Nvidia's GeForce RTX 40-series (Ada Lovelace) graphics cards were the first mainstream products to adopt the 16-pin (12VHPWR) power connector. However, Nvidia's insistence on the power connector may change with the upcoming GeForce RTX 4070. According to a new report from Igor Wallossek, only some GeForce RTX 4070 graphics cards will use the 16-pin power connector.
The GeForce RTX 4070 rumoredly launches in April. Wallossek's sources claim that the GeForce RTX 4070 has two categories. The premium or factory-overclocked models are reportedly in the first category with a 225W TBP. They will likely sport one 16-pin power connector or two standard 8-pin PCIe power connectors. The second category encompasses the standard GeForce RTX 4070 models; those are the MSRP level that adheres to a 200W TBP. Wallossek believes that the graphics cards in the second category will stick to a regular 8-pin PCIe power connector.
The 16-pin power connector has caused its fair share of havoc on the GeForce RTX 4090, one of the best graphics cards. Due to the more demanding power requirements, the power connector makes sense with GeForce RTX 4070 Ti and higher SKUs. However, there's no need for it with the lower-clocked GeForce RTX 4070 and inferior models.
Wallossek confirmed with his sources that the GeForce RTX 4060 Ti and GeForce RTX 4060 would employ a single 8-pin PCIe power connector. However, according to an early leak, the GeForce RTX 4060 may have a 115W TDP, so it only pulls a little power.
The 16-pin power connector is here to stay. However, if the rumors are accurate, the GeForce RTX 4070 Founders Edition and a few 225W custom models may be the last Ada-based graphics cards to use the 16-pin power connector for a while. So unless Nvidia launches a GeForce RTX 4080 Ti, GeForce RTX 4090 Ti, or the Titan RTX Ada, we likely won't see the 16-pin power connector until the next generation of GeForce gaming graphics cards.
AMD was reluctant to use the 16-pin power connector on the chipmaker's latest Radeon RX 7000-series (RDNA 3) graphics cards. But, surprisingly, graphics cards are becoming more power-hungry graphics instead of consuming less power. So it may be a matter of time before we find the 16-pin power connector on a Radeon graphics card.
Anyone who remembers early GPUs will remember cards with 'Molex'* connectors for external power input. Those vanished as soon as mini-fit Jr was adopted for GPU power. Even for cards there were only a few watts over the 75W card-edge power budget, they were saddled with a "no need" 6-pin 75W power connector, even if they could have worked just fine with a Molex connector. Similarly, standardising on the new 16-pin connector means the old 6-pin and 8-pin connectors can eventually be phased out. Personally I'd have loved to see a move to Micro-fit across the board with ATX12VO to further compact board area for cabling, but I'd have even more rather seen a move to 48V internal power rather than sticking with legacy 12V.
* Ironically, unlike almost all the other internal ATX power connectors, the Peripheral Power Connector was actually a Mate-n-Lok connector designed by Amphenol.
I miss those days and the full head of hair that came with them! Jiggling the wires at the base of the plastic housing to line up the pins as you finalized your connection...only to have one pin jam, bend or kink the connection forcing you to start all over. The memories....