Reviewed on: SoundStage! Solo, April 2022
I measured the HiFiMan Arya Stealth Magnet Version headphones using laboratory-grade equipment: a GRAS Model 43AG ear/cheek simulator/RA0402 ear simulator with KB5000/KB5001 simulated pinnae, and an Audiomatica Clio 12 QC audio analyzer. For isolation measurements, I used a laptop computer running TrueRTA software with an M-Audio MobilePre USB audio interface. For most measurements, the headphones were amplified using a Musical Fidelity V-CAN amplifier; I used a Schiit Magnius amplifier for distortion measurements. These are “flat” measurements; no diffuse-field or free-field compensation curve was employed. If you’d like to learn more about what our measurements mean, click here.
This chart shows the Aryas’ frequency response. This is very typical for HiFiMan headphones—a nearly flat bass response, with a lot of energy in the 3-to-5kHz range.
Here we can see the difference in the headphones’ response when a high-impedance (75 ohms) source is substituted for a typical low-impedance source (5 ohms). As usual with planar-magnetic headphones, the difference is insignificant.
This measurement shows how sensitive the headphones’ response is to the headphones’ position on the ears. The black line shows the curve I used above, which is the curve I got most often with the headphones optimally positioned on the ear/cheek simulator. For the other curves, I reseated the headphones a few times, and also shifted them about 5mm up, down, forward, and back on the cheek plate. The only range that differs significantly with the repositionings is between 6 and 8kHz, so positioning of these around the ears isn’t all that critical.
This chart shows the Aryas’ right-channel response, compared with some other open-back models—including the Dan Clark Audio Æon 2s with perforated earpads, which have a response similar to the Harman curve. The Aryas definitely have a lot of energy between 3 and 5kHz, but the HE6se’s have even more. The Dan Clark model is similar in the treble, but has more bass to balance out the sound.
The spectral-decay (waterfall) response of the Aryas has the usual “hash” between about 1 and 6kHz, which is a result of comb-filter effects caused by sound bouncing off the flat planar-magnetic diaphragm. These effects don’t seem to add any particular coloration to the sound, and they may create a greater sense of spaciousness.
This chart shows the total harmonic distortion, measured at 90dBA and 100dBA (both levels set with pink noise). As usual with planar-magnetic headphones, it’s very low. It does hit about 3% at 20Hz, but that level of distortion is not audible at such a low frequency, and it’s at the extremely loud level of 100dBA, which I doubt your ears could stand for long, anyway.
In this chart, the external noise level is 85dB SPL (the red trace), and numbers below that indicate the degree of attenuation of outside sounds. The lower the lines, the better the isolation. The Aryas’ isolation is typical for headphones in its class.
As with most planar-magnetic headphones, electrical impedance magnitude is just about dead-flat, in this case at 30 ohms, and impedance phase shift is close to zero.
Sensitivity of the Aryas, calculated for 32 ohms rated impedance and averaged from 300Hz to 3kHz, is 94.5dB, so using some kind of an external amp, DAC-amp, or high-quality portable music player with these is a good idea.
Bottom line: The frequency-response measurements confirm that these headphones will sound a little bright. I can find nothing in the measurements that should concern a potential buyer.
. . . Brent Butterworth
brentb@soundstagenetwork.com
Reviewed on: SoundStage! Solo, April 2022
I measured the PSB M4U 8 MKII headphones using laboratory-grade equipment: a GRAS Model 43AG ear/cheek simulator/RA0402 ear simulator with KB5000/KB5001 simulated pinnae, and an Audiomatica Clio 12 QC audio analyzer. A Reiyin WT-04 USB Bluetooth transmitter was used to send signals from the Clio 12 QC to the headphones. For isolation measurements, I used a laptop computer running TrueRTA software with an M-Audio MobilePre USB audio interface. For cabled measurements, the headphones were amplified using a Musical Fidelity V-CAN amplifier. These are “flat” measurements; no diffuse-field or free-field compensation curve was employed. If you’d like to learn more about what our measurements mean, click here.
This chart shows the M4U 8 MKIIs’ frequency response, using a Bluetooth connection with noise canceling off, which was my favorite listening mode. This is a fairly conventional response, except that the center frequency of the broad peak between 3 and 7kHz might normally be 1 to 2kHz lower in frequency.
Here we can see the difference in response between noise canceling on and off, and the Bluetooth connection and the wired connection (which can be either purely passive, or used with the internal amp on and noise canceling on or off). Noise canceling introduces an upper-bass dip (which is probably why I perceived it as sounding brighter), and a large peak centered at 32Hz, which won’t be very, if at all, audible with most music.
This measurement shows how sensitive the headphones’ response is to the headphones’ position on the ears. The black line shows the curve I found to be the most typical. For the other curves, I reseated the headphones a few times, and also shifted them about 5mm up, down, forward, and back on the cheek plate. Above 200Hz, the variance is negligible. Below 200Hz, the differences are large, which means the bass response of these headphones is especially sensitive to positioning—probably due to the small earpads, which make this more of a semi-on-ear design rather than a full over-ear design.
This chart shows the M4U 8 MKIIs’ right-channel response in Bluetooth mode with noise canceling off, compared with the DALI IO-6 (in the same mode), the Philips Fidelio L3 (ditto), and the AKG K371 headphones. The K371 are said to be very close to the Harman curve. Other than having comparatively less output around 2kHz, the M4U 8 MKIIs look pretty conventional.
The M4U 8 MKIIs’ right-channel spectral-decay plot (measured with the wired connection) shows no noteworthy resonances except just a bit in the upper bass, around 200Hz (the lowest frequency at which this particular measurement works).
Here’s the THD vs. frequency chart, measured using the wired connection (with the power off) at 90dBA and 100dBA (both levels set with pink noise). Distortion’s almost non-existent at 90dBA, but it gets pretty high at 100dBA. Doing the same measurement with Bluetooth pushed the distortion a little bit higher. Note that 100dBA is an extremely loud listening level; really, 90dBA is louder than most people can stand for long.
In this chart, the external noise level is 85dB SPL (the red trace), and numbers below that indicate the degree of attenuation of outside sounds. The lower the lines, the better the isolation. The M4U 8 MKIIs’ noise canceling is pretty competent, delivering a 15 to 20dB cancelation in the “airplane cabin band” between 100Hz and 1kHz, but it falls short of class leaders like the Bose N700 NC headphones.
Latency, measured with a standard SBC Bluetooth connection, typically ran in the range of about 170 milliseconds, although it occasionally dipped down as low as about 110ms.
The impedance magnitude, measured in passive (power off) mode, is flat, measuring 50 to 53 ohms, and the phase response is similarly flat. With the power on, the impedance rises to about 20kHz, which is typical of a preamp input.
Bottom line: Nothing particularly troubling here—there’s nothing extraordinary about the way these headphones measure, and nothing really anomalous outside of the sharp rise in distortion at 100dBA.
. . . Brent Butterworth
brentb@soundstagenetwork.com
Reviewed on: SoundStage! Solo, April 2022
I measured the Grell Audio TWS/1 earphones using laboratory-grade equipment: a GRAS RA0402 ear simulator and an Audiomatica Clio 12 QC audio analyzer. For isolation measurements, I used a laptop computer running TrueRTA software with an M-Audio MobilePre USB audio interface, and added the KB5000/KB5001 simulated pinnae with the full Model 43AG ear/cheek simulator. I used two different Bluetooth transmitters—a Reiyin WT-04 USB and a MEE Audio Connect—to send signals from the Clio 12 QC to the earphones. Due to the complications introduced by the ever-changing latency of the Bluetooth transmitter, I used FFT spectrum analysis for some measurements. This looks somewhat different. Because of the complications of measuring these, and the various unusual modes on them, this presentation will be somewhat different from the norm for this site. These are “flat” measurements; no diffuse-field or free-field compensation curve was employed. If you’d like to learn more about what our measurements mean, click here.
I started out by measuring the maximum volume of the TWS/1s, compared with the Denon AH-C830NCW and the Jabre Elite Active 65t earphones (two other true wireless models), using -10dBFS pink noise and A-weighted metering with Room EQ Wizard. The level was 92.2dBA with SoundID off and 99.6dBA with SoundID on. In comparison, I measured a maximum volume of 100.3dBA for the Jabras and 100.6dBA for the Denons.
This chart shows the frequency response of the TWS/1s, right channel, with SoundID on and off. You can see that the black line (SoundID) looks fairly typical for dynamic-driver earphones. SoundID—tuned for my ears—boosted the response a little at about 5.5kHz, and a lot at about 12kHz, which are both pretty good decisions, based on my most recent audiograms. They also add a lot of upper bass and lower midrange, between about 100 and 700Hz. I don’t have any hearing loss in this range (hearing loss in this range is rare), so I assume this is to sculpt the earphones’ sound to meet the SoundID profile.
But notice how much higher the volume is with SoundID on: an average of 5.97dB higher between 20Hz and 20kHz. This is bogus; it makes legitimate, level-matched comparisons between SoundID and the unprocessed sound practically impossible for average listeners. I wouldn’t use the word “bogus” if I weren’t convinced that every audio scientist alive would agree with me. I understand that level-matching two different EQ profiles is almost impossible to do with complete certainty, but clearly there was no intent here to make the comparison legitimate.
Here we can see the difference in frequency response with noise canceling on and off. There’s very little difference, which is admirable.
This chart shows the right-channel response of the TWS/1s (with SoundID off) compared with a couple of other true wireless models, and with the Sennheiser IE 300s, a set of passive earphones I like. The TWS/1s, in their native voicing, have the most “normal”-looking response, and it’s not too far off the Harman curve.
I thought I had heard some distortion in the TWS/1s, but it doesn’t look from the measurements like this is a real concern—the distortion is very low even at the very loud level of 99.3dBA (measured with pink noise). Normally I would use 100dBA but I couldn’t get the TWS/1s to play that loud.
In this chart, the external noise level is 85dB SPL (the red trace), and numbers below that indicate the degree of attenuation of outside sounds. The lower the lines, the better the isolation. We’ll start by showing the differences among the earphones’ listening modes. Interestingly, the NAR function—which according to my listening works very well in not-so-loud surroundings—doesn’t seem to do a lot in the presence of loud noise. But it probably shouldn’t, because the sounds it’s trying to eliminate would likely be masked. This performance is consistent with the way Grell describes the function—it seems intended primarily to reduce high-frequency noise in relatively quiet environments.
This chart shows how the TWS/1s’ noise canceling performs compared with some other leading true wireless models with noise canceling. It may not look all that impressive compared with some of the others, but in my experience, the roughly 15dB of noise canceling the TWS/1s offer is plenty enough to make listening to music much easier and more pleasant on airplane flights.
Whether I used the Reiyin or MEE transmitter—both of which have aptX Low Latency, but not aptX Adaptive, which the TWS/1s use—latency jumped around between about 220 and 270ms. So you’ll probably notice a bit of latency when watching movies or playing games on your phone, but it depends on the phone.
Bottom line: Other than the annoying level jump when SoundID is turned on, the Grell TWS/1s seem to have no notable technical issues.
. . . Brent Butterworth
brentb@soundstagenetwork.com
Reviewed on: SoundStage! Solo, March 2022
I measured the Denon AH-C830NCW earphones using laboratory-grade equipment: a GRAS RA0402 ear simulator and an Audiomatica Clio 12 QC audio analyzer. For isolation measurements, I used a laptop computer running TrueRTA software with an M-Audio MobilePre USB audio interface, and added the KB5000/KB5001 simulated pinnae with the full Model 43AG ear/cheek simulator. I wasn’t able to get my usual logarithmic chip frequency-response measurements using my Reiyin WT-04 USB Bluetooth transmitter to send signals from the Clio 12 QC to the headphones, so I used FFT spectrum analysis instead, which looks somewhat different. I was also unable to do distortion measurements. These are “flat” measurements; no diffuse-field or free-field compensation curve was employed. If you’d like to learn more about what our measurements mean, click here.
This chart shows the AH-C830NCW earphones’ frequency response. This is flatter than we usually see with earphones, but flat isn’t necessarily good with earphones or headphones. There’s more lower-midrange energy than we usually see, which my ears mistook for excess upper-bass energy.
Here we can see the difference in frequency response with noise canceling on and off, and with Ambient mode activated. Clearly, there’s very little difference, which is admirable—we still see too many headphones and earphones that sound substantially different when noise canceling is on.
This chart shows the right-channel response of the AH-C830NCWs compared with a couple of other true wireless models, and with the Sennheiser IE 300s, a set of passive earphones I like. (The Grell TWS/1 earphones’ response is shown without SoundID activated.) It’s obvious how much flatter (and therefore more midrangy) the Denons are than the others.
In this chart, the external noise level is 85dB SPL (the red trace), and numbers below that indicate the degree of attenuation of outside sounds. The lower the lines, the better the isolation. We’ll start by showing the differences among the earphones’ three listening modes: noise canceling on and off, and Ambient mode on.
This chart shows how the AH-C830NCWs’ noise canceling performs compared with some other leading true wireless models with noise canceling. It’s not Bose-class, but more than good enough to bring airplane cabin noise down to the point where it won’t interfere with your listening.
With the Reiyin transmitter, latency was usually around 120ms with the SBC codec. (The Reiyin doesn’t have AAC, so I couldn’t test that mode.) However, latency occasionally jumped to about 220ms.
Bottom line: The AH-C830NCWs have an unusually flat frequency response, which you may or may not like, and the noise canceling is very good.
. . . Brent Butterworth
brentb@soundstagenetwork.com
Reviewed on: SoundStage! Solo, March 2022
I measured the Monoprice Monolith M1570C headphones using laboratory-grade equipment: a GRAS Model 43AG ear/cheek simulator/RA0402 ear simulator with KB5000/KB5001 simulated pinnae, and an Audiomatica Clio 12 QC audio analyzer. For isolation measurements, I used a laptop computer running TrueRTA software with an M-Audio MobilePre USB audio interface. For most measurements, the headphones were amplified using a Musical Fidelity V-CAN amplifier; I used a Schiit Magnius amplifier for distortion measurements. These are “flat” measurements; no diffuse-field or free-field compensation curve was employed. If you’d like to learn more about what our measurements mean, click here.
This chart shows the M1570Cs’ frequency response with the lambskin pads. If that 3kHz peak were about 3dB lower, and the bass about 5dB higher, these would look not too far off the Harman curve. But that’s a big 3kHz peak.
Here’s the difference between the lambskin and velour pads. You can see that despite the extra 2dB or so of energy at 3kHz with the lambskin pads, the 5dB reduction in bass output with the velour pads makes the velour pads sound brighter.
Here we can see the difference in the headphones’ response when a high-impedance (75 ohms) source is substituted for a typical low-impedance source (5 ohms). As with most planar-magnetic headphones, the difference is negligible and probably inaudible.
This measurement shows how sensitive the headphones’ response is to their position on the ears, using the lambskin pads. The black line shows the curve I used above, which is the curve I got most often with the headphones optimally positioned on the ear/cheek simulator. For the other curves, I reseated the headphones a few times, and also shifted them about 5mm up, down, forward, and back on the cheek plate. There’s not much variance, so the sound isn’t all that fit-dependent. Results with the velour pads, which don’t create a tight seal, should be even more consistent.
This chart shows the M1570Cs’ right-channel response with the lambskin pads, compared with a few other closed-back models and the open-back M1570s. It’s clear that the M1570Cs’ ~3kHz peak is unusually strong, and that’s likely to make them a little blarey-sounding. It’s also apparent that the M1570s sound substantially different from the M1570Cs, which is disappointing to me—I’d have hoped the M1570Cs would basically be just the M1570s with closed backs, but they’re really a different set of headphones.
The spectral-decay (waterfall) response of the M1570Cs (measured with the velour pads) looks clean except for an unusual and strong, but narrow (i.e., high-Q), resonance at about 2.5kHz, which corresponds to that zig-zaggy peak/dip in the frequency response in this range. I’d guess it’s audible if whatever you’re listening to has significant content at that frequency. There’s also a smaller resonance at about 4.5kHz.
This chart shows the total harmonic distortion, measured at 90dBA and 100dBA (both levels set with pink noise, using the velour pads). It’s near zero—one of the best results I’ve seen for this measurement. Very crankable if used with an amp.
In this chart, the external noise level is 85dB SPL (the red trace), and numbers below that indicate the degree of attenuation of outside sounds. The lower the lines, the better the isolation. The M1570Cs’ isolation is typical for headphones in its class, and it’s a few dB better with the lambskin pads.
As with most planar-magnetic headphones, electrical impedance magnitude is dead-flat (in this case at 62 ohms) and impedance phase shift is close to zero.
Sensitivity of the M1570Cs, calculated for 60 ohms rated impedance and averaged from 300Hz to 3kHz, is 97.1dB with the velour pads, 96.8dB with the lambskin pads, so using some kind of an external amp, DAC-amp, or high-quality portable music player with these is a good idea.
Bottom line: These seem well-engineered, but that’s a big 3kHz peak. Those looking for headphones with a familial sonic relationship to the excellent M1570s better look elsewhere.
. . . Brent Butterworth
brentb@soundstagenetwork.com
Reviewed on: SoundStage! Solo, March 2022
I measured the Meze Audio Liric headphones using laboratory-grade equipment: a GRAS Model 43AG ear/cheek simulator/RA0402 ear simulator with KB5000/KB5001 simulated pinnae, and an Audiomatica Clio 12 QC audio analyzer. For isolation measurements, I used a laptop computer running TrueRTA software with an M-Audio MobilePre USB audio interface. For most measurements, the headphones were amplified using a Musical Fidelity V-CAN amplifier; I used a Schiit Magnius amplifier for distortion measurements. These are “flat” measurements; no diffuse-field or free-field compensation curve was employed. If you’d like to learn more about what our measurements mean, click here.
This chart shows the Lirics’ frequency response. Looks pretty normal, with maybe a little more midrange than usual, and with a little less bass than usual.
Here we can see the difference in the headphones’ response when a high-impedance (75 ohms) source is substituted for a typical low-impedance source (5 ohms). Normally with planar-magnetic headphones, the difference is at most 0.5dB, usually in the bass. But here the difference is substantial—a drop of about 1.5dB in the bass and lower midrange, and a boost of 1 to 3dB in the upper midrange, between 1.2 and 2.8kHz. So these headphones will sound noticeably brighter when used with high-impedance source devices, like some older (and even a few newer) solid-state amps, some pro amps, and many tube amps.
This measurement shows how sensitive the headphones’ response is to their position on the ears. The black line shows the curve I used above. Normally I pick a curve that’s somewhere in the middle, but it’s pretty clear that the one with the most treble response is “right.” Notice how the treble response falls a few dB when I moved the headphones off their optimal position? This indicates that the quasi-two-way design of the Rinaro diaphragm really works—although it also suggests that you should move the headphones around slightly when you put them on to make sure you’re getting optimal treble response.
This chart shows the Lirics’ right-channel response compared with a few other open-back models (including the Dan Clark Audio Æon 2s with perfed pads, which are the planar-magnetic headphones I’ve found come closest to the Harman curve). The Lirics are well in the ballpark of normal, with a little more midrange energy and a little less bass than Harman-curve headphones.
The spectral-decay (waterfall) response of the Lirics is very interesting. All those very-high-Q resonances you see are typical of open-back headphones, especially planar magnetics; I think it’s part of what gives them a more spacious sound. But here, the resonances cover a much broader part of the audio spectrum than usual, and are lower in magnitude than usual. Something interesting is going on here—maybe it’s that Phase-X thing at work.
This chart shows the total harmonic distortion, measured at 90dBA and 100dBA (both levels set with pink noise). The distortion is low—not the near-zero distortion I’ve measured with many planar-magnetic headphones, but it should be well below audible levels for headphones.
In this chart, the external noise level is 85dB SPL (the red trace), and numbers below that indicate the degree of attenuation of outside sounds. The lower the lines, the better the isolation. The Lirics’ isolation is typical for headphones in this class, although a little better than most below 1kHz.
Most planar-magnetic headphones have a dead-flat impedance curve, or close to it. Not these. The impedance is generally flat, at about 22 ohms, through most of the audio range, but there’s a huge peak up to 89 ohms at 1.9kHz, with an accompanying phase swing, too—confirming that these headphones’ sound will be unusually sensitive to the output impedance of the source device.
Sensitivity of the Lirics, calculated for 30 ohms rated impedance and averaged from 300Hz to 3kHz, is 95.9dB, so you’ll need something with a few dB more oomph than a smartphone or tablet to get good performance from them.
Bottom line: Other than that funky impedance curve, and the resulting tonal balance shift with high-impedance sources, the Lirics are well within the range of normal for high-end, closed-back headphones.
. . . Brent Butterworth
brentb@soundstagenetwork.com
Reviewed on: SoundStage! Solo, February 2022
I measured the Grado RS2x headphones using laboratory-grade equipment: a GRAS Model 43AG ear/cheek simulator/RA0402 ear simulator with KB5000/KB5001 simulated pinnae, and an Audiomatica Clio 12 QC audio analyzer. For isolation measurements, I used a laptop computer running TrueRTA software with an M-Audio MobilePre USB audio interface. For most measurements, the headphones were amplified using a Musical Fidelity V-CAN amplifier; I used a Schiit Magnius amplifier for distortion measurements. These are “flat” measurements; no diffuse-field or free-field compensation curve was employed. If you’d like to learn more about what our measurements mean, click here.
This chart shows the RS2xes’ frequency response. It’s quite unusual. There’s not much bass. There’s an unusual cancellation/reinforcement peak in the 250Hz region, although its bandwidth is narrow enough that it probably won’t be very audible. Now up to the treble—we’d normally see something like a broad peak centered somewhere around 2.5 to 3kHz, followed by a narrower peak centered at maybe 6 or 7kHz. But here we see narrow, high-Q peaks at 2.2 and 4kHz, followed by a broad lift between 7 and 10kHz. What’s that sound like? Hard to say without having heard more headphones that measure like this, but it’s definitely going to be trebly.
Here we can see the difference in the headphones’ response when a high-impedance (75 ohms) source is substituted for a typical low-impedance source (5 ohms). There’s about 4.5dB more bass in the 70Hz region with the high-impedance source, so I’m guessing these headphones would sound especially nice with a tube amp.
This new measurement shows how sensitive the headphones’ response is to the headphones’ position on the ears. The black line shows the curve I used above, which I found to be most typical; the other curves are taken with the headphones moved about 5mm up, down, and to each side. As expected with open-back headphones that use highly porous earpads, positioning of these on the ears isn’t very critical.
This chart shows the RS2xes’ right-channel response compared with a few other open-back models (including the Dan Clark Audio Æon 2s with perfed pads, which are the planar-magnetic headphones I’ve found come closest to the Harman curve). Clearly, the RS2x headphones have more overall treble energy and a lot less bass than the others.
That “hash”—a series of very high-Q resonances—between 2 and 4kHz is pretty standard for open-back headphones; I think it’s part of what gives them a more spacious sound. Notice also the big resonance around 250Hz—the same spot where that cancellation/reinforcement dip/peak appears in the frequency response. Maybe that peak and dip are more audible than I expected.
This chart shows the total harmonic distortion, measured at 90dBA and 100dBA (both levels set with pink noise). The distortion here is very high in the bass; I normally don’t get too concerned about distortion, but I wouldn’t recommend these for anyone who listens to music with lots of deep bass, or who likes to play their headphones loud.
In this chart, the external noise level is 85dB SPL (the red trace), and numbers below that indicate the degree of attenuation of outside sounds. The lower the lines, the better the isolation. Like most open-back headphones, the RS2xes don’t offer much isolation.
The impedance runs between 40 and 45 ohms through most of the audio range, but with a peak at the driver resonance at 82Hz. There’s a big phase swing at that frequency, too. That’s why the bass response changes so much with a high-impedance source.
Sensitivity of the RS2x headphones, calculated for 38 ohms rated impedance and averaged from 300Hz to 3kHz, is 107.1dB, so any source device should be more than ample to drive these.
Bottom line: If you’re technically minded when it comes to headphones, there are some red flags here—a treble-heavy, bass-shy frequency response and a lot of distortion.
. . . Brent Butterworth
brentb@soundstagenetwork.com
Reviewed on: SoundStage! Solo, February 2022
I measured the Sendy Audio Apollo headphones using laboratory-grade equipment: a GRAS Model 43AG ear/cheek simulator/RA0402 ear simulator with KB5000/KB5001 simulated pinnae, and an Audiomatica Clio 12 QC audio analyzer. For isolation measurements, I used a laptop computer running TrueRTA software with an M-Audio MobilePre USB audio interface. For most measurements, the headphones were amplified using a Musical Fidelity V-CAN amplifier; I used a Schiit Magnius amplifier for distortion measurements. These are “flat” measurements; no diffuse-field or free-field compensation curve was employed. If you’d like to learn more about what our measurements mean, click here.
This chart shows the Apollos’ frequency response. It’s extremely unusual; I don’t think I’ve seen a curve like this before with a set of headphones. There’s not much deep bass, the usual peak at 3kHz (which is generally thought to give headphones a sound more like that of speakers in a room) is entirely absent, and there’s a very strong, high-Q peak centered at 5.7kHz.
Here we can see the difference in the headphones’ response when a high-impedance (75 ohms) source is substituted for a typical low-impedance source (5 ohms). As is almost always the case with planar-magnetic drivers, there’s effectively no difference (maybe 0.5dB more bass with the high-impedance source) because the load is almost purely resistive.
This is a new measurement I’m adding, to show how sensitive the headphones’ response is to positioning on the ears. The black line shows the curve I used above, which I found to be most typical; the other curves are taken with the headphones moved about 5mm up, down, and to each side. Because of the wide, soft earpads, the seal of the headphones is consistent, so bass response really doesn’t vary at all. The variance in the higher frequencies is normal—high-Q resonances that shift in frequency with position, and aren’t very audible.
This chart shows the Apollos’ right-channel response compared with a few other open-back models (including the Dan Clark Audio Æon 2 headphones with perfed pads, which are the planar-magnetic headphones I’ve found come closest to the Harman curve). It’s obvious that the Apollos are way off the norm.
The “hash” above 1kHz in the Apollos’ right-channel spectral-decay plot is common for open-back planar-magnetic headphones, although its bandwidth is broader—which suggests to me these might have an unusually open and ambient sound. However, there’s a very strong resonance between 5 and 6kHz that corresponds precisely with the big peak in the frequency-response measurements, and I’d guess that’s audible, depending on the high-frequency content of whatever you’re listening to.
Here’s the THD vs. frequency chart, measured at 90dBA and 100dBA (both levels set with pink noise). The distortion here is very low and totally inaudible even at high levels.
In this chart, the external noise level is 85dB SPL (the red trace), and numbers below that indicate the degree of attenuation of outside sounds. The lower the lines, the better the isolation. You can see that the Apollos’ isolation is typical for large, open-back headphones. I included the Dan Clark Audio Æon RT Closed headphones so you can see how a closed-back model compares.
The rated impedance of the Apollo headphones wasn’t published at the time I did these measurements, but it’s basically dead-flat at 16 ohms with very little phase shift.
Sensitivity of the Apollos, calculated for 16 ohms impedance and averaged from 300Hz to 3kHz, is 95.8dB—high enough that you can get a useable volume from most sources, but low enough that you’ll want to use a DAC-amplifier or a portable music player to get the best performance from them.
Bottom line: The frequency-response plot shows that the Sendy Apollo headphones are going to have a distinctive and unusual sound. Otherwise, these measurements are fairly typical for open-back planar-magnetic headphones.
. . . Brent Butterworth
brentb@soundstagenetwork.com
The following categories containing listings of all product reviews published by the SoundStage! Network since 1995 from all of our online publications. The products are divided into categories and listed in descending order by date. There is no Search function within the listings, but you can search by bringing up the page with the appropriate list and using the "Find" command on your browser. (For Internet Explorer select: Edit > Find on this Page.)