Thursday, September 3rd 2026

AMD Ryzen 7 7800X3D Swells and Burns Again, This Time on an ASUS Crosshair X670E Hero

Another Ryzen 7 7800X3D has reportedly swollen and burned, while also damaging the ROG Crosshair X670E Hero motherboard in the process. UNIKO's Hardware flagged the case, citing a video posted to Bilibili as the original source. The video shows a Ryzen 7 7800X3D CPU with clear swelling on the underside where it made contact with the CPU socket. There is discoloration on both the CPU and the socket itself. Moreover, the socket pins are warped and damaged in the middle, presumably due to the pressure of the swelling. One peculiar finding is that one of the notches on the CPU socket looks slightly off. Perhaps more importantly, the board's BIOS version wasn't confirmed, which is directly relevant due to AMD's proposed fix.

This kind of failure first showed up in April 2023, when several 7800X3D and 7950X3D chips bulged and took their sockets with them on X670E boards. AMD pinned it on SoC voltage running too high and rolled out AGESA firmware that capped it at 1.30V across AM5 motherboards. This new case might be straightforward if the BIOS hadn't been updated at all. However, even after the BIOS fix, the reports never fully stopped. Just weeks ago, a 7800X3D reportedly burned out on an MSI PRO X870-P WIFI board running the latest BIOS, with EXPO enabled and nothing else touched. We have even seen reports of Ryzen 7 9800X3D CPUs burning in the exact same manner. If you're running a 7800X3D, it's probably worth checking that you're on the latest BIOS and keeping an eye on SoC voltage.
Source: UNIKO's Hardware on X
Add your own comment

160 Comments on AMD Ryzen 7 7800X3D Swells and Burns Again, This Time on an ASUS Crosshair X670E Hero

#51
Jtuck9
I noticed on the latest beta Agesa that auto VDDIO voltages won't even display in Zentimings anymore. 1.37 volts VDDIO set manually in BIOS corresponds to 1.408 volts read in HWinfo. The default DOCP used to be 1.4 in BIOS, which is surely pushing toward 1.45 > and beyond safe daily limits. I've had fluctuating memory voltages for a while, despite being set in a variety of places in BIOS. I started having the classic ASUS "drive disappearing" on reboot issue, which now doesn't boot at all, so the board (x870i) has gone back as RMA. We'll see what the result of that is...
Posted on Reply
#52
Paganstomp
Well it a good thing that CPU production run is EOL.
Posted on Reply
#53
Dragam1337
freeagentNo it was from a few revisions back. I kind of feel like a schmuck because this version is actually pretty decent.




I have it running at 1.2v with 4 sticks and tight timings, I will try a little lower..

Edit:

VDD and VDDQ are indeed 1.45v weird how it did that lol..
You can likely get your voltage lower without stability issues if you disable power down.

But generally speaking 3d chips are not for people who wanna push stuff to the limit - then stuff like the article happens. You wouid probably be happier with intel, where you have to fiddle with everything all the time to get decent performance, than a plug and play 3d chip :p
Posted on Reply
#54
AusWolf
kazuvikingMaybe there is a reason intel went with 2d cache for the new novalake and not on top of the silicon.
It's still strange that this only seems to happen with the 7800, and not the 9800 or 9850, or even 5700-5800X3D.
Posted on Reply
#55
Dragam1337
AusWolfIt's still strange that this only seems to happen with the 7800, and not the 9800 or 9850, or even 5700-5800X3D.
They probably introduced a weakness with 7800x3d, which they identified and fixed with 9800X3D.

But i am yet to see it happen on a 7800x3d running stock settings, always people running iffy settings.
Posted on Reply
#56
RinSenpai
AusWolfIt's still strange that this only seems to happen with the 7800, and not the 9800 or 9850, or even 5700-5800X3D.
It happened with the 9800X3D, especially with ASRock, but we don't have enough data/time since release about the 9850X3D. Most Zen5 burnt CPUs are on ASRock but it happened on pretty much every board within the 1st year.
Zen4 was most notorious with ASUS, though it's only happening on X3D for the most part.
5800X3D was the first version, so maybe their tolerance margins were much higher to ensure it works.
Posted on Reply
#57
kazuviking
AusWolfIt's still strange that this only seems to happen with the 7800, and not the 9800 or 9850
It happened to the 9800 and to the 9850 as its the exact same silicon just better binned. My hyphotesis is this is an issue with expo profiles not having correct voltage table in the spd chip that causes mobos to load different voltages and some sets the agesa maximum.
The x3d chip is extremely sensitive to voltage and to electron migration. I wouldnt be surprised if the cache gets destroyed in 6-8 years making the cpu an expensive paperweight.
Posted on Reply
#58
AusWolf
kazuvikingIt happened to the 9800 and to the 9850 as its the exact same silicon just better binned. My hyphotesis is this is an issue with expo profiles not having correct voltage table in the spd chip that causes mobos to load different voltages and some sets the agesa maximum.
The x3d chip is extremely sensitive to voltage and to electron migration. I wouldnt be surprised if the cache gets destroyed in 6-8 years making the cpu an expensive paperweight.
But EXPO doesn't touch the core voltage, and other voltages don't affect the CCD, right? If EXPO was really at fault, then I don't see why it only affects X3D CPUs.
Posted on Reply
#59
Dragam1337
AusWolfBut EXPO doesn't touch the core voltage, and other voltages don't affect the CCD, right? If EXPO was really at fault, then I don't see why it only affects X3D CPUs.
Expo changes vsoc voltage (if you have it on auto), which affects the entire chip.
Posted on Reply
#60
Grimling
This kind of failure first showed up in April 2023, when several 7800X3D and 7950X3D chips bulged and took their sockets with them on X670E boards. AMD pinned it on SoC voltage running too high and rolled out AGESA firmware that capped it at 1.30V across AM5 motherboards.
The images clearly show the damage is from the CCD and not related to the SOC.
The bulge is directly on the spot of the CCD chip.
Posted on Reply
#61
Jtuck9
GrimlingThe images clearly show the damage is from the CCD and not related to the SOC.
The bulge is directly on the spot of the CCD chip.
I wonder what VDDG voltages it was running?! Default on my board / gubbins was 1050mV for both IOD and CCD...
Posted on Reply
#62
freeagent
Dragam1337You can likely get your voltage lower without stability issues if you disable power down.

But generally speaking 3d chips are not for people who wanna push stuff to the limit - then stuff like the article happens. You wouid probably be happier with intel, where you have to fiddle with everything all the time to get decent performance, than a plug and play 3d chip :p
Intel chips dont die, but AMD does..

I dont knock you for running a stock system, why are you giving me a hard time for tuning my system?
Posted on Reply
#63
JustBenching
freeagentIntel chips dont die, but AMD does..

I dont knock you for running a stock system, why are you giving me a hard time for tuning my system?
Don't you think it has to do more with how bigger the marketshare of x3d is compared to normal ryzen chips? People have a hard time understanding statistics. If x3d outsells normal ryzen chips 10 to 1 then the melt cases will also be 10 to 1. I don't see why the added cache would make them extra spicy.
Posted on Reply
#64
Dragam1337
freeagentIntel chips dont die, but AMD does..

I dont knock you for running a stock system, why are you giving me a hard time for tuning my system?
It wasn't intended as a knock - simply saying that i know you like fiddling with your system alot, and that the 3d chips aren't very accommodating of that, where as the intel chips are. Hence you would probably have a better time with an intel chip.
Posted on Reply
#65
freeagent
I would have a better time with a chip that wont die..
Posted on Reply
#66
Dragam1337
freeagentI would have a better time with a chip that wont die..
Indeed, which is my point - 3d chips don't like being tuned. You are super hard into tuning, so probably worst chip ya could have gone with :P
Posted on Reply
#67
dtoxic
Must be some design problem in the DIE, ccd and the way it's arranged inside, combination of that and voltage, i don't see this as wide spread,so some "specific" combination that makes the cpu heat up at a specific place but can also indicate a small/large batch of defective cpu's?
Posted on Reply
#68
freeagent
Dragam1337Indeed, which is my point - 3d chips don't like being tuned. You are super hard into tuning, so probably worst chip ya could have gone with :P
My other X3D is just fine. You cant tune these chips very much, Dell life y0. Curve and memory speed is pretty much all you get. I don't bclk clock it.

Edit: My curve is -40 AC :D
Posted on Reply
#69
Dragam1337
freeagentMy other X3D is just fine. You cant tune these chips very much, Dell life y0. Curve and memory speed is pretty much all you get. I don't bclk clock it.

Edit: My curve is -40 AC :D
You're right, they can't be tuned as much as other cpu's, and while memory tuning may give more in benchmarks, it barely gives anything in games.

I just have expo enabled with trefi at 30k, and all voltages as low as they go, to ensure chip longevity.

As far as voltages go, i noticed your mem voltages are very high - is your vddio equally high? But impressively low curve, asssuming it's stable :)
Posted on Reply
#70
AusWolf
Dragam1337vsoc voltage (...) affects the entire chip.
Does it? I didn't know.
Dragam1337You're right, they can't be tuned as much as other cpu's, and while memory tuning may give more in benchmarks, it barely gives anything in games.

I just have expo enabled with trefi at 30k, and all voltages as low as they go, to ensure chip longevity.

As far as voltages go, i noticed your mem voltages are very high - is your vddio equally high? But impressively low curve, asssuming it's stable :)
While I had my 7800X3D, I had EXPO off with 1 VSOC, because the perceptible real-world performance difference was nothing.
Posted on Reply
#71
Dragam1337
AusWolfDoes it? I didn't know.


While I had my 7800X3D, I had EXPO off with 1 VSOC, because the perceptible real-world performance difference was nothing.
I might have misspoken there.

There are ram bandwidth heavy games like star citizen where it makes a difference, but in most games the difference is negible.

I would only use expo with a low voltage kit though. I would never ever run high voltage for a tiny performance difference. As i posted here, my voltages are quite close to stock.
Dragam1337 in AMD Ryzen 7 7800X3D Swells and Burns Again, This Time on an ASUS Crosshair X670E HeroSurely you can do better than 1.2v with 2 sticks. 457226
Posted on Reply
#72
AusWolf
Dragam1337I might have misspoken there.

There are ram bandwidth heavy games like star citizen where it makes a difference, but in most games the difference is negible.
Personally, I didn't notice any difference at all. Meanwhile, the difference in power consumption and heat was tremendous (5+ °C).
Dragam1337I would only use expo with a low voltage kit though. I would never ever run high voltage for a tiny performance difference. As i posted here, my voltages are quite close to stock.
Does RAM voltage affect SoC voltage, though?
Posted on Reply
#73
Dragam1337
AusWolfPersonally, I didn't notice any difference at all. Meanwhile, the difference in power consumption and heat was tremendous (5+ °C).


Does RAM voltage affect SoC voltage, though?
You probably had the voltages increased a fair bit. I use -30 curve aswell, so my heat and power is lower than stock.

Yes, you need a bit higher soc voltage with higher memory speeds. But mostly memory speed and voltage affects the needed vddio voltage (memory controller), which should ideally be the same as mem vddq.
Posted on Reply
#74
PixelTech
You know..... when I replaced my mobo's retention bracket with a Secure Frame, I did find that the screws were oddly loose, like looser then hand tight. Can it just be that AM5 CPUs aren't being pressed hard enough to make contact with the pins?
Has Intel not had these problems in the past for their way early generations?
Posted on Reply
#75
freeagent
Dragam1337As far as voltages go, i noticed your mem voltages are very high
Really?

They look stock to me.

Posted on Reply
Add your own comment
Nov 20th, 2026 13:09 PST change timezone

New Forum Posts

Popular Reviews

TPU on YouTube

Controversial News Posts