Gigabyte X570 Aorus Master won't power on - (temporary) solution

Hi everyone, joined this forum / thread because the same issue (ref: cold_steel) occurred in my 6 months old rig with X570 Aorus Master.

Can we pinpoint if this IS actually related to ErP on this and other X570 motherboards?

When this happened to me, this option was set to: ErP=Enabled

Can we all report the state of this setting when issue occurred?

Also can we focus on "no boot from cold with CMOS battery removal as only solution" case leaving out the situations with crashes and reboots from this thread? (We can create another dedicated tread for this.)

Cheers!
 
Last edited:
Just had the same issue. Tested backup PSU, had the board ready to RMA before I decided to check here and replaced the battery. Rebuilt the machine and it's currently working. All I can say is that at least in my case it wasn't related to anything else (I literally had the board outside the case with nothing in/attached but power, with CMOS cleared, etc) and further when I tested the original battery under load it was definitely on the low side. So I guess we'll see but I have to believe it's related to the dual BIOS, but that's just "IMHO." As someone else mentioned, forcing it to one BIOS may help (I'll test that if it happens again).
 
@Morph3R
lol, what?
the og poster wanted help, and as as long as him/her or a mod/admin does not state anything different,ppl can post more.
lots off ppl out there, that have a different problem, but might find this thread using a web search,
and if any of those might happen to fix any issues they experience, why not.

if you would have read my previous post, you would have found out ErP causes issues for some,
its a EU requirement, so its obviously not a setting that "has to be turned on", and technically,
could be missing on any hardware sold outside the EU, without causing problems.


@Maxx
use only one bios? how?
the chips are soldered on, and these boards (Master/Ultra) dont have any physical switches for that (just auto recovery if 1st is corrupted).
keeping a different bios version on the backup chip would only make sense to test a new release in case it doesnt work (properly).



@ALL
can anyone confirm that they still have issue after updating (both) bios using efiflash?

ever since the first time updating bios when i got the board, i noticed update speed fluctuating.
slow and irregular on first time, second time faster and smoother, but still "jumpy" (slows down/speeds up back and forth),
so i wanted to use efiflash to see more detailed info when updating (dos updates usually show writing to address),
and seeing that the read me contained info that the DMI info is NOT set to be cleared for "normal" update (has to be flagged),

so i used qflash to update and it was again inconsistent on write/update speed.
then i used the efiflash and it was smoother, but still not continuously going from 0-100%.
used efiflash again, this time setting /C and /DB flag (deletes DMI before update, and updates backup as well), this time update going at constant speed (0-100%).
using qflash after that had no hiccups anymore, constant speed like with using efiflash.

this again shows its most likely something with the (dual) bios.

gonna use identical settings and see if the sudden reboots still show.
so far, the cold boot is gone and not repeatable (which it was before, when i used the psu switch).
of course it could be the latest bios, not the way i flashed it, but im not gong thru reflashing everything with prior bios, just to see.
 
Last edited:
Oh cool, some BIOS activity. I'm not going to try it until it gets the "beta" letter off the end. I'm not having any operational issues and updating the BIOS can solve old problems while creating new ones so I'll wait until it's an offical version (without the letter on the end). Just would like to see some things resolved, for me it's faster boot times and things working right with CSM disabled.
 
Not sure what you want CSM off for (GB default IS on), as it causes at least minor issues on some boards (like lowering bios res resulting in a "zoomed in" appearance),
outside it doesnt need to be off even with 3xxx cpus and using 10.
 
Default is enabled. Disabling CSM causes a black screen longer on boot and lagginess when navigating through the BIOS menus. Others have reported this, but it could be something limited to the X570 Elite which is the board I'm running. Not the same board as the Pro, but takes the same BIOS.
 
rediscovered these forums after finding this thread via Google. I've got an Aorus X570 Master with a 3950X and 64GB RAM ... and I just experienced the same symptoms as the first. I had been using my system last night, powered off as normal. When I went to power up this afternoon- no luck. no fans no lighting no nothing. Tested PSU (good), tested power mains (good ... on UPS or off). Finally tried the workaround presented- removing and reinstall CMOS battery ... and that worked fine. I did not have to reset my CMOS settings though.

This is a fairly new build, maybe 2 months on it? this is the first stability issue I've had, so what gives?
 
CraigHB
i know what it does when turned off, i have the Ultra.
Thats why im asking: what do you want to have it off for, as it wont speed up boot (CSM on, is not like legacy mode).
 
I wanted to share my findings from yesterday.

Flashed bios F12g and setting up leaving most of the settings on optimized default.
Was excited to find out that F12g boots much faster than the F11 (<13 sec to Win 10 Pro login screen from cold).
Ran some tests and few games, everything rock stable!

Unfortunately after 3rd shut down machine won't boot!

My bios setup was as follows:
ErP = Disabled
Power Loading = Disabled
Power Supply Idle Control = Low Current Idle (my PSU supports C6 / C7 Power States )


So one thing is sure, F12g BIOS update on Aorus Master x570 doesn't solve the cold boot issue out of the box!

But I'll give it another chance.
Removed and reinserted CMOS battery (previously tested voltage which was spot on 3.0V)
Clear CMOS and load optimized defaults.

Applied the following bios settings:
ErP = Enabled
Power Loading = Enabled
Power Supply Idle Control = Low Current Idle


...and keeping my eye on it!
 
Last edited:
Morph3R
lol, really?
you tell ppl to go BTT, then completely ignore what i posted about the flashing procedure,
which FIXED MY COLD BOOT, when nothing else did.


Or turning on ErP, which is (almost 100%) not the reason for (boot ) issues.
If you think the cold boot IS power related, otherwise it makes no sense to change that setting in the first place,
then turning ErP on will actually archive the complete opposite (making the difference between power off and power on state even bigger than before).
 
Fry178
Sorry I missed your suggestion related to efiflash.
I will try this right away.

Update:
OK, I've done the efiflash re-flashing procedure using FreeDOS bootable flash drive (rufus):
1. > enter BIOS > Load optimized defaults
2. > boot from usb flash
3. > step 1. flash only main BIOS (efiflash X570AOMA.12g)
4. > step 2. flash both main & backup BIOS and Clear DMI data (efiflash X570AOMA.12g /C /DB)
5. > enter BIOS > Load optimized defaults > setup

I have set power related options to the values with which my motherboard worked flawlessly for 6 months:
ErP = Enabled
Power Loading = Disabled
Power Supply Idle Control = Low Current Idle


First cold boot after PC was shut down for 15 minutes with PSU switched off was successful!

Fingers crossed that this resolves my (and everyone else) cold boot issues.
Will report from time to time if any updates!

Cheers
 
Last edited:
Morph3R
lol, really?
you tell ppl to go BTT, then completely ignore what i posted about the flashing procedure,
which FIXED MY COLD BOOT, when nothing else did.


Or turning on ErP, which is (almost 100%) not the reason for (boot ) issues.
If you think the cold boot IS power related, otherwise it makes no sense to change that setting in the first place,
then turning ErP on will actually archive the complete opposite (making the difference between power off and power on state even bigger than before).

I hope you're right, bud. I've had the problem twice this week. How the hell did this happen all of a sudden to so many people? Is is some kind of ticking time bomb flaw? Could a flimsy CMOS battery have led to a Bios corruption that your EFIFLASH procedure fixes? Did a Windows update affect the shutdown procedure?

I Sent a support request to Gigabyte hoping they "get the point". I doubt they'll do anything about it though. I'm terrified of asking for an RMA after seeing other people's experience with Gigabyte. If they "don't know" about the issue, how the hell do they solve it? Is revision 1.1 any better? Will they just send me a board with the same issue?

So many questions...
 
Raisin
My thoughts exactly!

To be frank, I really have doubts that reflashing the same bios which worked flawlessly and then suddenly didn't can make any difference long term.
If BIOS IS corrupted, how do we explain that battery removal / reinsertion revives the MB?

I've done this efiflashing thing nevertheless as a desperate measure. Maybe DMI data gets corrupted and clearing it using /C helps.

I don't giving up on this board and I'm willing to do anything to help find the solution OR proper diagnosis of the issue.

When my first no-boot from cold issue occurred, my system was exactly the same as first day I assembled it in December last year
except for 2 things:
- hardware: month ago I changed my PSU from CM V850 (which worked flawlessly btw) with new CM V1000 Platinum
- software: Windows 10 2004 update, AMD chipset drivers 2.04.28.626 (6/3/2020), RGB Fusion B20.0430.1 ...

In order to rule out the possibility that software is somehow causing this I decided to do experiment and install the fresh OS on my old 850 Pro SSD
Will use the proven OS/driver combination: Windows 10 Pro 1909, previous AMD chipset drivers 2.03.12.0657 and will keep the OS clean without any OC tools like Ryzen Master or similar and will be using this lab rat OS as a daily driver.

If cold (no)boot issue remains, this would mean that we have pure hardware related problem.
 
Last edited:
Was excited to find out that F12g boots much faster than the F11

That's great to hear. Cold boot speed has always been a complaint with my X570 Elite. Though still I'm going to wait until the BIOS comes out of beta to load it.

Not sure what you want CSM off for

Mainly convenience, don't want to see MBR options in the boot menu or boot priority page in BIOS.

Also there's an obscure boot order bug that can affect me once in a while. If boot order has an MBR option ahead of a UEFI option on a UEFI only disk, boot will fail saying it can't find the boot disk. It fails to attempt the secondary which is UEFI in this case. I ran into this swapping out my boot drive, for some reason changing the drive causes MBR to become primary and UEFI secondary.

I can go into the BIOS and correct the boot order then everything works again, but I should not have to do that. First the BIOS should not change boot priority and second it should attempt the secondary if the primary fails. Disabling CSM eliminates the possiblity of an MBR option showing up in boot priority so it's a work around, but disabling CSM gives me other troubles.
 
Morph3R
lol, all good.
was just going "wth"

Maxx
and the other broads dont have it, so another thing taht can be excluded of making a difference.

CraigHB
thats why you limit boot devices to UEFI, doesnt need to be done over CSM.



@all
All the things that did NOT make any difference in board behavior (cold boot etc):
win 10 build 1803/1903/1909, last drivers (chipset) released by AMD since 2019, 1usmus power plan and/or ram calculator info,
prior bios updates.


In fact, i can actually blame the GB boards to 100%, as the same OS install that was running fine on a x570 MSI,
has issues on my Ultra.
Everything else HW wise stayed the same (case/ram/m2s/ssd/psu/gpu), and i can even narrow it down to being bios related.
Was late when i was trying to get some info about the reboots, and didnt give it much thought (and keep link etc),
but one guy actually saved the bios and looked at the file, and actually saw coed got corrupted , thus leading to issues,
fixable by removing battery/reflashing.
So far, nothing else impacted/fixed my problems, yet when i started doing the qflash+ THEN use qflash to update (both) bios,
it reduced my cold boot issue to only happen after psu switch was off, and now after efiflash even that is gone.
No other thing had as much impact as this, makes me exclude everything else (as cause) but HW (dual bios).
 
@CraigHB
thats why you limit boot devices to UEFI, doesnt need to be done over CSM.

Not sure what you mean by that. CSM is a function provided by the BIOS to boot from MBR which is the legacy method. It can be required if there's a specific need for it, but UEFI has been around long enough that MBR is pretty much deprecated. It should actually be disabled by default. It does not make sense for it to be enabled and I can't shut it off without side effects.

CSM enabled is an annoyance because I get additional MBR items in the boot menu and boot priority page in the BIOS. Of course those are dead entries since none of my stuff boots from MBR, I use UEFI on everything. Really gives me trouble with USB boot devices since I get two identical entries, one UEFI, one MBR. Can't tell them apart. Also gives me trouble with boot priority. It can change under unusual circumstances putting the MBR entry first and failing boot as it does not honor the secondary UEFI entry.
 
No its not.
Its to allow the UEFI to EMULATE a bios environment, nothing else. Enabling CSM then will allow for different options to appear/be selectable,
incl legacy/uefi oprom support.
As soon as you change/limit the type of boot rom that is allowed (storage) to only UEFI, you should not see any MBR drive, even if CSM is enabled.
If not limited to uefi only (legacy oprom allowed) seeing mbr drives is normal, and NOT a bug.

Format the usb sticks with rufus using gpt partitioning, not mbr.

Outside the fact GB recommended CSM on, at least for the Ultra, or i wouldnt even know about it (ran all other ryzen rigs with CSM off).
 
Last edited:
Add me to the list mine is a X570 AORUS Master 3900x Samsung 970 EVO Plus NVMe SSD 2TB with 64 gb ram built in February and I have this issue six times over that period. The only thing that will fix it is to remove the battery and in my build entails a fair bit of work and time.
 
@crankcasy
If you decide to go efiflash path, can you please do backup of your current BIOS before flashing?

efiflash backup.img /S

will save current BIOS image to 'backup.img' file
It could be useful if we can compare the file with official bios file.
 
Last edited:
As soon as you change/limit the type of boot rom that is allowed (storage) to only UEFI, you should not see any MBR drive, even if CSM is enabled.

You'd think so, but even if I set CSM options to UEFI only on the boot page I still see all the MBR entries in the boot menu and priorities. So I don't know, unless I'm unaware of a setting somewhere else the BIOS is disregarding that selection. The only way to get rid of MBR entries is to disable CSM altogether. So regardless of how it should or should not work, it's an annoyance with the Gigabyte BIOS of which I have many.
 
you need to (re) format the usb drives with gpt partition to NOT see them, that wont change.
 
All of my boot media is GPT formatted. Are you like trying to defend Gigabytes poor BIOS implementation just to be contrary?
 
No, but seen too many ppl claiming problems just because they are not familiar/dont know details about things (like bios settings),
and like to exclude things like settings, to see whats the cause.

E.g. i've seen a "...tech for over 30ys.." with +10K post and pretty extensive knowledge about stuff,
yet didnt know what "other OS" setting in bios meant, or that you want it switched to win 8/10 when installing/using any newer win than 7.

Outside the fact i have not posted a single positive thing about GB bios on this forum,
so yeah, im here to defend GB, after having to deal with an effed up 300$ board for +7 month, right.
But then again, you still bought another GB board after having issues with GB (bios) before?
To me, the only way i can have "many" ...
 
I've replaced the CMOS battery with a new one, and enabled ErP. Now running F12g.

Haven't had the issue since, but I'm not confident yet that it'll stay that way. Let's see, fingers crossed.
 
Hi, I am observing this issue since last days of December. At first I was thinking it was Linux firware issue, then erp, then battery. Nope - nothing helped. It was geting even more and more often - at first once 1,5 month, lately evey 2 weeks or even more often.
F12g is not helping.
Now - thanks to nanite84 - I think its bios issue, I believe he may be right when saying that F11 is broken - because it fits to my timeframe just right - as first failure happened ~2 weeks after installing F11, not before. So it happens on F11 and later bioses up to F12G here.
I am using F10 now (it is first which should work fine with 1usmus power plan, if i am right) and - as everyone here - keeping my fingers crossed. If fail will happen on F10 - I will downgrade to F7B as suggested by nanite84
 
Last edited:
Hi, I am observing this issue since last days of December. At first I was thinking it was Linux firware issue, then erp, then battery. Nope - nothing helped. It was geting even more and more often - at first once 1,5 month, lately evey 2 weeks or even more often.
F12g is not helping.
Now - thanks to nanite84 - I think its bios issue, I believe he may be right when saying that F11 is broken - because it fits to my timeframe just right - as first failure happened ~2 weeks after installing F11, not before. So it happens on F11 and later bioses up to F12G here.
I am using F10 now (it is first which should work fine with 1usmus power plan, if i am right) and - as everyone here - keeping my fingers crossed. If fail will happen on F10 - I will downgrade to F7B as suggested by nanite84

I too think it's related to the BIOS, system was working perfectly fine up until Jan I think, around the time I updated the BIOS. I've started to believe that this BIOS issue causes some other issue that drains the battery very rapidly.
 
@crankcasy
try efiflash. so far the only problem i have left is the random reboots.

So far, I've not encountered the issue since following Fry178's EFIFlash procedure. Is it dumb luck? Did it actually fix anything? Will it return? Only time will tell. Honestly, I'd like to flash the new bios that's out but I'm curious to let my freshly flashed and untouched F12G sit for a while to see if this power on issue comes back. I am not seeing random reboots like him though. Maybe the random reboot aren't related to this particular issue.
 
So far, I've not encountered the issue since following Fry178's EFIFlash procedure. Is it dumb luck? Did it actually fix anything? Will it return? Only time will tell. Honestly, I'd like to flash the new bios that's out but I'm curious to let my freshly flashed and untouched F12G sit for a while to see if this power on issue comes back. I am not seeing random reboots like him though. Maybe the random reboot aren't related to this particular issue.
Same here but I'm using my fresh windows 10 1909 installation constantly since efiflash with clear DMI.
Shutting down PC and completely powering down the UPS over the night.
Planning to update to 2004 and latest drivers over the weekend and proceed with observation.
I'm certain that random reboots has nothing to do with this cold boot soft-brick issue.
My system is rock stable with dynamic Vcore set to -0.07500V (downvolted)
 
Have experienced the same issue on two X570 Aorus Master boards. Did an RMA on my first board since I didn't know about the CMOS battery trick. Have had the new board since May 15, working flawless until yesterday. Default settings in BIOS, only changed the XMP to get correct speed on the memory. Had BIOS F11 on both the primary and backup BIOS when this happened again. Updated BIOS yesterday to version F20a using the efiflash. Hopefully it will solve this problem. I will update here if it should happen again, but since I seem to be lucky enough to only experience this issue with months apart it might take a while.

Oh. and yes I have registered a report at Gigabyte so it will be interesting to see what they say.
 
Have experienced the same issue on two X570 Aorus Master boards. Did an RMA on my first board since I didn't know about the CMOS battery trick. Have had the new board since May 15...

Out of curiosity, how was the rma process, and did they indicate they “fixed” anything on your board or did you just get a replacement?
 
Update:
This morning it happened again.
Exactly 2.5 weeks after efiflash-ing F12g

I also wrote report to Gigabyte and they suggested to try with latest F20a BIOS.

Also their suggestion: (quoting)
"it could be some OS failure (try to boot some live linux distribution, shutdown PC and check, if this power on failure appears next day, if yes, please claim your motherboard at your dealer)"

... a shot in the dark ...

My opinion is that Gigabyte is definitely having issues with revision 1.0 boards.
Hopefully the revision 1.1 or new 1.2 are done right!
https://www.gigabyte.com/Motherboard/X570-AORUS-MASTER-rev-11-12#kf
 
Last edited:
Have anyone experienced this issue with F20 bios?
RMA for my board (X570 Pro) has been confirmed by Gigabyte. I'm waiting for the replacement.
Specialties: I used my board (F11 bios) with Ryzen 7 3700x processor and G.Skill memory with RGB LED G.Skill F4-3600C16D-32GTZNC G.Skill Trid . Every night I switch off my PC from wall outlet until next evening.
 
Back
Top