Zenith II Extreme Alpha instability and restarts with BIOS 1603

[beastie]

My system has ben running perfectly stable with the following hardware:

MB: Zenith II Extreme Alpha
PROC: TR 3990X
RAM: 128GB G,Skill Trident Z 3600Mhz
GRAPH: EVGA RTX3090

I multiboot diferent OS's mainly Windows 11/ Proxmox 7.1 / VMware ESXi 7 and had been fine since I built the system almost two years ago and about 8 months with BIOS 1502.

Since the update to BIOS 1603 with AGESA 1.0.0.7 I'm having spontaneous shutdowns and restarts without any pattern and with any of the OS I boot, in Proxmox (Linux) I can see the following messages preceding the crash:

Apr 21 10:46:55 pve3 kernel: mce: [Hardware Error]: Machine check events logged
Apr 21 10:46:55 pve3 kernel: [Hardware Error]: Corrected error, no action required.
Apr 21 10:46:55 pve3 kernel: [Hardware Error]: CPU:0 (17:31:0) MC25_STATUS[-|CE|MiscV|-|-|-|-|CECC|-|-|-]: 0x98004000003e0000
Apr 21 10:46:55 pve3 kernel: [Hardware Error]: IPID: 0x000100ff03830400
Apr 21 10:46:55 pve3 kernel: [Hardware Error]: Platform Security Processor Ext. Error Code: 62
Apr 21 10:46:55 pve3 kernel: [Hardware Error]: cache level: RESV, tx: INSN


Reverting to BIOS 1502 makes the system rock solid and no unexpected shutdowns/restarts.

Someone have the same problem?

Is the any way to open a ticket with Asus to help in troubleshooting it?

Thanks!

[framm100]

I started experiencing TPM related issues after upgrading to 1603. Gonna revert soon I think!

[Karl-Johan]

No issues at all here.

[beastie]

Seems something related to the voltage control: undervolting a bit the SoC and the processor core, is stable like in the previous version.

[Mappi75]

Since updating to bios 1603 i got massive problems too!

MB: Zenith II Extreme Alpha bios 1603 incl. dedicated asus TPM 2.0 module
PROC: TR 3970X
RAM: G.Skill Trident Z Royal gold DIMM Kit 256GB, DDR4-3600, CL16-22-22-42
GRAPH: PNY RTX A6000
power supply: Corsair AX1600i
Asus Hyper Card with 4x Samsung 970pro 1TB / 2x 970pro 1TB on board

1. When installing Windows 11 Pro Workstation and entering the product key the system hangs (forever)
2. or later when i select the partitions in the setup using diskpart the system hangs too
3. sometimes during installation the pc reboots..
4. after installing win updates there are "red" errors in the reliability monitor in Windows
5. even installation of creative cloud gives errors in the reliability
6. sfc /scannow gives errors after fresh installed windows and all updates.
7. created a new windows 11 image via HP Zbook does not helped (workstation 100% stable)


First i thought it has something to do with the voltages of:
VSOC: 1,05v
VDDG IOD: 1,000v

but its NOT the solution!

I replaced the mobo (same model) using another graphics card PNY T600,
changed the power supply (same model), removed asus hyper card,
and the errors with bios 1603 are all replicable !

I got the system over two years and never got this hard issues before.

Even undervolting the memory can help here: 3200Mhz wont work stable, 2.666Mhz wont work stable
only at 2.133Mhz the system "began" to get stable.

Interessing on load the systems looks stable tested with:
tm5: 9 hours
Adia64 Stresstest: 4+ hours
Karhu RAM test: 12+ hours

But on low load or during installing windows 11 the system crashes/reboots and so on..

Going back to bios 1502 fixes all.

@Asus please fix bios 1603 for a newer agesa version which works,
thank you.

[lightknightrr]

Just like to chime in, @Mappi75, you are a ******* for having the very video card I was drooling over yesterday (an A6000). Here I am, stuck with an EVGA 3090 with 24GB of VRAM...

But more seriously, and I know this isn't the advice you want to hear, but buy this memory (256GB (8x32GB) DDR4-3200 Mhz PC4-25600 ECC UDIMM 2Rx8 1.2V Unbuffered Server Memory by NEMIX RAM) & use it: https://www.amazon.com/2x32GB-DDR4-3.../dp/B09GGX77TH. It's ECC, you own a Threadripper, show that motherboard some love and get it some ECC RAM. Works fine with the latest BIOS, don't even need to configure anything for it to run stock at 3200Mhz. And one guy on there, in the reviews, has apparently gotten it to run at 3600Mhz (on a Threadripper). I'm tempted to do the same, after I resolve that ACPI 15 problem (and after I order up some RAM heat spreaders, fans, and graphite thermal pads).

[Mappi75]

Sent you an pm.

Hope Asus will release a fixed 1603 bios!
Or better: with new agesa 1.2.0.7

A owner of 3x Zenith II Extreme Alpha boards.
Thank you.

[micboule]

My system has ben running perfectly stable with the following hardware:

MB: Zenith II Extreme Alpha
PROC: TR 3990X
RAM: 128GB G,Skill Trident Z 3600Mhz
GRAPH: EVGA RTX3090

Reverting to BIOS 1502 makes the system rock solid and no unexpected shutdowns/restarts.

Someone have the same problem?
Is the any way to open a ticket with Asus to help in troubleshooting it?

Thanks!

Hi Beastie and Mappi75,

Wow, I was about to quit and buy a new motherboard.

It started when I bought a new NVMe this week to install in my pc and install a fresh W11 Pro on it.

My pc:
MB: Zenith II Extreme NOT Alpha
PROC: TR 3970X
RAM: 64GB G,Skill Trident Z 3600Mhz
GRAPH: EVGA RTX3080TI
GRAPH: Asus STRIX 1080TI

At a moment I thought maybe it was the new NVMe or the RAM or even the Power Supply the problem, but it was not.

I was getting a lot of random errors during install.
Random errors I got durint the install of Windows 11 Pro.

• 0xc000000e
• 0x8007025d during Windows Install
• PC Reboot by itself randomly while installing Windows

Have removed a pair of the 4 x 16GB RAM and still the same problem.

I had reset to default the BIOS 1603 and set back my RAM to 3600MHz and got the same problem.

One thing got me curious is that if I don't set the RAM to 3600 and leave it to the default 2100Mhz, it was going fine for the install of the Windows.
But the memory was not at the good speed.

It's a that time I found you post and Oh yeah, I said to myself, why I didn't try to downgrade to 1502...

Thank you very much Beastie and Mappi75, after 7 hours of troubleshooting, you nailed it.

I have no more issue on BIOS 1502, stable memory, no random reboots and the NVMe goes find without any errors!

[Mappi75]

@micboule interesting that you have the problems too a non alpha board!

So bios 1603 is 100% faulty and asus should remove this bios from their website.
Before more people crash their systems!