This is just a heads up for everyone. If you want to take risks and say “its fine, it wont happen to me” then by all means go ahead and keep playing. Im just going to leave what happened to me below. New World is still killing GPUs and its not only 3090s from EVGA.
I have a 1660Ti from Galax and i played New World since the closed beta. I also played a little bit in the open beta and after launch i played for 60 hours with no issues. I played every second on low settings and with my FPS capped to 60 because i heard the issues people were having since day the first day of closed beta basically and i was scared to play with higher graphics/fps and also get my GPU fried.
However, 2 days ago it happened. While i was playing, out of nowhere my screen went black and fans went to 100%. My first thought was “well… rip my gpu”. But when i unplugged my PC from the wall (because that was the only way i could turn it off, pressing the button did nothing) and turned it on again, it was fine. The first thing i did after booting up again was uninstall NW. I was very lucky that my card didnt die on the first black screen like it happened to many others and i really cant risk killing my gpu just to play the game. I still want to play but its just not worth the risk.
Yesterday a friend told me that NW fried his RX580 (an AMD card), so its clearly an issue with the game, the issue is not happening only with one model from one manufacturer.
I still lurk in the forums/reddit to see if there is a solution or at least something by Amazon on this issue but its just the same old repeated posts. I guess i will wait a few months and see if people are still getting their gpus fried or not and then try again.
Can you link a screenshot of Event Viewer from the time frame this happend please.
This doesnt seem to be a game issue js. New World acts as a catalyst. Which your issue sounds like a driver crash that hasnt recovered( bad drivers) or the other thing is cheap VRMs that are overheating an the gpus VRM setup isnt designed for heavy loads of this gpu (with increased clocks from the factory),also is there any overclocks applied (by you) that have become unstable over time. Since New World seems to be acting as a stress test like Furmark not a lighter stress test as for example Heaven Benchmark ( which acts more like a video game).
If anyone alse has this exact issues can you post the event viewer log of when it happens please
Theres nothing abnormal on Event Viewer, its in portuguese but every error and critical event is just saying that the computer was turned off incorrectly (because i unplugged it from the wall since it wouldnt turn off in any other way).
I never overclocked by GPU, im using out of the box settings since the day i got it and it worked flawlessly in every game i tried (including the BF 2042 beta this weekend).
Hmm, yeah no log of driver crash , Is it possible for you to do a proper stress test an see if the instability comes back like furmark for a hour an see if the same issue occurs again ? and post temps aswell
What i have discovered is that when i use only low settings on everything, while running at 1440p or 1080, i’ve never had any crash. But when i start upping the graphics while running at 1440p its only a matter of minutes before i crash. With VSYNC on or off, with FPS capped thru the raedon software. Still crashing. And the strange thing is that the first days i had no troubles, but something happened when i did some adjustment with the video settings, that made the whole game unstable and i can’t seem to revert it, its like permanent.
Also, how should i share the dmp files? I can’t upload them here.
I have used stresstests before and never had a problem. Temps while playing New World are, for some reason, 2-3C higher than when stress testing and almost 10C when playing other very demanding games.
I played the BF2042 beta this weekend and my GPU was at 68C the whole time, other games i play get even lower temps but thats because they are way less demanding games. While playing New World it was 75C.
This game is a stress test, waiting in queue 90% gpu usage 210W power, for what? mining? ~71C my rx5700xt. When i lower manualy power to -30% ~60C 140W
What stress tests are you using specifically , some stress tests are more demanding then others. As different instruction sets that they are utilizing are different hence higher temps between different stress tests , you probably seen this more closely with certian cpus having to lower clock speeds because avx instruction set as it made them give extremely high temps on alot of cpus as it puts a different strain on a cpu compared to other instruction sets which arent as demanding.
If it was New World it self alot more people would have posts about gpus blowing up. I havent seen any gpus that are older than the gtx 10xx series so far and from amds rx 400/500 series and older .
The main concern is New World utilizing a different instruction set that is making poorly built graphics cards to show their flaws or are these Graphics card just poorly made from the factory. Or is their a instruction set that is making graphics cards fail. (Which if there was amd and nvidia (soon to be intel going into the gpu market) would have put protections in place by now and put a post out , nvidia iirc have even put advertising out for new world.
What i suspect is happening graphics card are in the last few years including my 1080ti are not being built to a quality they should be.
The brand i want to know if failing has occured is Asus. I havent seen any of their cards posted as of yet maybe i havent looked hard enough
I have used 3DMark (the free version), Furmark and Heaven in the past with no issues.
More people? Just look at the forums, there are hundreds of posts about gpus dying (or almost dying) while playing the game. You dont see that in any game. Yes, hundreds of dead cards amongst millions of players are not a huge percentage but that percentage should be at zero in the first place.
I think this is Amazon’s fault as much as it is the manufacturer’s fault. Their instruction set is so unoptimized or so different that not even the most demanding stress tests can reproduce the fault.
The game is pushing the cards beyond their safe limits (tests have shown 10-20% power draw OVER the limit set). This is the same thing as buying a bulletproof vest rated for pistols only, shooting it with a barrett 50 cal and wonder why the vest didnt stop the bullet.
Yea unfortunately the game runs ridiculously hot. Im playing on a laptop (yea notoriously hot) but i legit have to play this at 900p in lowest graphics limited to 49fps to run under 85c -- (can run bdo remastered 1080 120fps at 75c). Its a joke and has been a problem throughout all the alpha and betas etc, a little better now as i couldnt even run the game in first beta without being like 95c but i wouldnt plan on it being fixed anytime soon, just undervolt, fps limit and whatever else you can to keep shit cool. Game benches harder than bench software --.
