
Audiophiles generally pursue the "ultimate sound," which is interpreted as sound in the listening environment that most closely simulates live music. However, many producers and mixers like to "juice" the sound, enhancing bass and treble such that the music has a chance of some fidelity in less than optimal environments, such as cars and through MP3 players. Given that an audiophile is classically not interested in either of those listening venues, should they disregard seeking out components that faithfully reproduce inaccurate sound?
The objectivist crowd loves to point to graphs and charts, displaying which component is most faithful to the incoming sine wave, and the reality that even mediocre gear can reproduce sound nearly as well as the megabuck gear. If we are to assume that this test gear can measure sound that exceeds that of human hearing, the question remains - does that which humans cannot hear factor into relevant measurements for audio equipment?
Consumers require objective measurements to offer a comparative baseline for evaluating components, prior to taking the time to audition them. What matters in the end (subjectivists and centrists argue) however is what is heard, whether that's "accurate" to a test instrument or not is a matter of semantics. Many cannot be satisfied that what they are hearing is accurate if the graphs don't support it. What becomes interesting though is how the original mix on the recording medium compares to real music (remember, that's all audiophiles care about). Therefore, if the mix on the CD doesn't compare favorably to real music, but the device that plays back the inaccurate CD offers a "flavoring" that allows mimicking of that live event, which is more critical?
Double blind testing solves much of this issue. Presented with a bank of gear, speakers, and interconnects, listeners are encouraged to select which they prefer consistently, absent visual cues. Without knowledge of what is playing, listeners cannot pre-form biases based on brand or knowledge of price point. Interestingly, although this testing still relies on human hearing as opposed to the results of a test graph, if the test is to be successful, it still relies on consistent results above the mean (typically greater than 70 percent accuracy). Note however that this type of testing measures preference, since accuracy tends to be subjective when human ears are the only means of measurement.
The debate will never end, since both sides have fundamentally valid arguments. Who can argue that faithfulness to the source is not important? Equally, who can argue that it makes no sense to attempt to seek out gear that pushes the bar closer to real life? The consumer ultimately is the one making the decision here, but in the meantime, plenty of interesting debates are in the offing in many audio-focused websites and magazine publications.