Oppo Digital PM-1 Headphone Measurements

1
Oppo Digital PM-1 Frequency Response

FR.jpg
Brent Butterworth

I measured the performance of the Oppo Digital PM-1 the way I measure other over-ear headphones, using a G.R.A.S. 43AG ear/cheek simulator, a Clio FW audio analyzer, a laptop computer running TrueRTA software with an M-Audio MobilePre USB audio interface, and a Musical Fidelity V-Can headphone amplifier. I calibrated the measurements for ear reference point (ERP), roughly the point in space where your palm intersects with the axis of your ear canal when you press your hand against your ear. No compensation for EQ -- i.e., diffuse-field EQ -- was employed. All measurements were made with the supplied perforated leather earpads installed.

The chart above shows the frequency response of the PM-1 in the left (blue) and right (red) channels, with the test level referenced to 94 dB @ 500 Hz. There is no standard for what constitutes a "good" frequency response in headphones, but this measurement suggests a neutral sound. Most headphones have a response peak at 3 kHz or so (which is thought to make a headphone's sound more like that of speakers in a real room), and this one does, but its 3 kHz peak is mild at about +6 dB (a lot of them are more like +12 dB). There's another mild, and very narrow, peak centered at 8.8 kHz.

Sensitivity of the PM-1, measured between 300 Hz and 3 kHz with a 1 mW signal calculated for the rated 32 ohms impedance is 101.6 dB, which is pretty high for a planar magnetic headphone.

If you have any comments or questions regarding these measurements, please post them on the original blog referring to this article.

2
Oppo Digital PM-1 vs. Audeze LCD-X vs. HiFiMan HE-6

comparo.jpg
Brent Butterworth

This chart shows the right-channel frequency response of three high-end planar magnetic headphones: the Oppo Digital PM-1 (blue trace), the Audeze LCD-X (red trace) and the HiFiMan HE-6 (green trace). All three measure practically dead flat between 50 Hz and 1.5 kHz. Above that, the PM-1 basically splits the difference between the LCD-X and the HE-6, which suggests it may be the most neutral-sounding headphone in this bunch.

3
Oppo Digital PM-1 Frequency Resposne, 5 vs. 75 Ohms Sources

5v75.jpg
Brent Butterworth

This shows the frequency response of the PM-1 in the right channel when fed directly by the Musical Fidelity V-Can amp's 5-ohm output impedance (red trace), and with 70 ohms resistance added to create 75 ohms total output impedance (green trace). A perfect result here would be two lines that completely overlap, which would indicate that the PM-1's tonal balance doesn't change when you change source devices. And as you can see here, the PM-1's result on this test is pretty close to perfect.

4
Oppo Digital PM-1 Spectral Decay

waterfall.jpg
Brent Butterworth

Spectral decay (waterfall) plot of the PM-1, right channel. Long blue/green streaks indicate resonances, which are generally undesirable. This headphone shows no noteworthy resonances. (Yes, you do see longer decay in the bass, but that's normal.) Note that the original chart I posted showed a long decay across the entire audio band; on the original measurement I think I forgot to put damping material over the PM-1's open back, which I usually do with open-back headphones so that their sound doesn't reverberate in my lab.

5
Oppo Digital PM-1 Distortion vs. Frequency at 100 dBA

THD-2-copy.jpg
Brent Butterworth

Total harmonic distortion (THD) of the PM-1, left channel, measured at a test level of 100 dBA (orange trace) and 90 dBA (green trace). What you want to see here is a line that runs very low on the chart. The PM-1 doesn't have the near-zero distortion seen in the Audeze headphones, but the PM-1 reveals distortion only in a fairly narrow band between 220 and 300 Hz, rising to a maximum of 6 percent at 100 dBA and 2 percent at 90 dBA.

I've seen some commentary and speculation about this measurement on Internet forums, and I want to stress a few things that are necessary to understanding this measurement -- which, like most acoustical measurements, is easy to misinterpret.

First, 100 dBA is a very loud listening level. I choose that as my testing level not because it's a realistic listening level, but because it's a level that some headphones can reproduce without distortion and some can't. I used to measure every headphone at lower levels but I found that at normal listening levels, distortion almost never exists to any significant degree.

Second, when I was able to measure numerous headphones and compare the results of the measurement to the subjective impressions of the listening panelists I used, I learned how much and what kinds of distortion were more readily audible. In my measurements (174 headphones to date), I've found that listeners reported hearing distortion in only the most extreme cases, such as headphones that rise to 10 percent or higher THD in the bass.

Third, we are still in a primitive stage of understanding of distortion measurements of audio transducers. I think the industry has done reasonably well with the CEA-2010 subwoofer output/distortion measurements, but otherwise, distortion measurements of audio transducers are rarely performed. We do them with headphones because it's easy to isolate the transducers from the effects of environmental noise; with speakers, that would require an anechoic chamber. But just because we do the measurements doesn't mean we have a complete understanding of their implications.

Fourth, I know many people who do headphone measurements, and all of the ones I know are reluctant to draw specific conclusions from their measurements. (As all people practicing science should be.) Headphone measurement is still in its infancy; we're stuck with outdated and incomplete standards, so every technician is forced to follow his own judgment and best practices, and to adapt his techniques for whatever measurement equipment he owns. So if you've never done a headphone measurement in your life and you're drawing all sorts of specific, confident conclusions from a set of headphone measurements, you are overestimating your knowledge and expertise.

6
Oppo Digital PM-1 Impedance

imp.jpg
Brent Butterworth

Impedance magnitude (dark green trace) and phase (light green trace) of the PM-1, right channel. It's better if both of these lines look as flat as possible because impedance that's flat at all frequencies usually gives you a more consistent response when you change source devices. And indeed, the PM-1's about as flat as headphones get, with an impedance of 32 ohms (same as the rating) across the whole audio band, and negligible phase shift.

7
Oppo Digital PM-1 Isolation

iso.jpg
Brent Butterworth

Here's the weak spot of an open-back headphone. The chart here shows isolation of the PM-1 right channel, i.e., its ability to block external sound. Levels below 75 dB indicate attenuation of outside noise -- i.e., 65 dB on the chart means a -10 dB reduction in outside sounds at that sound frequency. The lower the line is on the chart, the better. The PM-1's isolation is actually better than average for an open-back planar magnetic headphone, although still, there's almost no isolation at frequencies below 3 kHz.