Multi-core and single-core performance seem to have a lot of importance focused on their distinction, enough for Cinebench to maintain two separate categories and for UserBenchmark to be ostracized for their focus on single-core performance being seen as a bias for Intel CPUs or at least as a deprecated way of thinking…
Depends on the software. Anyone sensible will run a SIM at least partly on the GPU to take advantage of the massively parallel vector math acceleration - it's orders of magnitude faster than doing it on the CPU for operations that can be vectorised (like physics) That doesn't mean that the software you use actually does…