Rtx 3080 Shutdown Issues

[Chomiq]

Do not use cheapo extensions. Especially with a 3080, especially from EVGA.
My 3080 Ti FE is doing perfectly fine with 750W Focus+ Gold, but the max I saw was 355W of total power for the GPU.

Are you perfectly sure your RAM is stable at whatever speed it's running?

Here's some guy on reddit that had similar issues on MSI board and a 3080 FTW3 and it turned out to be MSI's mystic light software:
https://www.reddit.com/r/EVGA/comments/jal75p

 

[wolf]

My 3080 runs great with a 600sfx gold PSU, has done since September 2020.The problem is unlikely to be wattage related and has zero to do with being Ampere specifically, it's a relatively power hungry gpu that demands appropriate, quality cabling and PSU, of which you might have an issue, or your exact 3080 example may be faulty, but from the description that seems less likely.

 

[ChugginHopps]

Update:

Removed the pcie extension cables and ran the Port Royal Stress Test. Previously I was only able to get to loop 6 before the test failed. This time it completed all 20 loops with a 99.1% passing percentage. Max wattage I saw pull was 380 watts and max temps were 68C.

I will continue to test more with different games that were having issues but so far the preliminary results are encouraging.

 

[Mussels]

Update:

Removed the pcie extension cables and ran the Port Royal Stress Test. Previously I was only able to get to loop 6 before the test failed. This time it completed all 20 loops with a 99.1% passing percentage. Max wattage I saw pull was 380 watts and max temps were 68C.

I will continue to test more with different games that were having issues but so far the preliminary results are encouraging.

Make sure you spread the word about extensions being bad for the new GPU's, too many people fail to believe it since they often run for a short time


Check the ends for damage, too - so you dont use them on another system

My XPG branded ones after 3-6 months:

 

[ChugginHopps]

Make sure you spread the word about extensions being bad for the new GPU's, too many people fail to believe it since they often run for a short time


Check the ends for damage, too - so you dont use them on another system

My XPG branded ones after 3-6 months:
View attachment 236007View attachment 236008

Unfortunately this is not the issue. The problem persists

 

[nguyen]

I have a very odd problem that I cannot figure out. Bought a new PSU and EVGA RTX 3080 FTW3 ultra graphics card. All other components previously worked in my previous build with a GTX 1080 without issue.

Started with an 850 watt PSU. Had frames drop down to like 2FPS and would never come back. Would have to shut down to get a game to play again. RMA'D the psu and got the same model 850 watt. Had the same issue and then the PSU started tripping. Only resetting the power switch on the PSU would allow it to work again. When this happens the RAM led's stayed illuminated.

Ordered a different brand PSU that is 1,000 watts. Installed and played a few hours and the PSU tripped. Decided to undervolt the 3080 and the issue seems to have gone away. Undervolt is set to around 850mv at 1830mhz

My question is why won't this card work when set to stock settings? Is it possible that the GPU draws that much power to trip a 1,000 watt PSU? Prior to installing the card I ran ddu so that no drivers from the 1080 were left. I feel like I am leaving performance on the table and not getting what I paid for by undervolting.

System Specs:

MB - msi x570 unify
CPU - Ryzen 7 3700x
Ram - 32gb Corsair vengeance RGB pro
GPU - EVGA Rtx 3080 FTW3 Ultra
PSU - Seasonic gx 1000
Case - O11 Dynamics
Cooling - Bitspower distribution plate with pump and 360 radiator. 9 fans total. 3 on radiator, 3 on the back side and 3 on the bottom
Drives - 1tb M.2 Ssd, 1 - 2TB 2.5" Hdd 1 - external 3tb backup drive
Windows 11

All GPU drivers are updated and am running 3 separate pcie cables from the PSU. Temps for GPU max out around 65C (when not undervolted) CPU temps hang around 50-55C

Thanks in advance

How about undervolting to 900mV or 950mV, combine with an Core Clock offset you should get the most out of your 3080. Undervolting should act as a safeguard against shoddy coded games like New World breaking your GPU too.

 

[cvaldes]

Unfortunately this is not the issue. The problem persists

Sadly this points to a bad GPU.

RMA it ASAP.

 

[ChugginHopps]

How about undervolting to 900mV or 950mV, combine with an Core Clock offset you should get the most out of your 3080. Undervolting should act as a safeguard against shoddy coded games like New World breaking your GPU too.

I was undervolting previously however I don't thing you should have to undervolt something that should work at stock. Doesn't sound right to me. Right now I am testing to see if afterburner/rivatuner may be the culprit. No shutdowns just frames dropping to 2fps and won't recover

 

[Mussels]

undervolting wont help solve the issue, and could bring in instability from another source


Just going back to basics... you have 100% tested the PC itself is stable, yes?
AFTER adding thew new GPU and its higher power draw?
Overclocks (even XMP) can require tweaking with time, and as parts change

 

[ChugginHopps]

undervolting wont help solve the issue, and could bring in instability from another source


Just going back to basics... you have 100% tested the PC itself is stable, yes?
AFTER adding thew new GPU and its higher power draw?
Overclocks (even XMP) can require tweaking with time, and as parts change

PC was in a previous build I used for 2 years without issue.

Added GPU, power supply, extension cables, 9 fans with commander core and custom water loop. Went through 2 EVGA power supplies and now on seasonic. Took out GPU extension cables. CPU EPS extension and 24 pin extension are still in. Only 1 EPS connected since I am not overclocking nor using PBO.

Only OC is on ram at 3200 with an fclock of 1600. Had that on my previous build as well.

Will continue testing with afterburner and mystic light not running again tomorrow.

 

[Shou Miko]

My MSI RTX 3090 Ventus 3x OC should even trip my Seasonic Prime 1200W even with original cables and running 2 cables directly from the PSU.

I had one trip with my current Asus ROG STRIX RTX 3070 8GB non-LHR but newer drivers and a monitor swap has actually fixed this issue. (My black screen was related to my Asus ROG Strix XG27UQ's G-Sync compatibility wasn't working correctly with RTX 30 series so Asus finally agreed to swap it since I have tested it on RTX 3090 for about 6months and now RTX 3070. When G-Sync was active the monitor would black screen and down to 640x4080 pixel which a 4K monitor cannot display nicely done Asus ).

 

[Mussels]

PC was in a previous build I used for 2 years without issue.

Added GPU, power supply, extension cables, 9 fans with commander core and custom water loop. Went through 2 EVGA power supplies and now on seasonic. Took out GPU extension cables. CPU EPS extension and 24 pin extension are still in. Only 1 EPS connected since I am not overclocking nor using PBO.

Only OC is on ram at 3200 with an fclock of 1600. Had that on my previous build as well.

Will continue testing with afterburner and mystic light not running again tomorrow.

Remember, i was the same. Added a 30 series card, all sorts of issues... because the extra load on the PSU screwed with everything. My overclock was unstable, riser, extensions, blah blah covered it all.

Basically, you changed something - so 'known good' is now "known good... with the older, lower power draw PCI-E 3.0 GPU"

 

[Chomiq]

Remember, i was the same. Added a 30 series card, all sorts of issues... because the extra load on the PSU screwed with everything. My overclock was unstable, riser, extensions, blah blah covered it all.

Basically, you changed something - so 'known good' is now "known good... with the older, lower power draw PCI-E 3.0 GPU"

Yup, too many variables changed at the same time.

 

[Shou Miko]

undervolting wont help solve the issue, and could bring in instability from another source


Just going back to basics... you have 100% tested the PC itself is stable, yes?
AFTER adding thew new GPU and its higher power draw?
Overclocks (even XMP) can require tweaking with time, and as parts change

My MSI RTX 3090 Ventus 3x OC wasn't happy with uv even in MSI's own Afterburner it didn't care voltage was stable no matter if I manage to luck it in the AB so I dropped it really.

 

[eidairaman1]

It did it to this PSU as well as 2 fr

It's plugged into a surge protector. I don't think you should have to have a power conditioner to get these to work properly.

Plug straight into wall or RMA Defective gpu

 

[Mac the Geek]

It's plugged into a surge protector. I don't think you should have to have a power conditioner to get these to work properly.

I once lived in a duplex with substandard wiring. When I (or my neighbor) turned on something with a substantial power draw, the lights would flicker for a second. That dirty power toasted at least two "surge-protected" motherboards, before I learned that surge protection doesn't do squat against power dips. Now all my systems live on UPSes with line conditioning, because the power in the wall is subject to too many factors beyond my control.

Hopefully this story is completely irrelevant to your situation; but if you're spending a couple grand on a new rig, I recommend spending a hundred or two on a decent UPS.

 

[Vayra86]

Boy am I glad I skip Ampere altogether.

Definitely the pandemic gen.

 

[oxrufiioxo]

Boy am I glad I skip Ampere altogether.

Definitely the pandemic gen.

Eh, I've done 3 3080ti based systems, 1 3080 based system, 1 3090 based system, 1 3060ti based system and only had issues with a FTW 3080ti shutting down systems. Besides that they've all been pretty great. I feel like Turing and RDNA 1 had way more issues the first year ish.

Small sample size I think I actually owned more pascal based cards with none having any issues but this gen minus the increased power consumption has been pretty good besides availability although 4 I've gotten for or near msrp without a bot so it is what it is I guess.

 

[Vayra86]

Eh, I've done 3 3080ti based systems, 1 3080 based system, 1 3090 based system, 1 3060ti based system and only had issues with a FTW 3080ti shutting down systems. Besides that they've all been pretty great. I feel like Turing and RDNA 1 had way more issues the first year ish.

Small sample size I think I actually owned more pascal based cards with none having any issues but this gen minus the increased power consumption has been pretty good besides availability although 4 I've gotten for or near msrp without a bot so it is what it is I guess.

I do guess the myriad of forum topics surrounding power and ampere has my view colored. Thanks for that input

 

[oxrufiioxo]

I do guess the myriad of forum topics surrounding power and ampere has my view colored. Thanks for that input

I don't blame people skipping this gen and really the more simple you can keep the systems the better imho..... I am a little worried about 4000/7000 cards and them needing 500+ watts though. I do hope the OP figures out his issues because gaming on these cards at 1440p/4k is just awesome and random shutdowns put a huge damper on that.

 

[ChugginHopps]

Update:

So I believe there were multiple issues going on with my build.

Issue 1 - pcie extension cables causing random shutdowns

Issue 2 - upgrading to windows 11 without doing a fresh install

So far no shut downs are occurring after several days of testing with the stock pcie cables from seasonic.

I also decided to wipe my OS drive and reinstall a fresh windows 11 OS. Previously upgraded windows 10 to 11 so there may have been issues there. Did not reinstall afterburner or dragon center just X1 Precision boost so I could controll the fan curve for all 3 fans on the GPU.

So right now my PC is running fine. I've had some game crashes specifically on theHunter call of the wild however others have had crashes in that game as well. No shut downs though.

MS flight sim 2020 seems to run just fine with no issues as well as death stranding and warzone. Temps are about 64C on the GPU at 97-99% load. Ambient is around 22C.

Looks like I had several issues with the build that looked like it could be a single issue. Will continue to test more but it's looking more stable than it previously was. I even have a slight OC on the GPU at 1980mhz and it's stable.

 

[Mussels]

dragon center is known to cause all sorts of issues, then again so is most of the bloated motherboard softwares - i suggest avoiding all of em