PC Drivers & Software

[williwilli01]

I believed a lot of people already know this issue but AMD still don't have any fix on the issue, which causing a lot of pcs' basically unusable as most or even all of the USB ports are not working, both internal or external are affected. 

For my issue, I RMAed a TUF B550M-Plus and the new one has the same issue. Tried every BIOS update and they did not fix it. Tried disabling C-state, forcing it to run in PCIE Gen 3 instead of Gen 4 also did not fix. All of the possible fix listed on reddit (https://www.reddit.com/r/Amd/comments/kbr6zl/usb_issues_on_b550x570_motherboards/) did not help at all.

I'm wondering whether anyone in this community has any recommendation on how to fix the issue as I'm running out of ideas now. 

If the issue still exist and AMD don't have a fix for it at the end of the January, I'll tell every customer to avoid AMD products at all cost.

[eagle12426]

hello, I'm about to do a build. About the 5000. Is what your talking about is it also the same with the 9 3950x? Are they effected also with the X570 MOBO?

[williwilli01]

---

@eagle12426 wrote:

hello, I'm about to do a build. About the 5000. Is what your talking about is it also the same with the 9 3950x? Are they effected also with the X570 MOBO?

---

This issue only appears on 5000 series so far (I've build around 100 pcs with AMD since the 5000 series launch) so I believe the issue only happens with 5000 series and B550/X570. However, the usb ports did happen when AMD first launch the 3000 series

[cww1]

Hi there,

i have experienced the same issue with my Ryzen 5 3400Ge Rog Strix B450-F Gaming II with latest BIOS 4201 on both OS Windows 10 Pro and Linux Mint 20.1 (Dual Boot)  especially using a Terratec USB DVB-C TV stick Since I unplugged it, the system went stable again. I used the stick in several other machines like  notebooks Dell Vostro 1000 with Athlon 64 and a HP Probook 6555b Tripple Core and a GA 870 UD3 with  DualCore. They were stable. The channel scan and the  H264 decoding used a lot resources but they used to stay stable.

I hope the fix this bug soon, either via BIOS update or replacement of the core on warranty,

yours sincerly

[goodplay]

Have you tried disabling 'usb selective suspend' ?

[williwilli01]

---

@goodplay wrote:

Have you tried disabling 'usb selective suspend' ?

---

Tried that one as well. I believe the issue is coming from the usb controller inside the CPU (it was built-in to the chipset in previous series of CPU, but this time they put it into the CPU for some reason, don't know what the heck they were thinking), and the usb controller is made from another company, might have a QC issue there 

[td47]

That is an interesting observation. Since the only device I get issues with in my Gigabyte B550M (with AMD 5600G CPU) is an external USB3 LaCie 4TB backup drive, is it worth me getting an internal USB3 controller card? Will that take the issue away from the clock/timing/bus issues in the 5600G USB internals, and make the USB3 connection and use stable again?

I was thinking of getting an add-in USB3.2 controller PCIE card, cold you recommend one?

By the way, is the USB selective suspend here below, or in the BIOS?

USB Power MGMT

[Amioran]

I opened the same exact thread myself pointing out this huge issue (that I believe at this point to be common knowledge) but in the "processor" section. For now  there's no solution to the issue of the USB ports not working correctly with Zen 3, we just have to wait for a (hopefully soon) solution from AMD, much probably in the form of a driver update for the chipset. 

[StephenG]

YES! Almost certainly this is an issue of the chipset driver. When I updated the chipset driver, it cured the problem (for me, at least - I obviously can't guarantee that it will work for everyone).

[xlox]

what's the link between usb and cpu ? you bought 100 motherboard of the same brand ?  maybe check asus too ? no ? bios update etc ?

[Amioran]

All brands show the same issue, and yes, there's a clear connection between USB ports and the CPU (in fact in many cases when you bend a pin on the CPU the first thing that starts working badly are the USB ports). For example by disabling the C-States in the CPU some of the issues (manly to do with the ports not working correctly anymore after being suspended) are resolved, but sadly those are only some of the problems the USB HUBs in the Motherboards have.

[xlox]

i have 5950x and x570 from asus, no usb problem detected at all since i have them (and had a 3900x on that board) ... so your issue doesn't seem related to cpu neither chipset,  in my view ... did you contact asus support or did you come directly to amd forum?

[Amioran]

It depends with which peripherals you use those ports (or how you use them); if they don't require full USB 3.x specs/power then you will probably not notice any problem in the majority of circumstances (for example I have no issues with my mouse and KB, or even my HOTAS or USB Wheel), however once you have something that does then you will have problems. Try for example attaching an external powered USB 3.0 Hub to a port with C-States enabled on your CPU and then you will see that after Windows puts the latter on a C-State the HUB will stop working completely and you will need a full power cycle for it to work again. This is just a single example, but there are many others (one of the most common is using the latest VR headset of HP that requires full power/specs from a 3.0 USB port and that, for this, has issues for EVERYONE that has a B550/x570 mobo with a Zen 3 CPU).

Just because it doesn't happen to you given that you probably don't use the ports you have to their fullest it doesn't mean the problem is not there and, btw, just a little search on Google would have dispelled your doubts on  the matter completely. I repeat: the issue happens with ALL brands and it is not specific to Gigabyte, Asus or whatever; it is a problem that has to do specifically with the CPU/Chipset and much probably AMD is already full aware of this (and, in fact, the latest Agesa BIOS are all targeted at USB ports issues, but they still have to resolve the matter fully).

[williwilli01]

In my case, it happens to almost all of my peripherals, including keyboard and mouse. The worst is VR headset, it will continuously crashing the headset that I'm afraid it will eventually kill it. Luckily, the pc is not mine but it's my friend's, but I still regretted to recommend him to go to AMD route, basically built a new pc that is unusable because of the stupid USB problem

[xlox]

..... i use several peripherals....

1 racing wheel , one 720p camera , one webcam

and have 3 external hard drive 3.1 (with 2 ssd), bluetooth dongle , 1 usb hub with 2 phone charging + usb mouse + 32Go usb key now + corsair RM1000i connected to usb + waterblock from corsair connected to usb ... so ....

 , no issue at all ( but no VR as well)  ...sorry for not searching over internet about problemes i don't have lol

in my view your issue doesn't seem to be a rule ... and VR headset of hp , is some precise case ...

mmh read a little of the reddit , seriously the guy has no idea where his problem is coming from, i confirm i don't have it  ... seriously i would have contacted asus first if i were you , that's my advice ... if the problem is as widespread that you say they surely have an answer for you

(and had 3 ryzen processor on that board 3700x , 3900x and now 5950x so clearly i can't believe it's a cpu problem ..)

[Amioran]

Ok, you clearly dont want to understand, and I really don't know how to tell you the matter more clearly that I already did. I don't care minimally if it doesn't  happen to you (I'm happy for you but anecdotal proof is a little different from empirical one), but before replying to a post insisting the issue is not there common sense should have told you to at least research a little on the matter before pretending to be an expert on the topic, is it not?

There are countless reports from countless people all with different Mobo brands all around the web on the matter, and everyone that uses peripherals that require FULL 3.x USB power/specifications (as the aforemtioned HP Reverb G2) encounter the same exact issues as everyone else (and if you want to try it out for yourself I even told you how you can replicate the problem with a simple externally powered USB 3.0 HUB for yourself, it's not so difficult). It can happen also with normal USB peripherals but the latest Agesa BIOS has resolved for many people some of the problems on the latter (especially concerning 2.x USB ports) but the full range of the problem is still there.

[xlox]

when people try to make rules , sometime maybe it can help to sort things out to see it's not exactly how they think it is , but clearly as i don't want to understand, i ve got nothing to do here anymore lol ! bye !

maybe YOU should read the reddit too , and maybe the answers to this reddit lol , not just these answers going your way ...that's good info too

[Amioran]

You aren't "sorting" anything out. You dont' even have a single peripheral that requires full 3.x USB standards, nor did you care to research on the matter minimally for yourself. The only thing you have proposed till now is "but it doesn't happen to me, hence it doesn't happen at all" that's the most ludicrous statement you could ever propose in rebuttal to a post highlighting a widespread problem, especially when your specific situation doesn't even cover the specific of the issue that covers ALL people involved.

[cww1]

Hi there,

i have experienced the same issue with my Ryzen 5 3400Ge Rog Strix B450-F Gaming II with latest BIOS 4201 on both OS Windows 10 Pro and Linux Mint 20.1 (Dual Boot)  especially using a Terratec USB DVB-C TV stick Since I unplugged it, the system went stable again. I used the stick in several other machines like  notebooks Dell Vostro 1000 with Athlon 64 and a HP Probook 6555b Tripple Core and a GA 870 UD3 with  DualCore. They were stable. The channel scan and the  H264 decoding used a lot resources but they used to stay stable.

I hope the fix this bug soon, either via BIOS update or replacement of the core on warranty,

yours sincerly

[StephenG]

I built an AMD PC in January and had a lot of problems at first with my mouse and keyboard, both connected through USB 2 ports on the rear I/O panel. These problems were greatly reduced when I updated the chipset driver. I think updating the UEFI BIOS has also helped. One thing I have done which significantly improved the stability of my system was to turn off the Windows 10 "fast boot" option (which is enabled by default). My boot drive is an NVMe M.2 SSD PCIe gen. 4 and is very fast. Turning off the fast boot option has not noticeably increased boot time, but it has definitely improved the overall stability of the system.

My build: Asus ROG Strix B550-XE gaming wifi, Ryzen 9 5900X, Scythe Mugen 5 ARGB Plus cooler, XFX MERC 319 Radeon RX 6800, 32 GB Corsair Vengeance RAM (4 x 8 GB, 3200 MHz), Silicon Power 1 TB NVMe SSD (PCIe gen. 4 x 4), Intel 2 TB 660p NVMe SSD (PCIe gen. 3), Corsair RM750 PSU, Enermax Saberay adv. case (with all preinstalled fans + 2 extra 120 mm case fans).

[Ray_AMD]

We would like to thank the community for its assistance with logs and reports as we investigated the intermittent USB connectivity you highlighted. With your help, we believe we have isolated the root cause and developed a solution that addresses a range of reported symptoms, including (but not limited to): USB port dropout, USB 2.0 audio crackling (e.g., DAC/AMP combos), and USB/PCIe Gen 4 exclusion.

AMD has prepared AGESA 1.2.0.2 to deploy this update, and we plan to distribute 1.2.0.2 to our motherboard partners for integration in about a week. Customers can expect downloadable BIOSes containing AGESA 1.2.0.2 to begin with beta updates in early April. The exact update schedule for your system will depend on the test and implementation schedule for your vendor and specific motherboard model. If you continue to experience intermittent USB connectivity issues after updating your system to AGESA 1.2.0.2, we encourage you to download the standalone AMD Bug Report Tool and open a ticket with AMD Customer Support.

Information on any further updates to this issue can be found on the following thread.

[td47]

Hello,

I have a Gigabyte B550M DSH3 (V1), running well with an AMD 5600G CPU.

However, even with the latest Chipset drivers, and the F19 BIOS, I still get issues with trying to use a USB3 LaCie 4TB drive directly connected to a USB 2 or 3 port. I forgot to unplug this external drive the other day, and after a fresh reboot, my system went really slow, and was behaving oddly. It would not shutdown, and when I unplugged the 4TB drive, it completed the shutdown, and was stable again.

With earlier BIOS and chip-set drivers, I had the BIOS splash screen just hang, until I unplugged the USB3 Lacie drive.

I am surprised that with USB3 being around since 2008, we are STILL getting issues like this!

That specific drive has been used prior to my current system, for many months, on an old ASROCK Z68 based desktop, and it always worked flawlessly, so I do NOT suspect any issue with the drive itself.

Having experienced this issue a few times, my  "gut feel"  is that the BIOS/UEFI initialization has timing issues with various routines being interrupted, and the USB INIT for USB3 is leaving something in a bad state, possibly due to a race-condition, and not completing properly.

Any advice appreciated.

[doehlmann]

Agree. I have a amd ryzen 7 5800X on a MPG X570 Board with Wifi. Ive got to say i need to start my computer with a hard reset and take out all my USB peripherals. I recently purchased a HP Reverb G2 VR headset and since it run on USB-C this is not usuable. I will be buying a new motherboard tomorrow to fix it. I should say that in this current age this is somewhat frustrating and I would think not very proffesional

[delevero]

I did read all of the above comments so sorry if its not 100% spot on.

However i like to say that i read somewhere today that AMD have a fix for the unstable USB's.. Its in the upcomming AGESA driver
and it will be send out to the motherboard firms in about a week from now and the first people can try it as beta in start of april month.. So i would say expect the fix to arrive after 14th april.

[cabits]

Sure hope its an easy fix. I have an ASUS laptop and a Ryzen 5-5600x on a Tuf Gaming X570PRO board that neither USB-C port work with devices like an iPhone 12 but can see and transfer data with an SSD.

Very frustrated..

[computeur]

AGESA V2 PI 1.2.0.1 is in bios but nothing change about instabillity

Asus b450 Tuf Gaming plus / Ryzen 3600

- Asus Rog Horus GK2000 v2 RGB keyboard 

I can't use my expensive keybord on my middle moba game.

Yesterday I ordered a PCI card with 4 USB3 ports, hope this will fix the problem. Because neither Asus nor AMD is trying to help me (or pretend).

As I wrote on reddit, my soon build will be an intel

[td47]

Hello,

I have a Gigabyte B550M DSH3 (V1), running well with an AMD 5600G CPU.

However, even with the latest Chipset drivers, and the F19 BIOS, I still get issues with trying to use a USB3 LaCie 4TB drive directly connected to a USB 2 or 3 port. I forgot to unplug mine the other day, and after a fresh reboot, my system went really slow, and was behaving oddly. It would not shutdown, and when I unplugged the 4TB drive, it completed the shutdown, and was stable again.

I am surprised that with USB3 being around since 2008, we are STILL getting issues like this!

I have an old TRUST USB2 mains  powered hub, but I think I may have had issues using that (I need to do some more testing).

Quick question: What Brand and model of USB3 PCI card did you try, and was it successful?

I have one in my previous Desktop PC now mothballed, so if it is similar to yours, I might give it a go.

[RayFroehlich]

FWIW, I seem to be another data point of a user still seeing this issue after having spent hours and hours trying *all* the aforementioned workarounds (disabling C-state et. al.) and fixes (i.e., BIOS update including AGESA).  My specs:

• Motherboard: B650 Auros Elite AX
• BIOS version: F31
• CPU: AMD Ryzen 5 7600X

[td47]

Thanks for your comments. I just noticed the other day, a newer AMD Chip-set driver suggested by the GCC (6.08.22.137), so I installed it. I tested a backup and a Create System Image to the 4TB (USB3) drive, and that worked OK, then did the safe removal.

I have NOT tested shutting the desktop PC down WITH it plugged in, then rebooting, as I have not had time yet. I will update this  thread if it proves any better, but my gut-feel is that it is a BIOS timing issue.

[td47]

UPDATE ON MY POST ABOVE:
I have recently applied the latest AMD Chipset Driver 6.08.22.137 to my desktop system, and done some testing, just using my single external USB3 device, a LaCie 4TB backup drive.

I left it connected for 3 reboot cycles over 3 days, and on the third reboot from cold, the system went unstable, and even crashed File Explorer.
I unplugged the drive, and the system started working OK, and so I rebooted again from cold.

I noticed that 4th reboot stayed MUCH longer in BIOS initialisation than is usually does (about 12 seconds rather then the usual 5 seconds).

This leads me to believe that over time, some stored USB parameters and internal registers/pointers are getting corrupted, either in the BIOS/UEFI or in the USB Controller section of my AMD 5600G CPU.

So, either this is a hardware design issue, or there are still serious timing/functional issues in the Chipset Driver and CPU Internal Controller chiplet.

I am going to get a PCIE USB3.2 expansion card to see if this takes the issue away from the CPU internal controller. Can anyone recommend a good brand?