diotalevi,
Remove what function calls? The ones that make the dispatch table a dispatch table and subsequently the exact thing I am benchmarking? While I value chromatic's opinion (and your's), there is no IO involved. He was making the point that it is silly to worry about a few microseconds here and there if your program was doing IO.

So what are you guys trying to say - it is pointless to Benchmark any program that does IO or has function calls? I specifically isolated the bench to the difference between an if/elsif chain and its equivalent dispatch table. I appreciate constructive critism but I am not seeing it here.