That's an interesting product. Wish I will have a chance to test it one day.
The reviewer does not seem to understand the basic rule of experimentation: change one variable, keep everything else constant.
I guess that depends on what question you are trying to answer. Sometimes one wants to know how complete systems compares,but it looked like further detail was provided in the third part:
"Just as I had compared individual components, I also compared the amps in isolation, that is, with one set system of components, speakers and cabling, and only the amps being switched."
I think it would be great if the author or other reviewers would continue using the comparator - perhaps in the same way a referee sometimes resort to video playback to ensure the right call is made.
BTW: What is the the best way to level match? SPL meter set to C-weighting with the microphone facing up towards the ceiling at ear height in the primary listening position?