Comment by ewzimm
2 days ago
Valve's only official statement as far as I know is that it will come when they see a significant enough hardware upgrade to warrant a new system. If they don't move to ARM, AMD's Medusa APUs are their next architecture with major upgrades, so I would guess that Valve would order another custom AMD chip but based on Medusa, which won't release until at least 2027. I would guess at least H2 2027 but probably early 2028 for an AMD-based Steam Deck 2.
I would be surprised if they moved to ARM any time soon, because even if the CPUs can punch that hard, they're definitely not competing on the GPU front, from what I understand of the state of the art outside of Apple, so they're gonna wind up with a dGPU anyway if they did.
Maybe my knowledge is out of date, but I'd be kind of surprised if a Snapdragon can get anywhere near competing with even the existing Steam Deck on GPU performance. Looking at [1] for a ballpark number on Snapdragon GPU performance doesn't seem encouraging.
[1] - https://chipsandcheese.com/p/the-snapdragon-x-elites-adreno-...
The X2 [1] is suppose to be more competitive now.
[1] - https://chipsandcheese.com/p/qualcomms-snapdragon-x2-elite
Does ARM support PCIe, by the way? Or has no one combined ARM with a dGPU because it's ridiculous to combine a low-voltage CPU with a space heater from Nvidia?
The PCI bus has nothing to do with the instruction set. Usually it is just a block a designer can add to a chip, and connect to an internal bus like AXI, give or take a few other adjustments on the chip. You can have PCIe buses without proper CPUs, even: it's quite common to find them paired with FPGAs.
For instance, Rasberry Pis have had a PCI bus for a few generations now, at first used for USB3. The Pi 5 breaks it out on a dedicated connector, making it easy to plug external devices: https://raspberrytips.com/pcie-raspberry-pi5/ (random link).
Of course, discrete GPUs are less ideal from a power efficiency perspective (duplicated memory controller, buses, and power circuits), so they wouldn't fit the Steam Deck. But write a big enough check, and I'm sure that AMD or Intel would be willing to share their iGPU designs. NVidia also makes Tegras.
AMD has an ARM chip in the works, and there's nothing about RDNA which makes it unsuitable for use alongside ARM.