If there were to be a case for more customised semi-custom silicon next generation, perhaps it could be along the lines of Nvidia's tensor cores and image flow processor gubbins.
The 3D pipeline and even GPU Compute (using shader hardware) seems fairly standardised-ish now (give or take), but upscaling and frame interpolation seems like the wild west. NV Tensor, Intel XMX, AMD .... nothing.
Not sure if there's be some kind of standardisation before next gen consoles roll around.
Edit: in other but not completely unrelated news, Toms have had a first look at XeSS on none Intel hardware with Shadow of the Tomb Raider:
Benefits Pascal and later, RDNA 2 and Arc — most of the time
www.tomshardware.com
As a few of us speculated, uses DP4a where available, emulates on SM 6.4 hardware where DP4a not available. A little faster in quality mode than native where DP4a is supported, slower where emulated (e.g. RX5700 XT).
@Dictator, would you mind dragging yourself over the hot coals and broken glass of image quality and performance once again to let us know about how the quality settings behave across different hardware? I know it's painful for you, but for the rest of us it's an interesting half hour over a brew!