[rod6700]

Actually, Microsoft delayed the release of XP64 for over a year to give the other half of the Wintel cabel to try and catch up. Itanium was supported from day 1 on server and workstation OSes

Itanium was a very unique processor co-developed by HP and Intel with different architecture and instruction sets than anything else before, then or since if I recall correctly. It was different enough that Microsoft had to develop a different OS just to support it. Never took off outside a few niche use cases. XP64 on the other hand was introduced when AMD put out the X86-64 instruction set with the Athlon 64 CPU's. Back then everything was 32 bit, but everyone knew that 64 bit was the future. Built my first box on AMD because at that time they were the only realistic 64 bit game around that was going to survive. Intel did not have anything in the general consumer space or enterprise outside of Itanium that could compete.Look at a Linux ISO download for example. For most, they are labeled AMD64 orX86-64 for 64 bit and X86 or X86-32 for 32 bit. AMD bootstrapped 64 bit into mainstream IMHO.

 

[rod6700]

How did you go back to F34? - i need help.

Use QFlash Plus to roll back the BIOS if available on your board.

 

[rod6700]

Hi, after last bios update (f36c) my Pro I Wi-Fi takes circa 4 mins to show Aorus logo and then boot.

I tried:
-to enable ultra fast booting
-to disable usb devices while booting
-to eliminate all other boot devices except the NVMe where OS is

No luck so far

What BIOS time is being reported by Windows? You can check the startup tab of Task Manager for this. I have a Pro WiFi and have noticed that any time I screw around in BIOS the BIOS time increases according to Windows. The fix seems to be clearing CMOS, loading optimized settings in BIOS, then setting up BIOS as I prefer. (Enabling XMP, fast boot, disable wake on LAN etc.) Seems to take a couple of boots for things to settle down after doing this. Watch the boot LED's to see which component is eating BIOS time also. I recently upgraded my GPU and by doing this shaved a full 5 seconds off of reported BIOS time. You can see this by watching how long each status LED is staying lit during boot.

 

[rod6700]

Tried with XMP disabled but still long boot.

Then I removed all external hd connected by USB, I’ve got some TBs there, and everything went back to normality: 13 sec to boot

So I’m assuming that with this bios version the option to disable usb storage while booting is not working properly

Seeing your latest reply on this makes me question if something hasn't been broken or buggy for awhile with exterrnal USB drives (specifically large drives) for some time. I bought a WD My Book 10 TB drive some time back and my boot times cratered until I disabled external USB drives during boot in BIOS. Cannot recall the exact BIOS version I was on at that point, but thinking either F20 or F30 versions. Curently running F34 with no issues to speak of that I have not found some sort of work around. Current BIOS time is ~7.0 seconds with 6 drives internally (1 NVME M.2 PCIEx4, 2 NVME M.2 PCIEx3 - 1 thru adapter card in PCIEx4 slot, 1 NVME SSD on a PCIEx1 slot on AIB, 1 SSD on SATA, 1 HDD on SATA and the one external with usb at full initilazation in BIOS.Considering I have never run a beta BIOS on this setup and the little issues that seem to occur random with "stable" release BIOS am reluctant to do so. I have had Asus, MSI boards in the past, but never seen a BIOS as finicky as on this current setup.

 

[rod6700]

Hi all

I just installed a external usb hard drive and noticed it slows my boot time right down as when it checks the drive it powers it up and hangs until its finished, is there a way to make this stop slowing down my boot time

its a gigabyte master motherboard,

thanks.

Look for a setting in BIOS related to USB under the boot tab. Should read in options - "Full Initial, "Partial Initial" or "Disabled depending on BIOS version. I had horrible boot times with a 10TB external USB hooked in with some prior BIOS versions on my X570 Pro WiFi. Older BIOS versions I disabled completely, but that option can render booting from a USB stick useless. I am currently running F34 with AGESA 1203B as anything newer screws up OC settings.

 

[rod6700]

[QUOTE="ManniX-ITA, post: 29039960, member: 538195

There's no acceptable AGESA since 1.2.0.3B for dual CCD, without VDDG bug or VID cap.
It's been 18 months now.
Which basically locks out of critical security fixes everyone that would like to keep its 5900X/5950X running with the performances they bought it for.
[/QUOTE]
Agreed on that point for AGESA 1.2.0.3> for overclocking here. Recently went from F34 AGESA 1.2.0.3B to F35 AGESA 1.2.0.5 on my X570 Pro Wi-Fi just to see what would happen. Overclock scores fell on every test I ran across it after. The main thing that really stuck out is the "security fix" portion of the BIOS update. The only change I could find is that Gigabyte has implemented IBRS (Indirect Branch Restricted Speculation) on the security fix. It is poorly labeled in the BIOS setup menu here and I suspect its common across the entire X570 lineup. The way it is labeled unless you know what to look for is they have labeled it just Indirect Branch Speculation instead of IBRS. Choices are Auto, Enabled, Disabled. If you set to enabled, thinking that it enables Branch Speculation, it actually enables the security mitigation. Some say this hurts performance that much more so. Any way you could double check this on your end?

 

[rod6700]

Your disk's partition-table was not set to GPT from the begning. You need to start from scratch and the easiest way to do is by using a live-cd Linux distro with gparted. Using gparted can give you a user interface so you can see what you are you doing, this is why I recommend it. A good Linux live-cd, out of many, is MX-Linux AHS with XFCE desktop environment. Link: Download Links – MX Linux

You can use also Windows installation USB and choose "Repair your Windows" and choose command prompt to configure the disk using a few simple commands.

Re-configuring your disk will delete all data on your disk using any of the methods so you need to backup your important files from the disk. Disconnecting other harddisks while configuring your main disk will eliminate user error and confusion.

Let us know which method you prefer and we will help you.

You can also download a standalone version of GParted to boot from and use RUFUS to burn to a CD/DVD or load on a bootable USB stick rather than an entire distro.
GParted -- Download

https://github.com/pbatard/rufus/releases/download/v3.20/rufus-3.20.exe

Remember that if booting UEFI rather than with CSM enabled is a no go this way. You need to enable CSM for this route to work. It is still a great way of partition manipulation as a general go to for disc erasing and changing partition and file system type. Just wondering as to how his system and boot partitions became separated at this point. 🤔
Aristotelian, could you post a screen of your full disk management screen that includes all drives and layouts please.

 

[rod6700]

What kind of a case do you have? I just did an experiment. Opened my side panel for 1h and the temps dropped by almost 10°C.
View attachment 2586454
The case i'm using is an NZXT h500 - poor airflow all around. Seems like i'd be better of replacing it. My gpu's also been idling at 57-58°C

The app mentioned earlier, Fan Control can be used to control PCH at least on my X570 Pro Wi-Fi but I dumped it due to other fan control options or lack of on peripherals. It's an option or adjust the fan in BIOS is all I have with what you are working with at the moment. As the case in use sucks for airflow, I would suggest going to a new case that supports vertical mount of the GPU if you can afford as last suggestion.

 

[rod6700]

Gigabyte have released beta F37a now for the Aorus master.

Only noted change:

1. Optimized performance

Ditto on Pro Wi-Fi: https://download.gigabyte.com/FileList/BIOS/mb_bios_x570-aorus-pro-wifi_f37a.zip SMU checker shows no change, and I am a noob with UEFITool for checking AGESA version. No mention of AGESA code tells me that current CVE with AM4 remains unfixed. AMD Product Security | AMD

 

[rod6700]

What is AMD's latest AGESA? On my Aorus Ultra I am on v2 1.2.0.7 and I had a quick look and MSI are on the same but I seen a couple of high end boards from ASUS are on v2 1.2.0.8

AMD latest AGESA on AM4 is 1.2.0.8 but it is pretty scarce at the moment. As it resolves some CVE issues, I would have expected quicker uptake as it was released by AMD latter half last year.

 

[rod6700]

This works as well if looking for TPM: Amazon.com: NewHail TPM2.0 Module LPC 12Pin Module with Infineon SLB9665 for Gigabyte Motherboard Compatible with GC-TPM2.0_S : Electronics Passes Windows 11 checking just fine.

 

[rod6700]

For those wondering if AGESA 1.2.0.7 fixes the fTPM stutter, it does for me on the X570 Aorus Xtreme and I've not had stutter for months, however, there is a catch.

There are two versions of AGESA 1.2.0.7 with differing fTPM firmware revisions, only the latest one will fix the issue and not the BETA.

For reference, please see the screenshot from my TPM settings page as this is the version that fixes the issue. View attachment 2597280

I've contacted Gigabyte regarding the availability of AGESA 1.2.0.8 as they are dragging their feet and they told me that it is currently in testing phase.

Seems all the OEM board makers are dragging their feet on 1.2.0.8 as it was made available second half last year. What version on the fTPM in Beta staging?

 

[rod6700]

Catching up with this thread...

My TPM version for fTPM is the same as rob-tech has listed above (on F36 final). I have not seen any issues/stutters since fTPM was enabled regardless of what the system is doing (General Use/Browsing/Videos, Building OS Images, Code Compiling, Virtual Machines, Gaming, etc.). So hopefully it will work for you as well.

The external module hasn't seen any issues either, though I am not actively sitting at that system most of the day... one interesting note is the external module seems to take significantly longer to initialize / Windows to run maintenance tasks against it.

Are you running Bitlocker enabled?

 

[rod6700]

Can we flash back to previous revisions from F37? I know there was talk about once you hit F35 you can't downgrade, but not sure if GB reversed that change or not?
[/QUOTE]
From what I understand it may be possible to roll back BIOS using QFlash+ but not QFlash in BIOS. Initially Gigabyte said no roll back after F35 period but removed the reference after initial posting.

I have the aorus pro. version 1.2 (not that it matters) and I can flash to any BIOS using Q-Flash. You can flash any BIOS using Q-flash as far as I know. It's not like flashing an ASUS motherboard which hinders downgrading the BIOS using EZ-Flash, I have not seen this restriction with Q-Flash.

This goes back to whether or not you use QFlash+ or QFlash. F35 introduced capsule BIOS support which Asus has done for quite a while now. That is why QFlash or EZ-Flash does not allow roll back below F35 as I understand it. Have you actually tried rolling back to a pre capsule support BIOS using either QFlash or QFlash+?

Most horrifying thing about capsule BIOS? Windows Update has the ability to update your BIOS if used as designed. Past track record on drivers and Windows Update gives me a huge case of heartburn in regard to using it to update BIOS.

 

[rod6700]

• Never update BIOS using Windows.

100% agree as I always advise flashing within BIOS rather than Windows no matter who makes the board. Only brand that I will do a BIOS flash from Windows is Dell as they seem to have gotten it correct and it is the only way they offer to upgrade BIOS. Thanks for showing that you can indeed use QFlash to roll back BIOS to a pre capsule support version. I had seen some reports that the only way to roll back after capsule support was QFlash+ and have not personally tried it. I know that HP was testing updating BIOS using Windows Update at one point in limited basis but cannot say if they ever went all in on doing so. Horrible idea IMHO, big difference between a dodgy driver update that causes BSOD and bricked hardware.

 

[rod6700]

I just noticed that Gigabyte has posted some new drivers (6.0.9373.1) for older boards that have the Realtek ALC1220-VB codec, both AMD and Intel platforms included. Currently I am using version 6.0.9313.1 from the Aorus Master X670E as it uses the same codec with no problems for several months now. I checked digital signatures, and the 9313 version is a newer signature date than 9373 but versioning level is lower. Gigabyte has also posted the 9373 version on the X670E Master support page, but it was not there yesterday and is posted to appear as if it was there before 9313. Zip file size on 9373 is roughly double of 9313 before extraction. 9313 is a DCH driver which may make a difference on the file size and versioning, but I cannot say with any certainty if this is why. I know that their web page maintenance is about as great as after sale support in general already. Anyone have any ideas on this version level and date difference?

Did some more digging and found that because of the 9313 being a DCH vs traditional is reasoning behind version and file sizes. DCH does not come with all the extra baggage required on a traditional driver install. Still does not explain the recent addition on the X670E page, just chalk it up as webpage maintenance. Sorry for the lame ass post.

 

[rod6700]

Honestly thanks for this post - I'd given up on seeing newer drivers for my x570 Master as well to the point of recently grabbing from an Intel board with the 1220 VB (I forget which Zx90 chipset it was).
Happy to have a fully matched driver again.
So far so good but I did have to disconnect from the internet to get out of a uninstall audio -> reboot - uninstall audio loop.


I also think that the 9373.1 and the 9235.1 drivers for my board are both DCH and the difference might be just new changes/devices and possibly less heavy compression on the larger file: 7z self extracting .exe vs a standard zip. Haven't checked the X670E pages.

I looked in the extracted files for both driver sets as well as within the .exe and 9313 carries far less data than 9373 which aligns with Microsoft's DCH driver implementation as designed. Both drivers carry roughly the same time/date stamp, so functionally they are probably the same. DCH drivers do not have some of the custom crap that the OEM tends to pack into a driver install. The Realtek Audio Console still installs and works with 9313, but that is a separate install from Microsoft Store and not actually included in the driver itself. You mentioned a uninstall loop? I had noticed someone else having the same problem in a Reddit post. I suggested DDU as it does have the ability to remove some Realtek Audio drivers as well as GPU drivers but would never have thought anything about disconnecting from the internet. Makes me wonder if it is trying to reach the Microsoft Store for Realtek Audio Console and if already installed is failing during this step in the install. I would try this but really don't feel the need to clean up a mess if it goes sideways.

 

[rod6700]

Yeah! Noticed this on the B550I Aorus R1.2 board too. Perhaps a fairly general rollout?

It is out for the Ultra and was briefly for the Pro Wi-Fi today. Pro had a placeholder this morning and later in the day was available, so I grabbed it. Looked again later on and the entire listing is gone. I have not flashed it and no way in hell am I going to if yanked that quickly.

 

[rod6700]

As far as I've seen any other version even like F37a and F37b don't have compatible bios profiles, which is a pain.

Ditto on this answer. The BIOS profiles are more for if you're doing several different OC profiles and need to switch between them for any reason. Gigabyte seems to change just enough of something with each BIOS that using them to save settings even if not OCing forces you to have to go through each setting after a flash.

 

[rod6700]

F37a on master v1 has been pretty great. This is literally the first bios update that is running pbo on my 3900x correctly.

What kind of differences? I gave up on PBO with my 3900X as all it did was add extra power usage with next to nothing in returns. Had better results using CTR, but it isn't in active development after Project Hydra came out and all the 💩 going on in the developer's world.