Gainward Reportedly Replaces RTX 4090's 12VHPWR Adapters

News
By
published

Will other AIB partners follow? And what about the RTX 4080?

12VHPWR Connector
(Image credit: Ricky TO/Facebook)

Gainward is reportedly delaying shipments of its GeForce RTX 4090 graphics cards in Australia, in a bid to replace Nvidia's 12VHPWR adapters, which have been seen to melt, overheat, or catch fire. Gainward is the only graphics card maker that does this, as far as we know, although we wouldn't be surprised if others follow suit.

"As you may have seen, issues have been reported by customers receiving the first batches of [GeForce RTX] 4090 cards, where these adapters have been exhibiting symptoms of overheating/melting under certain circumstances," Australian retailer TechFast wrote to a client who re-published the message on Reddit (via VideoCardz). "While investigations are still continuing and Nvidia has not yet released a public statement, Gainward has told us that the cables shipped with their cards will be being replaced. As a result, they are holding shipping of all cards until this has taken place." 

While reports about failed Nvidia's 12VHPWR adapters that come with today's best graphics card are now numerous and the issue seems to be very real, neither Nvidia, nor other makers of add-in-boards have officially started a recall of the adapters. Or at least none have stated so publically yet.  

Andreas Schilling from HardwareLuxx.de says that Gainward could be the only company to replace these adapters, at least for now. But TechFast says that Gainward will not be the only company to replace the adapter cables. 

"We also understand this cable replacement will not be limited to Gainward alone," the retailer wrote. "We have been given an estimate of the middle of November for this to occur, putting our receiving them around 2 weeks from now at this stage. […] We will be receiving all ordered cards at the same time." 

We could not verify with AIB makers in general or Gainward in particular that they are replacing Nvidia's 12VHPWR adapter with fixed ones at press time. So take this info with a grain of salt. 

What remains to be seen is whether the issue with 12VHPWR adapters will affect the launch schedule of GeForce RTX 4080 boards (due to launch on November 16th), which also come with a 12+4-pin power connector.

Anton Shilov
Anton Shilov
Freelance News Writer

Anton Shilov is a Freelance News Writer at Tom’s Hardware US. Over the past couple of decades, he has covered everything from CPUs and GPUs to supercomputers and from modern process technologies and latest fab tools to high-tech industry trends.

See more GPUs News
8 Comments Comment from the forums
  • Darkbreeze
    So does this mean, as our previous article stated, that all of Gainward's customers will have their warranty voided? I doubt it. Just like I very much doubt that using one of the adapters from Seasonic, Corsair or Cablemod will void the warranty. Pretty sure Magnussun-Moss covers that too.
    Reply
  • PlaneInTheSky
    The discussion at Gainward HQ that started it all:

    "Hey Huang, you think people's house catching on fire would be bad for our reputation?"
    Reply
  • geogan
    I predict this power connection standard will be DOA very soon. To much damage already inflicted to its reputation.
    Reply
  • Darkbreeze
    I would not argue against that opinion as being very probable. In the 35 years or so I've been working with computer systems, I think this is the very first time I've ever seen anything remotely like this happen and I think this is a direct result of companies trying too hard to rush things these days without a reasonable amount of real world testing in addition to any lab testing that's been done. I think PCI-SIG has to take some responsibility for this as well, not just Nvidia, since they apparently signed off on it and since we've now seen evidence that it is happening on native 12VHPWR connections from ATX 3.0 power supplies that are not using ANY adapters, as well. Seems like a major clusterfrack all the way around IMO.
    Reply
  • Sleepy_Hollowed
    This is the smart thing to do, not just for consumers, but for the company.
    This is the first time I’ve seen this and I was around for the pcie switch from AGP.
    Reply
  • RodroX
    If native PSU cable are also melting (not many reports yet compared to adapters, but it did happend), What will a different adapter change?
    Reply
  • Darkbreeze
    It probably won't, unless it is grossly over-engineered, and that might be exactly what needs to happen. The native connectors might not be using much better designs than these adapters and that might need to change. Thicker terminals, actual soldering and larger gauge wiring with far more thermally resistant plastic for the body might have to happen. And if that won't work, the whole design might have to be scrapped. Who knows? Better minds than mine are struggling to come up with solutions so who can say?
    Reply
  • RodroX
    Yeah I know, I was just wondering because from what I can understand after all I read and watched so far, main issue seems to be closer to the "tip" of the connector, where the male and female pins make contact with each other. Thats why I was wondering if theres really something to fix.

    You can use better soldering, larger gauge cables, but from all the pictures I have seen so far the issue is not much the cable or the soldering, but the contact between the pins.

    Perhaps only 6 pins for 450~600 watts is not enough? (Its 6 pins for 12V, 6 pins for ground, and the extra 4 "sense/data", right?).

    The old 8 pin pcie power cable, was 3x 12V + 3x ground + 2x sense, each of this cable was designed for 150 watts use, so you end up with 12x12volts pins for 600 watts, and 9x12volts for 450 watts (~ 50 watts x pin).
    I know the old 8 pin could handle more wattage (Can't remember the exact number) with the right gauge cable, but maybe for the new ATX V3 16 pin connecor, 2 times the wattage peer pin is too much for the actual pin to pin contact, and the closer each pin is to each other could also be a factor.

    I don't know.... maybe one day we will find out what is really happening.
    Reply