AMD is investigating USB problems with 500 series chipsets

USB4 is exactly that. A lot of mobo makers also sell TB3 addin cards for AM4.

Man wtf

 

I know hahaha, in PCI_E SATA card I have plugged another 2 HDD's(2*3TB) and plus another 2*NVMe (512GB and 1TB), I'm glad I have Caselabs M8 with pedestal

PC is used for rendering most of the time and some gaming if time allows

Hope this helps

Thanks, Jura

 

"haven't noticed any USB disconnections "
" also occasionally get the Windows dink-donk USB unplugged sound but never noticed what it could be"

Same problem but only when my monitor goes to sleep and sometimes on steam or just click on capslock on my keyboard.

x570 Aorus Elite + 3900x
F32 Bios

 

Very weird? I wanna say 'Propand..' hmm.

As you said, 'Bizarre' ?

 

I've an Asrock B450 Steel Legends with Zen 3 support. Same problem here! Loosing data in pendrives or external HDs.

 

I only have one extremely weird quirk with the USB and that is that I use a HUB to connect my Keyboard (Corsair K30), mouse (wireless Logitech) and bluetooth dongle.
When I restart the PC. The keyboard won't work on the BIOS while the mouse does.
Unless I connect it directly to the motherboard the keyboard just won't even have power.

Other than that, everything is going good USB wise.

Mobo is a ROG STRIX B550-F BIOS 1805

 

Thank you for bringing this up! I hope that AMD takes this issue seriously and provides a solution.

I've had 3 different Ryzen 5950x CPUs and all of them had such issues. I don't see how this is something related to a bad CPU.

My last CPU is from AMD RMA, a "Silver sample" and also has this issue.

I always have to set PCIE x16 to 3.0 instead of 4.0 and disable global c state. Then all is ok but that's a nasty workaround!!

 

AMD can't fix this issue with base code, the affected cpu's must be returned to AMD for replacement.

 

I recently "lost" one of two front USB 3.0 on my B550 board .... not sure yet if it's the cable, the case or the board itself, but it was certainly working a few months ago, and I haven't touched the inside of the PC.
Unable to test if the fault is at the board or the case/cable unless I try another board, which I don't have.

One of the ports on the back also has random disconnections (very rare, but still happening).

Hopefully it doesn't get any worse

 

the hub resets because of consistent bus errors (remember those pesky PCIE:AER errors?)

 

Have you tried flashing to a previous bios?

 

I haven't looked at USB4, but, the disadvantage of USB progress has been that it offers no advantages in terms of latency, only bandwidth. Whereas audio interfaces have benefitted from thunderbolt with it's pci-e connection which offers the lowest latencies currently possible.

 

Asus has chipset drivers dated from November of 2020, released a few weeks ago, for their x570 Tuff Gaming WiFi, and presumably other motherboards, available, and I read one poster say that they fixed their USB port situation.

I might have just had the issue as a few hours ago I got around to bringing over all my USB drives from an older, now dead, build over to my new one. I had stuttering on them, remembered the news about the AMD USB situation, and installed/reinstalled the new chipset drivers. I recently did a repair installation of Windows, so maybe that overwrote the newer AMD drivers.

I can't swear to having had the problem, or to having applied a fix. I'm just recounting my experience. I have a feeling that this wasn't a comprehensive solution, but for me it's good enough for now.

https://www.asus.com/us/Motherboard...TUF-GAMING-X570-PLUS-WI-FI/HelpDesk_Download/

 

Hah.... my reaction exactly

"with it's pci-e connection which offers the lowest latencies currently possible"
^this

 

There's never "enough" bandwidth between CPU and Chipset. That's entire idea behind chipsets.
And number of lanes is always same, because they are dedicated for chipset.

 

Wrong.

 

I was just watching a video and heard the disconnect sound, I have no usb devices connected other than keyboard & mouse & the xbox one wireless thingy, went backwards in the video to relisten make sure it was not the video making the sound and it wasn't...

Checked event log to see exact timestamp of disconnect sound, nothing in application, in security it has 3 entries for "Audit Success" event id 4790, all 3 the same, not sure what that means.

Nothing in setup, nothing in system.

Weird.

 

can we get a picture of that ?
and I thought I was crazy with 8 drives and 10 usb populated

 

USB4 is literally the open iteration of TB3.

 

you won't see this in event log, you will need usbdeview or to look at the event tabs of devices themselves..