Headphone Shootout: The Best and Worst Celebrity Cans Page 9


You can't throw a thumb drive these days without hitting a new headphone or headphone company. On a whole, this is a good thing. There's lots of choice, and for the most part, these new models sound better than the stock iPod earbuds.

People with built-in brand recognition obviously have an easier time selling product to non-technophile consumers and, of course to their fanbase. What this test has shown - to me anyway - is that you're paying for a name. I've reviewed many headphones at many different price points, and rarely have I come across such flawed (and often just plain bad) product as what we saw in this roundup. Even the ones we liked the most had serious problems. The winning Marleys, for example: Sure they sound great, but they're practically unwearable.

The real winners, in that regard, are the Skullcandy Aviators. Not because they were particularly good sounding or comfortable, but because they weren't particularly bad sounding or uncomfortable.

But take a step back from that. Why? Why would you buy any of these? There are so many good headphones on the market. Just because some guy with an album you like slaps his name on something, doesn't mean it's any good. Or that he's ever even seen the product.

But what do I know? I don't like buying name-brand anything. Maybe I should try selling something.

Now announcing, for the first time, the Geoff Morrison Pro Hair Gel Styling Cream. I promise, I've tested it myself.

Notes on the measurements

If you haven't seen headphone frequency measurements before, these curves may look strange. We know (to put it simply) that a speaker with a flat on-axis frequency response and a fairly flat off-axis response will generally sound best to most people.

But that's not the case for headphones, because heaphones don't work like speakers. When listening to headphones we miss out on what happens to sound moving through the air, reflected from surfaces, and transmitted through your body. So a standard headphone sounds best to most peope when it's voiced to make up for what's missing - with a substantial amount of bass boost, a dip in the midrange around 1 to 2 kHz, boosts in the midrange at about 2.5 kHz and 6 kHz, and a sharply rolled-off response above about 8 kHz. A headphone that actually offered flat, loudspeaker-like response would probably sound bad to most people.

But these rules are extremely loose, because the science isn't so certain. The experience of listening to headphones is very subjective - i.e., people don't respond as predictably to headphone response curves, because the sound is so heavily affected by the shape of the individual listener's ear. Even if you love the sound of a certain headphone, the next person might hate it. Even if you both know what you're talking about.

So you'll find plenty of headphones that measure very differently from that textbook response curve that get rave reviews - and vice versa. Just looking at the headphones tested here, the Beats Pro by Dr. Dre had the most "textbook" response - but while it scored well, it wasn't the panelists' favorite.

I measured the frequency response and sound isolating capabilities of all the headphones using a KEMAR ("Knowles Electronic Manikin for Acoustic Research") manikin, a device designed to simulate the human head for tests of headphones, earphones and microphones. I borrowed the KEMAR from the guys at Aurasound, a company that builds audio products for various companies on an OEM basis and also manufactures speaker drivers. The KEMAR has removable, rubbery fake ears with calibrated measurement microphones inside.

To measure frequency response, I drove the headphones with sine-wave test signals from the Clio using a Samson headphone amp. To measure isolation, I played pink noise from the Clio through a Genelec HT205 powered monitor and a Sunfire TS-SJ8 subwoofer. I placed each headphone on the KEMAR, measured the amount of sound getting through to the KEMAR's microphone, then divided this result by the measurement of the sound with no headphone in place. This showed me how much each headphone was cutting (or boosting) the environment noise at each frequency of sound.

You'll note that the Y-axis (left side vertical) scale for the frequency response measurements differs between headphones. This is because the sensitivity of the headphones varied so much. All of the headphones were driven at the same signal level, so you can see by the Y-axis scale how much output you can expect from each headphone relative to the others. - Brent Butterworth