Separate names with a comma.
Discussion in 'nVidia Flavor' started by FrgMstr, Nov 1, 2018.
Hug your 2080Ti's and tell them you love them, for you know not the date nor the hour...
Sorry if it was mentioned already (I didn't read pages 2-3) but Steve over at GN is currently working with owners of funked up cards to have them sent to him so they can do some further investigating. Could be interesting.
Zotac 2080Ti AMP and no issues at all.
im running 1809. I wonder if people not having the latest DX, which the RTX cards are wanting is any issue?
Registered a new account because this is relevant. (Have an old account, but young me was stupid and I can't recover it). Kyle is a cool dood.
TL;DR: Purchased two 2080TI FEs. Had to RMA one. Currently using the replacement card by itself and I've experienced one BSOD so far. 0323 serial numbers all around. Not feeling great.
I started out using both cards + NVLink, but experienced frequent crashes, especially in games that utilized SLI. Eventually, Windows was giving me the whole "this device is having issues" spiel (My resolution was low, etc.). I narrowed it down to one card, which I RMA'd. I've been using one card at a time ever since, and while I don't believe I've experienced any issues on the other card I originally ordered, I did just BSOD with the RMA replacement.
My case isn't great for these cards thermal-wise, but I've been leaving it open with a desk fan obnoxiously blowing directly on the card just to eliminate some variables. I ultimately want to put these cards under water in a new build, but I need a little more confidence in them before I get there. I've had these cards for nearly a month and I haven't even played with overclocking yet.
EDIT: It seems that there's a huge amount of complaints of BSODs in ARK specifically (which I've been playing). While the card I RMA'd definitely died, I am not sure what the true culprit my latest BSOD is.
I am on Windows version 1803, still good to go.
While looking for solutions for the FC5 crashing, I found a thread recommending putting power limit at 80%. Tried that for about 1 hour last night and I didn't see any crashing.
Granted, 1 hour is not a lot of testing, but I was usually getting crashes in the 15 - 30 minute range, so that is a positive sign.
Especially since changing PSUs helped a bunch, I wouldn't be surprised if there is some sort of power issue here with the cards.
At the end of the day it sounds like some are too finicky. I wouldn’t be surprised if it is the GDDR6 based on the reports.
If you get bored feel the ram from the back of the board and see if it burns your finger.
The other thing is these are clocked pretty darn high out of the box. I know with my 980 I volt modded it to lock voltage because if it throttled it would crash once OC’d... it would drop voltage faster than core mhz. I wonder if that’s why the 80% helps here. Or adding more voltage.
So is this just a Founders Edition problem? I can't seem to find any reports of dead EVGA and MSI cards. Digital Trends mentioned some Zotac and Asus though, but it seems like 95% of the dead cards are Founders Edition with a majority being from a specific batch 0323.
I'm curious about this too.
Spoke to soon....JUST after I logged into [H] for a status update on my replacement card....it just died around 30 seconds into AC Odyssey. I got the space aliens dancing from 1980 artifacts....oh joy....
Back to my 1080 for a bit and RMA time...again.......
My 2070 Armor from MSI is all good after 4 days and OC with nVidia Scanner (fingers crossed that it won't fail).
Good thing mining isn’t all the craze right now. If these cards have a significant failure rate at stock settings I can’t imagine how they’d do crunching numbers 24/7.
Anyone else find it ironic they push the envelope with price and what consumers will pay, and it’s the same ol’ Issues with launch cards that seem to plague every other or so generation launch?
At least if AMD decides to beta test with you it’s half the cost.
For me that’d be refund time. If I spend more time swapping cards... I’d take my money back and wait until this is all sorted.
Fury X and Vega 64 were both +$100 too high... the value wasn’t good at all.
Looks like the poll needs to let people vote for "it died" more than once
They really need to offer refunds......no reason why you should be stuck in the RMA cycle like that.
Wait, did you have two EVGA 2080 Ti die on you? These are not nVidia 2080 ti FE?
LOL refunds. Good 1.
Realize it’s a small sample size here, but if I am reading the numbers right, it looks like a good 25% of those here with FE 2080Tis may have had to RMA them... definitely not a good sign.
What I can say is that I've moved this card to two different systems. One is X99 one is X470. On both I get occasional issues where I have a black screen on boot and have to hard reset the system. It's not often, but it happens. This only happens with the 2080ti and not the 1080ti I had.
I didn't try that - Next time I will. It could just be a bug versus the actual card failing.
PCGamesN is is reporting that UK retailers have not seen above normal failure rates for the 2080ti cards that they sell.
It is notable that they don't sell Founders Edition cards.
Makes you wonder if this is a manufacturing problem at whoever Nvidia uses to contract manufacture these. (Foxconn?)
I had one EVGA 2080ti OC Edition die on me. I sent that back for refund. I then bought a 2080ti FE from Nvidia.com thinking that would be a safe bet....I didn't read all the forums on Nvidia.com or here before placing the order or I wouldn't have. The card arrived super quick too. This card died on day 4 and RMA'd
OMG. I'm so sorry.
Wow, talk about tough luck!
Some people being unaffected and other having multiple cards die immediately makes me wonder if there are some power supplies they don’t like. Prob a crazy thought and just random occurrence though.
Yes, I believe it's power related. I had nothing but problems with a Corsair AX1500i, and I swapped it for an EVGA 1000 G+ and it's been pretty solid since then.
Before, I was getting frequent blue screens as well as Windows not booting and all sorts of weird errors. After I put in the EVGA PSU it's been mostly solid.
Still got some crashing in FC5, but I think that is game/driver related and not a hardware issue. Though I did adjust the power limit on the card and that seemed to help.
I set it to 80% power limit and I haven't see a crash since then (probably put in about 4 - 5 hours now with no crash). Then I set to 120% to gain more performance, and it still seems working.
Spoke too soon. Got a crash and BSOD in FC5 just now.
This is really frustrating, especially for the cost of these cards.
Might have to do another RMA.
Just got crash to desktop in SotTR and Dying Light after about 20 minutes of gameplay. It's not just FC5.
EVGA 2080 ti FTW3...so far so good
I can't rule out driver issue for FC5 for me. I had a weird crash to desktop trying to leave a cave today, but on reload it was fine. Came back in the afternoon to play some more and found my computer rebooted itself. I'll have to monitor, but event logs didn't show anything conclusive. My crash from trying to leave the cave triggered a graphics error in the system log though. Subsequently I played 3 hrs non stop to no problems in FC5, 1 hr of Vermintide 2, and some OW to no problems.
PSU issues maybe if it random reboots though you aren't the first person to have issues with FC5 as well, I think Cyber is experience issues with FC5.
Random reboot is fairly indicative of a PSU problem.
Has anyone had issues with a RTX 2080? Have you been able to solve them?
I have the Evga 2080 Ti XC ULTRA, been great in games so far, using direct display port on my main monitor. And Evga Precision X oc for more aggressive fan profiles + the 130% power target.
While streaming to twitch, I usually have the broadcast and other junk on a second monitor, and I notice that stuff on the second monitor get's slight artifacts, while the game on my primary monitor is fine during the same time. My second monitor is using HDMI to DVI. Last time I saw artifacts like this but much much worse and on anything displayed was a single slot evga 8800 gt SSC that couldn't cool itself properly, they re released them with a bigger fan shroud hole for more air if I remember correctly.
anyway I game for hours on end, and as long as it doesn't impact a game I'm fine. I'm waiting for BFV next weekend, then I'll really put this sucker to work.
I would back down that power target and see if the artifacts disappear. You don't want to be running it like that.
Nope no issue but game casual what brand of 2080 do you have. I had crashes in game for COD Blops 4 though.
I need to see the random reboot in person to call it truly random and not chalk it up to yet something else bugging out. 1 unexplained reboot during non heavy gpu use in a span of 2 weeks is not a large enough sample size to work with. Also FC5 has had weird quirks since I got it in the summer with my 1080ti, so I can't rule out the game as being glitchy.2080ti is pulling 262W under load which is hitting its power reg limit at that point.
I've also installed the asus monitoring/oc program and have since prevented it from starting at start up now, so I'll keep an eye on it. GPU-Z is doing some logging for me now so I can see if maybe that catches something.
Mine is ok so far touch wood. Got mine last saturday. its a pailet one