What degree of OS-indepedence do you think is reasonable to expect from a consumer-facing benchmark suite like Geekbench?

Does that matter? I don't care of Apple's thread scheduler is faster or slower than Windows' or if their throttling algorithm behaves differently, I just want to see how it performs in a given task.

AFAIK it's mostly self-contained outside of basics like malloc. If you have any evidence to the contrary, post it.