Comment by cylemons

I assume to save on resources, even if your algorithm is not much more taxxing on silicon, maybe the designers at intel and amd just didn't think optimizing split locks was worth it