I know comments about “even though I am wrong because I still love my terrible measuring device sounds” are meant somewhat ironically, the irony is not necessary.
Generally, it is not possible for an individual’s subjective opinion that something sounds “good” (to them) or “bad” to be wrong. Same with the opinion that gear X sounds “better” than gear Y.
Even in the case when people perceive differences in sound quality between two pieces when measurements show no difference in objective sonic performance, it is not the case that they are “imagining” those sonic differences.
For example, if gear X comes with a high price, outstanding build quality and visual impact, compared to sonically identical but cheap ugly gear Y, it will usually “sound better” in sighted testing.
The experience of listening combines more senses than just hearing, and our hearing is “biased” by visual and other “out of band” factors. This is a real effect.
Hearing perception is highly variable, and even repeated listenings of exactly the same source will ”sound different.”
The bad reputation of sighted listening tests comes from the historical misuse of such tests to characterize the objective audio performance of audio equipment. What demonstrated this was the many examples of listener’s, “golden ears” or not, inability to distinguish between equipment in double blind testing.
There are only two reasons for justifiably claiming that better measuring
speakers “sound better” than competing models.
The first is that listening tests show that there is a strong correlation between
listener preferences and speakers with certain objective measurements.
The second is that these correlations persist across listener types and genres of music, and are repeatable.
This does not support an assertion that a speaker “sounds better” than another. The relationship between listener
preference and ”quality” needs to be more carefully defined.
With signal processing equipment like amps and converters, where objective sonic performance is easier to measure and simpler, the idea that equipment that is “more accurate” in objective signal measurement is “better” is likewise in need of thoughtful qualification.
If before a test you set a rule that we are going to set a criteria to be measured against, for example lowest THD+N , then we can declare a winner.
One thorn-in-side of this ill-defined conception is that much audio gear (non-speaker) is sonically indistinguishable in listening tests.
This precludes the kind of disciplined
preference testing from being performed like was done for speakers.
It’s incredible to me to read comments from readers deciding to get rid of equipment they previously thought sounded just great after reading one of Amir’s reviews!
This is overshooting the mark.
Manufacturers, audio pros, or others with interest in objective audio performance (like nut-jobs involved in “competitive car audio”
will have concerns above and beyond those of everyday users, or even “audiophiles.”
There is one area where I do think there is qualified value in having high objective performance on something like and amp of DAC for consumers, which is the psychological benefit of knowing that any perceived deficiencies in a listening experience are not attributable to the performance of that gear.
Overcoming this requires careful studies on blinded listener testing.
Such studies tend not to get far, because when it becomes obvious that the listers cannot distinguish between two pieces of equipmen, the more interesting competitive question of which is “better” becomes moot.
This is somewhat of a “hot topic” on ASR discussion when it comes to the performance of class D amps, because of their high levels of noise and distortion out of the audio band.
There has been some interesting discussion about whether class D amps might be unstable with certain loads, but this so far seems purely theoretical.
Controlled, blinded listening tests that showed well class D amps to be sonically indistinguishable from comparably spec’d class AB amps could put this question to rest.
This would allow other practical considerations lIke price, heat, efficiency, size, warranty, reliability, and all around fitness for purpose to come to the fore.