DOC TOR

THIS MONTH THE DOCTOR TACKLES...

↘ submit your questions to: [email protected]

> USB drive disaster

> PC overwhelmed

> Old password tool

PCIe 4.0 glitch

I’ve been building PCs since

the first 386 I put together,

so I’m no stranger to PC

problems. I recently bought

a new computer from

CyberPowerPC rather than

build my own, thanks to

the video card situation. I

had three Seagate external

drives connected to my old

computer—two 3.5-inch

drives with their own PSUs,

and a USB-powered Backup

Plus Portable. I connected

all three to the USB ports

directly on my MSI MPG X

Gaming Plus motherboard.

I had no problems with

the two larger drives, but

the portable drive kept

disconnecting every few

minutes. In Event Viewer, the

System logs were filled with

errors for that drive.

I didn’t think the drive

could go bad just plugging

it into a new computer, so I

updated the motherboard

BIOS and all the drivers.

Nothing helped. I chatted

online with Seagate support

and CyberPowerPC.

They both had the usual

suggestions—try a different

cable, plug it into a different

USB port, none of which

worked. I submitted an RMA

to replace the drive and

while waiting for the new

drive to arrive, I plugged

it into the front USB port,

which connects to the JUSB

header on the motherboard.

Astonishingly, the drive

didn’t experience any more

problems (I still don’t know

why this port worked).

I searched the internet

for any problems related to

my motherboard and, to my

surprise , I found out that

AMD has a USB connectivity

problem with its X

chipset. When it is set to use

PCIe 4.0 , that appears to

affect all motherboards using

this chipset, not just my MSI.

This problem has existed

for around a year from what

I’ve seen and there’s still no

fix. I have a Ryzen 5700G,

which only supports PCIe

3.0, so I just changed the

BIOS setting to use PCIe

3.0 and I’ve had no further

problems. But that’s not a

permanent fix if I ever want

to upgrade the CPU or my

M.2 drive. I’m wondering how

this problem has existed for

so long, but yet the support

people I chatted with had

no knowledge of it and AMD

doesn’t have a fix.

—Kevin Mahoney

THE DOCTOR RESPONDS:

First, thanks for sharing your

detailed troubleshooting

process and subsequent

discovery. AMD acknowledged

the problem back in February

and released an AGESA

firmware update (1.2.0.2) for

Ryzen 5000-series CPUs that

allegedly fixed the problem—

but not for all. Interestingly,

subsequent updates have

been rolled out, your X

board had two subsequent

patches (1.2.0.3b and 1.2.0.3c)

via BIOS updates but, in your

case, clearly to no effect.

The Doc can’t provide a

definitive explanation as to

why the front-mounted port

proved more stable for your

external self-powered USB

drive. The front-mounted

ports connect through the

X570 chipset rather than

directly to the AMD CPU (its

four dedicated ports are on

the back of your mobo), so

would be subject to the same

problems with PCIe, given that

the X570 chipset interfaces

with the AMD CPU through a

PCIe uplink/downlink.

That said, some confirm the

issue specifically manifests

itself with USB 2.0 ports,

which might explain why

shifting the drive to the JUSB

port (USB 3.2 Gen1 according

to your mobo specs) got it

to behave properly, but that

assumes you hadn’t tried

rear-mounted USB 3.2 ports

while troubleshooting.

However, the fact you’ve

found a port you can use

may mean you’ll be able to

upgrade your hardware and

re-enable PCIe 4.0 at a later

date, although you may then

subsequently experience

Try disabling PCIe power-saving to resolve PCIe 4.0 glitches.

quickstart

14 MAXIMU MPC FEB 2022