AMD has the crown until Intel decides to up core counts and PCIe lanes, and they've started doing the former.
On the flipside, many of the gains in the AMD camp are still enthusiest only. ECC support for instance doesn't matter to any of us, but it's still officially 'unsupported' (but it *might work* for a given app on a given board with a given CPU) and high bandwidth devices still have PCIe teething issues. Meaning there is chipset errata that might affect users of high bandwidth i/o cards, which is going to be based on EFI & board implementation rather than CPU. This again won't affect us...
So overall good news for audio users, at a time when Apple is throwing their own monkey wrench into things. AMD forcing them to move away from quadcore limited PCIe lane consumer implementations while Apple will hit them from the other direction with efficiency & power gains. Win-win in my book (and if RISCV ever gains traction, we might even get some pressure provided via the open sourcing of hardware designs and i/o busses).
My guess would be it means easier integration for partners, there will still have to be some sort of bridge chip because you can't have PCIe connections (trace) that long without it. I'll have to read up. I would say a TB based Xite might be a nice thing to see in the future as it's finally looking to be (almost) standard everywhere, but Apple might just muck it all up as usual.
