• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

NV 12VHPWR issues revisited

Page 3 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.
Who drove this crazy 600W connector as a requirement in the first place?

Not that it really matters I guess. No the only thing which matters is that for some reason Nvidia decided to go all out with 12VHPWR even on 200W cards like the 4070.

Then they placed it at the side so that unless you have 200mm+ wide case the connection would be under mechanical stress.

Multiple bad decisions led to this.

Still, this is Nvidia and they never admit to anything wrong (even way back look how they handled bumbgate by fobbing off the end users and the OEMs). Meanwhile their fans sweep it all under the carpet with the mantra "it was all user error" which is total nonsense on a $2k consumer item - sorry average consumer "you are pushing it wrong". That's Apple level of delusion.

The 3090?
That did not use the 12VHPWR connector, so unsure how it is relevant here?
 
Last edited:
The 3090?
That did not use the 12VHPWR connector, so unsure how it is relevant here?
Basically same 12 pin power connector minus the 4-pin sensors. In case it needs to be spelled out, the 12-pin is the relevant part and subject to the same 600w tolerances that apply to the 12VHPWR.
 
3 more in the wild, a consistent and recurring problem.

Code:
https://www.reddit.com/r/nvidia/comments/13p5jz9/4090_tuf_cable_burnt_fully_connected_cable_after/

https://www.reddit.com/r/nvidia/comments/13p2r3d/12vhpwr_adapter_melting_after_6_months/

https://www.reddit.com/r/nvidia/comments/13nmojh/4090fe_burned_corsair_sf750_platinum_corsair/
 
I have a 4090 and I'm scared 🙁
I'll sacrifice myself for your well being by allowing you to send me the card for free.

Seriously though, the worst that seems to happen so far is a destroyed connector, which means that you have to deal with an RMA or have to get the connector replaced by a shop.
 
Btw I wonder if the northridge guy really has over 250 cards with the molten connector issue, or did he mean 250 cards in total for various other repairs? Pretty sure he would have raised a much larger red flag over this by emphasizing the huge number of cards with this issue earlier. He hasnt mentioned "hundreds" of cards anywhere else, nor in his vid titles which would certainly be a very big story that would bring him a LOT of traffic and much more attention.
 
Bought a Corsair 12VHPWR cable for my AX1600i when I replaced my MSI 4090 Gaming Trio with a Gigabyte 4090 Waterforce.

Did it mainly for simplicity and cleaner aesthetics. Never worried about frying the connector. I have working hands and eyeballs to get things plugged in right.
 
Oh, you awesome dude, you. I mean clearly it's the rest of the plebes that "are holding" it wrong. Clearly, yo.
We all know that already. No need to point out the obvious.

My superiority is obviously cleanly clear with a clarity negating clarification.
 
Is that enough to cause the melting issues? I don't know ... but i don't think it's supposed TO BE ABLE to be pulled out that much WITH the lock thingy ON.
Yup. If it clicks into place, it should be immovable. And lock on top of that??? Wow. Monkeys were involved in this design. Guess everyone is cheaping out and not consulting senior engineers anymore.
 
NF already showed the cablemod adapter was fully seated and the melting still happened on the GPU side.

It's the 4090 connector at fault, with perhaps the shoddy design of the cable exacerbating it.
 
this might be the wrong thread and I don't know if i can link to tom's but they ran an article on a new version of the windforce 4090 with the 12vhpwr moved to the rear of the card. it'll be interesting to see if others follow suit later or in the future with other nvidia cards if nvidia doesn't drop this connector. it should reduce stress on the connector since there is no side panel bumping up against it but does pose another problem with shorter cases where there is not enough room for the cable or there is room but there's a thick radiator involved in the front business of the case
 
Colored plugs and moving the plug to the end are good things, but the fundamental problem is the same power from 3-4 PCIe going through a single 12VHPWR.

That and PCIe are solidly chunky and click into place with no movement in a very obvious way. This new one is smaller and can wiggle out with even minor force.
 
It's not actually that bad. It would be better if they used a reversed connector with the latch to the back of the card, and then they could have a reduced depth heatsink all the way to the top of the card. Coming out the back like that will be really nice for cable management though.
 
Back
Top