GPU power connector facts

Is 12VHPWR an NVIDIA-Proprietary Connector?

Because NVIDIA is nearly the only company actually shipping cards with it, it's easy to assume 12VHPWR is NVIDIA's own invention. The specification text says otherwise — it's an open industry standard anyone can use without paying NVIDIA anything. What's true is a narrower, more interesting claim: almost nobody else has chosen to.

Quick answer: No — 12VHPWR/12V-2x6 is published by PCI-SIG, the industry body that maintains the PCIe specifications, as part of the CEM 5.0 revision, co-sponsored by NVIDIA and Dell. Any GPU vendor can implement it without licensing anything from NVIDIA. In practice, adoption is almost entirely NVIDIA: AMD's RX 9000 series reference cards use traditional 8-pin, only SAPPHIRE's RX 9070 XT NITRO+ and ASRock's RX 9070 XT Taichi OC among non-reference cards use it, and Intel's Arc lineup doesn't use it at all. "Open standard" and "one company's connector in practice" are both true at once.

Download PowerDoze free Read the full spec origin story →

Who actually wrote and owns the spec

PCI-SIG is the industry consortium that maintains the PCIe specifications generally — the same body behind PCIe itself. 12VHPWR (later revised to 12V-2x6) is part of its CEM 5.0 revision, co-sponsored by NVIDIA and Dell. Being a PCI-SIG standard means membership and implementation are open: any vendor can build to the spec without paying NVIDIA a license fee for the connector shape itself.

PCI-SIG's own public statement on the matter draws a clear line: the specification "provide[s] necessary technical information for interoperability" and does not "address proper design, manufacturing methods, materials, safety testing, safety tolerances or workmanship" — each vendor member is responsible for its own product's design, manufacturing, and testing. In other words, the spec guarantees the plug fits; it doesn't dictate who has to use it, and it doesn't certify any individual vendor's implementation.

Who actually ships it — the adoption gap

Vendor / lineup Connector used
NVIDIA (RTX 40/50 series, most SKUs)12VHPWR / 12V-2x6
AMD RX 9000 series (reference cards)Traditional 8-pin
SAPPHIRE RX 9070 XT NITRO+ (non-reference)12VHPWR
ASRock RX 9070 XT Taichi OC (non-reference)12VHPWR
Intel Arc A7708-pin + 6-pin
Intel Arc B580One or two 8-pin

AMD doesn't require its board partners to adopt 12VHPWR, and only two non-reference cards have chosen to. Intel's Arc lineup skips it entirely. Whichever brand made the card in your case, the connector on it was built to the same open PCI-SIG spec — almost nobody besides NVIDIA has actually shipped a product using it at any real scale.

Myth vs. what's actually true

The claim: "NVIDIA invented and owns this connector"

Not supported. It's a PCI-SIG open standard, co-sponsored (not solely authored) by NVIDIA and Dell, and open for any member to implement without a license from NVIDIA.

Verdict: Doesn't hold up as a governance/ownership claim.

What's actually true: NVIDIA is nearly the only adopter

Correct, and the more useful version of this observation. The adoption data across AMD, Intel, and AIB partners backs it up clearly.

Verdict: Accurate — just a different claim from "NVIDIA-proprietary."

Why is a power management tool talking about spec governance?

Whether a connector standard is open or company-specific is a PCI-SIG governance question, not a software one — nothing here touches anything a Windows power tool could control. This page exists only because it's part of the same fact-check series as PowerDoze's coverage of GPU power limits and connector melting, since the questions come up in the same threads.

Honest note: You don't need PowerDoze for this question at all. See the full spec comparison and the connector melting timeline for the parts of this series that do involve something a power tool can affect.

Download PowerDoze free See all six myths checked →

Frequently asked questions

Is 12VHPWR NVIDIA's own proprietary connector design?

No. It's published by PCI-SIG, the industry body that maintains the PCIe specifications, as part of the CEM 5.0 revision, co-sponsored by NVIDIA and Dell. Any GPU vendor can implement it without licensing anything from NVIDIA.

Who actually publishes and maintains the 12VHPWR specification?

PCI-SIG. NVIDIA and Dell co-sponsored the CEM 5.0 proposal that introduced it, but PCI-SIG is the body that owns and maintains the published specification text, the same as it does for PCIe itself.

Does AMD use 12VHPWR on any of its graphics cards?

AMD's own reference-design RX 9000 series cards use traditional 8-pin connectors — AMD doesn't require partners to adopt 12VHPWR. Among non-reference cards, only SAPPHIRE's RX 9070 XT NITRO+ and ASRock's RX 9070 XT Taichi OC use it.

Does Intel Arc use 12VHPWR?

No. Intel's Arc lineup doesn't use it at all — the A770 uses an 8-pin plus a 6-pin connector, and the B580 uses one or two 8-pin connectors.

If it's an open standard, why does almost only NVIDIA use it?

The specification being open and adoption being narrow are two separate facts that are both true at the same time. NVIDIA co-sponsored the standard and builds nearly all its own cards around it; other vendors have simply chosen not to adopt it at scale, which is a business decision, not a licensing restriction.

Want your NVIDIA GPU's power limit put back for you instead of reset by the driver on every reboot? PowerDoze folds it into a power mode, and an all-day, every-day schedule rule pointed at that mode re-applies it at every startup — free, no account required.

Download free for Windows 10/11

See also: Full 8-pin vs 12VHPWR spec comparison · The GPU connector melting timeline · Is 12VHPWR corner-cut? · Is 8-pin really limited to 150W? · All features