There is definitely more going on with the named argument call, as more functions are called. On the other hand, the function calls are really simple and, if they are defined in the header file, and the entire implementation is visible to the compiler, there is no reason for the compiler not to inline all the Set() calls and eliminate the unnecessary temporary variables. With good optimization, we could expect similar performance from the named arguments idiom and the explicitly named aggregate object.
The appropriate tool to measure the performance of a single function call is the microbenchmark. We use the Google microbenchmark library for this purpose. While the benchmarks are usually written in ...
