AMD won't be adopting the 12VHPWR power adapter connector for its next-gen graphics cards, according to an IT industry veteran. It looks like Team Red has wisely — or luckily — dodged a white-hot tracer bullet, as the story of melting GeForce RTX 4090 power connectors erupted on Monday.
Ex-HardOCP editor and ex-Intel exec Kyle Bennett confirmed with multiple sources that neither AMD reference nor AIB custom designs using RDNA 3 GPUs have opted for the compact 16-pin power connector standard. AMD's Scott Herkelman, SVP and GM of AMD's Graphics Business Unit, has also chimed in to make it clear that it won't be using 12VHPWR on its boards.
After the @NVIDIAGeForce melting drama, I have verified through multiple sources that @Radeon Navi 31 ref cards will NOT use the 12VHPWR power adapter, and I could not verify any AIBs using 12VHPWR on N31 either. Picture in case you don't know what a smoked 12VHPWR looks like. pic.twitter.com/hLWhjqSm4vOctober 25, 2022
To recap the developing situation, we recently reported on the first few instances of melted RTX 4090 16-pin adapters. This left us pondering over whether it was a sign of things to come. Shortly after publishing, further reports of melted power connectors on RTX 4090 cards emerged.
CableMod also published an illuminating guide on how to correctly connect the 16-pin power connector from the PSU to the input on the graphics card. The cabling is apparently very particular and even modest bends can cause a less than perfect connection. Its advice is that the cables should be kept straight until the cable is at least 35mm away from the connector assembly. This is a big ask for practical installation in most PC cases.
AMD's most powerful upcoming Radeon RX 7000 graphics cards, which will use the Navi 31 GPU, apparently stick with the tried and true 8-pin PEG (PCI Express Graphics) power connectors. Bennett talked with his extensive contact list at AMD and Team Red AIB partners, and he claims that we won't be seeing RDNA 3 graphics cards integrating this emerging GPU power connector standard.
His language was a little more firm about the AMD reference designs not using the 16-pin power connector than for partner cards. However, that is the nature of the business. There are so many partners it is sometimes hard to keep track of their designs, especially at this secretive time in the process of preparing the next gen Radeon products.
The Radeon RX 6000 series and upcoming RDNA 3 GPUs will not use this power connector.October 25, 2022
Scott Herkelman's response seems to wash away any remaining doubt, with a definitive statement that, "...upcoming RDNA 3 GPUs will not use this power connector."
Some will naturally wonder why AMD hasn't gone down the 12VHPWR / 16-pin power connector route with its upcoming Radeon RX 7000-series flagship(s). It could be that the Navi 31 GPU powered cards won't need much more than currently existing designs, in which case a couple of 8-pin connectors should suffice — and partners could do triple 8-pin designs like we've seen with RX 6900 XT and 6950 XT models.
AMD may also have noticed in the early testing phases that there were issues with the compact 16-pin power connectors and decided to step back. While the 16-pin connector comes via PCI-SIG, it builds off the work Nvidia did with the 12-pin connector used on RTX 30-series Founders Edition cards — just with four extra sense pins for communicating telemetry data from the PSU. Given the Nvidia foundations of the standard, it's not too surprising to see Nvidia eying the result with some skepticism.
AMD will reveal its next-generation Radeon RX 7000-series graphics cards on November 3, and its anticipated that we'll see reviews and hardware availability later next month. Some models may not ship until December, and it's likely we'll only see a couple of GPUs this year, with the remaining RDNA 3 updates coming in 2023.
Either way, it's a Plug design issue and the lack of Strain Relief causing the internal Female pins to not be properly seated and making good contact with the male pins on the receptacle end.
I'm glad AMD is not using this 12VHPWR standard.
Being cautious & conservative about such big changes is good for AMD.
This is mostly a self inflicted wound from nVIDIA.
No connector likes having a hard bend at the plug. Copper work-hardens so sharp bends should be avoided as much as possible to avoid brittle failure over time. If you need a sharp bend, use cables with an angled plug or appropriate adapter.
Talk about irony.
Greater surface area leads to better connections. But the real problem I suspect is the cable gauge. While technically cables that thin can support the requested current, as the article states, there's a minimum radius before there is internal breakage of individual strands. These are stranded wires and not solid core as stranded it more flexible. Stranded is also better with high frequency pulses due to skin effect. But it's subject to breakage in tight corners. A lower gauge # (thicker cable) has more resistance to sharp bends. Also, the increase in the number of fibers means higher current and better natural strain relief as the strands support one another.
And DP 2.1 versus 1.4a
Off the top of my head, the general rule is 6-8 times the diameter of the conductor inclusive of jacket and insulation. ( . . dang those egg-head electrical engineers! . . )