GIGABYTE Latest Beta BIOS

[AMD718]

Myself and another forum member are currently testing a 10.49 beta driver provided by realtek. I haven't seen any issues so far, and I don't believe he has either, but of course we're only a sample size of 2. From this at least the take-away can be that Realtek is aware there are still more bugs to be worked out, and 10.49 will be forthcoming. No idea on timeframe.

10.49 Beta testing should now be complete. My understanding is that Realtek will be posting a new 10.50 driver for RTL8125 (hopefully) soon.

[D1g1talEntr0py]

Anyone else finding f33i on Auros Xtreme (or any other board) actually terrible? Worst BIOS I have tried I think. More WHEA errors than before and PBO/CO seem to do nothing? Actually I got worse performance than on f33h and worse performance than stock when using the exact same PBO/CO settings.

I have the Xtreme v.1.1 and I'm able to push my curve a little higher on some cores. I don't get any WHEA errors unless I push past 1900MHz IF, but that has always been the case on my 5950X. I'm not holding out hope that any BIOS will ever fix that. I didn't notice any performance regressions though. Maybe clear your CMOS and flash again?

[K900]

It seems like the issue is actually an interaction between newer firmware versions (F33 branch possibly? F33h and F33g are definitely broken, F32 is definitely not) and Windows Update. I've had a machine that was running fine on F33h, until Windows automatically downloaded and installed the RAID drivers, which made it start crashing with INACCESSIBLE_BOOT_DEVICE on the next boot. Manually removing the driver through DISM in WinRE makes the system boot normally until it gets automatically reinstalled again. It seems like the newer firmware makes Windows think it's running on a RAID setup even when it isn't for some reason...

Update: the firmware isn't actually to blame here, the RAID driver update only gets rolled out after you apply all the cumulative updates, which took a while for my new clean Windows install.

[jwsg]

I think one legitimate concern with Gigabyte's approach to this specifically is that they don't actually mark the beta versions as such.
Instead, a visitor needs to know that if the version number has a trailing letter it's a beta version (which is not explained on the page either).

Maybe other brands dont distinguish either - Asus have numbers for their BIOS and I'm not sure how you tell what are Beta.

In fact maybe there's isnt such a big difference between the suffix releases and ones without letters? The same rules apply to any version - like only update if you know what you are doing.

When a manufacturer doesnt say something then the internet loves to make up its own rules instead - there are always one or two on every forum who state things as fact that are really just myth.

[jwsg]

10.49 Beta testing should now be complete. My understanding is that Realtek will be posting a new 10.50 driver for RTL8125 (hopefully) soon.

What would be nice - and also for BIOS releases - would be if the faults that we all have are listed in the release notes and then the note says it's fixed.

The previous BIOS note mentioned Bluetooth and some users have said something's better but why cant the release note explicitly mention the Intel BT device disappearing issue?

With the LAN driver both 10.47 and 10.48 were supposed to be running great for testers but they still dont seem to be the fix we're waiting for.

Ive seen two failure modes with the more serious one failing an entire router port but these should be explicitly listed in the release note and marked as fixed or outstanding - otherwise were back to changing random settings in Device Manager like IM.

[hawk7000]

Maybe other brands dont distinguish either - Asus have numbers for their BIOS and I'm not sure how you tell what are Beta.

In fact maybe there's isnt such a big difference between the suffix releases and ones without letters? The same rules apply to any version - like only update if you know what you are doing.

When a manufacturer doesnt say something then the internet loves to make up its own rules instead - there are always one or two on every forum who state things as fact that are really just myth.

If we take Asus as an example, they do write "Beta Version" after the version number on the beta versions. (Note that I'm talking about how these versions are represented on the respective official web sites; I don't know if the version string embedded in the BIOS reflects that at all.)

As for whether the distinction is actually important/relevant, I suppose that is a separate question all in itself. I think it's a relevant distinction, but maybe I'm old fashioned.

[BagoBones]

Maybe other brands dont distinguish either - Asus have numbers for their BIOS and I'm not sure how you tell what are Beta.

Very easily:

https://www.asus.com/ca-en/Motherboards-Components/Motherboards/All-series/TUF-GAMING-X570-PLUS-WI-FI/HelpDesk_CPU/

[ZU476B72N4PC]

Hey guys. I seem to be having some issues with a Z590 Aorus Master and 10850k. I am getting constant PCIe bus errors in hwinfo no matter what BIOS settings are set. It trips error 17 in event viewer. It seems to point to a PCIe root controller, but doesn't tell me which one and so I've been scratching my head. I've tried F3 and F5b BIOS versions with no fix. Is this a known issue that is being worked on?

I finally managed to resolve something very similar to this by disabling PCI Express Link State Power Management in my Windows power plan advanced settings.

[K900]

Update: the firmware isn't actually to blame here, the RAID driver update only gets rolled out after you apply all the cumulative updates, which took a while for my new clean Windows install.

And another update: the issue is actually specific to Insider builds right now, including the Release Preview channel (!). Sticking to the stable channel (even with the 21H1 update already installed) allows you to not get the broken driver download. It's also mentioned on Reddit: https://www.reddit.com/r/windowsinsi...21370/gwcgj0q/

[Twdouglas]

[baldersz]

Yeah man same here, F13i has been awesome for me!

[ezflma]

stasio, for the Z490 Vision D, is it possible to request the latest posted F20 bios (4/19/21) with 0x3c microcode? I wasn't able to successfully flash the one you posted a week or two ago. I'd love to see if the updated microcode fixes my 0c sensor temp reporting, among other issues. Thank you.

[rubene66]

What SOC and VDDG voltages are you using?

[stasio]

stasio, for the Z490 Vision D, is it possible to request the latest posted F20 bios (4/19/21) with 0x3c microcode? I wasn't able to successfully flash the one you posted a week or two ago. I'd love to see if the updated microcode fixes my 0c sensor temp reporting, among other issues. Thank you.

Modded ..inside zip file is modded Efiflash...

https://www.mediafire.com/file/ntlonfw67etf7z4/mb_bios_z490-vision-d_f20_x3c.zip/file

[Twdouglas]

What SOC and VDDG voltages are you using?

I'm using 1.15 SOC, 1.05 CCD\IOD and 1.00 VDDP for 1866 fclk\3733mhz ram 15-15-15-30 1T GDM Disabled with very tight secondary and tertiary.

~55ns in AIDA.