Gigabyte Aorus X670E Master

Associate
Joined
30 Mar 2017
Posts
874
What a nightmare, but good to hear you've managed to sort it. Did you flash back to the previous bios or do F10 again?
I couldn't get into the UEFI bios as I had a black screen. So I put the F10 bios on a different USB stick and used Q Flash Plus by plugging it into the back of the motherboard. It only worked when I renamed the file gigabyte.bin, otherwise it would turn on for a second and then turn off. So that seemed to update the board normally to the new bios. Very slow boot times now with memory training, but I think you can disable that in bios. Not ready for another go at the 1.0.0.7 bios that just released!
 
Last edited:
Associate
Joined
30 Mar 2017
Posts
874
Last time I did a bios update on this board I thought I had killed it. It flashed but then went to a blank screen which I left on for about an hour then reset it. Of course it always says don’t reset during a flash but I didn’t know what was going on. LUCKILY it went ok. I will be honest I had similar weird things with a Gigabyte z690 board before.
Exactly what happened to me. I thought after 15 mins or so it was just boot looping anyway. So, I took more drastic action. Never usually had an issue with updating bios.
 
Associate
Joined
24 Jun 2016
Posts
845
Location
Hartlepool
Gigabyte throwing BIOS out like confetti at the moment with the release of F11a
  1. Checksum: 2DC9
  2. Update AMD AGESA 1.0.0.7 A
  3. Support DDR5 48GB/24GB high capacity DIMMs
  4. Follow AMD guidance to limit max. SOC voltage to 1.3V for Ryzen 7000 series X3D and non-X3D CPUs. Certain EXPO/XMP memory performance may be impacted due to this change.
As always, caution recommended with any beta bios. No other details on what changes are included in this version.
_
 
Associate
Joined
30 Mar 2017
Posts
874
Gigabyte throwing BIOS out like confetti at the moment with the release of F11a
  1. Checksum: 2DC9
  2. Update AMD AGESA 1.0.0.7 A
  3. Support DDR5 48GB/24GB high capacity DIMMs
  4. Follow AMD guidance to limit max. SOC voltage to 1.3V for Ryzen 7000 series X3D and non-X3D CPUs. Certain EXPO/XMP memory performance may be impacted due to this change.
As always, caution recommended with any beta bios. No other details on what changes are included in this version.
_
Won't be in a hurry to update to this one! :cry:
 
Associate
Joined
24 Jun 2016
Posts
845
Location
Hartlepool
More updated BIOS appearing, with the latest one, F11c, release to fix a security vulnerability...

F11c -
  1. Checksum : 44C3
  2. Addresses download assistant vulnerabilities reported by Eclypsium Research

F11b -
  1. Checksum: FE60
  2. Update AMD AGESA 1.0.0.7 A
  3. Support DDR5 48GB/24GB high capacity DIMMs
  4. Follow AMD guidance to limit max. SOC voltage to 1.3V for Ryzen 7000 series X3D and non-X3D CPUs. Certain EXPO/XMP memory performance may be impacted due to this change.

The links below have some more information about the potential vulnerability, which seems to impact more than just Gigabyte, and not just the B550 boards.



Official post from Gigabyte


-
 
Last edited:
Soldato
Joined
1 Nov 2002
Posts
10,181
Location
Sussex
Anyone know how to put a cpu (7700x) in to eco mode using this board? I was hoping it would be a bit more obvious, maybe I’m just being a plonker.
 
Soldato
Joined
1 Feb 2006
Posts
3,415
Anyone know how to put a cpu (7700x) in to eco mode using this board? I was hoping it would be a bit more obvious, maybe I’m just being a plonker.
Does the BIOS have a search? Think it will be in the Precision Boost Overdrive menu.
 
Last edited:
Associate
Joined
30 Mar 2017
Posts
874
So, I have been having nothing but nightmares with this board unfortunately, since updating from bios F8 to F10. F10 would just repeatedly crash and train memory, sometimes endlessly that required a full power down (that also wouldn't work sometimes.)

I tried updating to F12a and that was pretty much exactly the same.

So, I decided I'd flash back to the previous working F8 bios. It previously worked pretty much perfectly apart from giving code 76 (but didn't seem to affect anything.)

F8 doesn't seem to take any time doing any memory training and gets into windows every time. Unfortunately, now I am getting random crashes and restarts. Codes 00 (not in manual) and 76 (PCH DXE initialization (PCH module specific).

Have tried clearing the CMOS as well and that doesn't seem to have any impact.

Was working fine before I started investigating another bios, so not really sure what I can do now!
 
Last edited:
Associate
Joined
30 Mar 2017
Posts
874
I have had a real intense troubleshooting session trying to get my X670E Master, 7900X, RTX 4090, Corsair Dominator Platinum 6000 MHz (EXPO on QVL) system stable as it used to be before flashing a new bios over F8.

F8 bios is the only bios I have tried that is what could be described as anything like 'normal' operation.

As above post, so far I am always able to get into the bios and Windows with F8. Unfortunately, it crashes at random times (usually under low load conditions.) Often crashes with a black screen a few minutes into loading windows. But I always seem to be able to recover to bios, unlike other bios where it would just hang on memory training forever (I mean literally, it would never end. :cry: )

I have obviously done a lot of googling. These are some things I have noticed that seem to possibly be getting me back on track...

1. Windows high performance mode and disabling sleep. This may not be necessary, but keeping on just in case.
2. Memory Context Restore - Apparently designed to speed up booting by remembering previous ram settings. Set to disabled - to try and force it to train (I know this was my problem above, but it trains for about 15 seconds on this bios.)
3. Auto settings for IF and MCLK:UCLK in bios. It crashes with or without EXPO/XMP anyway. I know updated bios mention RAM may be affected by SOC voltage changes. Mine runs auto 1.308v F8 and new bios 1.296v.
4. Disabling PBO, even just for controlling thermal limit.
5. I noticed that when I crashed in Windows and returned to the bios, some settings such as IF and MCLK:UCLK were not saving as I programmed. So, any changes I made in bios and saved and exit, I needed to do another reset to get that setting to stick when it landed in Windows the second time.
6. Just for fun I disabled above 4G decoding and resizable BAR. I had issues with my 3090 with this on another platform.

As you can see, this is a bit of a saga. I won't say I have solved it yet, but I have removed the code 00 that displays when getting into Windows. I still get code 76, but I always used to get that and it never gave a problem.

I heard that some new bios settings are stored in the CPU and don't get removed when flashing to a previous bios. I imagine that might be a possible cause of my issues, even with a possible CMOS clear.

Will update on my progress and hope this helps anyone who has anything similar happen to them!
 
Last edited:
Associate
Joined
2 Jul 2023
Posts
22
Location
UK
Just came across your post and sorry to hear all your misfortunes with your motherboard.

I have the same board as you and I opened a thread on here because I had a pin in my CPU socket just disappeared on me, have no idea what happened but Gigabyte did sort me out in the end.

My experience with the Gigabyte X670e Mobo is totally the opposite to you, I am running much the same memory but the vengeance models and using tight timings and a 1.2v on SOC and every bios is stable, this was on my old board, the replacement is the same and runs like a dream.

Tested all bios’s and the best to me with my 7950X3D is F10.

I thought that Gigabyte may not replace the board so I brought a ASUS X670E Crosshair Hero and that was a nightmare to me, all voltages compared to the Gigabyte are all higher at stock and no way would my tighter timings work on that board and all bios’s was all unstable only 1303 worked just about, but Memtest5 would throw a error now again for no reason and that’s with higher Vsoc and Memory Voltage.

I know my memory and CPU are all good, so I sent that back as its just plain faulty, my Gigabyte is up and running and working fine.

Have you tested any other memory in your board and what CPU do you have?
 
Associate
Joined
30 Mar 2017
Posts
874
Just came across your post and sorry to hear all your misfortunes with your motherboard.

I have the same board as you and I opened a thread on here because I had a pin in my CPU socket just disappeared on me, have no idea what happened but Gigabyte did sort me out in the end.

My experience with the Gigabyte X670e Mobo is totally the opposite to you, I am running much the same memory but the vengeance models and using tight timings and a 1.2v on SOC and every bios is stable, this was on my old board, the replacement is the same and runs like a dream.

Tested all bios’s and the best to me with my 7950X3D is F10.

I thought that Gigabyte may not replace the board so I brought a ASUS X670E Crosshair Hero and that was a nightmare to me, all voltages compared to the Gigabyte are all higher at stock and no way would my tighter timings work on that board and all bios’s was all unstable only 1303 worked just about, but Memtest5 would throw a error now again for no reason and that’s with higher Vsoc and Memory Voltage.

I know my memory and CPU are all good, so I sent that back as its just plain faulty, my Gigabyte is up and running and working fine.

Have you tested any other memory in your board and what CPU do you have?
Hi, thanks for the reply and good to hear you have sorted your problem.

A CPU pin in the motherboard socket disappeared? As in physically or not recognised anymore or something?

I have a feeling F8 works for me as it seems to add just a little bit more SOC voltage and perhaps as it is over 1.3v the latest bios don't allow that and that affects stability. But my problems also occurred at stock RAM settings.

I have not tested any other memory yet, partly because before changing bios from F8 I didn't have an memory issues. So, I have just been trying to get it back working, with the same hardware. I have a 7900X.

I do seem to be making some progress though, but waiting to see if anymore crashes occur since making the above changes.
 
Associate
Joined
2 Jul 2023
Posts
22
Location
UK
Hi, thanks for the reply and good to hear you have sorted your problem.

A CPU pin in the motherboard socket disappeared? As in physically or not recognised anymore or something?

I have a feeling F8 works for me as it seems to add just a little bit more SOC voltage and perhaps as it is over 1.3v the latest bios don't allow that and that affects stability. But my problems also occurred at stock RAM settings.

I have not tested any other memory yet, partly because before changing bios from F8 I didn't have an memory issues. So, I have just been trying to get it back working, with the same hardware. I have a 7900X.

I do seem to be making some progress though, but waiting to see if anymore crashes occur since making the above changes.
Yes, the pin just disappeared from the socket after removing the CPU, the very top right corner one just a very clean snap, because they are so small I never noticed until channel B for the memory did not work and gave a error code C5.

I was testing CPU’s the 7950X3D because running the SHA3 via AIDA64 would BSOD if you use a negative CO, been through four CPU’s all the same but one would not work even at stock, crazy stuff, gave up now lol.

I see, bit strange, but I been thinking do you have 64GB of memory? As I have 32GB 2x16GB maybe its that, can’t be sure its when you said you needed a such a high VSOC voltage made me think as I just checked mine and in HWiNFO64 its reading 1.8V.

Hope you carry on crash free.
 
Associate
Joined
30 Mar 2017
Posts
874
Yes, the pin just disappeared from the socket after removing the CPU, the very top right corner one just a very clean snap, because they are so small I never noticed until channel B for the memory did not work and gave a error code C5.

I was testing CPU’s the 7950X3D because running the SHA3 via AIDA64 would BSOD if you use a negative CO, been through four CPU’s all the same but one would not work even at stock, crazy stuff, gave up now lol.

I see, bit strange, but I been thinking do you have 64GB of memory? As I have 32GB 2x16GB maybe its that, can’t be sure its when you said you needed a such a high VSOC voltage made me think as I just checked mine and in HWiNFO64 its reading 1.8V.

Hope you carry on crash free.
Ah OK, so it was a hardware issue with the pin. Can see that causing problems!

I have 32GB 2x16GB sticks. Not sure if the stability is related to VSOC but I have just noticed F8 bios adding a tiny bit higher voltage and it seems to be more stable than all the other bios, so leads me to believe that may be a possible reason why.

1.8V seems pretty high for VSOC. I am guessing that might be a different reading.

Still going fairly OK so far (fingers crossed) So, I am slowly re-enabling some settings one at a time, to see if I can keep stable.
 
Associate
Joined
2 Jul 2023
Posts
22
Location
UK
Ah OK, so it was a hardware issue with the pin. Can see that causing problems!

I have 32GB 2x16GB sticks. Not sure if the stability is related to VSOC but I have just noticed F8 bios adding a tiny bit higher voltage and it seems to be more stable than all the other bios, so leads me to believe that may be a possible reason why.

1.8V seems pretty high for VSOC. I am guessing that might be a different reading.

Still going fairly OK so far (fingers crossed) So, I am slowly re-enabling some settings one at a time, to see if I can keep stable.
Ah right, that makes sense now. Its how large the sticks are that is causing your problems, but it really shouldn’t, but we know what the 7000 series are like with memory.

No, that can’t be right, 1.8v and I reckon your CPU will be fried, if you look on HWiNFO64 and scroll down to CPUVDDCR SOC VOLATGE (SVI3 TFN) that will truly tell you what voltage it is.

So, now I know that we don’t have the same memory chips in the modules as mine uses Hynix M die, if yours is EXPO rated by corsair, I think they are Samsung Dies, maybe wrong, but if you could try some Hynix A or M die from the QVL list you may get lucky.

Good luck mate.
 
Associate
Joined
30 Mar 2017
Posts
874
Thanks for the info. Hope all goes well with your system too!

Under CPUVDDCR SOC VOLATGE (SVI3 TFN) I get 1.240 - 1.250V and under CPU VCORESOC I get 1.272-1.308V.

Not sure why there are two, but both don't sound that bad.
 
Last edited:
Associate
Joined
2 Jul 2023
Posts
22
Location
UK
Thanks for the info. Hope all goes well with your system too!

Under CPUVDDCR SOC VOLATGE (SVI3 TFN) I get 1.240 - 1.250V and under CPU VCORESOC I get 1.272-1.308V.

Not sure why there are two, but both don't sound that bad.
Both are fine mate.

You should try the new bios for your board that will be released in a few weeks, it as a major overhaul to how memory works on Ryzen 7000 hitting speeds of 8000Mhz.

Asrock have new bios’s out and ASUS have beta ones out and sorry I have no idea why I thought you was running a ASUS motherboard, what a long day, but we will have a new bios out in a few weeks here is the news.

 
Last edited:
Associate
Joined
2 Jul 2023
Posts
22
Location
UK
At last, we join the ranks of others lol, we have the new bios F13a AGESA 1.0.0.7 b, guess what I am going to do now and most likely get nowhere lol.

Here we go.

 
Last edited:
Associate
Joined
30 Mar 2017
Posts
874
Associate
Joined
2 Jul 2023
Posts
22
Location
UK
let us know how it goes if you update!
Well, tested and seems better, I wouldn’t say 100% better, but I can run 6200Mhz on my Hynix m die, but not stable, needs more tuning which I always had a problem with this memory running past 6000Mhz and don’t feel like replacing it, but it works.

What I can say that it works in gear down mode off due to nitro mode on, which makes first boot from bios why long, but seems stable and scores on benches seem better and now the Infinity fabric overclocks too, which it didn’t before.

Will post back when it ran for hours on Memtest5.
 
Back
Top Bottom