Thursday, July 16th 2020

The Curious Case of the 12-pin Power Connector: It's Real and Coming with NVIDIA Ampere GPUs

Over the past few days, we've heard chatter about a new 12-pin PCIe power connector for graphics cards being introduced, particularly from Chinese language publication FCPowerUp, including a picture of the connector itself. Igor's Lab also did an in-depth technical breakdown of the connector. TechPowerUp has some new information on this from a well placed industry source. The connector is real, and will be introduced with NVIDIA's next-generation "Ampere" graphics cards. The connector appears to be NVIDIA's brain-child, and not that of any other IP- or trading group, such as the PCI-SIG, Molex or Intel. The connector was designed in response to two market realities - that high-end graphics cards inevitably need two power connectors; and it would be neater for consumers to have a single cable than having to wrestle with two; and that lower-end (<225 W) graphics cards can make do with one 8-pin or 6-pin connector.

The new NVIDIA 12-pin connector has six 12 V and six ground pins. Its designers specify higher quality contacts both on the male and female ends, which can handle higher current than the pins on 8-pin/6-pin PCIe power connectors. Depending on the PSU vendor, the 12-pin connector can even split in the middle into two 6-pin, and could be marketed as "6+6 pin." The point of contact between the two 6-pin halves are kept leveled so they align seamlessly.
As for the power delivery, we have learned that the designers will also specify the cable gauge, and with the right combination of wire gauge and pins, the connector should be capable of delivering 600 Watts of power (so it's not 2*75 W = 150 W), and not a scaling of 6-pin. Igor's Lab published an investigative report yesterday with some numbers on cable gauge that helps explain how the connector could deliver a lot more power than a combination of two common 6-pin PCIe connectors.

Looking at the keying, we can see that it will not be possible to connect two classic six-pins to it. For example pin 1 is square on the PCIe 6-pin, but on NVIDIA's 12-pin is has one corner angled. It also won't be possible to use weird combinations like 8-pin + EPS 4 pin, or similar—NVIDIA made sure people won't be able to connect their cables the wrong way.

On topic of the connector's proliferation, in addition to PSU manufacturers launching new generations of products with 12-pin connectors, most prominent manufacturers are expected to release aftermarket modular cables that can plug in to their existing PSUs. Graphics card vendors will include ketchup-and-mustard adapters that convert 2x 8-pin to 1x 12-pin; while most case/power manufacturers will release fancy aftermarket adapters with better aesthetics.

Update 08:37 UTC: I made an image in Photoshop to show the new connector layout, keying and voltage lines in a single, easy to understand graphic.
Sources: FCPowerUp (photo), Igor's Lab
Add your own comment

175 Comments on The Curious Case of the 12-pin Power Connector: It's Real and Coming with NVIDIA Ampere GPUs

#151
Chrispy_
Fluffmeister
Turing is two years old, it's already died of boredom waiting for Big Navi to beat it to death.
Yeah, but Turing was beaten to death by Little Navi at under $400. Hell, the $375 or so that the 2060 is going for is a joke because it can't raytrace well enough to matter and it's no faster than a 5600XT which is two price tiers below it.

Even in 2019 there was no excuse for a card costing $350 to not have at least 8GB RAM, and the $350 founders edition was as rare as hen's teeth with all of the partner cards coming in at $370-$410. For a 6GB card? Ouch!

Without Navi, Nvidia would have continued to take the piss with such stellar values as the $600 2070FE. The "Super" lineup and price cuts were Nvidia's answer to Navi. They didn't upgrade all their shader counts and reduce prices out of the goodness of their hearts or charity, that's not how businesses run.... ;)
Posted on Reply
#153
medi01
BoboOOZ
But this time there are AMD APU's in all the consoles that matter, and the whole Radeon marketing team has been changed. So I reckon they might have something resembling a strategy.
Poor Volta was actually hilaroius.
It's not marketing team, but product development team that failed to deliver.

AMD seems to have stopped saying things just because some kids want to hear them.
Chrispy_
offered better performance/Watt and performance/Transistor
AMD was also cramming more transistors into the same area on the same process, but running them at lower clocks (e.g. Polaris). It has changed with Vega/Navi.
Fluffmeister
The joys of a free market
Is on other side when monopolies or, often, even duopolies are involved.
Posted on Reply
#154
cucker tarlson
4:50

AIBs are still going for 2x8-pin,3x8-pin for the overclocking versions
Posted on Reply
#155
Th3pwn3r
I should have worded things a bit differently. What I meant was in terms of current hardware generation wise. Nvidia's offerings are ancient while AMD are brand new. It would be cool if latest gen were released from Nvidia and AMD at the same time. I'm not biased, I buy whatever performs the best with no budget for my main rig and I buy whatever offers best performance per dollar for my other machines if I choose not to use what's in storage.
Posted on Reply
#156
Vayra86
Th3pwn3r
I should have worded things a bit differently. What I meant was in terms of current hardware generation wise. Nvidia's offerings are ancient while AMD are brand new. It would be cool if latest gen were released from Nvidia and AMD at the same time. I'm not biased, I buy whatever performs the best with no budget for my main rig and I buy whatever offers best performance per dollar for my other machines if I choose not to use what's in storage.
Well the latest is that they're at least going to be about a quarter apart in release schedule, but we can also count on AMD not having AIB versions ready at launch, and you can rest assured Nvidia will open the full barrage of that close to launch.
Chrispy_
Yeah, but Turing was beaten to death by Little Navi at under $400. Hell, the $375 or so that the 2060 is going for is a joke because it can't raytrace well enough to matter and it's no faster than a 5600XT which is two price tiers below it.

Even in 2019 there was no excuse for a card costing $350 to not have at least 8GB RAM, and the $350 founders edition was as rare as hen's teeth with all of the partner cards coming in at $370-$410. For a 6GB card? Ouch!

Without Navi, Nvidia would have continued to take the piss with such stellar values as the $600 2070FE. The "Super" lineup and price cuts were Nvidia's answer to Navi. They didn't upgrade all their shader counts and reduce prices out of the goodness of their hearts or charity, that's not how businesses run.... ;)
The whole Super line up was not so much a response to Navi, it was a response to shareholders and overall Turing reception. Initially, that is. The later tweaks were definitely responses in terms of price point. But to think Super exists because Navi is Navi... please. Nvidia doesn't need AMD to release updated cards. And SUPER is pretty competitive to Navi.

Also... let's turn this around. We can moan about 6GB, but what about die size, hm? Navi is what... 255mm2. A 2060 is twice as big. Why pay nearly 400 for something with much better yields?
Posted on Reply
#157
cucker tarlson
nvidia releases stuff early,amd releases stuff late.
it's amd what need to haul ass
Vayra86
Well the latest is that they're at least going to be about a quarter apart in release schedule, but we can also count on AMD not having AIB versions ready at launch, and you can rest assured Nvidia will open the full barrage of that close to launch.
and a reference that doesn't suck

I like what nvidia are doing with FE coolers on rtx2000
it drives aib prices down.

example:
my 2070s trio is the best cooler tpu ever tested.it cost me 2560 while fe was 2400.
meanwhile 5700xt reference was 1900pln,but decent aibs were 2200 pln.
see the markup ? percentage wise it's twice as big on amd partner cards
Posted on Reply
#158
Vayra86
cucker tarlson
example:
my 2070s trio is the best cooler tpu ever tested.it cost me 2560 while fe was 2400.
meanwhile 5700xt reference was 1900pln,but decent aibs were 2200 pln.
see the markup ? percentage wise it's twice as big on amd partner cards
Initially, FE was a tweak to push AIB prices up. Not sure what idea you are trying to sell here. The mark up exists and FE caused it. Now that the mark up is in effect we say 'oh look its almost the same price'... yeah :D Its also one LOCAL example of one generation. Let's go into wait mode on that one... because so far the net result of FE is certainly not that AIBs got cheaper. The price point for each tier has gone up. Maybe what you want to say is that FE sets a price guide as a hard item in the market, which keeps things somewhat under control.... except it never really did so far in any tangible way. But theoretically... yes.

All I read out of your example above here is that quality solutions are priced accordingly, both in Navi and Turing's cases. Navi stock cooler = shit value. AIB coolers = good value. Nvidia's FE which is also open air = good value, but not quite as good as the best tested one.
Posted on Reply
#159
cucker tarlson
Vayra86
Initially, FE was a tweak to push AIB prices up. Not sure what idea you are trying to sell here. The mark up exists and FE caused it. Now that the mark up is in effect we say 'oh look its almost the same price'... yeah :D Its also one LOCAL example of one generation. Let's go into wait mode on that one... because so far the net result of FE is certainly not that AIBs got cheaper. The price point for each tier has gone up. Maybe what you want to say is that FE sets a price guide as a hard item in the market, which keeps things somewhat under control.... except it never really did so far in any tangible way. But theoretically... yes.
not a local example
msrp on fe was 500 and on trio it was 515
entry level aibs were 500

you get the idea of "what I'm selling" ?
now check 5700xt msrp
$40 premium over a $400 card for a good aib
Posted on Reply
#160
Vayra86
cucker tarlson
not a local example
msrp on fe was 500 and on trio it was 515
entry level aibs were 500

you get the idea of "what I'm selling" ?
now check 5700xt msrp
$40 premium over a $400 card for a good aib
Yeah, so how do you defend the statement that AIB price levels are lower because of the FE? Their bottom end is priced like an FE now, which is pretty close to a top end card. So for most customers, the net result is their GPU got more expensive, not cheaper.
Posted on Reply
#161
cucker tarlson
Vayra86
Yeah, so how do you defend the statement that AIB price levels are lower because of the FE?
by looking at the numbers maybe
you pay a lower % markup on best 2070s cooler than best 5700xt coolers like nitro

15 bucks on a 500 card vs 20-40 dollars on a 400 card ?

how do I make it easier for you ?
Posted on Reply
#162
theoneandonlymrk
cucker tarlson
not a local example
msrp on fe was 500 and on trio it was 515
entry level aibs were 500

you get the idea of "what I'm selling" ?
now check 5700xt msrp
$40 premium over a $400 card for a good aib
2070super 489-699£ 5700Xt 369-469£. Scan UK now.

Since when did one price statement suit reality.
Posted on Reply
#163
cucker tarlson
theoneandonlymrk
2070super 489-699£ 5700Xt 369-469£. Scan UK now.

Since when did one price statement suit reality.
I said I was comparing msrps provided in tpu reviews you may wanna start reading before our conversation blows up into a 5 page long spat


www.techpowerup.com/review/powercolor-radeon-rx-5700-xt-red-devil/
www.techpowerup.com/review/msi-radeon-rx-5700-xt-gaming-x/
www.techpowerup.com/review/sapphire-radeon-rx-5700-xt-nitro/

40-50 dollar markup on a 400 card vs $15 for 2070 super trio
Posted on Reply
#164
theoneandonlymrk
cucker tarlson
I said I was comparing msrps provided in tpu reviews you may wanna start reading before our conversation blows up into a 5 page long spat


www.techpowerup.com/review/powercolor-radeon-rx-5700-xt-red-devil/
www.techpowerup.com/review/msi-radeon-rx-5700-xt-gaming-x/
www.techpowerup.com/review/sapphire-radeon-rx-5700-xt-nitro/

40-50 dollar markup on a 400 card vs $15 for 2070 super trio
You may wanna stop implying stuff about others you can't know.

I read your comments, but feel the price of one card on day one means nothing to anyone now.

So I checked up , provided facts, and there you go staying calm :)

Stop grabbing at one offs like they're the law too.
Posted on Reply
#165
cucker tarlson
theoneandonlymrk
You may wanna stop implying stuff about others you can't know.

I read your comments, but feel the price of one card on day one means nothing to anyone now.

So I checked up , provided facts, and there you go staying calm :)

Stop grabbing at one offs like they're the law too.
so official msrp is a one off
but one store in one country at one given time is "the law"

good grief
you can't be taught
Posted on Reply
#166
theoneandonlymrk
cucker tarlson
so official msrp is a one off
but one store in one country at one given time is "the law"

good grief
you can't be taught
Your use of language tut.

It was an example , your just pushing your same Nvidia best mantra with release price's, which I think ridiculous.

No need to teach or preach ,and what right anyway.
Posted on Reply
#167
Shatun_Bear
Nvidia card prices in shops are extortionate, anyone trying to argue they're not needs to get off the green koolaid.
Posted on Reply
#168
cucker tarlson
Shatun_Bear
Nvidia card prices in shops are extortionate, anyone trying to argue they're not needs to get off the green koolaid.
they are
wasn't my point at all

my point was,the better the reference model,the harder aibs have to try to sell theirs.
Posted on Reply
#170
Vayra86
cucker tarlson
my point was,the better the reference model,the harder aibs have to try to sell theirs.
Thát is what I said too, in fact. But its not what you said. AIB models didn't get better because of the FE. The FE just positions itself higher in the price stack. But initially (Pascal), the FE was simply a markup - and AIBs did position their cards below it, and comparable cards to a Trio were priced royally above an FE. That is why I said the single example says nothing, because it doesn't exist in isolation.

The comparison to Navi reference is also unfair because Navi reference is pretty shit, its comparable to the initial FE, more so than the current one. The two aren't the same thing.

So we can be all 'oh my its so cheap compared to XYZ' but what really happens is 'you get what you pay for'.
Posted on Reply
#171
cucker tarlson
Vayra86
The comparison to Navi reference is also unfair because Navi reference is pretty shit, its comparable to the initial FE, more so than the current one. The two aren't the same thing.
it is fair.the shittier the reference,the higher the markup on quality partner coolers.
Posted on Reply
#172
Vayra86
cucker tarlson
it is fair.the shittier the reference,the higher the markup on quality partner coolers.
Alright, then we are in full agreement after all, let's leave it there lol
Posted on Reply
#173
cucker tarlson
Vayra86
Alright, then we are in full agreement after all, let's leave it there lol
that was my point from the very beginning
Posted on Reply
#174
Saabjock
Most people do not have a lot of room between the forward end of their videocard and the front of the case. Positioning the connector there may pose a few issues with case fitment.
Posted on Reply
#175
RainingTacco
Saabjock
Most people do not have a lot of room between the forward end of their videocard and the front of the case. Positioning the connector there may pose a few issues with case fitment.
You are right. With a quality flexible cable, there's no problem to place it at the side, just as usual. Manufacturers should stop putting inline caps in power connectors.
Posted on Reply
Add your own comment