October 2014

Brent ButterworthReading old issues of hi-fi mags will tell you that audiophiles have always disagreed. But thanks to the anonymity afforded by the Internet, today’s disagreements seem more intense, particularly among headphone enthusiasts. I often see e-mails, comments, and tweets from people trashing me or a different reviewer or another enthusiast for liking some headphone they don’t, or disliking some headphone they love. I almost never read such criticism regarding speakers.

Often, the language goes something like, “Oh, so you like [insert example of bad sound: bloated bass, soft treble, colored midrange, etc.].”

Or the reverse: “Well, you just don’t like [insert key element of good sound: treble, bass, accuracy, etc.].”

Should you listen to these people? Might they be right, and might your taste in headphones be flawed, or downright wrong? Before you decide, consider a couple of things.

First, the science behind headphone sound is in a primitive state compared to the science of listening to speakers in a room. Thanks to decades of research, most notably at Canada’s National Research Council (where SoundStage! does its speaker measurements), we now have a clear idea of what a good speaker is. We know that a speaker with an essentially flat response on axis and a gradual, smooth treble rolloff as you move off axis will deliver what most listeners consider a high-quality, accurate representation of a music recording.

With headphones, we really don’t know. Not yet, at least. The standard for headphone testing is IEC 60268-7, now used only piecemeal because it’s so widely regarded as flawed and/or incomplete. Recently, audio researchers such as Harman International’s Sean Olive and Todd Welti have conducted new tests to try to find out what kind of frequency-response curve most listeners will perceive as accurate. Their work has resulted in a new target curve for headphone frequency response. However, this target curve is only now beginning to be used in actual products; it will take years to establish its merit (or lack of same).

If a headphone enthusiast of relatively meager experience claims to have answers that audio engineers and scientists don’t, it seems obvious that the enthusiast has overestimated his or her knowledge.

Second, there are well-established scientific reasons why listeners don’t agree as often about headphone sound as they do about the sounds of speakers.

In my experience of conducting comparative listening tests, which I’ve done since 1992, opinions about speakers seldom vary much. I may prefer speaker A and you may prefer speaker B, but I probably don’t hate B and you probably don’t hate A. With headphones, we often see much stronger disagreements, even among experienced listeners. Sometimes, their descriptions of the sound of the same headphone can vary substantially.

It’s tempting to think that someone whose perception of the sound of a set of headphones differs from yours has flawed taste in sound, but in many cases, they really are hearing things differently.

The shapes of ear canals vary radically. In fact, the way a headphone interacts acoustically with the same person’s ear canal can vary radically, depending on how the headphone fits. At frequencies above 10kHz, shifts of headphone placement of as little as 1mm can result in response differences as large as ±4dB -- enough to give someone a completely different sonic experience.

People’s head-related transfer functions (HRTFs) also vary radically. The HRTF is the way your brain interprets the acoustical effects of your head, shoulders, and pinnae (your outer ears) so that you can tell which direction a sound is coming from. Headphones bypass these acoustical effects, but your brain still applies HRTF to the sounds you’re hearing from the headphones. Because our bodies are all different, so are our HRTF curves -- and so is the sound we perceive when we listen to headphones.

Of course, there are plenty of other good reasons why you might hear a pair of headphones differently from someone else, including age-related hearing loss, differences in fit, differences in musical taste, etc. Enthusiasts who insist that their own perception of headphones’ sound are somehow more “right” than someone else’s probably know little or nothing of what I’ve explained above. All they have is their own experience and their own opinions, which may or may not be useful to you.

Whose advice should you take about headphones? Primarily, your own. To choose the best headphones for you, I recommend following these three steps:

Step 1: Learn what accurate audio reproduction sounds like by listening to well-engineered dynamic speakers in a good listening room. Thanks to the research cited above, we know that such a system delivers accurate sound. Without such a reference, you’re really just guessing as to what accurate sound is.

Step 2: Listen to as many good headphones as you can, perhaps at retailers or at audio shows, or by borrowing them from other enthusiasts. Use the same reference recordings with all of these headphones, and take notes. Then go back and listen to your speakers to hear how the sound compares. Obviously, this works best if you can bring the headphones home to make instant comparisons.

Step 3: Compare your notes with the descriptions provided by reviewers or other enthusiasts, and see whose opinions square with yours. Those are the people whose advice about headphones merits your consideration.

. . . Brent Butterworth