Tuesday, September 24th 2024
NVIDIA RTX 5090 "Blackwell" Could Feature Two 16-pin Power Connectors
NVIDIA CEO Jensen Huang never misses an opportunity to remind us that Moore's Law is cooked, and that future generations of logic hardware will only get larger and hotter, or hungrier for power. NVIDIA's next generation "Blackwell" graphics architecture promises to bring certain architecture-level performance/Watt improvements, coupled with the node-level performance/Watt improvements from the switch to the TSMC 4NP (4 nm-class) node. Even so, the GeForce RTX 5090, or the part that succeeds the current RTX 4090, will be a power hungry GPU, with rumors suggesting the need for two 16-pin power inputs.
TweakTown reports that the RTX 5090 could come with two 16-pin power connectors, which should give the card the theoretical ability to pull 1200 W (continuous). This doesn't mean that the GPU's total graphics power (TGP) is 1200 W, but a number close to or greater than 600 W, which calls for two of these connectors. Even if the TGP is exactly 600 W, NVIDIA would want to deploy two inputs, to spread the load among two connectors, and improve physical resilience of the connector. It's likely that both connectors will have 600 W input capability, so end-users don't mix up connectors should one of them be 600 W and the other keyed to 150 W or 300 W.
Above is a quick Photoshop job by TweakTown of how such a card could look like. The requirement of two 16-pin connectors should rule out older PSU types, and NVIDIA will likely only include one adapter that converts two or three 8-pin PCIe power connectors to a 16-pin, with the other input expected to be a native 600 W input from an ATX 3.0 or ATX 3.1 PSU. Most of the newer generation ATX 3.0 or ATX 3.1 PSUs in the market only have one native 16-pin connector, and three or four additional 8-pin PCIe power connectors. As for the connector itself, this could very likely be a 12V-2x6 with compatibility for 12VHPWR.
Some PSU manufacturers are beginning to release high-Wattage models with two native 12V-2x6 connectors. These would typically have a Wattage of over 1300 W. The Seasonic Prime PX-2200 W, released earlier this week, is an extreme example of this trend. Besides its high Wattage, this PSU puts out as many as four 12V-2x6 connectors. Another recent example would be the MSI MEG AI1600T PCIE5 (1600 W), with two native 600 W 12V-2x6.
Source:
TweakTown
TweakTown reports that the RTX 5090 could come with two 16-pin power connectors, which should give the card the theoretical ability to pull 1200 W (continuous). This doesn't mean that the GPU's total graphics power (TGP) is 1200 W, but a number close to or greater than 600 W, which calls for two of these connectors. Even if the TGP is exactly 600 W, NVIDIA would want to deploy two inputs, to spread the load among two connectors, and improve physical resilience of the connector. It's likely that both connectors will have 600 W input capability, so end-users don't mix up connectors should one of them be 600 W and the other keyed to 150 W or 300 W.
Some PSU manufacturers are beginning to release high-Wattage models with two native 12V-2x6 connectors. These would typically have a Wattage of over 1300 W. The Seasonic Prime PX-2200 W, released earlier this week, is an extreme example of this trend. Besides its high Wattage, this PSU puts out as many as four 12V-2x6 connectors. Another recent example would be the MSI MEG AI1600T PCIE5 (1600 W), with two native 600 W 12V-2x6.
110 Comments on NVIDIA RTX 5090 "Blackwell" Could Feature Two 16-pin Power Connectors
if using that PSU 1300-2200w just for i9 gen 14-15 series/r9 9000-10000 series with rtx 5090 48gb, how much pc gamer should use and what UPS type, for handling all power from that PSU ??
Are you so spoiled to want single power connector to the GPU? Where were you the past 15 years when almost every X60 class card were equipped with 2 8 pin connecters and the high end card were sometimes with 3x8, sometimes even 4x8? The 16pin connector is smaller than single 8 pin power connector, so what is the problem?
2030
"Nvidia have just announced their upcoming 7090 GPU, comes with 64pin connector to mean no more multiple cables, and even comes with its own 3000W PSU to ensure it has the power it needs. All for the great price $4000, The cooler has also been beefed to a 5 slot solution".
PSU vendors love Nvidia right now.
If you still cant imagine why 12V dont help to achieve these goals try to think what would happend if you tried to supply GPU cards by 5V voltage like it were in vintage days.
By the way, I never heard noises from the electricity of the GPU, you know, even if there is, when the GPU is loaded the fans are at high RPM and mute that. What trashy GPU you have to hear such noises?
Edit: One more think, currently the PSUs have 3.3, 5 and 12V. Do you want to know what will happen when you add 24V output to the PSUs? They will become more expensive, complex than now and your favorite - noisier
NVIDIA Updates its GeForce RTX Case Badge to Mention AI Acceleration Chops
:p
I guess if you can afford these top-end cards it wouldn't be a problem to get done though.
I expect the same to happen with this supposedly 600+W RTX5090. These high-end GPU don't really need to use that much power for great performance; there are commercial factors deciding that they just have to, primarily.
It is not a matter of trust - it is matter of desing constraints.
Do you have tiniest piece of idea why PMIC is mounted on each DDR5 module instead of power supply or on mobo itself ?
Shortly speaking 12V solution is archaic* these days and still exists as a matter of industry inertion.
Higher input supply voltage will force designers to spread higher power budget per more physically installed phases cos of higher ratio of conversion.
Now we have 12V --> ~1V so 12:1 ratio . In 24V supply case we will have 24:1 ratio instead.
For the matter of desing culture it is a game changer.
But some folks like you are unable to get it I see.
edit
Archaic for 400W+ power budgets. For egpu's power budget 80-150W 12V power voltage level is still fine and will be forever.
Nvidia can make efficient cards. Their RTX A2000 was phenomenal and now the RTX A4000 SFF is a little marvel out there, like a 3060ti @70W, fantastic efficiency.
Clearly they know how to create these gems.
I guess it's just not worth the effort plus the premium feeling you are gaming on a huge, gas-guzzler you can look through the glass and feel better about the money spent :P
In latest reviews it's :
14.6% in 1080p (192 vs 230 FPS)
22% at 1440p (148 vs 280.4)
27.8% at 4K (89.2 vs 114)
By "going green" you might be throwing away half the reason you spent 1750 EUR instead of 1000 EUR?
We got 600W from the 12V2x6
How in hell are they going to use more than that in a consumer-grade GPU ?
If you are pumping that much Watts, it's the intel way, it means the efficiency innovation is not fast enough and you need to pump wattage up to mark a clear improvement with Ada Lovelace.
In that case IMHO, there's no point gaming on Blackwell and I'd wait for the RTX6000, tired of companies using 300W CPUs, up to 1200W GPU because they want to use cheaper TSMC nodes that are more power hungry
I sure hope it's either a Titan, a 5090Ti or some OC AIB cards...I'm sticking with one 12V2x6 for my update (I really need to change that RTX2060)
The big issue here will also be bending. Much like the 30s didn't ever fail because the connector was angled, if this requires bending it will not be fault free but will be a step in the right direction.
Also, given the lack of card support we've seen on newer coolers and even backplates, coupled with this issue, I would recommend people only use these vertically mounted. Everything else is exceedingly problematic. They will keep breaking at the PCIe slot (or break the slot) and warranty will be denied, despite these being pretty much massive design flaws on cooler weight and proper support of it.