Der8auer Shares a Statement from ASUS in a Video Showing Damage to a Ryzen 7000 Non-X3D CPU

Peter_Brosdahl

Moderator
Staff member
Joined
May 28, 2019
Messages
9,032
Points
113
Well-known overclocker Der8auer has shared a statement from ASUS following reports about BIOS files mysteriously disappearing from its website. The discovery of the missing BIOSes happened while multiple users were reporting that their Ryzen 7000 X3D series were failing and when removed they could clearly see bulges on the PCB. There were numerous theories as to what happened and rather than add to them Der8auer reached out to ASUS on the matter. ASUS indicated that it implemented new protection and monitoring mechanisms in the latest BIOS and thus removed the older ones. It also said that manual Vcore control was available in the older BIOSes which was another reason they were removed.

Der8auer shares a discovery of his own towards the end of his video when reveals a Ryzen 9 7900X processor with the same damage. He too says that EXPO is likely the root cause but that this setting is not only common to ASUS but that the damaged 7900X had been installed on a GIGABYTE board. So in conclusion using EXPO with X3D and non-X3D Ryzen 7000 series CPUs could damage them if it increases the voltages too high.

See full article...
 
Yeah, if we were all honest with each other, especially to ourselves, *** we *** ALL know the TRUE reason to that... uhm... why! 😏
 
Asus does have a long history of "tweaks" to make their boards look better in benchmarks. Sending extra juice to the CPU, running the bus a few MHz higher, etc.
 
hubaduba said:
Asus does have a long history of "tweaks" to make their boards look better in benchmarks. Sending extra juice to the CPU, running the bus a few MHz higher, etc.
Click to expand...
I have an Asus B450-f Gaming with a Ryzen 2600x in it. When I first started running it I noticed temps seemed quite high. After some investigation it turns out the motherboard was pumping 1.2v into the SoC. After manually setting the voltage to 1.0v it ran a lot cooler with slightly better performance as the CPU could boost higher for longer periods under sustained all core loads. My son had the same board/CPU/RAM and his did the same thing. Since then I've been careful to check what all voltages are set at to make sure they are in a safe range and usually on the lower end of the safe range because I'm not trying to break overclocking world records.

I have a Gigabyte motherboard with a 5800x now and it has never set sky high voltages and the MSI Tomahawk and 5900x my son now has hasn't had any issues either.
 
I guess I should count it as blind luck I never tried an ASUS board. I haven't had anything against them it just that I've done well with the MSI and GIGABYTE boards I've had.
 
My current board is an ASUS TUF board and it's been rock solid. New chip new traces new limits... and I think it bit them in the ***.
 
I don't trust any of the major board manufacturers with firmware. Motherboard makers have been known to play fast and loose with voltages and default to overly aggressive settings that run afoul of official limits, and it wouldn't surprise me to learn of something similar happening here. The article also makes it seem like a plain old lack of QC is to blame. I don't think this would be the first time that the use of extended memory profiles (e.g. AMD EXPO/Intel XMP) to overclock RAM has been linked to premature failures, if that's indeed the cause.
 
hubaduba said:
Asus does have a long history of "tweaks" to make their boards look better in benchmarks. Sending extra juice to the CPU, running the bus a few MHz higher, etc.
Click to expand...
So does everyone else. Out of the "Big 3", MSI was the least guilty of doing this often running the bus at the CPU manufacturer's specifications. MSI still pulled the rest of the tricks to look better in benchmarks. All of the major manufacturers are guilty of cheating benchmarks and have done it for longer than I've been in the review business, which is 17 years at this point.

That being said, ASUS and GIGABYTE are the worst offenders. One trick they used to use is to take whatever your memory timings are and run them tighter than their XMP values despite reporting the correct timings listed in the SPD tables.
 
On the memory timing thing, I actually had ASUS admit that to me in person.
 
You must log in or register to reply here.
Become a Patron!
Back
Top