Tuesday, April 25th 2023

ASUS Releases Official Statement Regarding Ryzen 7000 Issues

ASUS has released an official statement regarding the recently discovered issues with AMD Ryzen 7000 series CPUs, especially the Ryzen 7000X3D series. ASUS has also released EFI updates on Friday, that include thermal monitoring mechanism to protect motherboards and CPUs and is working on new updates that should be available soon and define new rules for AMD Expo and SoC voltage, which appear to be the main issue related to the CPU VDDIO/MC voltages.

We have already covered the first reports of damaged Ryzen 7000X3D series CPUs that suffered physical damage, and some motherboard manufacturers have already released new BIOS updates, including MSI. In the meantime, Roman "Der8auer" Hartung has also discovered that the issue might not be just limited to the Ryzen 7000X3D series, but could also impact the Ryzen 7000 X-series CPUs. Although there were no earlier reports on such issues, AMD Expo appears to be the main source of the problem and users can either disable it or manually set the SoC voltage, at least until motherboard makers come up with new BIOS updates or we hear an official statement from AMD.
Sources: Roman Hartung (Youtube), via Videocardz
Add your own comment

37 Comments on ASUS Releases Official Statement Regarding Ryzen 7000 Issues

#1
Space Lynx
Astronaut
the fact that EXPO might have something to do with this burned chips is wild to me... all the Quality Assurance testing that had to be done to even make EXPO a thing, I just don't understand how this was possible. to my knowledge only happened to ASUS boards though, so I guess that is a good and bad, least it is not widespread
Posted on Reply
#2
Zubasa
Space Lynxthe fact that EXPO might have something to do with this burned chips is wild to me... all the Quality Assurance testing that had to be done to even make EXPO a thing, I just don't understand how this was possible. to my knowledge only happened to ASUS boards though, so I guess that is a good and bad, least it is not widespread
EXPO is still a form of overclocking, different DIY motherboards differs on the setting they run at.
There is quite a range of max voltage different brands apply, although anecdotally Asus tends to be more agressive with auto voltages.
Posted on Reply
#3
docnorth
At least they were fast this time.
Posted on Reply
#4
Space Lynx
Astronaut
ZubasaEXPO is still a form of overclocking, different DIY motherboards differs on the setting they run at.
There is quite a range of max voltage different brands apply, although anecdotally Asus tends to be more agressive with auto voltages.
yeah but for the last 20 years, if it said XMP and QVL for mobo, I knew like it was a fact of nature that i could turn that on and everything would be fine...

now you are telling me that is not the case... and my brain finds it very confusing. lol
Posted on Reply
#5
docnorth
Space Lynx...to my knowledge only happened to ASUS boards though, so I guess that is a good and bad, least it is not widespread
Let's hope so, but the usual pattern is that other vendors might be affected too.
Posted on Reply
#6
P4-630
So with other words, better buy intel if you plan to OC....
Posted on Reply
#7
Zubasa
Space Lynxyeah but for the last 20 years, if it said XMP and QVL for mobo, I knew like it was a fact of nature that i could turn that on and everything would be fine...

now you are telling me that is not the case... and my brain finds it very confusing. lol
Bascially both Intel and AMD has it in their warranty terms, that XMP / EXPO is considered OC an may void the warranty.
Of course neither of them actually tries to enforce those terms because it will be a PR disaster.
Most of the time the CS won't even ask, and if they do just play dumb and you be fine.
Posted on Reply
#9
muddymind
Space Lynxthe fact that EXPO might have something to do with this burned chips is wild to me... all the Quality Assurance testing that had to be done to even make EXPO a thing, I just don't understand how this was possible. to my knowledge only happened to ASUS boards though, so I guess that is a good and bad, least it is not widespread
No, it has happened with other motherboard vendors like gigabyte and MSI. See derbauer's video on it.
Posted on Reply
#10
piloponth
Why everybody is stating it is an SW problem (Expo, BIOS updates needed)?
To me it clearly looks like a problem with the LGA socket and poor engineering on AMD side with limited testing. These sockets starting to fail after some time has passed since shipping.
Posted on Reply
#11
Zubasa
piloponthWhy everybody is stating it is an SW problem (Expo, BIOS updates needed)?
To me it clearly looks like a problem with the LGA socket and poor engineering on AMD side with limited testing. These sockets starting to fail after some time has passed since shipping.
FYI AMD has been using LGA sockets in their server / workstation CPUs for a LONG time. All the way back since 2006.
So AMD all of a sudden forgot how to desgin an LGA socket is unlikely.
Posted on Reply
#12
Fouquin
Space Lynxonly happened to ASUS boards though
Nope, not only ASUS. Roman found damage to the CPU on a Gigabyte board as well.
Space Lynxyeah but for the last 20 years, if it said XMP and QVL for mobo, I knew like it was a fact of nature that i could turn that on and everything would be fine...
Oh boy is that ever not true. XMP has had its fair share of headaches (very flaky and consistently unstable on nForce 780/790i, VTT overvoltage on Z68, unstable timing profiles on early DDR3 XMP DIMMs, board-specific timing profiles, IMC overvoltage and chip damage on X99). XMP is also much newer than you think as it was developed for use with DDR3.
P4-630So with other words, better buy intel if you plan to OC....
Or don't panic due to a few failures, and keep enjoying your hardware. Just as nobody stopped buying 4090s due to the connector incinerating.
piloponthTo me it clearly looks like a problem with the LGA socket and poor engineering on AMD side with limited testing. These sockets starting to fail after some time has passed since shipping.
This is FUD and makes very little sense. AMD reused a successful socket design with slightly increased pin pitch. They've used this land array in some part or another since 2006 and it was never a problem. That's why these companies are looking at their UEFI setups for possible deviations in power allowance being at fault.
Posted on Reply
#13
Space Lynx
Astronaut
I never had a problem with my XMP on my 2500k setup back in the day. or any of my rigs until AM5, when I got unstable errors and gave up on AM5, even though the motherboard said it was QVL ram.

I guess I was just lucky until now, strange lol
Posted on Reply
#14
Zubasa
Space LynxI never had a problem with my XMP on my 2500k setup back in the day. or any of my rigs until AM5, when I got unstable errors and gave up on AM5, even though the motherboard said it was QVL ram.

I guess I was just lucky until now, strange lol
Yup an the QVL is written by the board vendors, an in the current gen price went up QA went down.
This is why I am pointing fingers at the board vendor this whole time, they seems to be the most likely curpits.
You even find some boards claiming "up to 6600+ support" while not having a single kit that is validate at that speed in the QVL.
Posted on Reply
#15
mb194dc
It's as simple as the small process node and cache on the chip can lead to massive current draw which melts the socket and the chip? Motherboard shouldn't have allowed to set voltage high enough to do that, so now they correct it.

The bigger point is why you'd bother overclocking 7800x3D anyway. Not much fun to be had with most of the latest higher end chips, they're pre overclocked pretty much.
Posted on Reply
#16
Karti
Space Lynxthe fact that EXPO might have something to do with this burned chips is wild to me... all the Quality Assurance testing that had to be done to even make EXPO a thing, I just don't understand how this was possible. to my knowledge only happened to ASUS boards though, so I guess that is a good and bad, least it is not widespread
Sadly not only to ASUS

IT was mentioned ASUS at first because first post of those 'Burned down" CPUs were on ASUS motherboards

but it also got confirmed on Gigabyte


While it was not confirmed on MSI - they did release a very "important" BIOS that... limits OC capabilities and limits voltage...
It feels more like board partners followed strictly steps from AMD - and those steps were hell wrong
Posted on Reply
#17
Bomby569
ZubasaEXPO is still a form of overclocking, different DIY motherboards differs on the setting they run at.
There is quite a range of max voltage different brands apply, although anecdotally Asus tends to be more agressive with auto voltages.
doesn't EXPO include all the relevant settings?
Posted on Reply
#18
Redwoodz
muddymindNo, it has happened with other motherboard vendors like gigabyte and MSI. See derbauer's video on it.
No. he doesn't count.
Posted on Reply
#19
Camm
I haven't seen but I wonder if this is an issue with the AMD AMP RAM kits & their associated profiles rather than the EXPO profiles for XMP / Mobo Vendor EXPO profiles.
Posted on Reply
#20
Fouquin
RedwoodzNo. he doesn't count.
His video is literally the source of this article.
Posted on Reply
#21
phanbuey
Some of these don't have EXPO enabled.



I don't think it's EXPO guys, the amount of heat/volts that needs to happen to bubble a chip like that is nuts, and should have kicked OCP/OTP on well before the chip had time to melt down like that. There's definitely more going on than just a buggy bios with expo.
Posted on Reply
#22
farmertrue
phanbueySome of these don't have EXPO enabled.



I don't think it's EXPO guys, the amount of heat/volts that needs to happen to bubble a chip like that is nuts, and should have kicked OCP/OTP on well before the chip had time to melt down like that. There's definitely more going on than just a buggy bios with expo.
DDR5 at speeds of 5800 would be considered either overclocking/EXPO or both.
Posted on Reply
#23
trparky
farmertrueDDR5 at speeds of 5800 would be considered either overclocking/EXPO or both.
What's base spec of DDR5?
Posted on Reply
#24
Suspecto
ZubasaFYI AMD has been using LGA sockets in their server / workstation CPUs for a LONG time. All the way back since 2006.
So AMD all of a sudden forgot how to desgin an LGA socket is unlikely.
Car manufacturers have been experiencing constant issues with their components and recalling them despite making cars for over a century. Every piece of hw is different and just because something worked in the past, doesn't mean that different new iterations will work without a problem in a future. This is a smaller socket in size, non-prosumer-focused with way higher frequencies and voltages and that alone is an entirely different challenge from the engineering standpoint. Making products without flaws is impossible and eventually, there will be a more severe one, that's just a matter of time and probability.
Posted on Reply
#25
R-T-B
phanbueySome of these don't have EXPO enabled.



I don't think it's EXPO guys, the amount of heat/volts that needs to happen to bubble a chip like that is nuts, and should have kicked OCP/OTP on well before the chip had time to melt down like that. There's definitely more going on than just a buggy bios with expo.
That AGESA is a version back. Wonder if it only happens on version 1.0.0.5 and below?
Posted on Reply
Add your own comment
Apr 26th, 2024 03:07 EDT change timezone

New Forum Posts

Popular Reviews

Controversial News Posts