Thats a wise decission.
Basically its allways smart to wait wenn a new bios popup,
and see if users are reporting any issues with it.
Next to that if you dont have any issues with the current bios version.
Then probablly its kinda better to stay on it for a little while longer.
1 Like
Exactly! Waiting to install BIOSes doesnāt hurt in general. That being said, on AM4 I still want new ones because of new AGESA updates and the related improvements 
I tried the 4.80 bios last week, bad things happenedā¦
For some reason it would not allow me to boot from the NVMe in the M2_2 slot.
I tried several things without luck- switched to M2_1, moved stick to an add-in card.
I even booted to a sata drive, formatted the stick and re-installed the OS (ubu mate) but no luck.
I had gone from 3.20 to 3.30 to 4.80.
I also changed a couple of other things at the time, including a new power supply (thought I had fried something on the main board with the new PSU. Was not fun)
It could be because I jumped a couple of bios versions, but rolling back to like 4.64(or whatever intermediate version there was) before throwing in the towel and going back to 3.30.
Would love to hear from others who could update who were booting from NVMe and had no issue?
Like I said, I changed a few things at the same time, so probably did something wrongā¦ But 3.3 works 
I just watched Buildzoidās PCB review and towards the end he states that basically everything except Level5 LLC isnāt really useable. Now, this vid is over a year old and I was wonderimg whether this is still the case? Has ASRock been able to alleviate this issue with a BIOS update?
In case ppl are still interested, there is a new BIOS for this board available:
As usually, Iāll wait a while before giving this one a try 
2 Likes
Yeah i tried 5.10 and it seems it started initializing the second slot PCIe GPU at boot.
Now i canāt use my second slot Vega64 for PCIe passthrough. Does anyone know how to disable this?
The new bios is better in almost all other ways.
Memory training bugs have disappeared, memory bandwidth bug is gone.
The only strange thing except for gpu initialization i have found is that it sets vcore to -0.10v by default witch which most often makes systems unstable. Easy to fix if you are aware of it.
2 Likes
Is it only me or is the CSM option gone from 5.10?
Thanks! Iāll still wait for a while thoughā¦
Interestingā¦
So have anyone successfully passed through a Vega with BIOS 5.10?
p state oc seems to work now
auto from main page
pstate 0 clock speed
offset voltage from main page
working 
both clock speed and voltage drops as they should, fucking finally
ā
left boost on, i think there was such setting? but since its working, ill let that be
Regarding my post earlier considering the mainboardās LLC levels I discovered a comment where a user states (and buildzoid recognises) that he made an error when measuring. Hence, the default LLC of 3 should be fine.
However, to you guys think keeping the LLC at 5 (the lowest setting) will increase the longevity/health of the CPU?
I seem to remember that when purchasing this mainboard there was something mentioned of an āASRock crashless BIOS featureā:
āThe ASRock Crashless BIOS feature is a safety feature that is used in the Instant Flash UEFI/BIOS update process. It allows an interrupted or failed UEFI/BIOS update process to be restarted over again, to be completed correctly. As long as you do not turn off the PC at any time during the UEFI/BIOS update process, and leave the storage media with the UEFI/BIOS update file in the PC, the Crashless BIOS feature should be able to recover from a failed or interrupted update.ā [1]
As of now, I am unable to find said feature listed anywhere. Is this feature removed from this board, e.g. is this chrasless BIOS update not so crashless after all?