May 27, 2010

Sony EX71

It's not fair. Why do these in-canal phones have such sweet bass but leave something to be desired at the higher frequencies? And why the anomalies in testing/listening?

In a head-to-head (heh) comparison with the Etymotic ER6i, the EX71s sounded richer and deeper in the bass, but that's not how my instrumentation saw it. It told me that the Etymotics went deeper and had flatter frequency response. Time for a closer look.

And so the frustration begins of trying to get meaningful, much less repeatable, information when doing audio testing. One of the biggest variables is how an in-canal transducer sits in your ear. It's never the same twice, and my testing involved a lot of switching back and forth, using only one rubber ear, because I didn't want another variable (a different microphone) to enter the equation.

This first run is smoothed response to pink noise from the ER6i in red, and the EX71 in blue:

Both phones are being driven from an Archos PMA430, a player with a strong output stage that's not intimidated by low-impedance headphones. Both the EX71s and the ER6is are nominally 16 ohms. I made both measurements without changing the volume level on the player or the gain on my instrumentation.

You can see that the ER6i response is very smooth, while the EX71 is kind of bumpy. Those little stairsteps on the top end of the ER6i curve are distortion artifacts, by the way. The Archos player is distortion-free, which is another reason to use it as a source for testing like this. The distortion appears to be harmonics generated in the earphone transducer. But it's unimportant in this test anyway.

Looking at these curves, the EX71s are clearly not as efficient, and the midrange hump is sure to color the music.

I took another shot at these measurements, turning off smoothing and reinserting the transducers in the measurement ear:

This time I didn't get distortion from the ER6i at the high end, but it picked up a little 800Hz bump, probably because of slightly different location in the ear. The Sony, meanwhile, looks pretty much the same, except that it's actually doing better in the high treble range. It's also picked up some 60Hz hum. I tried to eliminate it or figure out what appliance or piece of equipment was generating it, but no dice. It's just a bump, however; draw a mental line through it.

On this third run, I used my sound level meter to set the in-ear loudness the same on both earphones, figuring that if one was quieter than the other, you'd naturally turn it up to your preferred level of enjoyment. Both of these runs were at 78dB in-ear loudness. My onscreen instrumentation read the same input level for each, too.

Equalizing the volume tells a different story. Both earphones show a peak in their response curves, and now the Etymotic run has caught the 60Hz disease. But the important thing is that when equalized for loudness, the Sonys are stronger in the bass range.

Back here in the real world of listening, the EX71s definitely have more thump than the ER6is, and nowhere near the extended top end of the ER6is. So don't believe everything you see on a computer screen.

Blocking out noise
The EX71s are typical of most in-canal earphones, and my meter shows an average reduction of 10dB. You can see the attenuation of the actual frequencies here:

The blue line is white noise in my testing room. The red line is what my measurement ear hears with the EX71 inserted. They do a good job in the midbass, but flanged designs seem to have a penchant for letting 1KHz through. They get a lot better in the high end. This performance is somewhat surprising, considering that the EX71 is a single-flange design that doesn't go very far into the ear.

These are very comfortable earbuds; they rest lightly in the ear, but don't tend to fall out. They fit very small ears well, which is sometimes a problem for in-canal phones. This measurement ear was molded from a small woman, and is very petite:

The EX71s come with two different eartips for smaller and larger ears, a hardshell case, and a unique, smaller, cylindrical case that you snap the earbuds into.

The cylindrical case prevents the earbud leads from knotting up around one another and protects them. The left cord is shorter than the right, something that has become fashionable among some earphone manufacturers. I dislike it; I'd rather have them equal length so that the cord stays centered and doesn't tug on just one ear.

The Sony EX71s are a rock-'n'-roller's ear canal phone. I haven't encountered another in-canal phone with as much bass. They're very small, very comfortable, and don't attract the "mug me" attention of The White Earbuds. They don't exert the pressure on your ear canal that you get from the phones that go deeper. They also don't deliver the crisp, laser-sharp highs of the Etymotics, but at $50 or less, they're a good trade-up from your everyday earbuds.

Posted by machrone at 10:30 AM | Comments (7)

May 23, 2010

Etymotic ER6i

That's et-ee-MOE-tic, not etty-mottic, and the ER6i is Etymotic's most popular model, retuned with greater sensitivity and better low bass performance for the iPod and its ilk, which tend to have lower output than CD players and less oomph in the bass. But low-impedance earphones like the ER6i have, and as you'll see, can cause their own problems with bass.

In-canal earphones like the ER6i have two jobs: deliver good sound and seal out noise. People generally agree that in-canal phones do a better job than earbuds, but how well do they isolate you from outside noise?

Rather well, as it turns out. But I prefer the foam eartips to the three-flange rubber eartips (above) that you usually see on these phones. The three-flange eartips are fine; they're soft and comfortable, they seal out noise pretty well, and they fit a wide range of ears.

The foam eartips are more comfortable for extended wear, however, and they somewhat outperform the white rubber ones. I really like the closed-cell material Etymotic supplies with the ER6i--it compresses easily in your fingers, expands quickly to your ear canals, and doesn't exert undue pressure. Unlike several other foam eartips, the closed-cell foam is smooth. Some of the more open foams, as on the Shure E3c and E4c, are actually kind of rough and abrasive-feeling. They also collect dirt more readily and are harder to clean.

In this photo you can see how nicely molded the foam eartips are, and on the other driver you can see the filter at the end of the sound tube. The filter is a small foam-filled plug that does several things beyond the obvious one of keeping ear wax out of the tube. It provides a bit of backpressure on the diaphragm, which, along with the acoustical filtering properties of the filter, smooths out the frequency response. Without the filters, the ER6is sound harsh and brassy.

I test the frequency response in rubber ears that have been molded from human ears, with calibrated microphones behind them, mounted the proper distance into the ear canal. This is homemade equipment, not a $20,000 anatomically and acoustically correct binaural head. But I've been an electronics and audio hacker all my life, and I think my homegrown rig is pretty good, and definitely yields usable comparative results if not absolute, acoustics lab accuracy.

Etymotic rates the frequency response of the ER6is from 50Hz to 6KHz ±3dB and from 20Hz to 16KHz ±6dB. My measurements are consistent with the ±3dB rating, but depending on what you're driving them with, you may never hear a 20Hz tone from your ER6is (more on this later). My test rig attenuates both ends a bit. but is very consistent in the way it treats headphones, earbuds, and in-canal phones. At 50Hz, the response is at least 10dB weaker than my studio-grade Sennheiser HD280 headphones, but the ER6is catch up quickly and are dead flat across the spectrum. They tail off gracefully at the top end, but audio material with lots of highs is not the least bit constrained.

The ER6is do a decent job of blocking noise. Quoting a single dB figure is pointless since the amount of attenuation varies so much by frequency. So here's a picture of what they actually do:

The blue line is white noise in my testing room, measured through a rubber ear. It's coming from five speakers, and it measures an average 85dB. The red line is what the microphone in the ear with the ER6i equipped with the three-flange rubber eartip hears. As you can see, the attenuation is all over the place, and 1KHz lances right through. My equipment calls it a 10dB average reduction, but obviously it's much greater than that at the higher frequencies.

Here's the same test, but with the foam eartips. Notice how much better the high-frequency attenuation is, along with a couple of dB improvement in the midbass.

Looking at these charts, bear in mind that white noise has a degree of randomness in it, and these traces dance around a fair amount during testing. Your eye averages them out, but what you see when you grab a screen shot may not agree with what your eye was telling you. So I pull the trigger several times and use the one that most closely matches the active, average performance that I see on the screen.

Also, sound reaches your ears in paths other than your ear canals, and these tests don't account for it. Your skull, neck, and upper chest are all entry points.

The ER6is have a wonderful, open sound. They're a pleasure to listen to, but I admit that I'd like more thump in the bass. Piano, violin, and voices sound almost magical; they're clean and well-articulated. Brass is never strident. A jazz drumset is crisp, present, right up front. The lowest notes in the bass are more implied than forcefully stated, however. By contrast the Sony EX-71 in-canal phones have more bass but a lot less definition in the trebles and midrange.

If you wear the ER6is extensively, you'll eventually need to replace the filters. The sound gets dull as they become clogged with dirt and ear wax. Etymotic provides an extra pair, plus a little extraction tool that removes the old ones (destroying them in the process). Replacement filters are not exactly cheap at $12.95 for a pack of six; Q-tips are a wise investment.

In my testing, one of the filters came loose in the sound tube and created a buzzing rattle. I simply pressed it back into place, and didn't encounter any additional problems.

Etymotic increased the sensitivity of the ER6i over the original ER6 by reducing the impedance. I think that this is the wrong solution for most digital audio players, especially those with already-limited bass response. The limited power of digital audio players is a good reason to increase sensitivity--people want their loudness! But low impedance hurts bass response when there isn't power to spare because it just sucks more power out of the player than it can provide.

This quick-and-dirty Excel chart shows what I mean:

This is the performance of the player with two different sets of phones attached, sweeping from 20 to 400Hz. The player was a Creative Zen Micro, which has roughly the same bass performance as a white iPod, maybe a little better. As you can see, the ER6is cause a 5dB fall-off in the bass at 20Hz, down to 3 or so at 40Hz, 2dB at 80Hz, and fading into insignificance above that.

Irrespective of what the earphones can do, if the player can't deliver the bass because it's being loaded down, you're hardly going to hear the sweet lows; they'll be pushed far into the background.

I tried the same thing with an Archos PMA430, a multimedia player with prodigious amounts of audio power. The Etymotics still pulled down the response, but only by about 3db at the lowest frequency. I ran the tests flat, but the PMA430 can dial in large amounts of EQ without introducing distortion. The ER6is welcomed the bass boost.

This interaction between player and earphone is something you need to be aware of if you're looking for brain-compressing bass. Since so many players add distortion when you use their EQ settings, you need to be doubly cautious. The ER6is with a player of the PMA430's caliber are a real pleasure, but those low notes are noticeably "less there" on lesser players than I like.

The wires on the ER6is are very thin, and are prone to "sing" if you walk quickly outdoors or if there's a breeze. The tone is conducted right into your ears, so it's best to insert the drivers upside down, with the wires up, and loop them over and behind your ears if you're outdoors. Etymotic provides a clip so you can secure the wire to your clothing, as well as a nice little soft carrying case.

At their $129 street price, the Etymotics are more or less in the middle of the range for in-canal earphones. They aren't perfect, but they're the best in their price range. If you want to hear what they can really do, invest in or build a headphone amplifier. You'll be amazed.

Posted by machrone at 06:04 PM | Comments (2)