It's really what you would expect, which is why they designed it that way. All things being equal the wires should naturally load balance.Quick testing in OCCT @ 600w with a ASUS 5090 Astral
AMP spread over the pins work as it should 🧐
GPU 12volt drops to ~11.8v average
Glad i went with ASUS time time around which is the sole vendor that have taken these problems with the connector serious
(only card with software monitor for amp draw over the pins)
PSU is a dark power 1000w
Have also bought the later released 12v-2x6 / 12vhpwr 90° cable pci-e for this psu
As shown in Buildzoids video though even the Astral card won't load balance correctly.It's really what you would expect, which is why they designed it that way. All things being equal the wires should naturally load balance.
It'd be interesting to see if debauer redid his testing after removing the cable and cleaning each contact with some contact cleaner and abrasion.
Oh absolutely not. DetOx obviously didn't have to have a balanced load on his.As shown in Buildzoids video though even the Astral card won't load balance correctly.
And how many times have people cleaned and scuffed their 8 pin PCI cables? That should not be required to achieve a balanced load.
Quick testing in OCCT @ 600w with a ASUS 5090 Astral
AMP spread over the pins work as it should 🧐
GPU 12volt drops to ~11.8v average
View attachment 116857
Glad i went with ASUS time time around which is the sole vendor that have taken these problems with the connector serious
(only card with software monitor for amp draw over the pins)
PSU is a dark power 1000w
Have also bought the later released 12v-2x6 / 12vhpwr 90° cable pci-e for this psu
So, after some stories that surfaced online about the melting issues of the 12VHPWR power cables for the RTX 5090 and the RTX 4090, I’ve decided to check the one we’ve been using these two years for the RTX 4090.
Before continuing, I want to make it crystal clear that we’ve been using the official power cable for the RTX 4090 for over two years. We haven’t removed it when testing different GPUs. For instance, for benchmarking the RTX 3080, we’ve been using its own power cable. Same applies to the RTX 5080 and the RTX 5090. Moreover, the power cable is still working, and we haven’t encountered any stability issues.
These things are a legitimate fire hazard. We need to get away from PCIe slots and use a socket that can deliver sufficient power through the board, instead of shoving 500W through these flimsy cables.Look how they all come to play once someone else is willing to tank the giant.
Here's what's happened to the 12VHPWR power cable of our NVIDIA RTX 4090 after two years of continuous work
So, what has happened to the 12VHPWR power cable of our RTX 4090 after two years? Has it melted? Let's find out.www.dsogaming.com
View attachment 116971
Oh, they must not have inserted it all the way in properly. User error!
Yeah. Come to think of it, in the old days, a 500W PSU was a big thing. Most pre-builts would have 350W PSUs and you mostly needed a bigger PSU for making your own file server with a dozen hard drives.
All of this has happened before, all of this will happen again
It was present on previous gen cards prior to 4000 series. Shunt resistors cost literally sub 25c in reel quantities. Few $ to BOM to do it.While that current monitoring per pin is fantastic, I wouldn't want to pay Astral prices to get it.
It was present on previous gen cards prior to 4000 series. Shunt resistors cost literally sub 25c in reel quantities. Few $ to BOM to do it.
First 5080 reported burned now. 5000 series - so hot™
I did talk to some AIBs and they are well aware of the 12V-2x6 problems and wanted to get ahead of it. Colored plugs und LEDs can only mask the symptom. However, there was no second 12V-2x6 on the reference PCB and additional monitoring and protection circuits probably also fell victim to the design templates and time pressure.
The board partners have identified the problems, as the solution from ASUS shows. In some models, MSI also divides the six pins of the 12V-2x6 into two rails on their input, provides them with a shunt resistance, but then leads them together.