Keep wondering if Apple will do some kind of NUMA multi cpu setup for Mac Pros. Say four M1 like chips splitting CPU and GPU load each with their own on package memory.
Just seems elegant to do 4 M1's and get 64gb total memory and 4x gpu/cpu and seems hard to scale up a single chip to match current multi chip / discrete gpu systems.
Maybe even dimms for main memory and current on package memory acting just as a very large level of cache.