Discussion AGESA 1207 Beta BIOS Experience on AMD 300 Series and 400 Series Motherboards

[TechyInAZ]

Hey everyone! Since there is very limited data on AGESA 1207 reliability on 300 and 400 series boards, I thought i'd create this discussion piece to talk about potential issues from and great things from user expereince that have come about with AMD's latest agesa code working on 4 and 5-year-old boards.

If you have installed one of the brand new AGESA 1207 beta BIOS' on your 300 or 400 series motherboard, did it fix the fTPM stuttering issues and USB connectivity issues previously confirmed by AMD? Is it working well with your current CPU? If you did upgrade to Ryzen 5000, is it working well with that chip?

Here is my experience so far:

I installed AGESA 1207 through a beta BIOS just yesterday for my B450 Pro Carbon AC (the non max variant effectively), install process was smooth as butter. I put everything to defaults before upgrading, because I've found that can be optimal for some strange reason. Anyways, after upgrading I immediately plugged in all my settings back into the bios, including XMP and it worked great, and my memory timings remained identical as before when i was on AGESA 1006 from like 3 years ago -- even the subtimings remained the same between 1207 and 1006.

Booting into windows is where I got my first error however, right upon bootup I got a BSOD with an error I cannot remember. But on the flip side, upon the next reboot, it immediately fixed itself which I have no idea how or why that happened. My best guess is that Microsoft drivers responsible for the CPU, freaked out when going from a 3 year old AGESA code all the way to a brand new one released just a month ago. A couple of weird phenomena occurred as well, including my GPU driver settings being reset to default from the bios update and firefox glitching out in full screen mode (i fixed the latter with refreshing firefox). Just to be on the safe side, I reinstalled my CPU drivers and clean installed my GPU drivers to insure everything will run smoothly.

Besides this, everything has been rock solid. Boot times are faster, and I now have resizable bar in the BIOS and some more features with the new agesa code. Also, my Ryzen 5 3600 has apparently got a performance boost as well, gaining an extra 100 points in Cinebench R23 after a few consecutive runs vs my chip running on AGESA 1006.

I still need to test if I still have fTPM stuttering issues or USB issues, but it seems those are fixed for now.

 

[drea.drechsler]

...did it fix the fTPM stuttering issues and USB connectivity issues previously confirmed by AMD?...

We never experienced either of the issues on my son's B450/3700X system nor my B550/5800X system even before the update so can't say it helped since it was a non-problem before.

It installed smoothly for both of the systems, no problems booting into Windows after. But my practice is to un-install the chipset drivers before an update, do a CMOS reset, update the BIOS, do ANOTHER CMOS reset before booting into Windows on the new BIOS and then re-installing chipset drivers. That's always given me a smooth upate transition with no dramas.

Although...it does leave the major annoyance of re-entering all the BIOS setups for memory setup and whatnot. You can't reuse saved profiles, and probably shouldn't even if it let's you, so that's nothing really new. Memory setup in particular is tedious for me since I'm heavily overclocking memory rated at much lower clocks... even with XMP settings. So there are a lot of customized timings and for some reason it only goes well if I do it in an iterative process.

In your case, and as you suspected, things may have another complicating factor since I was updating (both) from relatively recent BIOS rev's. Your original BIOS was 3 years old so it wasn't compliant with Microsoft' latest requirements for Windows 11. That includes running UEFI mode (not CSM) as well as both FTPM and Secure Boot enabled by default. I imagine there are other things at work but if you didn't run those previously it may have played havoc with a first boot and until Windows configured itself for them.

 

[TechyInAZ]

I'm glad you didn't have issues drea! That's a great sign to see. I definitely understand the annoyance of going through memory timings and entering them one by one, since I used to have a full manual memory overclock on my previous memory kits and previous Ryzen configurations. I believe if you have a very extensive config, its best to enter some of them in groups then save and restart, since automatic sub-timings can cause issues if you don't.

Its worth mentioning that i did have UEFI mode enabled and FTPM and secure boot all enabled manually on the older BIOS because I did upgrade to windows 11.

 

[CountMike]

Newest BIOS beta 6042 with AGESA 1207 no problems with it, That stutter with TPM was apparently windows fault, since it was fixed in W11 although I never experienced it.
Restored custom BIOS settings from previous 6024 BIOS with AGESA1206 and everything's same as before.

 

[drea.drechsler]

....
Its worth mentioning that i did have UEFI mode enabled and FTPM and secure boot all enabled manually on the older BIOS because I did upgrade to windows 11.

That may not have been the source of your troubles then but I really intended it to be an example of how there can be a lot more going on with an update than we get clued in on.

It's a common issue and many people complain about the utter lack of release notes with BIOS updates. Even when they do provide some it's more often than not worthless babble like "improve compatibility" or "improve performance". The release notes for our boards did mention it included AGESA 1.2.0.7 but nothing about what it was supposed to 'fix' or 'help with'.

Also, many mfr's won't update across a wide span of BIOS revisions and require sequential or bridge updates to get to the latest. Possibly that's to preclude just the problem you experienced.

 

[TechyInAZ]

@drea.drechsler Hey I think I found out what happened. Look at Asus's official BIOS update recommendations in this post: View: https://www.reddit.com/r/Amd/comments/uhlo8t/asus_announces_uefi_bios_support_agesa_1207_for/


"Some updates will cause PCIe remapping and reinitialization of onboard controllers/devices. In these cases, you may need to reinstall drivers including your chipset drivers, graphics drivers or other PCIe or USB linked based devices. "

This sounds exactly like the problem I was having.

 

[drea.drechsler]

....
"Some updates will cause PCIe remapping and reinitialization of onboard controllers/devices. In these cases, you may need to reinstall drivers including your chipset drivers, graphics drivers or other PCIe or USB linked based devices. "

This sounds exactly like the problem I was having.

That's really very helpful information!

It might easily explain your case, and better yet explain why the updates in my case didn't show those issues since I always un-install first/ re-install after the chipset drivers. Also, the "in some cases" qualification suggests it could be the issue when bridging a wide gap in BIOS revision, which you had to and I didn't. This could also explain why some mfr's force intermediate revisions to bridge such a gap.

All in all, very helpful information to know.

One of our boards (the B550...with a 5800X cpu) was an Asus too, BTW.

 

[DSzymborski]

I have a Crosshair VIII Hero and experienced the issues with stuttering.

I upgraded the BIOS as soon as it was available and have had no issues with the occasional stutter since.

In my case, the update went fine; there were no problems applying the BIOS update nor were there persistent or new issues after the update was completed.

 

[TechyInAZ]

Glad everything going well DSzymborski!

Quick update, now it's day 3 since my BIOS update and basically, all the windows bugs are now gone. I did have a taskbar glitch out on me three times when i first enabled my 2nd monitor yesterday, but that has disappeared thanks to a reboot.

The important thing is that the actual stability of the hardware appears to be intact. So in a worst-case scenario, I'd just need to reinstall windows if the windows bugs were to get worse.

I will do a Prime95 test soon to verify stability though.

 

[TechyInAZ]

Update for day 5, ran Prime95 Blend with AVX and AVX 2 instructions enabled for 8.5 hours all night last night and system is perfectly stable with AGESA 1207.

This verify's that the BIOS and hardware itself is perfectly stable, and 1207's implementation in my B450 Pro Carbon beta BIOS is hardware stable.

 

[TechyInAZ]

Week 2 update, can fully confirm fTPM bugs have died with AGESA 1207 on my board. The system has been working absolutely flawlessly, and I can fully recommend using MSI's beta AGESA 1207 BIOS' if you either want to upgrade to Ryzen 5000 or have an older chip but need the stability improvements.

Just beware, that if you jump from a very old version, Windows will freak out. If that happens and you start getting glitches, you need to reinstall all system drivers including chipset, GPU, networking, wireless, and possibly audio drivers.

Next, you need to do a DISM health restore check on Windows 10 or 11 to make sure windows did not corrupt itself. Finally, I'd advise doing a stability test on the CPU and memory just to be safe.

It's still a beta BIOS, so there's still a chance very minor bugs could appear, but it seems that won't be the case.

 

[drea.drechsler]

....
It's still a beta BIOS, so there's still a chance very minor bugs could appear, but it seems that won't be the case.

For some reason the latest AM4 BIOS's are kept in BETA status across all the mfr's. They've been doing it for B4X0 and B3X0 boards for quite a while and only more lately for the B5X0 boards too.

 

[CountMike]

Switched to 5800x instead of 3700x few days ago. Bios 6042 is out of beta so I reflashed it. Run OCCT heavy test with BIOS at defaults (except XMP) for 6 hours, voltages and temps very decent (1,1v and 64c max). PBO is useless now but BIOS is of no help with Core Optimizer so tried with Ryzen Master with no good results. Previously with 3700x i had Firefox and only FF failing many times but no problems any more. In retrospective that apparently started with BIOS with AGESA 1006 and I was blaming FF for that.