rtx 3070 Random GPU(?) reboots? RTX 3070

[ofenfrisch]

Hello Guys,

 

So I got my hands on a 3070 from Palit and decided to build a new rig. 

Since using the rig my it randomly restarts. Weird thing is, it's not the entire PC, just the GPU, at least I believe so. The screens of both my monitors turn black and after a few seconds a kind of glitched out version of my desktop shows up(I can see the taskbar but everything is dark, no Icons and I can't click anything and have to manually restart the PC). This does not happen under load! I can let furmark run for as long as I want and nothing happens. It just happens randomly when gaming, sometimes with MSFS, sometimes with WoT. The temperatures are absolutely fine. Below 70C for the GPU and below 60C for the CPU.

This seems to be a common issue but Nvidia doesn't reply to any of the posts in the forum. In my case, it's weirdly not restarting my entire PC, that's why I'm making this post.

 

So far I have reinstalled all drivers via DDU, reinstalled Windows, switched the cables between PSU and GPU for new ones, tried with and without riser cable... nothing works it just randomly happens. Sometimes twice a day, sometimes once every two weeks but I can't put my finger on what's causing the crash... So far t has only happened when I was playing a game.

 

I don't have a second PSU nor a second GPU to swap my parts with. I have never done any overclocking on any parts. I do not get any error logs unfortunately. All drivers are up to date.

 

You guys are my last resort since the rest of the threads about this problem just don't find a solution or OPs just stop replying

 

Thanks in advance,

Tommi

 

Case: Fractal R6

PSU: EVGA 750 GQ, 80+ GOLD 750W

CPU: Ryzen 9 3900X

Cooling: several NZXT Fans and a Kraken Z73

RAM: 2x 16GB Vengeance RGB PRO DDR4 3200MHz XMP 2.0

Motherboard: TUF GAMING B550-PLUS

GPU: Palit GeForce RTX 3070 GamingPro (in a PCIe 4.0 Slot, hooked up with two separate power cables)

Storage: one M.2 SSD, one regular SSD, one HDD, basically all of them are empty, kinda overkilled it here

 

[ofenfrisch]

Sorry, forgot to mention: I also tried different power settings for the GPU and CPU, and also tried different DP ports on the GPU, also tried changing the HZ and resolution for my display. PCIE settings in the bios are correct. Reseated RAM and GPU. Im really clueless here

[ofenfrisch]

So Asus just released a beta BIOS for my MoBo and I just flashed it. Will report if that changed the error. 

Also since I installed it my idle CPU temps are like 10°C higher than before...what?

[ofenfrisch]

Update: did not fix my issue.....

I pulled the error logs from the minute the error occured and Ill attach them here.

Id be SUPER thankful if someone could analyze them for me!
 

crash.evtx

[Mark Kaine]

On 6/10/2021 at 11:16 AM, ofenfrisch said:

EVGA 750 GQ, 80+ GOLD 750W

Try another one and make sure its fully compatible with the 3070, such as Corsair RMX/I series.

 

 

[ofenfrisch]

33 minutes ago, Mark Kaine said:

Try another one and make sure its fully compatible with the 3070, such as Corsair RMX/I series.

 

 

I dont think it can be the PSU since the PC isnt completely rebooting when the error occurs. Also the error occurs just randomly and not under heavy load/benchmarking

Edited June 12, 2021 by ofenfrisch

[Hojunator]

Hey mate,

 

I have a somewhat similar problem with my RTX 3070..your not alone.

 

I'm running a..

 

Ryzen 2700x

16GB gskill 3200

Asus crosshair hero viii

M.2 samsung evo

Evga G2 1000w 

RTX 3070 Inno3d ichill x4

 

My issues only happen in gaming..I get black screen crash and reboot but I tend to get the crash to desktop more often.

 

Steps I've tried to fix are...

 

Reinstall Windows 10

DDU driver and installed latest from Nvidia.

Downclocking in msi afterburner by -200 

Voltage locking to 1v on the GPU with MSI Afterburner.

Reseated GPU.

Tried gpu scheduling on windows

Tried the control flow guard workaround in windows as I thought it might of been a issue with direct x12 

 

Temps are similar to yours aswell.

I'm really at a loss here, it's looking like an RMA unless I haven't tried something?

 

I just thought I'd share my steps incase you haven't tried any of those.

 

Good luck..

 

 

 

 

 

[ofenfrisch]

On 6/16/2021 at 4:32 PM, Hojunator said:

 

 

 

 

 

 

Thanks for contributing to the thread!

Ive actually tried everything you mentioned except for the underclocking because Ive read in several forums that this doesn’t fix it.

What’s really worrying to me is, that Nvidia doesn’t address this issue.

 

 

I really dont wanna RMA the card atm since Im afraid of not having a GPU for like 6 months

[ofenfrisch]

Just crashed again, took a screenshot this time of what mz desktop looks like afterwards. Can someone iodentifz the problem with this?

The taskbar is still there... just blacked out ...

[ofenfrisch]

So, I bought a 3090. Other specs didn’t change, not even the PSU. No issue anymore

[ofenfrisch]

Yup, I was wrong. Still randomly occurs with my new GPU