News Preventing Ryzen Burnout: Motherboard Makers Issue New Firmware

[Admin]

AMD motherboard partners have quickly implemented SoC voltage-limiting ups to keep your processor safe. Most vendors have new BIOS downloads available today.

Preventing Ryzen Burnout: Motherboard Makers Issue New Firmware : Read more

 

[zecoeco]

Unsurprising consequence when overclocking.
Just don't overclock and you're safe (hopefully).

 

[edzieba]

Unsurprising consequence when overclocking.
Just don't overclock and you're safe (hopefully).

There's a big difference between
"Intentionally overclocking and overvolting a CPU with a user-selected voltage above a certain threshold may damage your CPU"
and
"Enabling the DIMM-manufacturer-provided timings via the CPU-manufacturer-provided tool (EXPO) may silently set CPU voltage above a dangerous threshold without user intervention or knowledge"

Particularly as consumers have been used to a decade and a half of enabling an analogous feature (XMP) without the same CPU overvoltage risk.

 

[bourgeoisdude]

Asrock? They released some beta bios updates yesterday for this,

 

[hotaru251]

if it was just SOC I have a feeling AMD wouldn't of been so "we accept the fault" as EXPO/XMP is seen as OCing and void warranties.
I am type who always checks what setting is applied to pre-set stuff (as i know sometimes companies are way to generous on voltage or not generous enough)
Usually i'll see the EXPO/XMP settings, copy em down, manually tune em to that and lower voltage if its over what I think is needed.

 

[Brian D Smith]

They only ever refer to the new "3D" chips...

 

[RedBear87]

Unsurprising consequence when overclocking.
Just don't overclock and you're safe (hopefully).

It is fairly surprising when using one click solutions, normally they. I don't remember seeing CPUs literally exploding because of XMP. At worst I've read about people seeing faster than usual degradation because of PBO, but that's not really the same thing. Either AMD or its partners didn't do their homework well enough if they allowed something similar to happen.

 

[AdamB15]

ALL of Biostar's available BIOS files are from 4/13...So either they foresaw this and already limited the voltages, they did it by accident/chance, or their statement is false and their fixed bios is still incoming, right?
Imagine if they foresaw this issue lol

 

[laxman10100]

I guess ASRock hasn't been on the list of manufacturers to release a BIOS update addressing this issue because their boards didn't push the SoC_Voltage over 1.3V with EXPO enabled. In my experience with the Taichi Carrara, the EXPO settings only pushed SoC_Voltage to 1.3V. It does, however, push the Memory_Controller_Voltage to 1.35V or 1.4V with an EXPO profile enabled...

I wonder if this could cause the same problem - as AMD didn't state that the MC_Voltage being over 1.3V would cause this issue.... They just stated that the SoC_Voltage being over 1.3V would cause this issue...

 

[TerryLaze]

ALL of Biostar's available BIOS files are from 4/13...So either they foresaw this and already limited the voltages, they did it by accident/chance, or their statement is false and their fixed bios is still incoming, right?
Imagine if they foresaw this issue lol

The video of derbauer frying a x3d chip in asia is from 3/19 so the industry knows that there is something going on at least since then but realistically they should know stuff well before we find out.
Maybe they thought that it would be much rarer or something.

View: https://youtu.be/FVkTGq7brP4?t=1374

 

[TerryLaze]

I wonder if this could cause the same problem - as AMD didn't state that the MC_Voltage being over 1.3V would cause this issue.... They just stated that the SoC_Voltage being over 1.3V would cause this issue...

Isn't the SOC_Voltage the "all including" voltage and the mem controller a part of the soc?

 

[laxman10100]

Isn't the SOC_Voltage the "all including" voltage and the mem controller a part of the soc?

I suppose they are on motherboards from other companies. But the ASRock motherboards all have separate settings for this. They have SoC_Voltage and two SoC_MC_Voltage settings.

Perhaps ASRock delivers power to the vcore through a separate electrical path than what they use to deliver power to the memory controller? If that's the case, then I hope it would still be safe to allow for 1.4V to the MC on their motherboards... But I am not sure yet as I am not too familiar with ASRock boards. But, I will try lowering the MC_Voltage settings both to 1.3V tonight to see if my system still posts and works reliably with some memory and gaming tests.

Essentially, whenever I throw together a system on a platform I'm not too familiar with, I spend hours researching what the best motherboard would be for performance and reliability - and that is what lead me to purchasing the ASRock board that I recently did. I've been rocking with Intel stuff for the past decade and so I wasn't familiar with what would be best for AMD setups... I knew to stay away from Asus as they have had an absolute ton of manufacturing and reliability issues with their Intel boards for the past 2-3 years. MSI was creeping around the corner as the next big board company - but they were fresh to the turf with AMD boards as far as I knew at the time.

Anyway, my apologies for digressing here. I'm a huge geek over computer stuff and write too much stuff online because nobody understands this hobby IRL lmao.

Edit - I just reached out to ASRock support asking if their separate CPU SoC Voltage and Memory Controller SoC Voltage settings would prevent this issue from occurring. Granted, I have also noticed that they released a new BIOS version this morning (4/27/23) with a description stating that 7000X3D CPU users should upgrade to it... So, even though AMD didn't say so, perhaps the ASRock boards could've been affected by this issue as well... I suppose I'll find out tonight (once I flash the BIOS upgrade) or once someone from their support department responds to my support question.