Internet cafe owner claims 15 Ryzen 7 9800X3D CPUs failed out of 150 purchased, power supply to blame?
A Redditor who says they run an internet cafe (yes, they are still around) claims they have seen 15 dead AMD Ryzen 7 9800X3D CPUs out of 150 deployed since March 2025. That is a 10% failure rate, with one CPU reportedly failing every one to two weeks.
The same user says the PCs use the B650M-AYW WiFi motherboard and 850W 80+ Gold power supplies from Huntkey. He also says that he has not enabled PBO, and memory is running at DDR5-5600. In the same thread, the user confirms the CPUs are tray units.
What should be noted is that this is not another ASRock-only report. AYW is ASUS “As You Wish” motherboard series, designed as an affordable product line.
I purchased 150 Ryzen 9 9800X3D CPUs in March of this year. I’m using them in an internet cafe, and I’ve never had such a terrible experience. One CPU fails every one or two weeks. I don’t know whether to blame the ASUS motherboard or AMD. The motherboard is a B650M-AYW WiFi; it’s not a top-of-the-line board, but it shouldn’t be causing CPUs to burn out. The power supply is an 850W gold-rated unit, which is definitely not the problem. Although they are covered by a three-year warranty, what happens after three years? I’ve been using them conservatively; I haven’t even enabled PBO, and the memory is only running at 5600MHz, let alone overclocking. A 10% failure rate seems a bit excessive.
— RealisticLoad3327/Reddit
Source: Reddit
Obviously the first suspicions were put on the power supply. Apparently each system is equipped with a low-quality Huntkey 850W PSU. However, it is possible that the power delivery to the facility itself is an issue too. Modern PSUs can take a lot of imbalance in power, but they are not problem proof for everything.
We have not seen many reports about Ryzen 7 9800X3D issues on non-ASrock motherboards. But when these cases appeared, AMD described them as a complex issue and pointed to some boards not sticking to AMD-recommended values in BIOS, while also stressing BIOS updates. The cafe owner says they are on a September BIOS (3283), but it is not the newest release, and there is still no confirmed root cause that explains every failure report.
One thing that is missing is the photo of the alleged failed CPUs, or even a shred of evidence that this had happened. Should the user provides such proof, we will update this story with further details.
