I dont know where but recently(last 3/4 days ago) I read that the power draw was not even across all 3 cables, with one around 300w, and rest <= 150w. I will try to find out the article. If the power draw is uneven as indicated then you can expect a lot more burnt cables and gpus. And also the wire size and quality plays a important role. Hope the manufactures are taking note of this.
Article Link:
https://wccftech.com/atx-3-0-12vhpw...r-safety-risk-using-adapter-confirms-pci-sig/
Edit 1: Found the Article and added <= sign
First of all this article is more than a month old. After launch we haven't seen any widespread failures reported by users.
As usual Wccftech hypes a seemingly general message from PCI-SIG and creates panic. From the article comments quoting "johnnyguru" whom every-knows as one of the most knowledgeable in the PSU industry. There is a recent video where he explains this very well to a youtuber
JonnyGuru
•
a month ago • edited
So, just to make sure people know where I'm coming from so those who failed maths class and/or cannot be bothered to look up a data sheet while using disqus from their mom's basement can stop needlessly trolling....
Look up the data sheet for a mini-fit jr. terminal and do the math. Even do worst case scenario math, if you wish. The "spec" for the 8-pin mini-fit jr. connector is not 150W as the article states. In fact, it's potentially twice that. The 150W specification came down from Nvidia about 10 years ago and was not based on the power delivery capability of the connector. Think about it. What's the difference between a 6-pin and 8-pin in "power rating"? The 6-pin is for 75W and the 8-pin is for 150W. What's the difference between the two? The 8-pin has two sense wires. Two wires that don't even participate in power delivery, yet somehow magically doubles the capability in power delivery? Think about that for a moment.
In fact, the wire to board current rating for a dual row 8-pin connector using HCS 16g terminals is HIGHER than the rating of the terminals used in the 12VHPWR connector (10A vs. 9A). The truncated, redacted, whatever letter received by wccftech failed to mention/show that the melting connectors were actually the 12VHPWR connectors and not 8-pin mini-fit jrs. And you'll probably note that the letter does not make any mention of mini-fit jr., splitters, adapters, etc. Only that they have witness thermal variance.
As for the part about variance between terminals: You can take ANY multi-conductor terminal, micro-fit, mini-fit, etc., and find a variance in resistance, voltage and thermals; especially at higher loads like this.
This is why I take issue with the article. T
here is NO problem with using the Seasonic dual 8-pin to 12VHPWR shown here: https://www.tomshardware.co..., for example, and the "concerns of thermal variance" does not apply to it whatsoever.