I’m not about to ask reviewers to provide audio grams, but testing my own abilities in regards to distortion and having a custom curve based on my hearing ability has made me far more aware of what, exactly, will impact my enjoyment.
I, too, have taken the testing for distortion and found myself unable to detect differences with distortion levels much higher than most would suppose.
However, I can still detect very slight spectral tilts and the presence of timbre-modifying harmonic distortions. These are generally not present in the electronics of today (at least unless intentionally colored) but they are still present in transducers and in the vintage equipment that interests me.
But I can easily hear distortion with test tones to levels much lower than with music. The difference between the -60 dB distortion in one signal generator and the -96 dB distortion in another was subtle but easy to hear, when playing a 1 KHz sine wave. The distortion at -45 from yet another signal generator was not subtle in comparison. While it is true that the music of the tests I took masks harmonic distortions at those levels, I do expect those distortions to become audible in certain situations, even if rare.
And the distortion related to clipping, or the strategies taken by different equipment to manage clipping, is likewise audible enough to detect or evaluate with training and experience. Of course, in the presence of audible clipping, distortion levels rise quickly to levels easily detected. The experience helps to detect it on dynamic percussion or sharp musical attacks where the clipping occurs for only a brief period. I don't find many measurement-based reviews to cover that aspect of performance, and it is still not obvious as to whether one amplifier (or speaker) will play realistic dynamic music louder than another amplifier (or speaker) based on specifications or the reports in reviews, even Amir's reviews.
Another reason there is still room for expert reviewers, in my view, is that they have an opportunity to audition a wide range of equipment in listening environments where they have a lot of listening experience. Few of us have that opportunity, and there seems to me enough variability in the performance of parts of the system to give value to the comparisons.
And one more thing: Reviewers for magazines, if they are any good, review usability aspects that Amir, for example, specifically does not address beyond perhaps a sentence or two about terminal spacings on the rear panel and maybe the impression of the overall aesthetics. Nor does he provide much description of the features and how they are manipulated by the user, but I see those aspects in good professional reviews, too. Amir specifically avoids those precisely because, as he said it, they are already provided in the audio press. For me to want to buy something requires not only excellent measurements but also the right set of features and controls.
All that said, too many reviews suffer from English-Major Disease and look for flowery descriptions and obscure references instead of plain, clear writing. And--with connection back to the subject of this thread--too many reviewers think their role is to be the antidote for measurements, rather than to supplement measurements with aspects the measurements didn't evaluate.
Rick "no problem with opinions when expressed as such" Denney
