BIOS Updates on Ryzen

[LunarMist]

So I updated to prepare for the ZEN 5, and now it is messed up. I have no clue which USB device this could be, but it must be casued by the new BIOS. What can I do?


Device Status

Windows has stopped this device because it has reported problems. (Code 43)

The SuperSpeed link to the USB device keeps going to an error state Compliance. If the device is removable, remove the device and then disable/enable from device manager to recover.

Device USB\VID_0000&PID_0008\b&23bb30c0&0&3 was not migrated due to partial or ambiguous match.

Last Device Instance Id: PCI\VEN_1022&DEV_149C&SUBSYS_148C1022&REV_00\6&2ad73c09&0&0340000A
Class Guid: {36fc9e60-c465-11cf-8056-444553540000}
Location Path:
Migration Rank: 0xF000FFFFFFFFF122
Present: false
Status: 0xC0000719

 

[Mercutio]

You might want to move your USB devices around to different ports. Start with the keyboard and mouse on a non-superspeed port. The device in question could be something internal and non-removable. If you have nothing installed and still have a problem, it's something on the motherboard and you just have to wait for a fix. It could also be something funky like USB C on a video card, but for now, direct-attach what you need one device at a time and see what happens.

My threadripper gives non-stop USB errors if I connect a Logitech unifying adapter to a Superspeed port even though the devices clearly work fine.

 

[LunarMist]

It must be the BIOS since nothing else changed and I did not even crack the case at the time.

 

[Mercutio]

It might still be helpful to narrow down which ports are impacted. Maybe there's a virtual device (TPM?) causing the problem.

 

[LunarMist]

The USB cables are dangling or devices are powered off. I do have a Logitech wireless mouse. Keyboards are wired.

 

[LunarMist]

Security devices: AMD PSP 10.0 Device
I removed the TPM function via BIOS to defend against the Win 11.

I am using the most recent AMI Beta BIOS, with AGESA ComboPI 1.2.0.0a. Who knows what that means, but it works now.
Will there be a new BIOS when the final CPUs are released?

 

[sedrosken]

Do AM5 Zen boards still have the issue where some manufacturers skimped on the flash for the firmware so their solutions ended up being making people with older CPUs stay stuck on the older versions of the UEFI? I know my B450M-Pro4 was stuck on a years-old revision for a long, long time because the newest versions didn't have microcode for the older Zen+ chips baked in.

 

[LunarMist]

It has a NUVOTON with 256Mb, whatever that means. The BIOS file size is exactly 32MB, so it may be that they are using all of it. All of the unpacked BIOS files since I bought it are the same size. Just about all X670E or B650 boards are claiming 9000 series support, so the differences from 7000 series must be minor.

 

[LunarMist]

How long does a BIOS remain a BETA and do you continue to use BETAs months later?
I wonder if they are waiting for the retail 9900X and 9950X to drop or perhaps there a certain timeframe.

 

[Mercutio]

Do AM5 Zen boards still have the issue where some manufacturers skimped on the flash for the firmware so their solutions ended up being making people with older CPUs stay stuck on the older versions of the UEFI? I know my B450M-Pro4 was stuck on a years-old revision for a long, long time because the newest versions didn't have microcode for the older Zen+ chips baked in.

I've only found boards where pre-Zen 2 CPU compatibility is dumped in favor of getting Zen 3/3+. Which is fine. I wasn't planning to use an Athlon 3000 for anything, anyway.

 

[LunarMist]

How long does a BIOS remain a BETA and do you continue to use BETAs months later?
I wonder if they are waiting for the retail 9900X and 9950X to drop or perhaps there a certain timeframe.

Another BIOS was released today and it is the final (non-BETA) version of that sequence.

 

[LunarMist]

I discovered that there were BIOS updates for all my Ryzen mainboards. They already included the CV fix for the low level microcode to prevent potential hackers. As far as I can determine there is no performance impact and all my testing was done after they were applied.

 

[Mercutio]

Something I've learned about Ryzen Flashback updates is that motherboards are very picky about what drives can be used for them. Drives that are too big don't work and drives that are too slow don't work. Since I don't have any thumb drives smaller than 32GB any longer, I wound up using an 8GB microsd card to update my Asrock x670e Lightning, and I had to try a couple before I found one that worked. The winner was an old Lexar card in a Sandisk microsd reader.

 

[LunarMist]

I update the MSI BISO with a 32GB WD Easystore USB flash drive in any mode. I don't have the Ryzen Flashback of Asus.
Has AMD mitigated the 9900X and 9950X defects with the new BIOSes?

 

[Mercutio]

There was an AGESA update that supposedly caused the inter-CCX latencies to be reported correctly. There's also a Windows 11 update that improves performance by 3 - 5% but so far the new Ryzen 9s haven't had any special attention.

 

[LunarMist]

I found this recent update:
AMI BIOS 2024-09-24 (Beta version)

Description:
- AGESA PI 1.2.0.2 released for trial use.
- Improved CCD delay and gaining better performance for Ryzen 9000-series.

Is this what you mean? I'm not so excited about a Beta BIOS.

 

[Mercutio]

That is exactly what I mean.