PCI standards group deviates and assigns responsibility for melting GPU power connectors

Expand / Nvidia's 12VHPWR power cable adapter, as it comes with an RTX 4080. Andrew Cunningham

Nvidia's new RTX 4090 and 4080 GPUs both use a new connector called 12VHPWR to supply power to satisfy ever more power-hungry graphics cards without having to reserve the physical space required for three or four power supplies with 8-pin connectors. But this power connector and its specifications weren't created by Nvidia alone - to ensure interoperability, the specifications were jointly developed by the PCI Express Special Interest Group (PCI-SIG), an organization that includes Nvidia, AMD, Intel, Arm, IBM, Qualcomm and others.

But recent overheating and melting issues experienced by some RTX 4090 owners have apparently prompted PCI-SIG to clarify exactly which parts of the spec it is responsible for and which it is not responsible for. is not responsible. In a statement reported by Tom's Hardware, the group reminded its members that they, not PCI-SIG, were responsible for safety testing of products using connector specifications like 12VHPWR.

"Members are reminded that the PCI-GIS specifications provide the technical information necessary for interoperability and do not attempt to address design, manufacturing methods, materials, safety testing, safety or workmanship," the statement read. "When implementing a PCI-GIS specification, members are responsible for the design, manufacture, and testing, including security testing, of their products."

In other words: PCI-SIG tells companies how to make their 12VHPWR products work with other companies' 12VHPWR products, but they are not involved in the manufacturing or safety testing of any 12VHPWR < em>specific product. This statement appears designed to absolve the PCI-SIG of any blame in the power connector fusion saga in light of a lawsuit that has been filed against Nvidia over this issue.

Third-party and Nvidia testing suggests that overheating power connectors on RTX 4090 GPUs are mostly due to user error rather than a fundamental design issue . In most cases, overheating and melting of connectors has shown evidence. that they had not been fully inserted into the GPUs 12VHPWR connector. This issue seems to affect all 12VHPWR cables, whether you're using an adapter for 8-pin PCIe power cables or a native 12VHPWR cable that runs directly off your power supply, but it also seems to be relatively rare .

However, manufacturers could incorporate more safeguards to detect incorrect connections or make it more obvious when a cable is fully inserted and when it is not; a larger connector with stronger, more easily visible retention clips has helped 6- and 8-pin PCIe power connectors prevent this kind of problem for years. Hopefully we will see advancements that will help prevent this type of user error, whether from manufacturers or in the form of a PCI-SIG specification change. The 12VHPWR connector is likely here to stay, thanks to its inclusion in Intel's ATX 3.0 specification for PC power supplies.