GIGABYTE Latest Beta BIOS

[Dr.PaWel]

After weeks now using F31, on a x570 master with a 5900x, PBO disabled and man. OC without any WHEA I gave the F33a a shot. F33a is not possible for me due to F9 errors, sometimes I am able to get into bios (all default, also I did a cmos). I did a role back to my old F31 and performed a second trial and the same problem using F33a. Following F33a I updated to F32 Bios from the GB page. This one works for me quit similar to F31, meaning absolutly unstable without PBO disabled (meaning WHEA errors when gaming) but stable with PBO disabled. All three Bios (F31, F32a and F33a) still have the M.2 bug, meaning i cannot boot from it as long as no other Windows is installed on another drive. F32 also seems to have a bit higher CPU temps compared to F31.
In summary for me nothing has changed, still the MB is not able to be stable in default mode!

[Waltc]

Maybe in the default xmp the flag for geardown mode isn't set. Check in the advanced ram setting. Geardown mode. If disable, set to enable. Geardown allows odd timings for amd CPUs.

Actually, Gear down results in only even numbered Cas latencies...;) My XMP ram @ 3733MHz has standard Cas latency setting of 17--but if Gear down is enabled Cas 18 is what the system sets. Disabling Gear down allows me to run with the XMP standard profile of Cas 17. Disable Geardown but manually set 1T and you should be fine.

@Dr. PaWei I've been booting my x570 Master Zen2 for the last year (at least) with only one Win10 installation on an NVMe .M2 drive--no other Win10 installation anywhere--I have six physical drives installed and 10 partitions. No problems with F31 or F32. Your bug theory is wrong and I cannot see how it could be related to the immaturity of the Zen 3 AGESAs atm. Also, running with PBO enabled and with custom settings presents no problems at all--with Zen 2, however.

Hopefully, you installed Win10 in UEFI mode and you are not trying to run the OS in legacy mode with CSM enabled! If you are then that is likely the source of your highly unusual problems with you x570 Master. I have never heard of the ".M2" bug. But I am on a 3900X!

It is true that you guys with Zen 3 CPUs are enduring teething pains with Zen 3 right now, unfortunately. It's not a lot of fun, but in a couple of months it should stabilize as the Zen2 AGESAs have stabilized--of course, Zen 2 has been shipping for 18 months +. Hope it doesn't take much longer for you Zen 3 early adopters...I went through it with Zen 2--had no desire to repeat it with Zen 3--I'll wait until the AGESAs and bios revisions for Zen 3 mature--and then I'll buy...;)

[lmarcelus]

Anyone have the B550 Aorus Pro AC F12 AGESA 1.2.0.0 that leaked? PM me please, i`m tired of waiting.

[DivesMori]

am i the only who gets random reboots with the last few Bios Versions?
F11 was fine for me, a bit hotter on the CPU (5800x) but ok.. it never gave me the random reboots. (temps got way better after lapping the CPU :) )
the unofficial F12 (Alpha Version with Agesa 1.2.0.0) gave me 1-2 reboots per Day so i flashed the newer F12 with Agesa 1.1.0.0 from the global Gigabyte Site just to get a random reboot after 2-3 hours after flashing.

Cheers

[h3x1e]

The Realtek 2.5GbE 10.46.1231.2020 driver is still borked btw, not that you can do anything about it.

[timeDrapery]

Same BIOS with AGESA 1.2.0.0.

I'm only seeing f12 on the support site for my B550 AORUS MASTER... Yet you've posted f12a, which I'm running currently
For whatever reason I can't view the AGESA version though so which is it? How'd the finalized f12 go up on the support site with no f13a or whatever is next in the versioning sequence?

[XTROnETS]

Hello i Have issuse with official f12 bios for B550 Aorus Elite, on F11 bios be ok
If pcie 4.0 dds drive write lot of data then hangs my gtx 1070ti driver , any idea ?

[GuzLightyear]

The Realtek 2.5GbE 10.46.1231.2020 driver is still borked btw, not that you can do anything about it.

The only version that works for me is 10.38.1118.2019, all the rest have the hang up bug. What a shame.

[Marucins]

Ok this feels like an AGESA bug... I have to amend my previous results. I can run everything EXCEPT 1900 FLCK.
5900X + 3800CL14 (B-Die) RAM + Aorus Xtreme F33a

1833 OK
1867 OK
1900 NO BOOT
1933 OK

Wtf! I now have the RAM running 3866MHz @ 1.5V and:
VSOC 1.1V
VDDG CCD 1.1V
VDDG IOD 1.1V

...running perfect. I feel like those voltages could probably be lower too.
No WHEA or other errors, 45min into Karhu and so far so good, this is crazy!

What's wrong with 1900 FCLK?

Edit: this did eventually crash in Prime95, although suspect the CCD voltage was too high. Since reduced it to 940mv and retesting.

The difference between the voltages must be greater

Vsoc > VDDG.... !

[NATO]

I'm having an issue with the latest F32 and F33a BIOS updates for my X570 AORUS Pro with 5800X.

Seems to have no issues running benchmarks and games but when I've been playing a game and then quit out of it back to the desktop I'm getting a hard reboot not long after (even once when I just pressed ESC in the game where it will suddenly drop the CPU/GPU load), it's like someone just hit the restart button. The weird thing is the system doesn't restart, it becomes unresponsive and doesn't respond to the power or restart buttons and it looks like it's stuck in a reboot cycle (can see the little motherboard debug lights changing) and I have to physically switch off the power to the system.

When I switch the power back on and try to boot the system I get the BIOS Splash Screen and then the screen goes black and it's back in the weird reboot cycle I mention above. The only thing I can do at this stage is pull the power once again, power up and go into the BIOS where it displays a message saying the BIOS settings have been reset, so I have to set it all up again.

I thought perhaps because this is a low-power/idle issue I tried setting the Power Supply Idle Control setting to 'Typical' and that doesn't change anything.

I do have an undervolt defined in Curve Optimiser but I dialled the settings away back in case it was undervolted too far.

When I revert back to F31, the issue seems to have gone again.

My Specs are:
AMD Ryzen 5800X
X570 AORUS PRO
32GB (4x8GB) Crucial Ballistix DDR4 CL16
GeForce RTX 3090 FE

Is there a way of reporting this up to Gigabyte as a potential issue?

[h3x1e]

The only version that works for me is 10.38.1118.2019, all the rest have the hang up bug. What a shame.

Is there any archive for the older versions? Can't seem to find it anywhere and my driver list only goes back to 10.39 it seems.

[Rock3t]

The Realtek 2.5GbE 10.46.1231.2020 driver is still borked btw, not that you can do anything about it.

I use f12 bios and 10.046 driver. I tested 2.5G lan some days. I disabled ip v6 , and everything in adapter properties except interrupt moderation and rx tx offload and flow control. Also i set QSS to 1. And i have no problem for a long time since f11 bios and 10.045 driver. Before this my adapter power off every time i open qbittorent (during some time). Try to disable ipv6 and set QSS to 1. I was ready to buy usb lan adapter from aliexpress , but now it works fine (by the way i disabled wake on lan in bios)

[NATO]

Actually scrub that, having the same issue in F31 - random freeze (or reboot) when load is suddenly reduced, resets the BIOS.

[h3x1e]

I use f12 bios and 10.046 driver. I tested 2.5G lan some days. I disabled ip v6 , and everything in adapter properties except interrupt moderation and rx tx offload and flow control. Also i set QSS to 1. And i have no problem for a long time since f11 bios and 10.045 driver. Before this my adapter power off every time i open qbittorent (during some time). Try to disable ipv6 and set QSS to 1. I was ready to buy usb lan adapter from aliexpress , but now it works fine (by the way i disabled wake on lan in bios)

Hmm the only thing I didn't turn off previously was IPv6 so I guess i'll try that, and if it doesnt work, I'll see if turning the routers IPv6 off.

The adapter was randomly losing internet connection for me, and previously would completly destroy my whole network's connection, ie, the router would explode and not be able to give internet to any of the other devices on the network. I fixed this by giving my PC a static LAN IP, so now only my own PC loses connection and the rest of the networked devices are fine.

[Schildkroete09]

Problem with GigaByte X570 AORUS ULTRA with SAM and too high SOC voltage! I got the problem with the Ultra under control thanks to users from the Luxx forum. As I already wrote, my board with several BIOS versions from GigaByte already had problems with PCIe 4.0 and the graphics card connected to it. After every 3-4 PC restart from Windows, the screen remained dark, but the PC booted, which was clearly audible from the Windows startup sound. After that, it was only possible to get an image output on the monitor again with a CMOS reset.

Those who have black screens with different GPUs: take a look at the SoC voltage. I think Gigabyte sets it too high if everything is on cars. Set manually to 1.05V or 1.075V and report times. 1.1x V ex works is definitely too much. No other manufacturer makes that so high.

It turns out that the SOC voltage in the AUTO mode of the mainboard supplies a voltage that is much too high. For me, this was 1.231V. The standard value in the BIOS suggested 1.100V to me.

Then I manually changed the SOC voltage to 1.050V and what can I say? These problems no longer occur. The board is now holding in the range of 1.01-1.05V



No more dark screen, no more CMOS RESET! And SAM does what SAM is supposed to do. The test was carried out with BIOS version F32!