Ryzen 3000 series support

Glad to help

A detail that nobody reported (despite of asking about it) is that L3.54 is AGESA 1.0.0.4 according to CPU-Z

I just installed this bios and I have this:
https://imgur.com/gX9BrNE

Sleep issues are very difficult to resolve because there are so many factors
that effect it. My Ryzen 1600x system would not wake from sleep for over 6
months after I got it, I had given up trying to fix it and then one day it just
started working. It turned out a windows update fixed the problem. My X570
system also won't wake from sleep, it just reboots and loads CMOS defaults.
After talking to ASRock Support for several days we came to the conclusion it
must be something to do with my RAM (two different Corsair kits) and the X570
chipset (my 3600 wakes from sleep fine on my X370 boards with the same RAM
kits). I eventually just gave up on sleep and use hibernate instead or just
switch off my system when I am not using it. Saves a lot of headaches,
especially when you can never figure out what is causing the damn sleep issues,
it could be anything from AMD and their AGESA to the motherboard to the RAM and
worse, Windows itself.

Yeah, had a lot of issues with sleep too, on multiple platforms with multiple brands.

Edited by Xaltar - 11 Oct 2019 at 1:04pm

Some facts about L3.54 beta bios (B450 ITX) in my rig:

1) CPU Stability: Prime 95 max power 16 threads stress test fails instantly (instead of after a few seconds like in previous BIOS). This is with stock settings and seems to be fixed by increasing the VCORE OFFSET by 50mv (could be my sample of 3700X).
> I had some crashes playing "Code Vein" something I never had before playing this game or any other game. But again this is probably related to P95 failing in 0 seconds instead 30-200 seconds.

2) dGPU is working at 8x (1.0/2.0/3.0) instead 16x (1.0/2.0/3.0)

3) Full UEFI: If I disable CSM (something I'd like to do but haven't been doing) I need to reset the CMOS or remove the NVME SSD drive because it fails to finish the post (screen shows properly though).
In previous BIOS i was able to boot without CSM but sometimes I was getting a frozen screen in the ASRock logo / POST.

Is still a beta BIOS of course. So maybe this will be fixed in the future.

I'm mostly worried about point 1. I don't know if my 3700X sample is bad or what. Anybody knows if raising VCORE +50mv is risky?

Ah ok, then I'll keep +50 VCORE Offset 24/7 (I'm not doing OC nor enabling PBO).

If I still have this issue with future AGESA in December or January...
Then I'll proceed to RMA with AMD telling them that my 3700X was never stable, that fails instantly with any stress test, and waited a few months before RMA to see if it was a BIOS issue or something.

Stock voltage = Prime 95 Max Power 16 threads fails in 0 seconds.
+50mv = Same test doesn't fails. At least tried for 5 min, I don't need more than that.