RAM Issues depending on AGESA version (1.2.0.5 to 1.2.0.7)

[GoLLuM13]

Hello,
I would like to report an issue that started with AGESA 1.2.0.5 and continues with AGESA 1.2.0.6 and 1.2.0.7 for Ryzen 3000 CPU owners.
Currently, I have BIOS 3801 based on AGESA 1.2.0.3 and SMU version for Matisse (Ryzen 3000) processors is 46.71.0.
On this BIOS version everything works as it should (apart from the fTPM problem with Windows 11 which causes a stutter supposed to be fixed with AGESA 1.2.0.7).
Since BIOS version 3904 (i.e. 3904 - 4002 - 4006 and 4201) the AGESA has gone from 1.2.0.3 to 1.2.0.5 then 1.2.0.6 and finally 1.2.0.7 and the SMU for the Matisse CPU (Ryzen 3000) is increased from 46.71. 0 to 46.72.0 and from this version (AGESA and/or SMU) it is almost impossible to start or restart the computer without Q-Code error (FA mostly), black screen, BSOD, as soon as the DOCP profile is activated or the memory frequency is set manually above 2733 Mhz. The only way to make the BIOS based on these AGESA versions work is to leave the RAM frequency on "default" which is not ideal because in my case for example it means 2133 Mhz or set it manually to a max of 2733 Mhz.
Of course, reverting to an earlier BIOS like the 3801 (which I'm currently using) fixes the issues instantly.
I did some memtests and other tests, like increasing or decreasing vcore, RAM voltage... etc the problem is not with my hardware. I also tried several times to fresh install Windows 11 on the latest BIOS version and couldn't install it, I had Q-Codes, a message of "corrupted files"..., and the second I flashback the 3801 (AGESA 1.2.0.3) BIOS it installs Windows flawlessly on the first attempt.
In all my tests the second I flash a BIOS on AGESA 1.2.0.5 or later problems starts, and as soon I flash back BIOS 3801 (AGESA 1.2.0.3) everything goes back to normal.

I researched various forums (including Asus and MSI) and found that this is a common problem among owners of Ryzen 3000 processors and an X570 motherboard or B550. I even saw persons having the same problem, who "solved" the problem by changing their Ryzen 3000 processor to a Ryzen 5000.

PS: I have already contacted Asus France support, they have already offered me several solutions, none have worked, and none will work as long as it is a problem within the AGESA.
I also contacted AMD France support they refused to escalate it to a competent team, and they refused to do anything about it, for them, if it works at 2133 Mhz or 2666 Mhz there's no problem and everything's fine, which is clearly not

 

The Build :

Asus RoG Crosshair VIII Hero (Wi-Fi) - AMD Ryzen 9 3900x - 4* Corsair Vengeance RGB Pro 8 GB 3000 C15 - Noctua NH-D15S Chromax.Black - nVidia RTX 3090 Founders Edition - Corsair AX1500i - Lian Li O11 XL RoG Certified

 

EDIT :

This looks like a confirmation to me : https://www.reddit.com/r/Amd/comments/uquyu1/comment/i92su02/?utm_source=reddit&utm_medium=web2x&context=3

[Guest]

It is your hardware.

 

The cpu memory controller hates running 4 sticks of cheap Vengeance memory with cheap SK Hynix modules.

[Levent]

I swear those Corsair Vengeance rams are the bane of a lot peoples existence. Nowadays every time I see a ram issue, its always with corsair rams. I got a B450 Aorus M and a B450M S2H V2 and both of them are running just fine on 1.2.0.7 using Kingston rams with Micron and Hynix DRAM chips. I wouldnt be surprised if your issue was due to those RAMs instead of AGESA version.

[GoLLuM13]

8 minutes ago, Levent said:

I swear those Corsair Vengeance rams are the bane of a lot peoples existence. Nowadays every time I see a ram issue, its always with corsair rams. I got a B450 Aorus M and a B450M S2H V2 and both of them are running just fine on 1.2.0.7 using Kingston rams with Micron and Hynix DRAM chips. I wouldnt be surprised if your issue was due to those RAMs instead of AGESA version.

9 minutes ago, Guest 5150 said:

It is your hardware.

 

The cpu memory controller hates running 4 sticks of cheap Vengeance memory with cheap SK Hynix modules.

It is not a hardware / RAM issue as you're saying, like I said it works perfectly fine up to BIOS 3801, and it still works on this version flawlessly

Problems are only with BIOS 3904 and later (simple and logical to know what's the problem here)

@Leventwhich CPU do you have ?

[191x7]

Are all 4 models the same RAM (the same OEM with same parts and latencies)?

If not, try switching the pairs in the slot pairs.

[DoctorNick]

3 minutes ago, GoLLuM13 said:

It is not a hardware / RAM issue as you're saying, like I said it works perfectly fine up to BIOS 3801, and it still works on this version flawlessly

Problems are only with BIOS 3904 and later (simple and logical to know what's the problem here)

@Leventwhich CPU do you have ?

Sure it isn't just Asus and there weird BIOS versions? I have had issues on both my recent rog board. Basically always run very old version which allow for a stable OC. Was the same on my Crosshair VII Hero, same on Maximus XI hero. If you havent disable ASUS armoury crate in bios and uninstall it using Armoury crate uninstallation tool: https://www.asus.com/supportonly/Armoury Crate/HelpDesk_Download/ 

[Levent]

13 minutes ago, GoLLuM13 said:

It is not a hardware / RAM issue as you're saying, like I said it works perfectly fine up to BIOS 3801, and it still works on this version flawlessly

Problems are only with BIOS 3904 and later (simple and logical to know what's the problem here)

@Leventwhich CPU do you have ?

3100 and 3600. Considering the number of Corsair ram related issues I saw over here, I would definitely consider rams being the problem. I am not saying rams are faulty, I am just saying they are not great for Ryzen CPUs.

[GoLLuM13]

12 minutes ago, 191x7 said:

Are all 4 models the same RAM (the same OEM with same parts and latencies)?

If not, try switching the pairs in the slot pairs.

They're all 4 part of the same kit and came in the same box, can't do better

8 minutes ago, DoctorNick said:

Sure it isn't just Asus and there weird BIOS versions? I have had issues on both my recent rog board. Basically always run very old version which allow for a stable OC. Was the same on my Crosshair VII Hero, same on Maximus XI hero. If you havent disable ASUS armoury crate in bios and uninstall it using Armoury crate uninstallation tool: https://www.asus.com/supportonly/Armoury Crate/HelpDesk_Download/ 

It's not only Asus, I saw the same with MSI boards, and I doubt it's Armoury Crate (as buggy as it can be), I couldn't even install Windows with these recent BIOS versions (AGESA 1.2.0.5 and later) while everything went fine with BIOS 3801 (AGESA 1.2.0.3)

[herhey]

36 minutes ago, GoLLuM13 said:

Hello,
I would like to report an issue that started with AGESA 1.2.0.5 and continues with AGESA 1.2.0.6 and 1.2.0.7 for Ryzen 3000 CPU owners.
Currently, I have BIOS 3801 based on AGESA 1.2.0.3 and SMU version for Matisse (Ryzen 3000) processors is 46.71.0.
On this BIOS version everything works as it should (apart from the fTPM problem with Windows 11 which causes a stutter supposed to be fixed with AGESA 1.2.0.7).
Since BIOS version 3904 (i.e. 3904 - 4002 - 4006 and 4201) the AGESA has gone from 1.2.0.3 to 1.2.0.5 then 1.2.0.6 and finally 1.2.0.7 and the SMU for the Matisse CPU (Ryzen 3000) is increased from 46.71. 0 to 46.72.0 and from this version (AGESA and/or SMU) it is almost impossible to start or restart the computer without Q-Code error (FA mostly), black screen, BSOD, as soon as the DOCP profile is activated or the memory frequency is set manually above 2733 Mhz. The only way to make the BIOS based on these AGESA versions work is to leave the RAM frequency on "default" which is not ideal because in my case for example it means 2133 Mhz or set it manually to a max of 2733 Mhz.
Of course, reverting to an earlier BIOS like the 3801 (which I'm currently using) fixes the issues instantly.
I did some memtests and other tests, like trying with 1 or 2 or 4 sticks (the same), increasing or decreasing vcore... etc the problem is not with my hardware. I also tried several times to fresh install Windows 11 on the latest BIOS version and couldn't install it, I had Q-Codes, a message of "corrupted files"..., and the second I flashback the 3801 (AGESA 1.2.0.3) BIOS it installs Windows flawlessly on the first attempt.
In all my tests the second I flash a BIOS on AGESA 1.2.0.5 or later problems starts, and as soon I flash back BIOS 3801 (AGESA 1.2.0.3) everything goes back to normal. 

I researched various forums (including Asus and MSI) and found that this is a common problem among owners of Ryzen 3000 processors and an X570 motherboard or B550. I even saw persons having the same problem, who "solved" the problem by changing their Ryzen 3000 processor to a Ryzen 5000.

PS: I have already contacted Asus France support, they have already offered me several solutions, none have worked, and none will work as long as it is a problem within the AGESA.
I also contacted AMD France support they refused to escalate it to a competent team, and they refused to do anything about it, for them, if it works at 2133 Mhz or 2666 Mhz there's no problem and everything's fine, which is clearly not

 

The Build :

Asus RoG Crosshair VIII Hero (Wi-Fi) - AMD Ryzen 9 3900x - 4* Corsair Vengeance RGB Pro 8 GB 3000 C15 - Noctua NH-D15S Chromax.Black - nVidia RTX 3090 Founders Edition - Corsair AX1500i - Lian Li O11 XL RoG Certified

I Have the exact same issues and seen on other forums people experiencing the same problems. My specs are:

 

- Asus ROG Crosshair VIII Dark Hero

- Ryzen 7 3700x

- 32GB (4x8) Thermaltake Toughram Z-One RGB (3200mhz)

[DoctorNick]

1 minute ago, GoLLuM13 said:

They're all 4 part of the same kit and came in the same box, can't do better

It's not only Asus, I saw the same with MSI boards, and I doubt it's Armoury Crate (as buggy as it can be), I couldn't even install Windows with these recent BIOS versions (AGESA 1.2.0.5 and later) while everything went fine with BIOS 3801 (AGESA 1.2.0.3)

Yeah must be an AGESA issue then. Why not just run 3801 if it's stable? You can try another version later?

[GoLLuM13]

2 minutes ago, Levent said:

3100 and 3600. Considering the number of Corsair ram related issues I saw over here, I would definitely consider rams being the problem. I am not saying rams are faulty, I am just saying they are not great for Ryzen CPUs.

The thing is that they worked and still work flawlessly since 2019 and the first BIOS version of this motherboard and up to BIOS 3801 included. They're even in the QVL list for this motherboard, problems only occur with 3904 and later, you gotta admit that's weird

[GoLLuM13]

1 minute ago, DoctorNick said:

Yeah must be an AGESA issue then. Why not just run 3801 if it's stable? You can try another version later?

I'm on Windows 11 and there's a stutter issue caused by the fTPM on BIOS based on an AGESA version older than 1.2.0.7 (like mentioned in my first message)

[herhey]

27 minutes ago, 191x7 said:

Are all 4 models the same RAM (the same OEM with same parts and latencies)?

If not, try switching the pairs in the slot pairs.

I have the exact same issues.

 

• Asus ROG Crosshair VIII Dark Hero
• Ryzen 7 3700x
• 32GB (4x8) Thermaltake Toughram Z-One RGB (3200mhz)

[renesdk]

I can chime in on this.

I'm on an MSI X570-A PRO with a AMD Ryzen 7 3700X and this is my exact experience also.
I've had my system since July 2019 when the X570 was released.

• On the AGESA 1.2.0.3 (and earlier) it is completely stable.
• On AGESA 1.2.0.5 it started doing weird BSOD's at random times, with or without XMP.
• On AGESA 1.2.0.6c it seems to be running fine. I could restart the computer no problem, however when I shut down the PC and powered it back on, I would get a BSOD during boot to do a repair of the Windows install, after every single power off.

Going back to AGESA 1.2.0.3 fixed it all and everything was completely stable.

All drivers have been updated to the latest drivers from AMD.

I'm running  a 32 GB Corsair Vengeance DDR-3200 kit (4x8 GB) and it's a supported model by MSI on their supported list, and have been since the release of the board.

I am experiencing the issues both with and without XMP enabled.

I've tested the RAM and there are no hardware errors.

[renesdk]

2 hours ago, DoctorNick said:

Yeah must be an AGESA issue then. Why not just run 3801 if it's stable? You can try another version later?

The issue is that there are micro stutters with TPM enabled under Windows 11 on most AMD systems with Ryzen CPU's on anything less than AGESA 1.2.0.7, which is actually programmed to fix the stutters.
So we are forced to either downgrade to Windows 10, or accept to not be able to fully utilize the performance of our systems with AGESA 1.2.0.7 under Windows 11.

 

I personally like Windows 11 and have since it launched, and I have no plans to downgrade, and with how AMD is acting regarding this I'll just have to accept to loose performance on something that has worked fine up until now.

[191x7]

Why not use Memory Taiphonn and Ryzen DRAM Calculator to set the timings manually?

[GoLLuM13]

2 hours ago, herhey said:

I have the exact same issues.

 

• Asus ROG Crosshair VIII Dark Hero
• Ryzen 7 3700x
• 32GB (4x8) Thermaltake Toughram Z-One RGB (3200mhz)

 

2 hours ago, renesdk said:

I can chime in on this.

I'm on an MSI X570-A PRO with a AMD Ryzen 7 3700X and this is my exact experience also.
I've had my system since July 2019 when the X570 was released.

• On the AGESA 1.2.0.3 (and earlier) it is completely stable.
• On AGESA 1.2.0.5 it started doing weird BSOD's at random times, with or without XMP.
• On AGESA 1.2.0.6c it seems to be running fine. I could restart the computer no problem, however when I shut down the PC and powered it back on, I would get a BSOD during boot to do a repair of the Windows install, after every single power off.

Going back to AGESA 1.2.0.3 fixed it all and everything was completely stable.

All drivers have been updated to the latest drivers from AMD.

I'm running a 32 GB Corsair Vengeance DDR-3200 kit (4x8 GB) and it's a supported model by MSI on their supported list, and have been since the release of the board.

I am experiencing the issues both with and without XMP enabled.

I've tested the RAM and there are no hardware errors.

 

9 minutes ago, renesdk said:

The issue is that there are micro stutters with TPM enabled under Windows 11 on most AMD systems with Ryzen CPU's on anything less than AGESA 1.2.0.7, which is actually programmed to fix the stutters.
So we are forced to either downgrade to Windows 10, or accept to not be able to fully utilize the performance of our systems with AGESA 1.2.0.7 under Windows 11.

 

I personally like Windows 11 and have since it launched, and I have no plans to downgrade, and with how AMD is acting regarding this I'll just have to accept to loose performance on something that has worked fine up until now.

Thank you both for your feedback, much appreciated.

I'd like to introduce this screenshot that probably explains better the SMU "story" I was talking about on the first message

 

[GoLLuM13]

3 minutes ago, 191x7 said:

Why not use Memory Taiphonn and Ryzen DRAM Calculator to set the timings manually?

Because even though I set memory frequency to 3000 Mhz (my ram frequency speed) with auto timings or something equivalent to CL30 and a high enough voltage (for example) it doesn't work on these specific versions of BIOS it just refuses to work above 2733 Mhz whatever I do.
But as soon as I go back to 3801 everything works fine @ 3000 CL15 1.35v, I can even OC RAM (did it for tests only) I tried 3200 CL16 and 3600 CL18 effortlessly, and it worked (and as you said I used "Thaiphoon" and "DRAM Calculator for Ryzen" to do so)

[renesdk]

This is my SMU versions on the 3 BIOS'es released for my board

 

[GoLLuM13]

42 minutes ago, renesdk said:

This is my SMU versions on the 3 BIOS'es released for my board

 

Thanks I like this interface better, it's easier to read so I upload again mine so it will be easier to compare (it doesn't detect AGESA version so I added it manually)

 

It appears to me that everything's the same except SMU version on AGESA 1.2.0.3c for Ryzen 5000 CPU
And I can see that :
Ryzen 1000 : looks abandoned (incompatible with W11 anyway)
Ryzen 2000 CPU : a new update on AGESA 1.2.0.6 and nothing since
Ryzen 2000 APU - Ryzen 30x0 APU (Raven Ridge 2) - Ryzen 4000 APU : two new updates one on 1.2.0.5 and the other on 1.2.0.6
Ryzen 3x00 APU (Picasso) - Ryzen 5000 CPU - Ryzen 5000 APU : a new version of SMU for each version of AGESA
Ryzen 3000 CPU : a new update on AGESA 1.2.0.5 and nothing since

[GoLLuM13]

I just found an article on Tom's Hardware saying that AGESA 1.2.0.5 introduced problems, amongst them CPU Performance, memory instability, overclocking errors and more. Also keep in mind while reading that article, SMU version for Ryzen 3000 CPUs hasn't changed since AGESA 1.2.0.5. Which means any issue introduced in 1.2.0.5 is still here.

 

Toms's Hardware article here : https://www.tomshardware.com/news/agesa-1205-bugs-amd-ryzen
SMU version number evolution (same as previous table above) : https://i.imgur.com/L2Ny6Gs.png

[Guest]

I had the same issues, 3700x with 4 sticks of DDR4-3600 CL16, gigabyte x570ud

 

was perfectly stable before AGESA 1.2.0.5.

After agesa 1.2.0.5, the system wouldn't boot 4/5 times. To get is stable I had to increase several voltages in ryzen master (VDDG, VDDP, and RAM I think)... but I didn't need to mess with any voltages in previous BIOS versions.

 

ended up upgrading to a 5800x3d and have had zero issues with the new processor (just flipped on XMP and was good to go)

[GoLLuM13]

1 hour ago, bwat47 said:

I had the same issues, 3700x with 4 sticks of DDR4-3600 CL16, gigabyte x570ud

 

was perfectly stable before AGESA 1.2.0.5.

After agesa 1.2.0.5, the system wouldn't boot 4/5 times. To get is stable I had to increase several voltages in ryzen master (VDDG, VDDP, and RAM I think)... but I didn't need to mess with any voltages in previous BIOS versions.

 

ended up upgrading to a 5800x3d and have had zero issues with the new processor (just flipped on XMP and was good to go)

Thank you for your feedback, and I guess if I'm right I remember you from reddit, so thank you one more time

[renesdk]

Just had to downgrade to AGESA 1.2.0.3c again.

I kept getting corrupted files in Steam when downloading. Downgraded and everything verified perfectly first try. So not even at standard speeds of 2666 MHz does my current RAM modules work with AGESA 1.2.0.5 to 1.2.0.7.

I'm tired of upgrading or downgrading. I'll just have to live with the micro stutters and I'm staying on 1.2.0.3c for the foreseeable future.

 

I've tried writing MSI technical support about this to see if there is anything they can do to help.

[freeagent]

That’s weird about the stutters. I have two AMD systems and none with micro stutters that people talk about. One is a Strix -F with 5600X and 2x8, the other is a Strix -XE with a 5900X and 4x8.