Monday, April 24th 2023

AMD Ryzen 7000X3D Processors Prone to Physical Damage with Voltage-assisted Overclocking, Motherboard Vendors Rush BIOS Updates with Voltage Limiters

AMD Ryzen 7000X3D processors are prone to irreversible physical damage if CPU overclocking is attempted at some of the higher VDDCR voltages (the main power domain for the CPU cores). A Redditor who goes by Speedrookie, attempted to overclock their Ryzen 7 7800X3D, leading to an irreversible failure. The motherboard socket and the processor's land-grid contacts, show signs of overheating damage caused by the contacts melting from too much current draw.

A Ryzen 7000X3D processor features a special CPU complex die (CCD) with stacked 3D Vertical Cache memory. This cache die is located in the central region over the CCD where its 32 MB on-die L3 cache is located, while the difference in Z-height of the stacked die is filled up by structural silicon, which sit over the regions of the CCD with the 8 "Zen 4" CPU cores. It stands to reason that besides having an inferior thermal transfer setup to conventional "Zen 4" CCDs (without the 3DV cache), the CCD itself has a higher power-draw at any given clock-speed than a conventional CCD (since it's also powering the L3D). This is the main reason why overclocking capabilities on the 7000X3D processors are almost non-existent, and the processor's power limits are generally lower than their regular Ryzen 7000X counterparts. Attempting to dial up voltage kicks up the perfect storm for these processors.
Igor's Lab posted a detailed analysis of the region of the Socket AM5 land-grid most susceptible to a burn-out in the above scenario. The central region of the LGA has 93 pins dedicated to the VDDCR power domain, dispersed in a mostly checkered pattern, toward the center of the land-grid. Igor isolated 6 of these VDDCR pins in particular, which are most prone to physical damage, as they are located in a region below the CCD that sees it sandwiched between the L3D (stacked 3D Vertical cache die), and the fiberglass substrate below. Apparently, AMD's thermal and electrical protection mechanisms aren't able to prevent a runaway overheating of the pins that causes the substrate to melt, deform, and bulge outward, resulting in irreversible damage to both the processor and the socket.

Meanwhile, AMD's motherboard partners are rushing to release UEFI BIOS updates for their entire lineups of motherboards, which enforce tighter limits on the VDDCR voltage. MSI is the first motherboard manufacturer with such updates. MSI, in a press statement, stated that it has redesigned automated overclocking for 7000X3D processors. "The BIOS now only supports negative offset voltage settings, which can reduce the CPU voltage only," the MSI statement to Tom's Hardware reads. "MSI Center also restricts any direct voltage and frequency adjustments, ensuring that the CPU won't be damaged due to over-voltage." On the other hand, the update introduces an automated overclocking feature called Enhanced Mode Boost, which optimizes PBO settings to improve boost frequency residency, without any manual voltage adjustments.
Sources: Tom's Hardware 1, 2, Igor's Lab, Speedrookie (Reddit)
Add your own comment

258 Comments on AMD Ryzen 7000X3D Processors Prone to Physical Damage with Voltage-assisted Overclocking, Motherboard Vendors Rush BIOS Updates with Voltage Limiters

#76
mahoney
oh look now even derbauer is reporting about the damage on his cpu

Posted on Reply
#77
Outback Bronze
mahoneyoh look now even derbauer is reporting about the damage on his cpu

Yeah, not just X3D's or Asus boards.

Edit: He seems to think it's coming mostly from running AMD Expo - SoC voltage.
Posted on Reply
#78
Why_Me
After reading about all the bugs that came with AM4 this latest news doesn't surprise me in the least. AMD is known for releasing unfinished products.
Posted on Reply
#79
trparky
Outback BronzeYeah, not just X3D's or Asus boards.
Then I might want to turn off EXPO mode on my 7700X.
Posted on Reply
#81
Dwarden
let's hope ASUS (and others) aren't removing ability to underclock / undervolt alongside, for X3Ds
that would be very sad,bad and facepalmish
Posted on Reply
#82
Outback Bronze
trparkyThen I might want to turn off EXPO mode on my 7700X.
How long have you been running it for?

He did test 1.5v on the SoC for half n hour and his CPU didn't kill itself but he's waiting for GN for more detailed info on this matter.
Posted on Reply
#83
Mahboi
MarsM4NI did read recently a post from a PC builder who said the same thing. :D Had tons of RMA's for ASUS boards, he's now only using mainly MSI boards, AsRock (2nd pick) & some GigaByte boards for his builds.

He also mentioned that MindFactory (a major German reseller) isn't selling ASUS boards anymore because dealing with all the RMA's was way too costly for them. Couldn't believe it, checked their homepage and sure enough, there is no ASUS mainboard to be seen, LOL.
How the mighty fall...I always saw ASUS as the highest tier of mobo makers. Never had any problems with them, and been using ASUS and Gigabyte since pretty much forever.
Posted on Reply
#84
Why_Me
MahboiHow the mighty fall...I always saw ASUS as the highest tier of mobo makers. Never had any problems with them, and been using ASUS and Gigabyte since pretty much forever.
It isn't limited to Asus boards ^^
Posted on Reply
#85
Mahboi
Why_MeIt isn't limited to Asus boards ^^
It's clearly them far more than others.

Now I have to find a new mobo master for my future builds. MSI maybe...?
Posted on Reply
#87
trparky
Outback BronzeHow long have you been running it for?

He did test 1.5v on the SoC for half n hour and his CPU didn't kill itself but he's waiting for GN for more detailed info on this matter.
I've been running EXPO mode on my system for about three months, ever since I built the system.
Posted on Reply
#88
Gmr_Chick
GicaBecause, in others, it is under an octocore (7700X and even 7700).
Because you know they have protections that prevent destruction.

A much better question: did AMD provide the correct technical data to motherboard manufacturers? It seems not.
I believe AMD did provide the correct tech data to the mobo manufacturers. The mobo manufacturers just don't give a shit. All they care about is that their board "won" in benchmarks/reviews.
Posted on Reply
#89
Outback Bronze
trparkyI've been running EXPO mode on my system for about three months, ever since I built the system.
6000Mhz on the mem?

If your keen i would take the cpu out and have a look see for the so called “burn mark” on the cpu . If you cant see anything your probably ok and this could be due to just bad batches or maybe even mobo related. More info is needed at this time.
Posted on Reply
#90
trparky
Outback BronzeIf your keen i would take the cpu out and have a look see for the so called “burn mark” on the cpu .
I really don't want to do that. That would involve cleaning the CPU and heatsink and applying new paste. Not at all something that I want to do on a Monday evening.

I've since disabled EXPO on my memory and from here on out, hope that nothing bad happened.
Posted on Reply
#91
Super Firm Tofu
trparkyI really don't want to do that. That would involve cleaning the CPU and heatsink and applying new paste. Not at all something that I want to do on a Monday evening.

I've since disabled EXPO on my memory and from here on out, hope that nothing bad happened.
You could have also just checked what EXPO had set your SoC voltage to and adjusted down if needed.

Setting EXPO for my 7700x (at launch) on an MSI X670 pushed SoC to 1.35v. I adjusted it back down to default (1.02v) and ran that way without issue until I sold it.
Posted on Reply
#92
trparky
Super Firm TofuSetting EXPO for my 7700x (at launch) on an MSI X670 pushed SoC to 1.35v. I adjusted it back down to default (1.02v) and ran that way without issue until I sold it.
I will check that. Be back in a few.
Posted on Reply
#94
trparky
I’m going to assume that the setting I’m looking for is VCORE SOC?

Edit: Posting from my phone.
Posted on Reply
#95
Super Firm Tofu
trparkyI’m going to assume that the setting I’m looking for is VCORE SOC?

Edit: Posting from my phone.
I'm not sure what Gigabyte calls it, but that sounds right.

I'd set EXPO, boot into windows and see what SoC shows using HWinfo. Then compare to what disabling EXPO shows.

*edit*

Found one of my old posts. Here's where it shows up on an MSI board in HWinfo

Posted on Reply
#96
trparky
OK, this is with EXPO enabled.
Posted on Reply
#97
Outback Bronze
trparkyNot at all something that I want to do on a Monday evening.
Yes ofc, in your leisurely time. Sounds like your cpu is fine.

Would just be interesting to have a look see that's all.
Posted on Reply
#98
trparky
And this is after putting the CPU through its paces with a C/C++ project compile using Microsoft's Visual Studio C++ Compiler. Yes, EXPO was enabled.

It looks like VDDCR_VDD reached a maximum voltage of 1.377 volts while VDDCR_SOC only reached 1.245 volts.

A ran a Cinebench R23 test and the only change was that the VDDCR_SOC only reached 1.246 volts.
Posted on Reply
#99
tabascosauz
Outback BronzeHow long have you been running it for?

He did test 1.5v on the SoC for half n hour and his CPU didn't kill itself but he's waiting for GN for more detailed info on this matter.
The problem with derbauer drawing conclusions from that "test" is that you actually need to be running current through the appropriate rail to see whether VSOC is safe for VDDCR_SOC. All Cinebench is doing is running current through VDDCR_CPU. I can just as easily set a ridiculous VSOC on my APUs and call it tentatively "safe" by running CPU-only loads for an hour.

Same idea why 1.5V Vcore on Zen 2 and later is "safe" in the hands of the stock Precision Boost algorithm, but certainly unsafe once you manually punch in the same 1.5V in the static Vcore box.

Traditionally the only Zen CPUs that can actually run a shitload of current through SOC are the APUs. On AM5 you can also get close since it has an iGPU, but that involves OCing the shit out of the iGPU and running heavy 3D load on the iGPU, something derbauer definitely didn't do by running Cinebench. Through normal memory controller loads alone, the most you can hope for is maybe 20W-ish flowing through SOC.

Given the time he spends around these platforms I thought he'd know some basic fundamentals of Ryzen topology.

1.35V is pretty standard for an EXPO VSOC auto-rule and people have been one-click set-and-forget for months with no problems; I honestly doubt that's what's going on, probably something else in the firmware given Asus' statement and pulling all their old BIOSes.
Posted on Reply
#100
Hxx
sethmatrix7What facts? All you’ve said is it’s ok to void the warranty. It’s ok to fuck up the CPU by using settings the end user doesn’t have any grasp of.

Here’s a fact you can’t understand- modifying the CPU outside of manufacturer specifications voids the warranty. This includes EXPO overclocking.
Except the fact is that you dont know which BIOS options put this chip outside the manufacturer specifications LMAO . And why would you ? Cause they dont tell you in BIOS. You have to figure that out yourself - reading forums, footnotes, fine print in random places, etc. Sure you can figure out at some point hopefully before you fry your chip. Thats the mammoth in the room here lol. The whole point of this whole conversation that there has to be clear guidance to the end user . Also dont exclude EXPO.
Posted on Reply
Add your own comment
Jun 17th, 2024 14:43 EDT change timezone

New Forum Posts

Popular Reviews

Controversial News Posts