News Amazon's New World Is Out of Beta, Still Killing RTX GPUs

[Admin]

New reports have come in that Nvidia's GeForce RTX 30-series graphics cards are still dying after playing New World.

Amazon's New World Is Out of Beta, Still Killing RTX GPUs : Read more

 

[hotaru.hino]

Either Amazon found a killer poke by accident, or video card manufacturers are getting a rude awakening on skimping the build quality.

 

[Phaaze88]

...
Oops again?

Seriously now. I knew it wasn't exclusive to EVGA.
New World also isn't the only game out there that was killing them either.
There shouldn't be any more excuse at this point. These Ampere gpus are broken on the hardware level.


Now, it's quite possible that only the FEs, and the aftermarket models that followed their design, or did even better are safe.
But until more news surfaces, that's only speculation for now.

 

[2Be_or_Not2Be]

I'm hoping these are just examples of hardware quality being exposed when running. I hope it's not an example of software somehow getting around hardware safeguard protections (like pulling too much power, getting around fps limits, etc.).

What would be scary is if this was a software issue that could be weaponized to brick almost any GPU. That would be some nasty malware!

 

[2Be_or_Not2Be]

...
Oops again?

Seriously now. I knew it wasn't exclusive to EVGA.
New World also isn't the only game out there that was killing them either.
There shouldn't be any more excuse at this point. These Ampere gpus are broken on the hardware level.


Now, it's quite possible that only the FEs, and the aftermarket models that followed their design, or did even better are safe.
But until more news surfaces, that's only speculation for now.

So I've heard of high-end Gigabyte & EVGA (note: EVGA said some models had bad soldering) models biting the dust, but from what you've seen, no FE has had a problem - even the 3090FE?

 

[Phaaze88]

So I've heard of high-end Gigabyte & EVGA (note: EVGA said some models had bad soldering) models biting the dust, but from what you've seen, no FE has had a problem - even the 3090FE?

Now, I'm just one person, so take what I say with some skepticism.

No, haven't heard of any FEs getting bricked.
Nvidia used some high quality components in their FEs. The cooler isn't the most effective out there, but I wouldn't be surprised if the rest of the gpu isn't plain designed and built better than most of the aftermarket models.

The partners didn't comply with everything laid before them - they tried to cut corners to save profit margins - because Nvidia is a dick to its partners.
They stood to lose money - perhaps make money slowly is more accurate - if they applied the FE design and bill of materials to all their models.


Once again, somebody, or someone, gets screwed over, because money.

 

[Derren001]

Meanwhile my 980 Ti is laughing at the dead 3090's as it struggles to hit 50 fps.

 

[helper800]

My EVGA 3080 FTW3 has been fine so far. I wonder if it has something to do with the memory density on the 3090's and 3080 ti's. Who knows...

 

[VforV]

This is both sad and funny at the same time...

 

[Alvar "Miles" Udell]

If it happens even with the frame limiter in the nVidia Control Panel set to 60, it sounds to me like there's something in the game's code which can, under certain circumstances, either cause a feedback loop, or other similar mechanic which will cause enough stress to break the GPU, or interact with certain nVidia driver code which controls voltage levels resulting in a fried GPU.

 

[NightHawkRMX]

I guess $2000 isn't enough to afford functioning over current protection.

 

[plateLunch]

Is nVidia involved in any investigation of the failures? They probably have more resources than any board manufacturer to figure out what is going on. They also have a lot to lose if a defective chip design is the problem.

 

[alceryes]

No game or program should be able to do something outside of driver specifications. Driver specifications should ALWAYS prevent the GPU from doing something that is detrimental to its ability to function (overclocking and power limits included).

Until I hear of an FE card suffering this fate, I put the responsibility for these failed cards soley on the AIB manufacturers. If it turns out to affect FE cards as well, I put the responsibility on the drivers AND AIB manufacturers.

At no point will I say that the game is at fault. Drivers AND hardware should protect against poor programming - period.

 

[Phaaze88]

My EVGA 3080 FTW3 has been fine so far.

Is it OC'ed/undervolted, or do you not bother with either?

 

[hotaru.hino]

Is nVidia involved in any investigation of the failures? They probably have more resources than any board manufacturer to figure out what is going on. They also have a lot to lose if a defective chip design is the problem.

Considering that only 3090s have been reported to be fried by this game, this sounds more like manufacturers are at fault here for not building their 3090s substantially enough to handle the amount of power the GPU can suck up.

No game or program should be able to do something outside of driver specifications. Driver specifications should ALWAYS prevent the GPU from doing something that is detrimental to its ability to function (overclocking and power limits included).

Until I hear of an FE card suffering this fate, I put the responsibility for these failed cards soley on the AIB manufacturers. If it turns out to affect FE cards as well, I put the responsibility on the drivers AND AIB manufacturers.

At no point will I say that the game is at fault. Drivers AND hardware should protect against poor programming - period.

Drivers, and software in general, can't do anything to hardware that is shoddily built or isn't reporting things correctly. If the GPU reports its power limit is 350W even though it really can't support that because the power delivery circuitry isn't up to snuff, then how does the driver know that? At best all NVIDIA can do at that point is add a driver imposed power limiter and/or a firmware update so the GPU properly reports what it can handle.

 

[NightHawkRMX]

Drivers, and software in general, can't do anything to hardware that is shoddily built or isn't reporting things correctly. If the GPU reports its power limit is 350W even though it really can't support that because the power delivery circuitry isn't up to snuff, then how does the driver know that? At best all NVIDIA can do at that point is add a driver imposed power limiter and/or a firmware update so the GPU properly reports what it can handle.

The issue is even deeper than that. Even with a power limit set well beyond what the vrm is capable of, the card shouldn't fail. The VRM should simply trip ocp and shut down safely.

Either ocp has been disabled via a modified vbios, which I doubt, or ocp is set improperly. Ooooor, the issue is a hardware fault.

 

[alceryes]

Ooooor, the issue is a hardware fault.

Until we see FE cards suffering the 'new world death,' I'm going to say it's the hardware (AIB manufacturers).

Good point about the onboard OCP. That even takes the drivers out of the equation. It's just hardware and firmware at that point. (if that's really what's causing the card's death)

 

[brandonjclark]

This might be anecdotal but I run an Alienware R11 (10-9800kf and RTX 3090) and I have zero issues with the game.

 

[alceryes]

This might be anecdotal but I run an Alienware R11 (10-9800kf and RTX 3090) and I have zero issues with the game.

Dell's version of these cards is a custom OEM design. I've seen reports of them being made by multiple AIB manufacturers, including MSI and ECS.

It could be that Dell followed the specs close enough to where they don't suffer the 'new world death' but it's all conjecture until a true cause is determined.

 

[10tacle]

My EVGA 3080 FTW3 has been fine so far. I wonder if it has something to do with the memory density on the 3090's and 3080 ti's. Who knows...

I have a 3080 Ti FTW3 and have been overclocking it using Afterburner (never liked EVGA's own Precision) with zero issues for nearly two months now. I also keep an aggressive fan curve in AB where temps never get above 60C under load. Fans never run above 70% for that in typical ambient 24C indoor temps. I'm curious how hot all those 3090 GPU owners with bricked ones were allowing their cards to get under load.

 

[ingtar33]

My 2070 super has been hitting 60fps full setting, no problem, no temp issues, no dead gpu.

There have been problems with the 3000 line since they released, only thing is, no one could get their hands on them so no one really noticed. I think the core problem is Nvidia learned last minute how good AMD's RDNA2 was, and they simply overclocked the 3000 series past the point of safety in response; combined with some piss poor coding by amazon in the loading screen (the fps isn't capped, no matter what your ingame, or driver settings are) and you get gpus combusting.

 

[helper800]

Is it OC'ed/undervolted, or do you not bother with either?

I have the 500 watt BIOS and the max power limit with a +120 on the core and a +1200 on the memory. I definitely bothered...

 

[Sleepy_Hollowed]

This at least makes me happy to have jumped to the red team, though I mostly did it due to not caring one bit about ray tracing.

On the more serious side, what on earth is causing this? AMD video cards overall run with less power, but that can't be it, can it? It's just like a 50 W difference at most.

 

[jamiewhite]

“It's unlikely that the Gigabyte models are also affected by poor soldering, unless you believe in coincidences”

You do realise that these are often all made at the exact same factories, based on the exact same Nvidia reference designs, just with a custom designed cooler laid atop. The difference usually comes in the little tweaks to the numbers each vendor makes and the cooling solution.
I’m baffled why I’m explaining this to “Tom’s Hardware” of all sites…

 

[jasonkaler]

Either Amazon found a killer poke by accident, or video card manufacturers are getting a rude awakening on skimping the build quality.

Either way, it's up to the manufacturer to fix the problem.
Stop blaming Amazon's game