• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

AGESA 1.0.0.7a /1.0.0.7 BIOS (AM5), system stability and VSOC

Soldato
Joined
30 Jun 2019
Posts
8,098
If my AM5 system is unstable with an AGESA 1.0.0.7a BIOS when the DRAM runs at 6000 MT/s, is that likely to be caused by the SOC voltage being limited <1.3v on the new BIOS versions?

If so, does enabling the liquid nitrogen mode (LN2 mode) allow the VSOC to be set at 1.3v or higher, and is that likely to result in a stable system?

Do we have any idea yet at which point the SOC voltage can cause Ryzen 7000 series CPUs to burn out /break?

My current AGESA 1.0.0.4 (beta) BIOS defaults to ~1.25 SOC when EXPO @ 6000 MT/s is configured. I've been running it at 1.3v for ages now (LLC at default), as I'm fairly sure this was needed for stability.

EDIT - I'm stability testing 1.25v as Anandtech said this was enough for their memory testing benchmarks (they tested with the latest BIOS version), with DRAM running @6000 MT/s "We eventually settled on 1.25 V on the SoC for this kit and our Ryzen 9 7950X3D, and we found stability in memory-intensive benchmarks was solid". Link:

Gaming Nexus also said that "one thing remains abundantly clear: the issue is not just one that relates to SoC voltage" - Regarding CPUs getting burned up.
 
Last edited:
A key takeaway from Anandtech's article is that the SOC current is significantly higher with EXPO profiles enabled on all AGESA BIOS versions, and the lowest EXPO SOC (average) current was possible on AGESA 1.0.0.7 (both with the SOC at default and at 1.25v):

ASRock%20X670E%20Taichi%20AGESA%20Average%20Current%20ALL%201.png


Here is their advice for Ryzen 7000 series CPUs "the large current and heat are accelerating dielectric degradation and are not only weakening the integrity of the silicon and the internals but it's effectively damaging them beyond a point of no return. This is why it's important to operate with lower voltages which in turn lowers current, lowers total power output, and in turn, lowers temperatures".
 
Last edited:
I'm curious, has anyone else had stability issues with the newer AGESA versions (e.g. 1.0.0.5 -1.0.0.7a)? But had no issues with older versions such as 1.0.0.4? At the time I upgraded to this version, somebody on another forum told me that they'd made changes to how CPUs were powered with each BIOS version, so I assume that's that reason, but still don't know exactly why :confused:

Did increasing the CPU voltage via PBO + curve optimiser make any difference to your system stability? I think it's possible to increase it by 50-100mv on all cores.
 
Last edited:
SOC should not exceed 1.30 MAX ( CPU VDDSR_SOC ( SVI3 TFN ) on HWINFO 64 ), anything more and you are into 'dangerous' territory.

That is the reason all mobo vendors and AGESA reduced it to that, following on from the testing of damaged cpus'.

AGESA 1.0.0.7a ( the current one ) has been stable for me.
 
Last edited:
I suppose the issues I've had could be caused be an incompatibility issue with the RAM (Fury Renegade 6000 MT/s CL 32), but for some reason it's fine on AGESA 1.0.0.4.

1.25v SOC voltage works fine also.
 
Last edited:
AGESA 1.0.0.7a means I can no longer run EXPO Tweaked on my X670E Hero :( just have stability issues. Was absolutely solid before that.
 
The RAM I have is specced for CL32 with the 6000 MT/s XMP profile enabled, I wonder if that is what caused the issue on 1.0.0.7a?
 
1.0.0.6 was still solid with EXPO Tweaked. Never tried 1.0.0.7 as I was waiting for it to come out of beta then it ended up being straight to 1.0.0.7a
I suppose this might be worth a try, but at this point, I might as well wait for when I next upgrade my processor! Not sure yet if I will upgrade to an Zen 5 CPU in 2024, I imagine I will stick with a 7700X as long as it can handle 60 FPS without performance dips.

I regret thinking that it would be a good idea to upgrade to the latest stable BIOS version, before upgrading my graphics card (I thought this would give me the best possible performance).

There doesn't seem to be a risk involved with just continuing to use AGESA 1.0.0.4, especially with VSOC at 1.25v.
 
Last edited:
This might explain some of the initial problems I had when I initially set the system up.

The memory is called Kingston FURY Renegade 32GB (16GB x 2) 6000MT/s, and the model code is:
KF560C32RSK2-32

According to the current QVL list, this RAM has been supported since BIOS 1.01, for both XMP and OC:

I'm sure that ASRock said it wasn't supported at the time though, and the BIOS version would've been greater than 1.01.
 
Last edited:
Well there's a possibility that the new AGESA may solve the issues I faced when I tested other recent BIOS versions (but not the 1.0.0.4 beta AGESA version), which only occurred when running at 6000 MT/s:


Maybe I'll give it a try at some point. There seems to be a new BIOS for my board that "Update(s) AGESA to 1007b".

It will be interesting to see if they can add official support for higher speed DDR5 modules also (> 6000 MT/s), my dad will be interested in buying the highest speed DDR5 modules for a new system.
 
Last edited:
Back
Top Bottom