X670E Ace Bios Problems

B

BiscuitBomber

Active member
Joined
Oct 19, 2022
Messages
167
Hardware:
Processor: 7950X
Motherboard: MSI X670E Ace
GPU: EVGA 3080 TI FTW3
MEMORY: G.SKILL Trident Z5 Neo RGB Series AMD EXPO 32GB (2 x 16GB) 288-Pin PC RAM DDR5 6000 (PC5 48000)
PSU: EVGA 1000W P6
Cooler: 360 Phanteks AIO MPH

After updating my BIOS today on my PC. When booting up I was getting BIOS Code A6 and the RAM light was staying Amber. This didn't occur on the previous BIOS 7D69v13. I attempted to clear CMOS but same issue. I ended up downgrading my BIOS back to 7D69v13. I opened a ticket with MSI, but just a heads up in case anyone else runs into this. Before 7D69v142 I had zero issues with RAM and bootup problems.


----------------
EDIT Title
Alan J T
 
Last edited by a moderator:
Sort by date Sort by votes
not a fan of pbo boost
plain -25PBO for me
or 103-105 eclk if you want more mhz
maybe also learning how to tweak curve shaper

BTW, have you tested thoroughly while using boost if the windows scheduler is alright? cause on 7950x3d it wasnt if you remember
I'd just stick with the pbo negative curve only (not for this particular scheduler reason) cause the performance improvement coming from just the pbo boost is negligible
Increasing the clock with eclk is much more impactful in terms of performance gains per mhz
maybe look at skatterbencher site to start with
 
Last edited:
Upvote 0 Downvote
My PC is very very very stable now.
I have tried everything for days, it works, not a crash, not an error, not a failed boot but...

if fails to boot if I restart the windows in UEFI with the command:
C:\Windows\System32\shutdown.exe /r /fw /f /t 0

or if I try to choose the boot device at boot by pressing F11.
I generally start pressing F11 when I see the B4 stage, I noticed that if I press the F11 button too much times it hangs at boot.
It hangs with F3 error code.

This is so weird, it hangs only if I try to reboot to UEFI or if I try to select the boot device at boot.

Never seen something like this.

This happens randomly.
 
Upvote 0 Downvote
hangs without any code in particular?
did you already forced disabled mcr/pd?
if it's sporadic, are you using any particular usb drive/device connected from time to time?

seems like a variant of the expo/reboot problem
 
Upvote 0 Downvote
Michele P said:
hangs without any code in particular?
did you already forced disabled mcr/pd?
if it's sporadic, are you using any particular usb drive/device connected from time to time?

seems like a variant of the expo/reboot problem
Click to expand...
No I haven't disabled MCR/PD because the boot seems perfect if I don't try to access UEFI.
When the problem happen it hangs on F3 error code with the green light turned on
 
Upvote 0 Downvote
sblantipodi said:
No I haven't disabled MCR/PD because the boot seems perfect if I don't try to access UEFI.
Click to expand...
F3 seems an early stage error, very easily caused by memory, but I've never seen that code on my system
I've been using the F11 boot menu pretty often recently (spamming the key a lot), but not the windows to uefi shortcut.
Also curious to know if it happens with or without a previous code 15 training phase

uhm, so no excess time stuck on 64 (caused by usb devices) or the typical failed training trinity 15/C5/Ab
I would try that, both those settings disabled and all usb devices removed, except mouse/kb.
The worst that can happen is losing few seconds in boot times: everything else should improve, in particular stability and latency (you can compare with your previous AIDA latency tests)
The fact you see PD disabled in zentimings could mean the mb is not using MCR most of the times, even if you say the boot feels already fast
But leaving Auto may trigger MCR under certain circumstances (a reboot to uefi from windows seems a valid trigger for example)
 
Last edited:
Upvote 0 Downvote
Guys.
You shouldn't use Latency Killer.
Tests have shown, that it only gets you better latency in AIDA, but outside of benchmark environments, the performance decreases.

Only the efficiency suffers, and you even get slightly worse FPS in games.
 
Upvote 0 Downvote
Lubix said:
Guys.
You shouldn't use Latency Killer.
Tests have shown, that it only gets you better latency in AIDA, but outside of benchmark environments, the performance decreases.

Only the efficiency suffers, and you even get slightly worse FPS in games.
Click to expand...
can you show us those tests please?
 
Upvote 0 Downvote
I can confirm that latency killer kills performance.
www.pcgameshardware.de

Latency Killer im Test mit Ryzen 7 9800X3D: Finger weg!

Der Latency Killer soll die Leistung verbessern, doch stimmt das überhaupt? Wir testen mit 19 Benchmarks und Ryzen 7 9800X3D!
www.pcgameshardware.de www.pcgameshardware.de
My system seems pretty stable and fast but I still have a weird issue.

I can't control Power Down, no matter if I enable memory context restore or disable it.
Power down is always disabled (in both zen timings and ryzen master),
even if I manually enable it.

That's very weird. How can this be possible? Am I the only one with this problem with ACE and 9000 series?
 
Upvote 0 Downvote
Any test in combination with the x3d gaming mode?

I suspect the latency killer option affects smt in some ways (cause smt influences a lot latency in amd cpus)

It’s not unexpected that you lose something in pure raw power when you decide to prioritize just the latency.

I would search for latency killer improvements in scenarios/titles where also the x3d gaming mode shows positive effects (remember that x3d mode disables smt entirely)

I agree it’s better to leave those options untouched, unless you are 100% sure it affects positively your typical pc workload (rare)
Or you just want to take a nice aida latency screenshot.

sblantipodi said:
That's very weird. How can this be possible? Am I the only one with this problem with ACE and 9000 series?
Click to expand...

Most likely it's a zentiming bug with recent agesa updates

EDIT: yes it is, in their discord you can find multiple reports of PD not being detected correctly since 1.2.0.2 agesa
Probably they need to update some registry readings (and ryzen master too)
 
Last edited:
Upvote 0 Downvote
yes, probably it needs a fix
just like the mem VDD voltage after enabling high memory voltage support (shows a wrong 0.0000V)
or you can try a beta release, always in their discord dev channel

F1LQyS9.png
 
Upvote 0 Downvote
The weird thing is that memory context restore works well with power down disabled now, but Gear down mode disabled does not boot at all.
The complete opposite of 7950X3D
 
Upvote 0 Downvote
Probabily GDM off needs few voltage adjustments
You are running different and generally lower voltages on the new cpu (vdd/vddq/vddio/vddp/soc, but maybe more vddp dependent in this case)

MCR/PD correlation could also have changed cause AGESA refinements.
Previosly mcr on with pd off was a no go: windows bsod was almost cartain with that combo (maybe somebody was luckier than others)
But things are constantly evolving
I still think the optimal way to do things is mcr and pd disabled, for more stability, robust training, consistence, and better latency.

If PD is no longer needed for mcr that’s a good news, cause the latency penalty comes mostly from PD being enabled

But without zentimings or other updated tools being able to detect the correct status, idk
 
Last edited:
Upvote 0 Downvote
Probably it's AMD that mixed the cards with new agesa and the old registry entry where was stored PD status now shows useless info.
While the real PD is stored elsewhere
When you have 2 layers of menu (Vendor/AMD) it's something that may happen
 
Last edited:
Upvote 0 Downvote
hi there, the system is very stable here but from Windows events I can see this error at boot:
The l1vhlwf service was not started due to the following error:
Hypervisor feature not available for the user.
Click to expand...

do you have the same error? what is it?

I see even this error on boot:
Resource Publication Service for Discovery stopped with the error:
One or more arguments are invalid.
Click to expand...
 
Upvote 0 Downvote
Back
Top