Cache size and memory bandwidth can outweigh multi-core speed in 4K video work

[Samuel.M]

I'm building a new workstation for 4K video editing and 3D rendering, and I'm trying to decide between the latest high-core-count CPUs from AMD and Intel. I've been looking at CPU benchmarks on various review sites, but the results seem to vary wildly depending on the specific software used for testing. For a real-world workload in DaVinci Resolve and Blender, is raw multi-core performance the only metric that matters, or are there other architectural factors like cache size and memory bandwidth that make a bigger difference in practice than the synthetic benchmarks suggest?