The failure chatter is now giving Intel's 13th and 14th-gen high-end mess a run for its money, which is not the sort of competition AMD should want.

Most reports previously pointed the finger at ASRock motherboards, but this batch is nastier because every death happened on Asus kit.

The cafe owner says the systems use an Asus B650M-AYW WiFi motherboard, paired with 850W 80+ Gold power supplies from Huntkey, which is not exactly a confidence booster.

The PSU gets called “pretty bad” in the report, so it sits right there in the suspect lineup, even if past cases suggest it is not always the smoking gun.

Of the 150 units deployed from March, the owner says failures occur at random intervals, usually every 1-2 weeks, working out to roughly 1.6 dead CPUs a month.

The cafe owner says they have never had such a “terrible” experience before, and it is hard to argue when you are basically running a chip graveyard.

They say they did not enable PBO or any overclocking, so this is not the usual tale of someone cranking sliders then acting surprised.

On firmware, the systems are reportedly on BIOS 3283 (98.81.0) from September, which is new-ish but not the latest, and it was meant to “improve” system stability.

There is no clarity on how many chips died before or after that update, but running an older BIOS does not magically explain why processors dropped dead.

The write-up even suggests that it could be the CPU drawing more voltage than it should, which is a grim prospect for a part sold as premium.

AMD and its board partners still have not pinned down a cause, and nearly a year in, the 9800X3D death watch is still getting fresh entries.