As titled I am experiencing constant crashes on my 3080. It’s crashed the game sometimes 15 minutes apart, sometimes 5 seconds. It has also restarted my PC and blue screened. I’ve read that some people have had their cards destroyed.
I didn’t experience this problem during the beta at all, maybe I was just lucky? Ive played around with graphics settings and lowered the VRAM usage but nothing worked. I’m powerless to fix this.
I’ve now stopped using my 3080 and have swapped in my old 1080 to play the game. This is ridiculous. I shouldn’t have to do this, what’s going on? This needs to be fixed as soon as possible.
This is an absolutely zero percent chance that this is a hardware issue. I have had no issues with Diablo 4 running under maximum settings with uncapped FPS during both betas and launch day. POST UPDATE. Constant crashing. game lasts for 30 to 60 minutes at most. This is a BLIZZARD PROGRAMMER ISSUE that they messed up real big.
That is a driver/hardware error. Caused either by a software conflict in the driver, or a hardware fault.
Is that card crashing?
BTW.
Have you thought about clearing the LocalPrefs file from the Documents\Diablo IV folder? Especially if you played in beta/slam? It is not removed or updated from those times, and might cause setting problems. Of course they haven’t said as much, but a simple rename of the file and reopen of the game creates a new default settings file.
I love how instead of playing the game as defaulted when installed and making sure it plays ok, everyone wants to crank it to Ultra then don’t understand why its having issues.
I know its frustrating when these things happen.
But just because you think it is something else, doesn’t mean its something else. But don’t get me wrong here. I am not putting Blizzard off the hook either. There are issues in play and they have more work to do, but they cannot physically damage the card with the game. There has to already been a problem in order to have it show up.
Is the game optimized and running well on various cards? No. I just spent time on another thread doing a simple 10 minute standing in place test at the armor vendor. It makes the card run hotter than if you are running around on the map fighting monsters. So yeah, they have issues. But their software can’t just damage hardware. It can however stress the card to the point where something will give.
Same here, I’ve got a 3080 as well (MSI production) and I’m getting the same crashes, minus the blue screen - my monitors just go black after the GPU fan shoots up to 100%. Sometimes stable for hours, sometimes stable for 5 minutes before the crash.
No, the 1080 does not crash at all. Not even once.
Using my 3080 however I had gone through all of the motions because swapping it out for a lesser card is one of the most annoying things imaginable. I had:
Messed around with many graphics settings and monitored it - crashed on any setting regardless of VRAM usage.
Repaired the game a few times
Made sure it wasn’t my internet by being hardwired
Made sure drivers were up to date
Only thing I haven’t done is reinstalled the game. I have uninstalled the server slam.
This is a blizzard issue, it’s nothing more than that.
My card is also a gigabyte card, yes, but this shouldn’t matter. Why can I play literally every other game on whatever setting I want but running D4 on lesser graphics than what my 1080 can handle, causes it to ‘overload’ ? Nah man, doesn’t sound right.
Because Gigabyte is known for having a flawed BIOS, bad Caps, and a few other issues, that’s why.
But that’s ok. Its easier to blame the game, I get it.
Tell you what.
Do a search for Gigabyte 3080 and bad capacitors. Tell me how many videos you find about it.
There is even a post on here about how the fan curve set by the BIOS ignores some of the temperature monitoring and doesn’t speed up the fans correctly. Setting a custom curve helps though.
I use an RTX 3080 with no issues. I can play all day with all settings maxed out at 1440P/165 FPS.
I would make sure to install the latest game ready Nvidia driver, as that is the one specifically updated to support Diablo IV.
If that still causes issues afterwards, use MSI afterburner and lower the power limit to 80%. It sounds to me like you might have a heat issue, or inadequate ventilation. Check temperatures while gaming while you are at it.
Yeah i can confirm about the same issue. MSI 4090 suprim crashes every 5 minutes. tried every solution given.
Fresh game install
Scan and repair via bnet
New drivers
Fresh windows install
Moved game onto HDD from SDD/M.2
Removed HD textures
Lowered settings
Ran a memory diagnostics
A few others i cant remember.
Put in my 2080 TI and the game works fine. Also i had ZERO issues on the 4090 during the 1st beta on the same setup. Blizz needs to fix this.
So explain why other card manufacturers (not gigabyte) are also having problems? Possibly gigabytes are more plentiful by default so statistically they’re likely to be the ones having the most problems?
I have an MSI RTX 3080 Gaming Z Trio, latest drivers, latest bios, clean driver install with a 5800X cpu. Temps on the hotspot never got above 70c as I’m liquid cooled. GPU core never above 50c. Memory around 60c. Can confirm my sytem is rock solid - ran stress test for 5 hours gpu pegged at 99% while also pegging cpu at 90% to eliminate power supply failure possibility. Since this seems to be a memory intensive game I ran gpu vram checker for 4 30 min cycles - not a single error. No issue in any other games.
System black screens anywhere from 2 min to 2 hr ONLY in d4, happend about 5 times until I gave up. Luckily, after a hard reboot things were always fine. Capped at 170 fps, 1440p ultra setting, DLSS on quality mode. I am aware of some solutions stated to cap fps to 60 and lower settings. I worked hard to get my pc and games - so I wish to play them at whatever settings I want. Those solutions are only a stopgap.
Staying away from this game untill this is fixed due the reports of bricking cards. Yes, these are sympotms of hardware failure, but it’s 100% possible for software to cause hardware to malfunction. It’s time for this to be addressed
If your computer is blackscreen restarting it is 100% a hardware issue. D4 might be incredibly unoptimized and because of that it might trigger the hardware issue that is otherwise undetected, but it’s still a hardware issue. Software can’t directly kill hardware unless you’re specifically being malicious and even then incredibly rare. It can trigger hardware defects or firmware issues that can kill cards but that’s hardware side. The things that will kill hardware are heat and power, both things that the software isn’t in control of.
Trust me, I felt the same way. It HAD to be the game. I was having black screen crashes in WoW, specifically in certain zones/dungeons. No other game I played had this issue at all and I could run stress tests all day without issue. I tried all the normal stuff, clean driver install, reinstalled game, reinstalled windows, etc. I swapped GPUS, not the issue. I RMA’ed memory not the issue. RMA’d the PSU, nope(Seasonic CSR was awesome). RMA’d motherboard(gigabyte SUCKS), RMA’d the CPU(Lots of hoops but quick shipping), still crashed. RMA’ed everything again.
Want to know what the cause of the issue was? CPU
Early stepping AMD 5xxx CPU’s have a relatively common issue with blackscreen restarts. I got 2 bad CPUS, I had never had a single bad CPU in 20+ years of building computers before that. Not sure if the 7xxxx ones have the same issue, but something to keep in mind is that the CPU, even intel, can cause this issue. Check your event viewer for errors, iirc it’s usually WHEA error 18 and 19.
Another culprit of black screen reboots is GPU transient power spikes, ESPECIALLY the 3800/3900. Basically the GPU suddenly demands 2x+ the max power for 1ms. If your PSU isn’t big enough or built to handle such a large spike the system restarts. If you’re running under a 1000w power supply or an older PSU that can absolutely be an issue with those cards. Watch the video on gamers nexus on this issue. The better way to test this is to use a stress test that CYCLES your card, I think OCCT can do that iirc. Having it pegged at 99% isn’t producing the consecutive/variable power spikes that might trigger the issue.
Occt is one of the software that I use in my test suite. System is still solid. Also did check the event viewer - totally clean. I have a 1300w psu, but the unit is kinda old so I swapped it for an 1000w unit today - same problem. As you said - it may be the CPU but after building 100s of computers over the years, I’ve also never seen one fail. At worst, a bent pin on the socket or CPU.
And yes, it is very very rare for software to outright kill hardware nowadays- it is not so rare for broken code to crash a system. Vram and ram overflows (mostly) just crash the software. But as others have shown it seems like a power issue in which the GPU is bailing out due to a failsafe. This would mimick a hardware failure even though it is the software that is causing it. It’s hard to track because I haven’t gotten the logs in hwinfo to persist through a crash.
Luckily, nothing is bricked, but it sucks to consistently crash regardless if the damage is permanent or not. It’s something that either Blizzard and/or Nvidia should address and it seems like enough a problem that us, the consumer, shouldn’t just be scrambling to test and fix every part in their computer when this is the only software that gives us issues
4070ti here and same issue. Sometimes I can play for a couple hours, but most times I crash to desktop or blue screen of death right after character selection.
I love people blaming the hardware as if all the people having the issue just have bad hardware. With all different types of hardware the only common issue is the one game. I can run any other game max settings all day no issues. Diablo 4 will sometimes run for hours with no problems and then sometimes crashes before it even gets through login. I’ve even tried running hwinfo with logging going to see if there is some issue with temps or power draw and I don’t see anything happening, it’s just an instant power off of the whole system.