I see that you excluded that from your opening quote.

I see that you excluded part of what I said from your opening quote:

Note that none of this is meant as much of a criticism of what you wrote.

But you certainly seemed to have taken it that way. Understandable, though, despite the disclaimer.

But I will object to one new thing you added:

Anything, no matter how small, when multiplied enough times will become a bigger thing: that bigger thing "may be significant".

That math doesn't actually work very well when talking about code optimization. The more you multiply the code, the more the significance ends up being divided.

Heck, you can't even multiply the results from Benchmark.pm by just 1. Benchmark.pm starts by telling you that something takes 150% (or 100%) more time and then I multiply the code with a 10,000-iteration loop and the difference is divided by 2 or 3 (down close to the maximum possible difference you can actually achieve even with completely contrived code, because real code doesn't have the option of ignoring overhead, like Benchmark.pm tries to do).