This is a review, listening test, detailed measurements and equalization of ZMF Atrium Closed headphone. It is on kind loan from a member and costs US $2,500.
The headphone feels and looks high-end with the unique wood cups and leather work. It was comfortable to wear but is extremely difficult to adjust the headband height. It takes a lot of force to adjust it at which point, it jumps a number of detents.
I was surprised that despite the large cups, it was rather difficult to get consistent measurements. Difficulty of adjusting the headband didn't help in this regard. Looking online, I found differing measurements. Mine below correlates best with the company measurements published in a video.
FYI I sent the frequency response a couple of days ago to the designer for input but did not hear back.
ZMF Atrium Closed Headphone Measurements
As usual, we start with our frequency response measurement and comparison to target using GRAS 45CA fixture:
I had to spend nearly half hour to get the right channel to be close to the left channel and at maximum bass output. I like that there is ample bass but very disappointed in that huge trough centered around 1.8 kHz. This number will keep coming up as you will see below. Such a suck-out will take the life out of vocals and spatial qualities from my experience.
Relative target shows that exact compensating EQ by eye will be challenging:
But we will try in the next section.
Distortion measurement shows sharply increasing levels at you guessed it, 1.8 kHz:
I marked it as a resonance but looking at it now, it is wide so my have other causes. Regardless, such high distortion where our hearing starts to get very sensitive is not a welcome addition by the headphone. Distortion measurements may not be predictive of preference but they are a wonderful tool in the way they show defects in design.
Group delay is more messy than I am used to seeing in bass frequency and also shows a dip where our 1.8 kHz disturbance lives:
Headphone is high so you want a headphone amp with high voltage drive:
And yes, we see a disturbance in the impedance graph at 1.8 kHz which means it is a major acoustic event.
Sensitively is below average so you want a good headphone amp to drive it if you want to play loud:
ZMF Atrium Closed Listening Tests and Equalization
Immediate impression is that of bass which is not so bad. But eq that trough and you realize what you are missing in upper midrange/lower treble:
Band 3 is doing all the heavy lifting. The rest are complimentary but may depend on your taste. The triple filters in bass tightened its response which I prefer. Some may want the warmth without.
Tailoring the 1.8 kHz filter to the frequency response with the other two was high and miss. But I tried.
Conclusions
It is very obvious that the company has its own ideas of what tonality people like. And that is something that has little to do with the research over a decade that says otherwise. Shame is, company has extensive measurement tools. What it is used for, I don't know.
Further, it seems sometimes that I am the only one caring about proper engineering through my distortion tests. The same measurement that produces the frequency response also outputs THD (both absolute and relative). So the designer is a click away from discovering anomalies that need to be investigated and remedied.
And it is not just objective measurements: listening tests clearly show massive response errors if you perform AB with and without correction. To be sure, as is response is not annoying. It just muffles the heck out of good part of the music spectrum.
I know I am being a bit harsh with my rating here but at some point, we need to stomp our feet and get companies to either follow the research or prove the research is wrong. Creating random response in such an expensive headphone is just wrong.
Needless to say, I can't recommend the ZMF Atrium Closed. As a woodworker, it aches my heart to have written that but I have no choice.
------------
As always, questions, comments, recommendations, etc. are welcome.
Any donations are much appreciated using: https://www.audiosciencereview.com/forum/index.php?threads/how-to-support-audio-science-review.8150/
The headphone feels and looks high-end with the unique wood cups and leather work. It was comfortable to wear but is extremely difficult to adjust the headband height. It takes a lot of force to adjust it at which point, it jumps a number of detents.
I was surprised that despite the large cups, it was rather difficult to get consistent measurements. Difficulty of adjusting the headband didn't help in this regard. Looking online, I found differing measurements. Mine below correlates best with the company measurements published in a video.
FYI I sent the frequency response a couple of days ago to the designer for input but did not hear back.
ZMF Atrium Closed Headphone Measurements
As usual, we start with our frequency response measurement and comparison to target using GRAS 45CA fixture:
I had to spend nearly half hour to get the right channel to be close to the left channel and at maximum bass output. I like that there is ample bass but very disappointed in that huge trough centered around 1.8 kHz. This number will keep coming up as you will see below. Such a suck-out will take the life out of vocals and spatial qualities from my experience.
Relative target shows that exact compensating EQ by eye will be challenging:
But we will try in the next section.
Distortion measurement shows sharply increasing levels at you guessed it, 1.8 kHz:
I marked it as a resonance but looking at it now, it is wide so my have other causes. Regardless, such high distortion where our hearing starts to get very sensitive is not a welcome addition by the headphone. Distortion measurements may not be predictive of preference but they are a wonderful tool in the way they show defects in design.
Group delay is more messy than I am used to seeing in bass frequency and also shows a dip where our 1.8 kHz disturbance lives:
Headphone is high so you want a headphone amp with high voltage drive:
And yes, we see a disturbance in the impedance graph at 1.8 kHz which means it is a major acoustic event.
Sensitively is below average so you want a good headphone amp to drive it if you want to play loud:
ZMF Atrium Closed Listening Tests and Equalization
Immediate impression is that of bass which is not so bad. But eq that trough and you realize what you are missing in upper midrange/lower treble:
Band 3 is doing all the heavy lifting. The rest are complimentary but may depend on your taste. The triple filters in bass tightened its response which I prefer. Some may want the warmth without.
Tailoring the 1.8 kHz filter to the frequency response with the other two was high and miss. But I tried.
Conclusions
It is very obvious that the company has its own ideas of what tonality people like. And that is something that has little to do with the research over a decade that says otherwise. Shame is, company has extensive measurement tools. What it is used for, I don't know.
Further, it seems sometimes that I am the only one caring about proper engineering through my distortion tests. The same measurement that produces the frequency response also outputs THD (both absolute and relative). So the designer is a click away from discovering anomalies that need to be investigated and remedied.
And it is not just objective measurements: listening tests clearly show massive response errors if you perform AB with and without correction. To be sure, as is response is not annoying. It just muffles the heck out of good part of the music spectrum.
I know I am being a bit harsh with my rating here but at some point, we need to stomp our feet and get companies to either follow the research or prove the research is wrong. Creating random response in such an expensive headphone is just wrong.
Needless to say, I can't recommend the ZMF Atrium Closed. As a woodworker, it aches my heart to have written that but I have no choice.
------------
As always, questions, comments, recommendations, etc. are welcome.
Any donations are much appreciated using: https://www.audiosciencereview.com/forum/index.php?threads/how-to-support-audio-science-review.8150/
