Found that I am able to run 90 FPS using medium settings and using MSI afterburner with power at 80% and fans between 80-100. Overall it’s been smooth but now I’m having this issue of in game dialogue lines/audio either not playing at all or lagging then all playing simultaneously. Almost like they are all loading in at once. Have been having a blast otherwise though! If anyone knows how to fix the dialogue thing lmk.
1 Like
I got the same problem here, after 3hrs playing… Gigabyte 3080 Ti just black screened and got bricked. I know the card shoudn’t die because of the game, but still sad tho. Now I have to wait until my 3080ti got fixed and i don’t know if I shoud play D4 again and risk to have the same problem.
I’d suggest selling this card as soon as you get is back, and buy some other manufacturer and card model. There is nothing like being scared of playing a game because GPU may not survive its intended use.
2 Likes
Im having similar issues with lockup and reboots that I didnt have during the betas at all. The betas ran super smooth for me without any crashes.
Surprise … Im not running a Gigabyte card. EVGA 3080 TI.
My Nephew has a 3080 from EVGA.
His was full of dust and pet hair causing it to run too hot and reboot to save itself.
Not saying that’s your issue, but something to check for… lol
However, we also adjusted his settings. He is running a 1440p G-Sync monitor. We capped the FPS to 120. Lowered the settings to High. Turned off Reflex.
But we also setup Afterburner and set a custom fan curve to increase the cooling a tad.
Since all that and cleaning it out, its been running fine and he has been playing for hours at a stretch.
1 Like
Adding another Gigabyte 3080ti to the graveyard here. Hoping that the RMA process will be kind.
2 Likes
Yikes.
Hate to hear about it. What brand was it?
This.
To everyone with a Gigabyte RTX 3080/3080Ti/3090 (and not only Gigabyte tbh, but that’s just me), I’d suggest selling their cards and getting an RTX 4000 series card, or an AMD RX 6000/7000 series card.
Personally I already sold my Gigabyte RTX 3080 Ti and I’ll be getting an AMD RX 6950XT/7900XT as a replacement, probably next month, as gpu prices are about to drop.
2 Likes
Amazing move, mate. I am happy in Radeon camp and I had zero crashes or overheating or anything. Playing in 4K 144Hz non-stop, no FPS limits. 
1 Like
Same issue here. Playing on Diablo 4 and of course it is 99% utilization and well above the nominal thermal ranges. Screen goes black, I force reboot the cpu and my 3080 no longer powers on. Not able to get it to initialize from reseating it into a different GPU slot or anything.
Sent it back to GIGABYTE and their RMA identified a broken component (of something?) and I am just waiting for it to get back to see if it is actually fixed (probably don’t need to mention how bad GIGABYTE’s RMA service is).
I had frames capped 120 fps for foreground and 10 fps background via the in-game settings however I don’t think they were capped in the nvidia control panel. This go around I am going to be much more closely monitoring to the GPU voltage ripples, thermals etc and constantly record in case I run into another brick. The ironic part is my backup GPU (1070) more or less plays at the same quality settings as my 3080, so… that’s raises a good question why a 3080 is sitting at 99% utilization?
Anyhow, I would assume an engine being unoptimized would just result in crappy FPS as it’s primarily just sitting ontop doing DirectX calls, and DirectX calls are working with the Nvidia driver suite. Nvidia driver sits ontop of the GPU firmware and that’s likely the core issue on why they’re committing suicide is just bad voltage/thermal control management from being pushed from up top because of poor optimization/clugy implementation somewhere.
You had better hope it isn’t this:
I posted about this a week ago. Its possible that the cards have this crack and the stress of the game makes them finally fail. And Gigabyte refuses to warranty this.
1 Like
Yikes, well that is scary. I will say these 3080s had absolutely the most amateur power interface you could imagine.
They didn’t even bother to just glue in the contacts to the PCI-E power molex connector and it has some horrible pig tail going to a board mount connector (which should’ve been what had the PCI-E power pins in the first place) so I assumed it was just something in there that finally died. You could actually break the card (sort of, push the pins out requiring you to open it up to push them back in) just by inserting in the PCI-E power if you didn’t have ridiculous oversized sockets.
Good to know though and add another reason why it’s worth spending money on a different brand than GIGABYTE. It’s too bad EVGA cut ties with NVIDIA as they seemed to have been the premium brand to go with.
1 Like
My last 7 cards were:
EVGA 580 (blew up one day playing a game, sold on ebay for parts)
Asus 980 Strix (still have, still works, in a Win7 system)
MSI 1660 (used for a while and sold. Ran D4 closed beta on this card)
MSI 2060 (had it for 2 weeks, sold on ebay; also ran closed beta)
MSI 3060 12GB (just sold this last week, used it for 10 months)
Asus 2070 Super (this is for sale, it was an impulse buy at good price, but don’t need)
MSI 3070 Ti (current)
Not a Gigabyte among them. MSI has been solid for me, the Asus weren’t bad either.
However, up till the past few years, I worked with a LOT of Gigabyte motherboards. But my last two desktops were MSI boards. Been happy with their stuff so far. *shrug
MSI definitely might be the game plan. I have heard ASUS RMA has gotten worse, I use a MSI mobo and never had an issue at all. My 1070 is a GIGABYTE card (windforce, I feel like everyone had this one) but that might’ve been during a period of time before they took a nosedive on quality?
One thing for sure, maybe it is time to get a case like the MSI MPG Velox 100R where it comes out of the box with vertical GPU mounting, that way sagging should be no concern on breaking the PCB.
Great breakdown. You got a point there about the fans not spinning up below 41%. I have a Zotac 3090 Trinity OC (eyeroll over OC) I just got a couple days ago and it was on fire right out of the box. Regular gaming temps were over 70C with thermal junctions reaching over 90C. Anyway, I knew this was not safe, so I undervolted it and made a custom fan curve. Gaming temps are now in the high 60s with an occasional leap to over 70 with the same performance. So safe overall, but after reading this post, I decided to look at the fans as my card was idle at 35C. Sure enough fans were cycling on/off. I had my curve base at 35% which is where I usually start so fans don’t stop. I bumped the base of the curve up to 42%. Still cycling on/off. I bumped them up to 45% and now they are on continuously which is what I was aiming for. Glad I read this post otherwise I wouldn’t have known. I believe this has to do with the “energy efficient” crap these manufacturer’s have been doing with cards. Letting them cook at low temps and passive cooling, then ramp up when the card loads voltage. Never liked this, but it seems this may be a BIOS feature as fans do not normally take 45% voltage to power on. So even though Gigabyte cards are frying, I expect to see Zotac and many others have this same problem as they are not the only ones doing this. My thoughts on this are that these chips are heating up too fast because they are relying on passive cooling at the die and thermal junctions, but when a serious load is placed on the whole board, there are other hotspots that are already heated up (probably solder joints) that immediately melt and fail if a high enough voltage is loaded on the line with only passive cooling as a stop gap. Manufacturer’s need to ditch this “energy efficient fan” crap. People don’t buy 3-400 watt cards for “energy efficiency.” They can get away with that for bottom end chips that are pushing a quarter of the watts. But these high end internal combustion-like cards need MASSIVE cooling ALL the time. My suggestion is people start all their gaming card fan curves at 45-50% to override this horrible feature. That’s my 2 cents. Spend it or save it.
2 Likes
Man, it just happened to me right now… absolutelly the way you described. Even the crash before the final death of the card was the same… and the fans. My Card is Gygabyte Aorus Master 3080Ti 12gb. I just opened a RMA on Gigabyte right now, since the warranty on them is 4 years. I hope you have a good outcome.
So after sending my Gigabyte 3080ti in after D4 bricked it. I got it back and it works. No explanation, not even a confirmation on it even being repaired. The ticket didn’t specify much so I figured I’d just plug it in and see if it worked. It works. Gigabyte honored the warranty. Now do I try to play D4 on it or not… lol
That’s not going to work.
And besides, as some have posted, the RMA replaced the card with a good one. So some have been getting repaired, which means they apparently had an issue to begin with.
why you such a simp for Blizz dude? I see you white knighting them all over the forum. I know this because I’ve been frantically searching for a solution to the VRAM leak, constant lagging, and frequent DC’s. I also see you posting that it’s on the client and not Blizzard… ok, my guy.
If you have been, then you been missing the point, which is sad to be honest.
I am not simping for Blizzard at all.
Absolutely not.
I have seen two primary issues:
-
They have a serious issue with how the VRAM is handled.
-
They have a problem with the Diablo IV network traffic.
This much I have been fairly certain about.
The first is pretty obvious. It affects EVERY SINGLE SYSTEM. However it doesn’t DEBILITATE EVERY SINGLE SYSTEM.
As for cards being bricked, I’m at the point where I’m pointing my finger directly at the card, the setup, the care of it, the install, and luck in general. It sucks to lose a card, but honestly, this isn’t a new concept that cards can go bad when pushed. The game isn’t bricking your cards.
The second I see happening a bit more scattered. But their traffic is causing some people’s network hardware to crash, or its being blocked by some ISPs. (or appears to be). WHY is the question. What is going on with their traffic that only their traffic is being affected?
NOW.
At the SAME TIME.
There are also several of you that have had issues with their hardware, how they have things setup, and/or issues with their ISP or network connection.
So you have BOTH sides causing issues, and some people confusing one for the other.
So no “my guy” I’m not simping for Blizzard. But I’m also not going to sit here and blow smoke up everyone’s butt when they have an issue on their system or their internet either.
Does Blizzard need to address these two primary glaring issues? Absolutely.
Are they to blame for every single issue someone might have on their setup? Of course not.
That’s it and that’s all. Game on.
1 Like