Yes — in fact, it's arguably more natural in most cases to write your performance-critical stuff (e.g. math in loops) in plain C. There's rarely any reason to start boxing values up in NSNumbers and then unboxing them again.

In addition, Apple's Accelerate framework makes it easy and natural to take advantage of SIMD for further speedups — you can get powerful vector operations with single function calls. This seems to have no Android equivalent.