It's the outcome of trying to optimize for cost while trying to maintain good performance on two very different categories of workload, and hurriedly throwing together a product based on the components they had on hand (the Core and Atom families of CPU core designs).

For well scaling workloads optimizing for efficiency is optimizing for performance. For serial workloads even one P-core is enough. There aren't too many workloads in the middle that scale but only to a few cores.