News Intel's laptop version of its Nova Lake processors will use Panther Lake-HX BGA2540 packaging — will smooth over transition for notebook makers mov...

[Admin]

Intel's upcoming Nova Lake-HX processors will share the same BGA2540 packaging as Panther Lake-HX, simplifying platform transitions for notebook makers.

Intel's laptop version of its Nova Lake processors will use Panther Lake-HX BGA2540 packaging — will smooth over transition for notebook makers mov... : Read more

 

[edzieba]

If we can move from the stopgap of ATX12VO to full 48V power distribution (as is already commonplace for server parts) on the desktop and for laptops, that would be outstanding news. Effectively quadrupling the power capacity for the same gauge wiring or PCB traces, or quartering the current required for the same power draw. The power-hungry devices in any given system - that is, the CPU and GPU - already have their own power supplies to step down 12V to the ~1v required, so changing to a 48v to 1v stepdown is not a dramatic change, can draw designs from existing production hardware, and is slightly more efficient (lower VRM power loss).

Though more boringly, "48V EPP PD AIC" probably just means a minor typo and it supports drawing and/or supplying 48V under the USB-PD EPR standard.

 

[abufrejoval]

If we can move from the stopgap of ATX12VO to full 48V power distribution (as is already commonplace for server parts) on the desktop and for laptops, that would be outstanding news. Effectively quadrupling the power capacity for the same gauge wiring or PCB traces, or quartering the current required for the same power draw. The power-hungry devices in any given system - that is, the CPU and GPU - already have their own power supplies to step down 12V to the ~1v required, so changing to a 48v to 1v stepdown is not a dramatic change, can draw designs from existing production hardware, and is slightly more efficient (lower VRM power loss).

48V for laptops? You must be thinking about the 'luggable' category, because transforming battery voltages to 48V to then transform again in the various stages to reach the ~1V range these chips seem to consume, sound like a bad idea to me.

On the desktop? Again, it's funny you even mention that in the context of a CPU topic. Because, unless you're again talking say about a ~100 core workstation, the only power consumer worth debating is the GPU.

But for me the PC desktop is about the ability to transform it constantly, change, upgrade or rebuild from parts you largely already have it as the use cases change.

And in that context backard compatibility still rules, while max efficiency is for what I'd call appliances, even if they are built on PC parts. And that includes most laptops today, as the ability to re-configure/re-build those is largely gone.

So no, I'd say, please don't give them the wrong ideas. Those new Nvidia power connectors are bad enough already. Let them tinker with consoles or handhelds, although I think 4.8V would be more appreciated there.

 

[Amdlova]

Will be great if they are powered direct from the wall... A world without power supply...

I'am joking

 

[edzieba]

48V for laptops? You must be thinking about the 'luggable' category, because transforming battery voltages to 48V to then transform again in the various stages to reach the ~1V range these chips seem to consume, sound like a bad idea to me.

On the desktop? Again, it's funny you even mention that in the context of a CPU topic. Because, unless you're again talking say about a ~100 core workstation, the only power consumer worth debating is the GPU.

That already occurs from the 12V norm of ATX, and from the 19V norm for laptop chargers (which often means 19v > 12V > ~1V in some cases).
Converting from a higher voltage is more efficient using standard switch-mode voltage conversion, not les.

So no, I'd say, please don't give them the wrong ideas. Those new Nvidia power connectors are bad enough already. Let them tinker with consoles or handhelds, although I think 4.8V would be more appreciated there.

Moving to lower voltage distribution would be a catastrophically poor choice for efficiency and for cost. It would mean greater conversion losses, require much greater masses of copper due to the higher currents for the same power, and more heating from resistive losses. 3.3v and 5v supplies were effectively abandoned within the ATX standard (remaining as a few mA for support and standby components) for good reason.

 

[Notton]

It took some 4 years, but we now have a 240W USB-C PD3.1 charger and laptop to go with it.

The only 240W PD3.1 charger.
Sadly, they don't have the Framework 16 to test out 240W with it.

Review of Delta's World's First PD3.1 240W USB-C Power Adapter

ChargerLAB got our hands on this product as soon as possible. It supports multiple voltage outputs such as 28, 36, and 48V, and can meet the needs of a variety of high-power devices, especially the upcoming laptops equipped with RTX 50 series graphics cards. Next, we will evaluate this Delta PD3.1 2

www.chargerlab.com

The only laptop capable of 240W PD3.1 input.
https://frame.work/products/laptop16-diy-amd-7040

But if the conversion efficiency is 94~95%, then it's roughly 12~16W wasted heat output at 240W.