Friday, February 28th 2025

Custom AMD Radeon RX 9070 XT Appears Powered by 12V-2x6 Connector

Days before releasing, Chinese leakers on Chiphell are showcasing a custom variant of AMD's upcoming Radeon RX 9070 XT with what appears to be a 12V-2x6 power connector. The custom AIB model is Sapphire Radeon RX 9070 XT NITRO+, which features a triple-fan cooling configuration and a departure from Sapphire's older GPUs that used eight-pin power connectors. Despite proving to be problematic for NVIDIA, this power connector could do well with AMD's Navi 48 XT GPU SKU due to its power consumption envelope. With an entire GPU projected to use 304 W total board power, the 12V-2x6 connector could easily power this configuration without the need to overheat and possibly melt. If it manages to power 600 W TGP NVIDIA GeForce RTX 5090, then powering a custom Radeon RX 9070 XT shouldn't be an issue.

Even if the card experiences a heavy overclock, power limits will remain within a usable range so as not to cause any trouble. Sapphire's reason for ditching the older, reliable eight-pin power connectors is unknown, but the use of the new 12V-2x6 here isn't expected to be problematic either. Some previous VBIOS records in our GPU DB indicate that the GPU could boost the Navi 48 XT SKU to 2,520/2,518 MHz, which is about 120 MHz higher than the stock AMD Radeon RX 9070 XT boost clock.
Sources: Chiphell, via HXL on X
Add your own comment

47 Comments on Custom AMD Radeon RX 9070 XT Appears Powered by 12V-2x6 Connector

#1
usiname
How you insert the cable under the backplate?
Posted on Reply
#2
Cheeseball
Not a Potato
Nothing new here. The ASRock Creator Radeon RX 7900 XTX also had a 12V-2x6 albeit it is a blower type and did run loud at 5000+ RPM. My only problem with it is that it was long, which made sense since it fits in 1U racks pretty snug.
Posted on Reply
#3
outlw6669
Is it just me, or does having a covering on the cooler fins seem to defeat the purpose of having a flow through cooler :confused:

I am not against anyone using the 12V-2x6 connector, but I really feel it was badly specced.
As I recall, it has only a 10% safety overhead for the connector, which means it is running overly close to the limit, when supplying 600W.
On a card like this, where power draw is kept under 450W, it should be more than fine.
Anything higher than a 450W TBP, though, and I strongly feel 2x connectors should be required...
Posted on Reply
#4
TheDeeGee
usinameHow you insert the cable under the backplate?
Look at the pictures...
Posted on Reply
#5
remekra
outlw6669Is it just me, or does having a covering on the cooler fins seem to defeat the purpose of having a flow through cooler :confused:

I am not against anyone using the 12V-2x6 connector, but I really feel it was badly specced.
As I recall, it has only a 10% safety overhead for the connector, which means it is running overly close to the limit, when supplying 600W.
On a card like this, where power draw is kept under 450W, it should be more than fine.
Anything higher than a 450W TBP, though, and I strongly feel 2x connectors should be required...
It's not like they completely blocked the flow through, one would assume they tested it with and without and found the difference to be small or not existent at all.
And Sapphire knows how to make GPU coolers.
Posted on Reply
#6
TheDeeGee
outlw6669Is it just me, or does having a covering on the cooler fins seem to defeat the purpose of having a flow through cooler :confused:

I am not against anyone using the 12V-2x6 connector, but I really feel it was badly specced.
As I recall, it has only a 10% safety overhead for the connector, which means it is running overly close to the limit, when supplying 600W.
On a card like this, where power draw is kept under 450W, it should be more than fine.
Anything higher than a 450W TBP, though, and I strongly feel 2x connectors should be required...
The spec itself is "alright", it's the crimp connectors that are the issue. They need to switch from 3-dimple to 4-spring crimp connectors, that would solve atleast the poor contact issues.
Posted on Reply
#7
Chomiq
TheDeeGeeLook at the pictures...
Or:

I do wonder how they solved the bending issue, with connector located in this position you're bound to run into some hard turns:
Posted on Reply
#8
TheDeeGee
ChomiqOr:

I do wonder how they solved the bending issue, with connector located in this position you're bound to run into some hard turns.
Owning such a connector there is plenty of space to route it in the area, but beats me why you would want to blow hot air from the card over it.
Posted on Reply
#9
usiname
TheDeeGeeLook at the pictures...
I am looking at the picture and the only possible way to use the card is to remove the back cover, which lead to the question, what is the point in the cover in first place?
Posted on Reply
#10
Cheeseball
Not a Potato
ChomiqOr:

I do wonder how they solved the bending issue, with connector located in this position you're bound to run into some hard turns.
That's already enough space for the 12V-2x6 connector to curve right (or left depending on where you're looking). Most of the RTX 4090 burning issues are because that connector was pulled hard upwards or downwards which then disconnected slightly.
usinameI am looking at the picture and the only possible way to use the card is to remove the back cover, which lead to the question, what is the point in the cover in first place?
That back cover still has vents:


They look open enough for air to still pass through despite a cable crossing through it.
Posted on Reply
#11
Hecate91
CheeseballThat's already enough space for the 12V-2x6 connector to curve right (or left depending on where you're looking). Most of the RTX 4090 burning issues are because that connector was pulled hard upwards or downwards which then disconnected slightly.
The connector being pulled hard at an angle can't be great either. I don't see point of 12v2x6 being used just to hide it under a backplate, good thing other AIBs can use 2x 8 pin molex.
Posted on Reply
#13
Cheeseball
Not a Potato
Hecate91The connector being pulled hard at an angle can't be great either. I don't see point of 12v2x6 being used just to hide it under a backplate, good thing other AIBs can use 2x 8 pin molex.
Why would it be pulled hard at that angle though? Unless the 12V-2x6 cable is really short, it should have enough free play to flow toward the motherboard and to the back or bottom of the case.

Most of those NVIDIA AIB cards were so tall they ended up slamming the connector straight into the tempered glass (or solid/meshed) case panel. Not sure how those with vertical GPU setups with risers had burning issues but I'm assuming they probably had it slammed against an air cooler tower or such.
Posted on Reply
#14
Hecate91
CheeseballWhy would it be pulled hard at that angle though? Unless the 12V-2x6 cable is really short, it should have enough free play to flow toward the motherboard and to the back or bottom of the case.

Most of those NVIDIA AIB cards were so tall they ended up slamming the connector straight into the tempered glass (or solid/meshed) case panel. Not sure how those with vertical GPU setups with risers had burning issues but I'm assuming they probably had it slammed against an air cooler tower or such.
The cable looks like it bends towards the pci-e slot side of the card.
I don't trust the connector being bent at all since the crimped in pins are still an issue causing uneven power loading per wire. Unless Sapphire took account for it in their design, using 12v2x6 seems unnecessary IMO.
Posted on Reply
#15
INSTG8R
Vanguard Beta Tester
This makes my love for Sapphire go right out the window…it’s a 300W card it’s not even necessary…shame on you Sapphir…
Posted on Reply
#16
Vayra86
TheDeeGeeOwning such a connector there is plenty of space to route it in the area, but beats me why you would want to blow hot air from the card over it.
They are simply beta testing this cable. 'What happens if we innovate in this silly way?' /s
Posted on Reply
#17
FoulOnWhite
Looks good to me. The weight of the cable is supported by the cooler so no heavy cable/s hanging off the GPU pulling on that connector.
Posted on Reply
#18
Cheeseball
Not a Potato
Hecate91The cable looks like it bends towards the pci-e slot side of the card.
I don't trust the connector being bent at all since the crimped in pins are still an issue causing uneven power loading per wire. Unless Sapphire took account for it in their design, using 12v2x6 seems unnecessary IMO.
This is actually the best part about this layout. Just one cable and it sinks toward the motherboard instead of coming out from the other directions. It should look seamless if cable management was done correctly. I'm good with the older two PCI-E 150W connectors (hence my MBA XTX) but having three for an overclocked model is just a waste of space IMO.

In a vertical GPU setup with a riser, as long as the case has a slot or hole that goes straight into the PSU cage, you might not even notice the cable unless you look for it.

Besides, that burning issue mostly happened with the RTX 4090s and those 5090s which all pushed 450W+.

And the fact that almost all the ATX 3.1 PSUs coming out have a 12V-2x6 connector with cable now, so might as well use it and reduce the amount of cables in a setup.

I still hate this thing's size though. Looks good but too big. But I think this can still fit in a HYTE Revolt 3 but definitely not a Fractal Ridge. AMD caused some disappointment by not having their own MBA model.
Posted on Reply
#19
wNotyarD
CheeseballAMD caused some disappointment by not having their own MBA model.
Especially when their renders looked damn good.
Posted on Reply
#20
wolf
Better Than Native
interesting concept, heat the already controversial cable with the heat from the flow through cooler.

Visually this is nice, I hope that functionally it is too, but wanting for <400w power I'd imagine low issues with the connector itself.
Posted on Reply
#21
john_
Obviously they target those who have a PSU with that connector.

At least they could add a cable clip or something on the back plate, 5 centimeters away from the socket and in line with the socket, to help the cable stay in straight line and not bend at the socket.
On the other hand, imagine AMD using this socket on all it's cards and we had no incidence with AMD cards. Tech press would have a difficult task to push the "user error" narrative.
Posted on Reply
#22
Bomby569
the cable and that plate with the holes that are there i assume because of the cable, are really a stupid way to solve a problem that doesn't exist. Why use that stupid connector in the 1st place.
Posted on Reply
#23
Chomiq
TheDeeGeeOwning such a connector there is plenty of space to route it in the area, but beats me why you would want to blow hot air from the card over it.
How hot do you expect that air to be? 200C?
Posted on Reply
#24
Bomby569
ChomiqHow how do you expect that air to be? 200C?
the connector and the cable are plastic, and that plate thing is impeding airflow, and the connector is stuck inside the card and the PCB, not sure any of that is a good idea
If a fan goes out for example, if can get very toasty
Posted on Reply
#25
user556
There never was a bending issue. It was always bad quality plugs on the cables. And those cables all need recalled.
Posted on Reply
Add your own comment
Feb 28th, 2025 13:20 EST change timezone

New Forum Posts

Popular Reviews

Controversial News Posts