a userspace application can fry a card.

well if the 3090 kingpin and strix is fine then its just that the power delivery is under specd for the power surges ampere generate.(ampere has really high power spike compared to prev arch). well then people will argue, why not all ftw3 are dying? well it depends on Silicon quality, good silicon will pull less voltage/power for the same boost algo nvidea has, so the power spike is a lot less. also ftw3 has a load balancing problem where the pcie power needs to be raised to pull more power through the 8pins(i dont know whether it has changed or not). this is a stupid decision to prevent kingpin bios flashes. thats why xoc bioses dont do anything on ftw3. so if someone had gotten a black screen running new world on their card. try to under volt by a lot as well as underclock and make sure the power spike doesnt go above the bios limit. because the power limit isnt on the vrm controller so if u have an oscilloscope and the bios power limit is 500w u can see that the power shoots up to 600w(depends on set voltage and load). the bios power target does not regulate spikes due to refresh rate(polling) being low. the bios power target polling rate is way slower than the voltage controller. (edit: it looks like it might be ampere problems)

even still there are ocp/ovp/opp/otp on the card. how can it die....
 

babayega

varunpriolkar
Anyone knows what exactly is triggering this? It's a bit insane that a userspace application can fry a card.

well if the 3090 kingpin and strix is fine then its just that the power delivery is under specd for the power surges ampere generate.(ampere has really high power spike compared to prev arch). well then people will argue, why not all ftw3 are dying? well it depends on Silicon quality, good silicon will pull less voltage/power for the same boost algo nvidea has, so the power spike is a lot less. also ftw3 has a load balancing problem where the pcie power needs to be raised to pull more power through the 8pins(i dont know whether it has changed or not). this is a stupid decision to prevent kingpin bios flashes. thats why xoc bioses dont do anything on ftw3. so if someone had gotten a black screen running new world on their card. try to under volt by a lot as well as underclock and make sure the power spike doesnt go above the bios limit. because the power limit isnt on the vrm controller so if u have an oscilloscope and the bios power limit is 500w u can see that the power shoots up to 600w(depends on set voltage and load). the bios power target does not regulate spikes due to refresh rate(polling) being low. the bios power target polling rate is way slower than the voltage controller. (edit: it looks like it might be ampere problems)

 
Even if I could get a kingpin it would not fit in my current configuration.
 
Good points though and likely the only fix is for New World to optimize and revise their game.
 
 

kevinc313

flg2010
 
 
You know.. I am no superfan of Jay however occasionally he says something that makes sense.
 
Take a look if you have not yet at his recent video on this:
 
https://www.youtube.com/watch?v=kxoXbfzP5BU&t=629s
 

Jayz latest video on this:
 
https://youtu.be/6A0sLVgJ7qU

So poor build quality in some ftw3’s, and now a vbios issue as well on the ftw3. 👎