Some NVIDIA GPU Users Are Complaining of BSODs with Ryzen 3000 Chips

Tsing

The FPS Review
Staff member
Joined
May 6, 2019
Messages
11,235
Points
83
The curse of early adoption has struck a number of NVIDIA fans. A growing number of green team members say they have been experiencing WHEA or BSOD errors after upgrading to a Ryzen 3000 chip, which are crashing their systems. It seems to be triggered by gaming and other graphics-intensive tasks.

The good news is that NVIDIA has already replicated the problem, which may have something to do with PCIe 4.0. (Some users said that switching to PCIe 3.0 solved the issue.)

NVIDIA hasn't mentioned when the hotfix will be available, however.
 
Of course someone has to be an early adopter, but this is why I always prefer to wait at least a couple of months to make sure everything is ironed out before buying stuff.
 
It's not too uncommon for me to be an early adopter with GPU's. Pretty much try to get them as soon as the one I want is released, and in stock, or I can afford it. I've grown accustomed that from DirectX 10 to RTX that some features are not always readily available and performance won't usually be optimal in most games in the beginning. Other than NV borking something later on with a update I haven't had any memorable issues but that one usually happens about once a year. It does sound like NV still has all their eggs in one basket with PCIe 3.0 and haven't really put much thought into 4.0 yet.

As far as new ports or iterations like SATA, PCIe, USB, etc., I usually will wait a little while to see it all play out. I've seen too many of these kind of stories over the years.
 
So someone tries to run PCIE 4.0 on a driver optimized for PCIE 3.0 and gets a BSOD?! How unexpected!
 
is there a setting in the BIOS to change to PCIe 3.0 with the new X570 mobo's? ..or how does that work exactly?
 
ThreeDee said:
is there a setting in the BIOS to change to PCIe 3.0 with the new X570 mobo's? ..or how does that work exactly?
Click to expand...

This right here is a good question. Anyone have a 570 that can explain?
 
ThreeDee said:
is there a setting in the BIOS to change to PCIe 3.0 with the new X570 mobo's? ..or how does that work exactly?
Click to expand...

It's just a setting. It works the same way it does on every other motherboard that has settings to adjust between PCIe Gen 1.0, 2.0, and 3.0. The only difference is that X570 has an additional PCIe 4.0 setting.
 
Dan_D said:
It's just a setting. It works the same way it does on every other motherboard that has settings to adjust between PCIe Gen 1.0, 2.0, and 3.0. The only difference is that X570 has an additional PCIe 4.0 setting.
Click to expand...
I guess I don't ever recall seeing this particular setting .. I'll have to look at my x470 a little closer now, just out of curiosity
 
I have yet to see this on the test bench using either the ASUS Crosshair VIII Hero or the MSI MEG X570 GODLIKE.
 
Dan_D said:
I have yet to see this on the test bench using either the ASUS Crosshair VIII Hero or the MSI MEG X570 GODLIKE.
Click to expand...

You're more than welcome to come to NC and evaluate my build. I can make it crash on demand. Already contacted Gigabyte and did the Nvidia survey about the issue.

The strange thing is I can extend the time before a driver or WHEA crash by reducing the FCLK below 1500. I've tried all kinds of things to stop the crashing. Turned PBO and boost completely off, clocked RAM at 2133, turned all power savings features off, booted from a SATA drive instead of PCIe M.2. Nothing gets rid of the problem.

Definitely not temps. GPU never gets above 52ish and CPU barely hits 60.

All I know is that if it's not fixed in a week this entire build is getting boxed up and sent back. I'll buy a 9900K and be done with it.
 
Riccochet said:
You're more than welcome to come to NC and evaluate my build. I can make it crash on demand. Already contacted Gigabyte and did the Nvidia survey about the issue.

The strange thing is I can extend the time before a driver or WHEA crash by reducing the FCLK below 1500. I've tried all kinds of things to stop the crashing. Turned PBO and boost completely off, clocked RAM at 2133, turned all power savings features off, booted from a SATA drive instead of PCIe M.2. Nothing gets rid of the problem.

Definitely not temps. GPU never gets above 52ish and CPU barely hits 60.

All I know is that if it's not fixed in a week this entire build is getting boxed up and sent back. I'll buy a 9900K and be done with it.
Click to expand...

I wasn't doubting that it was happening. It's well known that it does. I'm simply stating, that so far, I have yet to see this problem for myself. Are there specific circumstances in which this occurs?
 
Riccochet said:
You're more than welcome to come to NC and evaluate my build. I can make it crash on demand. Already contacted Gigabyte and did the Nvidia survey about the issue.

The strange thing is I can extend the time before a driver or WHEA crash by reducing the FCLK below 1500. I've tried all kinds of things to stop the crashing. Turned PBO and boost completely off, clocked RAM at 2133, turned all power savings features off, booted from a SATA drive instead of PCIe M.2. Nothing gets rid of the problem.

Definitely not temps. GPU never gets above 52ish and CPU barely hits 60.

All I know is that if it's not fixed in a week this entire build is getting boxed up and sent back. I'll buy a 9900K and be done with it.
Click to expand...


So what pieces of this puzzle if any have you swapped out/replaced? Bad power module on the motherboard? Is it under specific loads? Swapped video card?

Just curious, if this is an upgrade and you still have some older parts might be able to swap some around to see if those are the problem. (If it's motherboard Ugh... that can be a pan to isolate.)

I feel for your pain here and wish you the very best of luck!
 
Dan_D said:
I wasn't doubting that it was happening. It's well known that it does. I'm simply stating, that so far, I have yet to see this problem for myself. Are there specific circumstances in which this occurs?
Click to expand...

Any kind of heavy load on the PCIe bus. For instance running Furmark. Sometimes it'll run for 5-6 minutes, others 1-2 minutes, before a crash. Joining a match in PUBG lasts about 2 minutes before Nvidia driver crash and WHEA error. Same for GTAV and WoW. Pretty much can't play any game right now.

I can let P95, AIDA64, OCCT, whatever run indefinitely without issue. I've ran memtest86 for hours, so it's not the RAM.

Definitely something with how these 3000 series CPU's interact with the PCIe bus that's causing hardware errors. Strange that it's not happening for everyone, and mobo make/model has no bearing either. X470 mobo's are effected as well when running 3000 series CPU. The common denominator is the CPU.
 
Weird. I have actually played a couple hours of Destiny 2 on the Ryzen 3000 series CPU's without issue. I've run game benchmarks without issue either.
 
wish I could play even an hour. Do you know any tricks I could try to get this thing working?
 
I haven't a clue since I haven't ever encountered the problem.
 
On my mobo I had to manually set to GEN 3 (PCIe 3.0) for pcie otherwise my RX 5700XT would confuse my mobo when set to auto because it's a PCIe 4.0 card. Maybe something similar happening but backward and you have to set it to force PCIe 3.0 for your video card .. or something of the sort because it's not auto negotiating correctly for whatever reasons like my ASRock mobo was doing prior to a BIOS update
 
You must log in or register to reply here.
Become a Patron!
Back
Top