ZMF Atrium Closed Headphone Review

[CedarX]

It reminded me that, if something costs nothing, or costs way beyond what it should… the consumer is the product.

 

[Curupira]

I think customers are not very well aware of how these headphones measure.

I think most of their fanbase simply don't care, viewing tonality as an artistic interpretation of music, where the manufacturer is the interpreter and each headphone is different.

 

[RogerSmith]

I think most of their fanbase simply don't care, viewing tonality as an artistic interpretation of music, where the manufacturer is the interpreter and each headphone is different.

I guess that is really what many audiophools think, sometimes called "tastefull coloration", often accompanied by tube sound and R2R DACs (distortion). GoldenSound is a promiment proponent of this 'ideas', although he should and in fact does know better. I think everybody who is clear in his mind realizes that this completely contradicts the concept of high fidelity, but arguments, education or a discussion are mostly in vain on adherents to these views.

 

[imagidominc]

I think customers are not very well aware of how these headphones measure. So little independent reporting. And even those (reviewers) that measure, are motivated to say everything expensive is good. So purchases are made and companies think they are doing the right thing.

As someone who ~nearly~ bought a ZMF headphone, I can say that the aesthetics play a huge part in the overall quality judgement.

 

[amirm]

As someone who ~nearly~ bought a ZMF headphone, I can say that the aesthetics play a huge part in the overall quality judgement.

It does for me as well. I really like to have such products do well.....

 

[Maiky76]

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.

View attachment 459836
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:
View attachment 459837
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:
View attachment 459838
But we will try in the next section.

Distortion measurement shows sharply increasing levels at you guessed it, 1.8 kHz:
View attachment 459839
View attachment 459840
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:
View attachment 459841

Headphone is high so you want a headphone amp with high voltage drive:
View attachment 459842
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:
View attachment 459843

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:
View attachment 459844
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/

Here are some thoughts about the EQ.
Please report your findings, positive or negative!

Notes about the EQ design:

• The average L/R is used to calculate the score.
• The resolution is 12 points per octave interpolated from the raw data (provided by @amirm)
• A Genetic Algorithm is used to optimize the EQ.
• The EQ Score is designed to MAXIMIZE the Score WHILE fitting the Harman target curve (and other constraints) with a fixed complexity.
  This will avoid weird results if one only optimizes for the Score, start your journey here or there.
  There is a presentation by S. Olive here.
  It will probably flatten the Error regression doing so, the tonal balance should be therefore more neutral.
• The EQs are starting point and may require tuning (certainly at LF and maybe at HF).
• The range around and above 10kHz is usually not EQed unless smooth enough to do so.
• I am using PEQ (PK) as from my experience the definition is more consistent across different DSP/platform implementations than shelves.
• With some HP/amp combo, the boosts and preamp gain (loss of Dynamic range) need to be carefully considered to avoid issues with, amongst other things, too low a Max SPL or damaging your device. You have beed warned.
• Not all units of the same product are made equal. The EQ is based on the measurements of a single unit. YMMV with regard to the very unit you are trying this EQ on.
• I sometimes use variations of the Harman curve for some reasons. See rational here and here
• NOTE: the score then calculated is not comparable to the scores derived from the default Harman target curve if not otherwise noted.

average L/R match.

I have generated one EQ, the APO config file is attached.
For once same number of BQ for both my EQ and Amirm's.

Score no EQ: 38.4
Score Amirm: 54.0
Score with EQ: 65.4

ZMF Atrium Closed APO EQ Score Flat@HF 96000Hz
July032025-151120

Preamp: -9.70 dB

Filter 1: ON PK Fc 196.5 Hz Gain 6.91 dB Q 1.90
Filter 2: ON PK Fc 210.5 Hz Gain -5.78 dB Q 0.73
Filter 3: ON PK Fc 645.3 Hz Gain 2.24 dB Q 2.88
Filter 4: ON PK Fc 1856.2 Hz Gain 9.74 dB Q 1.65
Filter 5: ON PK Fc 3210.6 Hz Gain 3.16 dB Q 4.97
Filter 6: ON PK Fc 5223.7 Hz Gain -5.18 dB Q 5.00

 

[Egoist]

Was always curious about the Atrium Open (never liked closed headphones). Could it be something similar to this one?

 

[RogerSmith]

Here are some thoughts about the EQ.
Please report your findings, positive or negative!

Notes about the EQ design:

• The average L/R is used to calculate the score.
• The resolution is 12 points per octave interpolated from the raw data (provided by @amirm)
• A Genetic Algorithm is used to optimize the EQ.
• The EQ Score is designed to MAXIMIZE the Score WHILE fitting the Harman target curve (and other constraints) with a fixed complexity.
  This will avoid weird results if one only optimizes for the Score, start your journey here or there.
  There is a presentation by S. Olive here.
  It will probably flatten the Error regression doing so, the tonal balance should be therefore more neutral.
• The EQs are starting point and may require tuning (certainly at LF and maybe at HF).
• The range around and above 10kHz is usually not EQed unless smooth enough to do so.
• I am using PEQ (PK) as from my experience the definition is more consistent across different DSP/platform implementations than shelves.
• With some HP/amp combo, the boosts and preamp gain (loss of Dynamic range) need to be carefully considered to avoid issues with, amongst other things, too low a Max SPL or damaging your device. You have beed warned.
• Not all units of the same product are made equal. The EQ is based on the measurements of a single unit. YMMV with regard to the very unit you are trying this EQ on.
• I sometimes use variations of the Harman curve for some reasons. See rational here and here
• NOTE: the score then calculated is not comparable to the scores derived from the default Harman target curve if not otherwise noted.

average L/R match.

I have generated one EQ, the APO config file is attached.
For once same number of BQ for both my EQ and Amirm's.

Score no EQ: 38.4
Score Amirm: 54.0
Score with EQ: 65.4

ZMF Atrium Closed APO EQ Score Flat@HF 96000Hz
July032025-151120

Preamp: -9.70 dB

Filter 1: ON PK Fc 196.5 Hz Gain 6.91 dB Q 1.90
Filter 2: ON PK Fc 210.5 Hz Gain -5.78 dB Q 0.73
Filter 3: ON PK Fc 645.3 Hz Gain 2.24 dB Q 2.88
Filter 4: ON PK Fc 1856.2 Hz Gain 9.74 dB Q 1.65
Filter 5: ON PK Fc 3210.6 Hz Gain 3.16 dB Q 4.97
Filter 6: ON PK Fc 5223.7 Hz Gain -5.18 dB Q 5.00

View attachment 461004

Wow, I don't remember any headphone that is even after EQ so distant from the target, so it is actually beyond EQ salvation, oh boy...

 

[Wavey22]

I loved this headphone for my hip/hop music taste. This was without looking at measurements.

I think measurements are important to see what works for you especially if you cannot hear it directly. But I honestly would just listen first before passing final judgment.

I liked this better than the DCA stealth by a large margin.

I loved the DCA norie better than the Atticus or the eikon.

I like the caldera open better than any planar I've heard for my metal needs aside from Grado headphones, which probably goes down as number 2 favorite metal headphones of all time, LCD 2 probably number 3.

There are people who like it and people who don't. Nobody is right here imo. ‍

 

[Somafunk]

I like the caldera open better than any planar I've heard for my metal needs aside from Grado headphones, which probably goes down as number 2 favorite metal headphones of all time.

There’s no scenario where Grado are good, and to listen to shouty, peaky heavy metal on such headphones must be tortuous unless you suffer from partial hearing loss

 

[RogerSmith]

I loved this headphone for my hip/hop music taste. This was without looking at measurements.

I think measurements are important to see what works for you and maybe as a guideline. But I honestly would if possibly listen first before passing final judgment.

I liked this better than the DCA stealth by a large margin.

I loved the DCA norie better than the Atticus or the eikon.

I like the caldera open better than any planar I've heard for my metal needs aside from Grado headphones, which probably goes down as number 2 favorite metal headphones of all time.

There are people who like it and people who don't. Nobody is right here imo. ‍

Maybe one day make a test if you would pass as a skilled or trained listener. My guess, with all the due respect, is no.

 

[Wavey22]

Maybe one day make a test if you would pass as a skilled or trained listener. My guess, with all the due respect, is no.

Does it matter if I'm a "skilled listener"? If I put it on my head and it sounds awesome? What more skill do I need?

 

[Wavey22]

There’s no scenario where Grado are good, and to listen to shouty, peaky heavy metal on such headphones must be tortuous unless you suffer from partial hearing loss

Nope. I actually started with the sr-80 and was blown away. I have sense moved on. But might return with the rs1x for my Americana needs.

What do you listen to for metal?

 

[unpluggged]

If I put it on my head and it sounds awesome?

At least listen to other, better headphones. I described my impressions on the previous page, for instance. And I don't think they sound "awesome" or just good.

 

[Wavey22]

At least listen to other, better headphones. I described my impressions on the previous page, for instance. And I don't think they sound "awesome" or just good.

I just read your impressions. I think it's totally valid. You did not like the same thing I liked and there is nothing wrong with that . I've tried many headphones over the years. For instance the susvara really is a headphone I was not impressed by.

Which other "better" headphones do you suggest?

 

[solderdude]

Let's remember that tastes differ and not everyone prefers the Harman target (which is abundantly clear from the Harman research).
Let's also keep in mind that the way certain genres are recorded certain colored headphones may even sound 'more pleasant'.

When it comes to measured performance on specific test fixtures and adhering to a specific target there is little room for discussion (science).
When it comes to taste/preference and (subjective) found differences and using EQ or not (and the correct EQ) there is certainly room for divergence in opinions.

 

[Wavey22]

Let's remember that tastes differ and not everyone prefers the Harman target.
Let's also keep in mind that the way certain genres are recorded certain colored headphones may even sound 'more pleasant'.

When it comes to measured performance on specific test fixtures and adhering to a specific target there is little room for discussion.
When it comes to taste/preference and subjective found differences and using EQ or not (and the correct EQ) there is certainly room for divergence in opinions.

Thank you and I believe this is a balanced response. I found the thread interesting and decided to just state my experience with the headphone. Lol. Not sure why these people are so upset and telling me I'm wrong. I did not make any outlandish claims or go on with flowery language about buttery midrange for example.

 

[solderdude]

Not sure why these people are so upset and telling me I'm wrong

here's why:

When it comes to measured performance on specific test fixtures and adhering to a specific target there is little room for discussion (science)

And the measurements is what respondents base their opinion on. Most people never get to actually hear/feel/handle ZMF headphones.
Let's face it... most of Zach's creations do not measure well and won't adhere to 'Harman research science'.
Don't get me wrong... @zach915m can certainly measure headphones but seems to want to create things with a certain 'sound' rather than push out models with the same sound signature so people get a choice. He has to make a living from it.
Owners usually could not care less about measurements. They just order one or hear it somewhere, like it or don't and move on to something else.
They usually don't EQ these headphones to some target/measurement, they just use it as it is.
This is not the 'ASR way' as it were and is why certain (mostly quite expensive) creations (not just Zach's) get a lot of flak.

 

[Wavey22]

I get it. It's really the only way to get some form of idea of what works. But in my mind there's room for other stuff.

I've seen this play out with zu audio, Grado, zmf and others. I'm all for people preferring the harmen curve and EQ'ing. I'm not much into EQ because there's no base measurements I would want to follow. Maybe I can play around with suggestions though.

 

[solderdude]

There is always a cult-following for every brand and item in all things produced on the planet.
The ASR 'cult' is mostly about measurements (electronics, speakers and headphone) and a bit about psycho acoustics, humor and ...... UFO's
But there is also room for listening to music which is preferred over 'plot-gazing'.

I think everyone has a right to their own opinion but a lot of people confuse their opinions with facts (measured science in this case) and they usually get the full load even when it was not merely to troll (which happens a lot with new members coming on hear to voice their dismay with ASR).

Stick around ....