Dear community, headphone enthusiasts, audiophiles and mixing professionals,
I am searching for a good pair of open-back headphones to use for:
- mainly studio mixing (in combo with my Neumann KH150 for checking bass/details), this is the number 1 priority
- mixing/producing binaural content
- also sometimes mixing, recording and producing when I am on the road (yes, also rarely tracking)
- listening (music, films, …) when i can’t use loudspeakers
Well - it’s a journey and I just began mine, having had cheap headphones until now (DT990, sometimes I used DT1990 Pro, wasn’t really a huge fan of those due to treble/highs) and having not a lot of experience with dedicated headphone mixing.
This post is quite detailed - ehmm that's a euphemism - this post is long, I know, I’m sorry, but I can’t transport the info and questions any other way as I am both still learning and many factors are interlinked. I'm aware not a lot of people will read this through and some questions have been discussed somewhere probably. But I read a lot, and things still don't come together. I'm happy for every answer, every correction, every hint on good sources/reads, and also every comment on maybe a lack of experience or a wrong approach in some way. I’m glad about recommendations on what to test/buy, but even more so if somebody can clarify some of my irritations/misunderstandings. And I see this post as a mix of personal experience, in some rare case potentially a bit of a guide for buyers' decisions, and primarily as a headphone amateur's pool of questions.
So I decided it should be a pair of open-back headphones. Even if I will occasionally do some recording/tracking, I could for some quieter sources still use open-backs and would have the advantage of a more natural and linear sound, and for most cases I still have a cheap pair of DT990 and some in-ear for tracking, too.
Important: I am also questioning whether I am having some problems with my ears considering stereo imaging. I am writing about it a bit below, it could change everything about testing and judging of course, but it’s almost impossible for me to really know if that is true, as measurements in the labs of doctors tell me my hearing should be okay. It’s a thought in the back of my head I’m getting to later again.
I tested the following ones, listening to each one for only some 30mins when I didn’t like them and hours to days when I liked them. I listened to all of them un-EQ’ed. Talking about these cans from now on I’m referring to them un-EQ’ed if not stated otherwise.
I ranked them to my personal liking like this:
1. Audeze LCD-X 2021 (1250 €)
> sounded the most pleasing with the highest “wow” factor while still having loads of detail
> treble/air is there and nice but it’s not an overly bright headphone (I do like sound on the “warmer” side)
> best “soundstage” and impressions of the rooms/reverbs on most tracks (not all!)
> very natural bass but a little muddy and imprecise at times (not as transient as other headphones, "longer decays")
> natural+precise, but they also seem to add a bit of nice stage/depth/room when other don’t (good or bad!?)
> heavier and bulkier than the others, and with the bulky cable adaptors not perfectly suited for traveling
> very comfortable although a little bit heavy (they move when looking down)
2. Neumann NDH 30 Black (550 €)
> in lack of better words: very dry sounding headphones. fast transients, seem very precise
> bass sometimes very dry and fast and revealing, but sometimes lacking upper bass/low mids a little bit
> smooth treble (wouldn’t say dark but warm, still precise treble). I like the tuning (I also mix on KH150)
> not as pleasing and “wow” as the LCD-X, but then again this can also be a good thing (mixing)
> on good organ church recordings or percussion in real rooms or binaural: best soundstage/natural
> difference in sound heavily dependent on positioning (front/back axis, angle of head band)
> on first use: very uncomfortable. It seems my headshape is not perfect, it really hurt on top of my head
> comfort: I stretched them a bit using a headphone stand. Now more comfy, but less clamping force
> less clamping force could now possibly lead to less sealing and therefore sometimes bass issues?
> not so heavy, they look sturdy, very beautiful and simple design, foldable, very handy, portable
> yes, cable is on the right side, and it makes some sounds, but it's manageable
3. HifiMan Sundara Black (earlier edition, supposedly the same as the new, the one I got for 249,- €)
> good bass (not amazing but very good), good treble (but a bit too harsh/strong for me)
> very precise but transients/punch not as dry as NDH30 and sound not as natural as NDH30/LCD-X
> sound just very slightly more phasey or just little tad more weird than NDH30/LCD-X
> looks like a very nice build quality (although people say sometimes the earpads loosen), nice design
> very comfortable! Most comfortable together with the HD 490 Pro, at least for me
4. Sennheiser HD 490 Pro (430 €)
> very nice sounding actually, I liked the treble, soundstage (if I even know what it is) was also okay
> the bass just rarely seemed a tiny bit distorted or it felt as if the energy wasn’t enough to reproduce
> amazing comfort! They’re very light, I could almost forget they are on my head! improves immersion!
> would have loved to keep them but I couldn’t shake the unprecise feeling of a hint of “weakness”
5. Ollo Audio X1 1.0 2024 (540 €)
> very good headphones, I I got used to them I could really work with them, but I just don’t like them
> if I had to describe their negatives for me: feeling of slightly (!) resonant, slightly weak?
> I used their custom sample-specific EQ-curve in Realphones, but it didn't really help
> very comfy, nice simple design
6. Beyerdynamic DT1990 Pro 250 Ω (400 €)
> the treble doesn’t work for me. too bright! Don’t like them. Still very good headphones
7. Sony MDR-MV1 (395 €)
> didn’t like the hyped bass/treble sound and the build quality, they were the first to sort out
I quickly sorted out headphones 5-7 and listened more to 1-4, especially to 1 and 2. What I found was:
LCD-X and NDH 30:
With some tracks I can hear details with the LCD-X that I can’t hear as much with the NDH 30, on other tracks it’s completely the opposite and I hear details with the NDH 30 that the LCD-X won’t show me. I’m talking about treble details, hidden percussive elements, room reverbs, depth, movements, …
Is a boosted treble/air band something where you automatically get more articulation from and then can also have more detail, or is detail and precision more about the transients, or both? I mean, probably transients are the most important factor here, you can’t EQ transients but you can EQ after the transients do what they do. But transients (and distortion) should be great with the NDH30 and not a problem, I guess. Bass sounds super dry.
It’s hard to say which one wins to my liking. On some sources I think NDH 30 does the job better, but then I listen again to LCD-X and now they suddenly both do the job, or I hear a room reverb on the NDH 30 that I didn’t hear with the LCD-X (finally! A winner! And it’s the cheaper headphones!), but when listening again on the LCD-X I find that yes, the reverberation may be not as good, but still there, but then the bass sounds more three-dimensional or there’s another detail I was missing on the NDH 30.
Or I listen to a track on NDH 30 and have the feeling that the instruments (especially amped clean electric guitar, but in conjunction with bass and drums) are exactly (!) how they’re supposed to sound, like they are so right and pleasing that I get the feeling I want to eat them (if that’s understandable). Then, listening to LCD-X, I am missing this impression, this feeling of compactness, but then the voice sounds more natural (sometimes even much more natural and full) and correct on the LCD-X - which would be due to more present lower mids, while on paper (on frequency responses found online) at least it shouldn't?
And what is that thing about the LCD-X, that it sounds so full-bodied and so roomy? Some online review said that compared to NDH 30 the LCD-X, especially on dark sounds, sounded "more tangible and clearer... and with an impressive breadth, depth and clarity". This sounds also very pleasing, but when assuming the NDH 30 is as linear as claimed I get a bit sceptical if it's actually showing me the original content, or if it is adding to it. Or, a wild and probably nonsense theory: if the build quality, comfort and weight of the headphone that I can feel on my head add to the feeling that the sound had weight, strength, was full-bodied.
So… I can’t decide. Judging speakers seemed an easier task. And I can’t really seem to understand the parameters that would make my judgement jump back and forth from one to the other and my attention jumping between different aspects that I wanted to judge isolated.
For example, NDH 30 are indeed warmer or darker than all the other headphones. I generally prefer a warmer sound to a very trebley sound and hate harshness (of course). At a specific position on my head, in my case with the front part of my ears (the tragus) close to the cushion of the ear pads, their treble sounds best. Other people report that here, on this specific position the treble sounds too harsh even, while most say it’s generally too dark in all positions (or used it with a wrong positioning?).Okay, so the treble sounds nice to me at specific position. And Neumann aims at a rather warm/smooth sound that emulates speakers in a room. Makes sense. But compared to LCD-X I am missing some details sometimes, which probably is because of the rather tamed and soft treble.
And if I want to use headphone equalisation software (I can EQ them to sound closer to LCD-X or other headphones, which could be a good thing for some tasks) and these software (like Sonarworks) also use curves that simulate a bit of that, I would basically have double/added darkness, right?
Also, this position-dependency has me really worried. It’s not that I don’t find a position where I like the sound or that I can’t correct the position once I find that there’s lacking some treble. It’s more about very slight differences and mainly about these two things:
1. Hearing is so subjective and also so dependent on so many factors, that I might get it wrong every other day when they are not positioned exactly the same as when I mixed a track the day before, and I could make mistakes while mixing or judging a specific part of a mix.
2. The bass will be dependent on how the cushions seal. Sometimes now, after having bent the headphones a little bit having them sit on a headphone stand for 1-2 days, they feel less clamping and more comfortable, but also I imagine (mustn’t be true though) that the bass is lacking a bit now, especially as opposed to the LCD-X.
But looking at the manufacturer's frequency responses of both, this couldn't really be true: the LCD-X on paper should have less bass, and while it’s hard to judge stuff going on above 2-3 kHz from the frequency response, it should be okay to judge it in the bass by looking at the graphs, no? The data says that the NDH 30 had more bass. What I hear is the opposite. How can that be? Have I bent the bass away from my NDH 30? Are they never really positioned correctly on my head? When do I actually know it’s positioned correctly? When do I know the bass is how it’s supposed to be? Do I have weird ears (my right ear touches the headphone, my left ear doesn’t)? The NDH30 hurt at the beginning. Now less, since I had them on the headphone stand, now they also have less clamping force. Now, I sometimes imagine (as I am becoming more paranoid for problems since testing headphones haha) the bass got less. Less clamping force = less bass?
Well - I asked Audeze for a frequency response curve for my sample of LCD-X (2021) and this actually explains a bit! It doesn't explain why it seems I lost some of the bass with the NDH 30 (although it's hard to say because I don't have measurement equipment at home and I can only say so from memory).
But what it shows is that there's quite a difference in oratory1990's measurement of LCD-X (2021) and what the Quality Control department at Audeze sent me.
This is Audeze QC:
This is oratory1990 (measured with GRAS 45BC) from autoeq.app:
The mean level between 1.5k and 10k on the Audeze graph is down about 6dB compared to the mean level between 30 Hz and 1.5kHz! This is a lot of bass-emphasis, no?
On oratory1990's measurement, however, it's quite the opposite, I would say the mean level below 1.5kHz is even lower than the mean level above 1.5kHz.
This would mean, assuming the measurements of both parties are approaching ideal measurement, my sample of LCD-X had more than 6 dB (maybe 7 or 8 even) stronger bass and lower mids than the one oratory1990 had? Or does it mean that instead the measurements are just very different? Because if this is the case: Doesn't this mean we can't really rely on measurements as found on autoeq.app?
When looking at Innerfidelity's (measured with HMS II.3) measurements they also look completely different. Very broad wide peak of +6dB from 2 kHz to 5 kHz (serial number 7456406)...?!?
Okay, putting found and received frequency response curves to the side, and back to choice making: So the LCD-X are a bit above my budget right now, but more importantly, they are a bit too large and/or heavy for traveling and not as versatile for a mobile setup, especially if in some cases I don’t use pro audio gear with solid headphone amplifiers but for example my MacBooks audio output (I have to admit: I guess my RME Fireface UCX headphone preamp sounds better than my MacBook Pro's audio output, but differences are quite small and I couldn't vouch for it!) or even use them with my Sound Devices MixPre device (nice headphone amplifier, but 3.5mm) and then have to put the two adaptors that are quite heavy, it’s really not really an option for me, looking for something more versatile and mobile. Also, what is speaking against LCD-X is the slight feeling that they make the sound more beautiful (more warm muddy roomy bass) and less neutral/true.
Still a tough decision to make. If I ruled out LCD-X for the reasons above, there is still HifiMan Sundara. Hmm, they didn’t really work for me. If I wanted to have planar drivers, I would probably go back to the LCD-X which were in my opinion just capable of more. Although the HifiMan Sundara is crazy good value for the money!
HD 490 Pro - also an option. Especially for comfort! How I would love to have a headphone that you can forget about while it’s still on your head. Especially for immersive stuff like binaural or video games or whatever, I have to say that the impression of a good sound is not always just what you hear but also what you think you hear, want to hear, or believe you hear. And also, how comfortable you are. And with lightweight airy and not bass-heavy music they seemed very natural, I guess forgetting I even wore them added to that. But still, they couldn’t compete.
Having a hard time finding the right headphones, I read a lot in forums. And measurement data. NDH30 was very much praised on its low THD here, and in forums mostly by a person named indigogo who was crazy about it. So crazy even that sometimes, the user would write stuff I both couldn’t believe and found very over-the-top. Quite the anti-ad, in my opinion, and while I don’t believe Neumann or someone paid the user, I don’t believe there is the perfect headphone, while I would really like to believe that the NDH30 is close to perfect, it would help make a decision and save some money (opposed to more expensive headphones). There has been some praise for the NDH 30 by 3-4 other persons in forums, but 95% of praises in different forums came from indigo. Well… probably won’t impact my decision. I will have to like it and I did prefer the NDH30 still above all the others for some reasons, but two things were still big question marks: listening pleasure and soundstage.
Listening pleasure:
Just a short question here: Are there headphones that are perfectly suited for both, mixing and listening pleasure? Or can you really say that for example if LCD-X sounds much more 3-dimensional and pleasing and "wow", it will automatically be less neutral and precise? If it sounds more 3-dimensional and pleasing in the bass and lower mids, isn’t that also coloration?
Soundstage:
To be honest - I really don’t know exactly what soundstage is. Is soundstage the depth and width of the recording? Is it how good reverberation can be heard? Does good imaging contradict good soundstage or can there be both AND good linearity? What is the science behind soundstage? Can somebody maybe elaborate on that a bit more in detail? Some users even speak of "height", and while I as a human would know what that sounded like, I couldn't find it in any headphones. Is my hearing too imprecise?
I’ve read in some forums that some mixing professionals who’ve listened to a lot of headphones actually don’t have a clear answer to "soundstage", admitting they didn’t know what the reasons for soundstage were technically and that they couldn’t predict if a headphone had good soundstage, and that they didn’t know if it was something desirable in terms of mixing, and some even claiming they can’t tell you exactly what soundstage even is - while a lot of audiophiles/enthusiasts seem to know perfectly what it is. How is that possible?
I even read some user (https://www.head-fi.org/threads/the-science-of-soundstage.959791/) explaining (without any sources on research/statistical tests/physiological/psychoacoustic explanations) where soundstage came from (they said a certain dip in a frequency range in the treble region) and that it could improve from a bass and upper treble enhancement and that they could read a frequency response graph and tell you wether some cans had soundstage. A lot of users disagreed. It seemed to me this frequency response explanation was a bit over-the-top and not necessarily true, however I stumbled upon a lot of threads where people claim headphones having a soundstage are some that have some more present bass and certain highs.
Some say that headphones with a huge soundstage and a very pleasing listening experience with “wow” effect can even be bad for mixing (good would be linearity and precise imaging). I guess so, If “pleasing” means delivering more bass or treble than a track might be mixed for on “flat” loudspeakers in a well treated room. But then again, what if soundstage is not about frequency response but about showing the results of clever mixing (when for example psychoacoustic tricks were used). What if I’m missing these details with headphones that have less soundstage? After all people listen to mixes on headphones, and as a mixer I would take soundstage into consideration. So I should hear it. Or, to dive further, I could ask the question, how mixing engineers produce soundstage. Is it just about reverb, early reflections and pre-delay (in the digital/artifical reverb domain) or very good stereo recordings in the real-world-recording-domain?
I’m asking these questions obviously to understand if soundstage is even an important aspect to consider when choosing for a pair of mixing headphones, and I am starting to suspect that cans with an amazing soundstage are not necessarily linear anymore. Or that soundstage was a tradeoff with imaging/precision on the other side, or with linearity. But is that true?
The Neumann NDH 30 don’t seem to have a very deep or wide soundstage, LCD-X often seemed to have more soundstage, but that was more of a feeling. But, like I said in the notes about NDH 30 and LCD-X above, soundstage also differed from track to track. In some good binaural recordings and some stereo church organ recordings (Toccata and fugue in D Minor BWV 565 Fugue, Kay Johansson) NDH 30 sounded more natural to me and I good hear a deeper reverberation in the room and had a feeling of being more immersed, a stronger feeling of a deep and wide soundstage. Or in Buena Vista social clubs number-one-loudspeaker-and-headphones-test-track “Chan Chan”, where I could hear a very spatial reverb in the first minute, caused by a percussive instrument that was sitting to the mid-left with beautiful echoes/reverb in the mids and highs, reverberating in the right and mid-right position that I was missing in this precision with the LCD-X. But then, for other tracks like Jazz or RnB, the LCD-X sounded more realistic/externalizing. In a lot of pop mixes, like D’Angelo’s “Send it on” for example, LCD-X’s soundstage seemed better (background voices were further away, the horns to the sides left and right seem more exterior and further away from the ears). Some people say the “further away” externalisation also depends on how close the drivers are to the ears. Is that so?
What about reverb and reverb details? Shouldn’t cans that show you all the details of room reverberation also add to the most realistic room feeling? In recordings like “Meus 26” by Tim Bernandes, for example. Here, by the way, I heard the reverb on the vocals mainly to the right - what a weird decision in mixing! Flipping my NDH 30 and wearing their right pad on my left hear to prove it's not my hearing didn't work, they just sounded weird (angled drivers etc.). Anyway, I could hear more details of this reverb on the NDH 30 than on the LCD-X, though, but also it sounded a little bit more isolated, taking a bit of the soundstage immersion, and the LCD-X sounded like having more soundstage, also because the bass and lower mids were more smooth and organically blending into the room reverb, making it more of a round thing and adding to an impression of soundstage and immersion more.
Also quite a nice track for room reverb and depth: Papajazz - “Money Speaks”. Both LCD-X and NDH 30 sound amazing on this, with LCD-X sounding more deep and round and organic and warm and effortless, while NDH 30 sounding a bit more precise and revealing the room differently. Is any of these better? It’s so hard to tell.
I read this article of Diego Estan (https://www.soundstageaccess.com/in...ic-i-use-to-evaluate-soundstaging-and-imaging), which is about speakers, not about headphones! But he listed a lot of songs that would have extraordinary good imaging and soundstage. I listened back on my Neumann KH system, and I could hear what he would mean, but I was not impressed as he was - probably because my room is not treated as well as his was (also a reason why I need open back headphones for checking for some mix anomalies due to room treatment problems).I wondered if these would translate to headphones, too. But I was really missing the deep externalisation he was talking about, both on my speaker system and on my headphones. When he talks about elements sounding like coming from behind the walls of his room - no, I can’t hear that. I remember listening to a recording of percussion in a medium reverberating room on medium quality studio speakers in an acoustically treated room (but not really with bass traps), that was a lot of years ago, and I was amazed on how 3D I could hear the percussive elements - some pianissimo to piano sounds felt like they were actually played super close to my nose (!) - that was really holographic! Unfortunately, I never had that impression again and when some years later I listened to a lot of studio monitors and found the Neumann KH 150 perfectly suited for me, did quite some room treatment (but for mainly mids and treble, no huge bass traps) and used the MA-1 room correction DSP system, I still couldn’t get anywhere as close to what I heard years back then. The only time when I heard a very precise imaging or something close to a very deep soundstage was, if I remember correctly, on coaxial speakers (Geithain and KS Audio), something I couldn’t hear on non-coaxial. Friends who listen on my system also seem to hear isolated mono sources centred better than I do. Then again, they don't always have trained ears. I am saying all this because I am kind of speculating on the fact that...
...my hearing may impaired:
and I may not be able to actually hear (anymore) what good stereo imaging is. Let me quickly summarise:
When listening to speakers in a well treated room, a mono source appears sharp and in the center only on a few rare occasions. It mostly moves a little to the left and the right jumping around a slight-left center, but most importantly: It mainly sits somewhere to the slight left. This may be bad loudspeaker correction or bad room acoustics, so listening with headphones should do different, no? Well, only sometimes. Sometimes I hear a voice in the center, but mostly just slightly to the left, regardless of the headphone orientation. I am not talking 9 o clock, I am more talking 11h45. In fact, in most of the mixes that I am listening to, the majority of content seems to be heard very slightly to the left, they don’t seem completely balanced (there is a thread somewhere where users discuss that music is actually mixed more to the left - is it music or is it my ears?). Listening to some good stereo recordings of pianos that seem to make use of stereo in the sense that lower register is panned more to the left and higher register to the right, a weird thing happens sometimes in the mids: they seem to jump around, and even if a player played a chromatic or diatonal series of notes going up in pitch in middle register, they wouldn’t move from left to right but jump around from centre-slight-left to left and to center (I don’t know specifically how most of these were recorded and unfortunately I can’t do my own piano recordings to further investigate this right now). What did the doctors say? Well, they said that 1. my hearing is still very good and my ear canals are quite clean and there is no earwax on my eardrums and 2. that there are some dips at the 4k octave (typical for some damages of loud music in concerts/clubs etc) and 3. that my left ear is a bit worse than my right ear (if I remember correctly, I hear just slightly less on the left ear and maybe here the dip around 4k was a bit deeper, probably due to 11 years of violin playing). But shouldn’t that result in me hearing sources more to the right because my right ear hears better? Or could it be that my location cues are actually affected and my brain makes up for the differences and unfortunately does the job too well, shifting things to the left? My right ear seems to be slightly different to my left ear also in physiognomy, as sometimes in-ear headphones will hurt a bit more in the right ear and with the NDH 30, the helix of my right ear always touches the fabric on the driver, when the left ear's helix doesn't. And why is it, that sometimes I actually do hear things very precisely in the center, but I can’t seem to find a specific reason that would be reproducible? Is it micro-details in the 2k and 4k octave for location cues? Can I re-train my brain? I've asked ear specialists about this specific problem, but they said they can't find a reason and don't know the answer and recommended me to google it. Well... haven't been successful so far
Ear canals:
One thing to throw in to complicate this even more:
I found this video link in some headphone forum:
David Griesinger is talking at the Tonmeister convention about how hard it is to really find linear headphones for binaural reproduction because the shape of the ear canal is usually overlooked and is a huge factor in shaping the sound. If I understood it correctly, there is a difference in listening to a linear loudspeaker in a well-treated room, because you can measure and adjust it to be flat and then hear it being flat, knowing your ears, while you can’t do that with headphones, because they interact with the shape of your ear canal so strongly, that what would sound flat for one person would sound heavily filtered (especially in upper mids and treble) for another. While the research here is mainly for binaural recording and reproduction and therefore should take into consideration also the ear canal, does this also have an impact on choosing headphones for mixing? Or would you have learned your individual ear canal frequency response in that way that a "true linear headphone" would sound linear to everybody (and take into consideration ear canal shape only for measuring and recording binaurally)?
Speaking of "linearising" headphones to your "preference": Does anybody know how to get a grip on something like his headphone app, only not for Windows but either for MacOS or platform-independent, or the basics of what the software is doing to write a little program (for example in Cycling 74's Max) that does the same as what his headphone app is doing? He says what his app does differently than others is it tests frequency bands while always interrupting the test band with a 500 Hz tone, because a) the ear canal has least resonances at 500 Hz and b) the human hearing always adapts and needs to be "recalibrated" all the time by a reference tone. I would like to try to take this into consideration and build a little program to calibrate my headphones, but I neither know how this app really works nor do I know if that would even make sense for listening/mixing on headphones. Here is his website: http://www.davidgriesinger.com
Here are some of the graphs from his video, he says that different persons would hear a very different frequency content using the same (HD600 in that case) headphones, you can see that one would hear almost linearly while another not at all:
EQing:
If taking your personal hearing into consideration (and knowing it well, I guess), you could of course always EQ your headphones to your personal liking. But this strikes me as a problem, not having listened to very linear headphones a lot and not really KNOWING exactly what flat would be. I could of course always rely on measurement curves from autoeq.app (oratory1990’s curves for example), but these are just curves of his measurement equipment with a specific sample of a headphone model. How can I know that EQ-ing my headphone is actually not making it worse somewhere? I think it’s impossible to say, no?
There was a reddit of a user who had the LCD-X for 6 months and only then could adjust EQ to preference, the online frequency responses were not good enough. This is the thread: But what to do if you can only listen to the headphones for some weeks and then have to decide?
Speaking of EQ: I tried sonarworks (didn’t sound nice in no single case, always weirdly changed the sound),
I tried equalising with Fab Filter using curves from for example oratory1990 on autoeq.app. (how long it took to find out the Q values are calculated differently… ugh… I can try to find the source if someone needs this, also went crazy about Ableton Live’s Eight band EQ, again different Q values than the standard…)
And I tried Realphones 2. Realphones sounded best in some cases, Fabfilter EQ sounded always too strong, but when I dialed down the Fabfilter correction curve in amplitude, it sounded always quite good. Dialing down the sonarworks dry/wet: weird. Dialing down realphones: okay-ish. Applying Fabfilter EQ seemed the most reasonable thing to do because you would really know what you were doing as opposed to some algorithms you don’t really know about (now, is there a "room simulation" curve applied or not, and am I sure I really know what the plugins are doing?). EQing the NDH 30 with oratory1990’s curve (bass boost, treble boost) made them sound more like LCD-X, but more like LCD-X un-EQed! Interesting. Now - is that getting closer to linearity or closer to hyped sound? I guess in the treble a little boosting is a good idea, but then again this a very unpredictable area. And as I will be using the NDH30 also in situations where I am listening directly without any EQ/DSP, I still should be comfortable with how my headphones sound as-are.
Speaking of EQ/frequency response curves you find online: are they really useable? They are so different! You'll find different measurements with different data for treble above 2…3kHz and you simply won’t know which to choose and if to use the treble part of the curve or not, right? Some say there's some measurement influences around (if I remember correctly) 10kHz you should ignore and above a certain frequency the data is unreliable, but how do I know if the correction curves on autoeq.app or elsewhere took that into consideration or not? and not EQing NDH30 in the upper treble - maybe I should do it just a bit, its quite warm.
Even more so: autoEQ.app seems to do weird calculation stuff: I downloaded the EQ settings for the LCD-X three times on three different days and I got three slightly different curves (frequencies would differ by up to 10 Hz and dB values in some cases by up to something like 3...6 dB).
Also: I got the EQ curve for the specific LCD-X pair I had from Audeze’s quality control. Tried to use it for EQ, I don’t know, maybe helped a bit? But it sure was quite different from oratory1990’s on autoeq.app. So which one to use?
How can I know if the curves I found were actually the right ones? I talked about it earlier in this post, where I also put the frequency response curves. Correction, yes please, but based on what exactly? There's so many different curves. And even if I put a curve with an EQ, switch it on, compare it A/B back and forth, find that "wow this curve does the trick", go for a coffee, come back, only to find that "what? Yeah maybe the curve added some missing sparkle and detail in the upper treble region, but where are my nice natural mids? They're all gone! Back to no EQ, it was better!" etc etc... My point/question here is: We don't know how Sonarworks or other companies measured, we don't always know what additional algorithms or curve they add, and most importantly we don't know the frequency response of our specific set of headphones. Do people really install Sonarworks or download the autoeq.app curve, and that's it, "boom. Perfection." (this is actually a quote from a NDH 30 review)? I mean I'd love that, but I can't say that it worked for me in any case...
I also called a person at Neumann, asking them for a quality control frequency curve of my specific NDH30 pair. Surprise: two employees independently both said no, both explained it similarly: handing out frequency response curves wouldn’t make any sense. They would be highly dependent on measurement equipment and the measurement adaptors that were used. Every company had different techniques/adaptors. And it would only make sense to compare different headphones from the same company using the same measurement technique. I didn’t fully understand though how that was linked to completely unusable curves, I mean the curves should show at least something, no? They are saying: no, we won’t give these away, they will show up in forums and people will discuss them like crazy, but they won’t be representative - does that maybe also show some kind of honesty/responsibility/more scientific-based than marketing oriented positions? Or I can ask the question differently: Why are Ollo Audio (in their case they send you a measurement of your sample, it's part of their marketing even!) and Audeze (I'm not the only one who got a sample-specific measured frequency curve from their quality control department) handing out their frequency response measurements for the individual headphones, whereas Neumann don't, arguing that it wouldn't make any sense. Neumann are almost sounding like in an ideal world they would love to not give any frequency response for their headphones at all?
(The person at Neumann said I should try the HD600 if the NDH30 was not comfortable enough, by the way. Haven’t tried them yet. I am afraid of missing the bass, as some people claim.)
so...???
To be honest - I would really like to keep the NDH30 (for different reasons: portability, design/looks, simlilarity to KH120/KH150, …), just get used to them and get to love them as a precise and good working tool. Change the headband using a HD5xx or HD6xx headband for more comfort. And try to get to know exactly how they sound and apply some EQ for some work situations maybe… But I have been reading so much online and made so many listening tests that I don’t really know anymore what I really hear and what to think and some things really irritate me. Maybe this post can give some insights on the complexity of the topic (for an audio person/professional, but a headphone beginner like me), and maybe you are having some ideas concerning my irritations, misunderstandings and maybe even some theories on my hearing irritations in stereo/depth/stable phantom center - I would like to find out whether I should trust my ears and think of them as very precise and able to hear imprecise imaging, or if I should mistrust them and think of my hearing as being slightly impaired and my brain of trying to make up for it a bit too much - which would again really complicate the journey on finding a matching pair of headphones. What do you think?
Thanks a lot for your patience and time!
I am searching for a good pair of open-back headphones to use for:
- mainly studio mixing (in combo with my Neumann KH150 for checking bass/details), this is the number 1 priority
- mixing/producing binaural content
- also sometimes mixing, recording and producing when I am on the road (yes, also rarely tracking)
- listening (music, films, …) when i can’t use loudspeakers
Well - it’s a journey and I just began mine, having had cheap headphones until now (DT990, sometimes I used DT1990 Pro, wasn’t really a huge fan of those due to treble/highs) and having not a lot of experience with dedicated headphone mixing.
This post is quite detailed - ehmm that's a euphemism - this post is long, I know, I’m sorry, but I can’t transport the info and questions any other way as I am both still learning and many factors are interlinked. I'm aware not a lot of people will read this through and some questions have been discussed somewhere probably. But I read a lot, and things still don't come together. I'm happy for every answer, every correction, every hint on good sources/reads, and also every comment on maybe a lack of experience or a wrong approach in some way. I’m glad about recommendations on what to test/buy, but even more so if somebody can clarify some of my irritations/misunderstandings. And I see this post as a mix of personal experience, in some rare case potentially a bit of a guide for buyers' decisions, and primarily as a headphone amateur's pool of questions.
So I decided it should be a pair of open-back headphones. Even if I will occasionally do some recording/tracking, I could for some quieter sources still use open-backs and would have the advantage of a more natural and linear sound, and for most cases I still have a cheap pair of DT990 and some in-ear for tracking, too.
Important: I am also questioning whether I am having some problems with my ears considering stereo imaging. I am writing about it a bit below, it could change everything about testing and judging of course, but it’s almost impossible for me to really know if that is true, as measurements in the labs of doctors tell me my hearing should be okay. It’s a thought in the back of my head I’m getting to later again.
I tested the following ones, listening to each one for only some 30mins when I didn’t like them and hours to days when I liked them. I listened to all of them un-EQ’ed. Talking about these cans from now on I’m referring to them un-EQ’ed if not stated otherwise.
I ranked them to my personal liking like this:
1. Audeze LCD-X 2021 (1250 €)
> sounded the most pleasing with the highest “wow” factor while still having loads of detail
> treble/air is there and nice but it’s not an overly bright headphone (I do like sound on the “warmer” side)
> best “soundstage” and impressions of the rooms/reverbs on most tracks (not all!)
> very natural bass but a little muddy and imprecise at times (not as transient as other headphones, "longer decays")
> natural+precise, but they also seem to add a bit of nice stage/depth/room when other don’t (good or bad!?)
> heavier and bulkier than the others, and with the bulky cable adaptors not perfectly suited for traveling
> very comfortable although a little bit heavy (they move when looking down)
2. Neumann NDH 30 Black (550 €)
> in lack of better words: very dry sounding headphones. fast transients, seem very precise
> bass sometimes very dry and fast and revealing, but sometimes lacking upper bass/low mids a little bit
> smooth treble (wouldn’t say dark but warm, still precise treble). I like the tuning (I also mix on KH150)
> not as pleasing and “wow” as the LCD-X, but then again this can also be a good thing (mixing)
> on good organ church recordings or percussion in real rooms or binaural: best soundstage/natural
> difference in sound heavily dependent on positioning (front/back axis, angle of head band)
> on first use: very uncomfortable. It seems my headshape is not perfect, it really hurt on top of my head
> comfort: I stretched them a bit using a headphone stand. Now more comfy, but less clamping force
> less clamping force could now possibly lead to less sealing and therefore sometimes bass issues?
> not so heavy, they look sturdy, very beautiful and simple design, foldable, very handy, portable
> yes, cable is on the right side, and it makes some sounds, but it's manageable
3. HifiMan Sundara Black (earlier edition, supposedly the same as the new, the one I got for 249,- €)
> good bass (not amazing but very good), good treble (but a bit too harsh/strong for me)
> very precise but transients/punch not as dry as NDH30 and sound not as natural as NDH30/LCD-X
> sound just very slightly more phasey or just little tad more weird than NDH30/LCD-X
> looks like a very nice build quality (although people say sometimes the earpads loosen), nice design
> very comfortable! Most comfortable together with the HD 490 Pro, at least for me
4. Sennheiser HD 490 Pro (430 €)
> very nice sounding actually, I liked the treble, soundstage (if I even know what it is) was also okay
> the bass just rarely seemed a tiny bit distorted or it felt as if the energy wasn’t enough to reproduce
> amazing comfort! They’re very light, I could almost forget they are on my head! improves immersion!
> would have loved to keep them but I couldn’t shake the unprecise feeling of a hint of “weakness”
5. Ollo Audio X1 1.0 2024 (540 €)
> very good headphones, I I got used to them I could really work with them, but I just don’t like them
> if I had to describe their negatives for me: feeling of slightly (!) resonant, slightly weak?
> I used their custom sample-specific EQ-curve in Realphones, but it didn't really help
> very comfy, nice simple design
6. Beyerdynamic DT1990 Pro 250 Ω (400 €)
> the treble doesn’t work for me. too bright! Don’t like them. Still very good headphones
7. Sony MDR-MV1 (395 €)
> didn’t like the hyped bass/treble sound and the build quality, they were the first to sort out
I quickly sorted out headphones 5-7 and listened more to 1-4, especially to 1 and 2. What I found was:
LCD-X and NDH 30:
With some tracks I can hear details with the LCD-X that I can’t hear as much with the NDH 30, on other tracks it’s completely the opposite and I hear details with the NDH 30 that the LCD-X won’t show me. I’m talking about treble details, hidden percussive elements, room reverbs, depth, movements, …
Is a boosted treble/air band something where you automatically get more articulation from and then can also have more detail, or is detail and precision more about the transients, or both? I mean, probably transients are the most important factor here, you can’t EQ transients but you can EQ after the transients do what they do. But transients (and distortion) should be great with the NDH30 and not a problem, I guess. Bass sounds super dry.
It’s hard to say which one wins to my liking. On some sources I think NDH 30 does the job better, but then I listen again to LCD-X and now they suddenly both do the job, or I hear a room reverb on the NDH 30 that I didn’t hear with the LCD-X (finally! A winner! And it’s the cheaper headphones!), but when listening again on the LCD-X I find that yes, the reverberation may be not as good, but still there, but then the bass sounds more three-dimensional or there’s another detail I was missing on the NDH 30.
Or I listen to a track on NDH 30 and have the feeling that the instruments (especially amped clean electric guitar, but in conjunction with bass and drums) are exactly (!) how they’re supposed to sound, like they are so right and pleasing that I get the feeling I want to eat them (if that’s understandable). Then, listening to LCD-X, I am missing this impression, this feeling of compactness, but then the voice sounds more natural (sometimes even much more natural and full) and correct on the LCD-X - which would be due to more present lower mids, while on paper (on frequency responses found online) at least it shouldn't?
And what is that thing about the LCD-X, that it sounds so full-bodied and so roomy? Some online review said that compared to NDH 30 the LCD-X, especially on dark sounds, sounded "more tangible and clearer... and with an impressive breadth, depth and clarity". This sounds also very pleasing, but when assuming the NDH 30 is as linear as claimed I get a bit sceptical if it's actually showing me the original content, or if it is adding to it. Or, a wild and probably nonsense theory: if the build quality, comfort and weight of the headphone that I can feel on my head add to the feeling that the sound had weight, strength, was full-bodied.
So… I can’t decide. Judging speakers seemed an easier task. And I can’t really seem to understand the parameters that would make my judgement jump back and forth from one to the other and my attention jumping between different aspects that I wanted to judge isolated.
For example, NDH 30 are indeed warmer or darker than all the other headphones. I generally prefer a warmer sound to a very trebley sound and hate harshness (of course). At a specific position on my head, in my case with the front part of my ears (the tragus) close to the cushion of the ear pads, their treble sounds best. Other people report that here, on this specific position the treble sounds too harsh even, while most say it’s generally too dark in all positions (or used it with a wrong positioning?).Okay, so the treble sounds nice to me at specific position. And Neumann aims at a rather warm/smooth sound that emulates speakers in a room. Makes sense. But compared to LCD-X I am missing some details sometimes, which probably is because of the rather tamed and soft treble.
And if I want to use headphone equalisation software (I can EQ them to sound closer to LCD-X or other headphones, which could be a good thing for some tasks) and these software (like Sonarworks) also use curves that simulate a bit of that, I would basically have double/added darkness, right?
Also, this position-dependency has me really worried. It’s not that I don’t find a position where I like the sound or that I can’t correct the position once I find that there’s lacking some treble. It’s more about very slight differences and mainly about these two things:
1. Hearing is so subjective and also so dependent on so many factors, that I might get it wrong every other day when they are not positioned exactly the same as when I mixed a track the day before, and I could make mistakes while mixing or judging a specific part of a mix.
2. The bass will be dependent on how the cushions seal. Sometimes now, after having bent the headphones a little bit having them sit on a headphone stand for 1-2 days, they feel less clamping and more comfortable, but also I imagine (mustn’t be true though) that the bass is lacking a bit now, especially as opposed to the LCD-X.
But looking at the manufacturer's frequency responses of both, this couldn't really be true: the LCD-X on paper should have less bass, and while it’s hard to judge stuff going on above 2-3 kHz from the frequency response, it should be okay to judge it in the bass by looking at the graphs, no? The data says that the NDH 30 had more bass. What I hear is the opposite. How can that be? Have I bent the bass away from my NDH 30? Are they never really positioned correctly on my head? When do I actually know it’s positioned correctly? When do I know the bass is how it’s supposed to be? Do I have weird ears (my right ear touches the headphone, my left ear doesn’t)? The NDH30 hurt at the beginning. Now less, since I had them on the headphone stand, now they also have less clamping force. Now, I sometimes imagine (as I am becoming more paranoid for problems since testing headphones haha) the bass got less. Less clamping force = less bass?
Well - I asked Audeze for a frequency response curve for my sample of LCD-X (2021) and this actually explains a bit! It doesn't explain why it seems I lost some of the bass with the NDH 30 (although it's hard to say because I don't have measurement equipment at home and I can only say so from memory).
But what it shows is that there's quite a difference in oratory1990's measurement of LCD-X (2021) and what the Quality Control department at Audeze sent me.
This is Audeze QC:
This is oratory1990 (measured with GRAS 45BC) from autoeq.app:
The mean level between 1.5k and 10k on the Audeze graph is down about 6dB compared to the mean level between 30 Hz and 1.5kHz! This is a lot of bass-emphasis, no?
On oratory1990's measurement, however, it's quite the opposite, I would say the mean level below 1.5kHz is even lower than the mean level above 1.5kHz.
This would mean, assuming the measurements of both parties are approaching ideal measurement, my sample of LCD-X had more than 6 dB (maybe 7 or 8 even) stronger bass and lower mids than the one oratory1990 had? Or does it mean that instead the measurements are just very different? Because if this is the case: Doesn't this mean we can't really rely on measurements as found on autoeq.app?
When looking at Innerfidelity's (measured with HMS II.3) measurements they also look completely different. Very broad wide peak of +6dB from 2 kHz to 5 kHz (serial number 7456406)...?!?
Okay, putting found and received frequency response curves to the side, and back to choice making: So the LCD-X are a bit above my budget right now, but more importantly, they are a bit too large and/or heavy for traveling and not as versatile for a mobile setup, especially if in some cases I don’t use pro audio gear with solid headphone amplifiers but for example my MacBooks audio output (I have to admit: I guess my RME Fireface UCX headphone preamp sounds better than my MacBook Pro's audio output, but differences are quite small and I couldn't vouch for it!) or even use them with my Sound Devices MixPre device (nice headphone amplifier, but 3.5mm) and then have to put the two adaptors that are quite heavy, it’s really not really an option for me, looking for something more versatile and mobile. Also, what is speaking against LCD-X is the slight feeling that they make the sound more beautiful (more warm muddy roomy bass) and less neutral/true.
Still a tough decision to make. If I ruled out LCD-X for the reasons above, there is still HifiMan Sundara. Hmm, they didn’t really work for me. If I wanted to have planar drivers, I would probably go back to the LCD-X which were in my opinion just capable of more. Although the HifiMan Sundara is crazy good value for the money!
HD 490 Pro - also an option. Especially for comfort! How I would love to have a headphone that you can forget about while it’s still on your head. Especially for immersive stuff like binaural or video games or whatever, I have to say that the impression of a good sound is not always just what you hear but also what you think you hear, want to hear, or believe you hear. And also, how comfortable you are. And with lightweight airy and not bass-heavy music they seemed very natural, I guess forgetting I even wore them added to that. But still, they couldn’t compete.
Having a hard time finding the right headphones, I read a lot in forums. And measurement data. NDH30 was very much praised on its low THD here, and in forums mostly by a person named indigogo who was crazy about it. So crazy even that sometimes, the user would write stuff I both couldn’t believe and found very over-the-top. Quite the anti-ad, in my opinion, and while I don’t believe Neumann or someone paid the user, I don’t believe there is the perfect headphone, while I would really like to believe that the NDH30 is close to perfect, it would help make a decision and save some money (opposed to more expensive headphones). There has been some praise for the NDH 30 by 3-4 other persons in forums, but 95% of praises in different forums came from indigo. Well… probably won’t impact my decision. I will have to like it and I did prefer the NDH30 still above all the others for some reasons, but two things were still big question marks: listening pleasure and soundstage.
Listening pleasure:
Just a short question here: Are there headphones that are perfectly suited for both, mixing and listening pleasure? Or can you really say that for example if LCD-X sounds much more 3-dimensional and pleasing and "wow", it will automatically be less neutral and precise? If it sounds more 3-dimensional and pleasing in the bass and lower mids, isn’t that also coloration?
Soundstage:
To be honest - I really don’t know exactly what soundstage is. Is soundstage the depth and width of the recording? Is it how good reverberation can be heard? Does good imaging contradict good soundstage or can there be both AND good linearity? What is the science behind soundstage? Can somebody maybe elaborate on that a bit more in detail? Some users even speak of "height", and while I as a human would know what that sounded like, I couldn't find it in any headphones. Is my hearing too imprecise?
I’ve read in some forums that some mixing professionals who’ve listened to a lot of headphones actually don’t have a clear answer to "soundstage", admitting they didn’t know what the reasons for soundstage were technically and that they couldn’t predict if a headphone had good soundstage, and that they didn’t know if it was something desirable in terms of mixing, and some even claiming they can’t tell you exactly what soundstage even is - while a lot of audiophiles/enthusiasts seem to know perfectly what it is. How is that possible?
I even read some user (https://www.head-fi.org/threads/the-science-of-soundstage.959791/) explaining (without any sources on research/statistical tests/physiological/psychoacoustic explanations) where soundstage came from (they said a certain dip in a frequency range in the treble region) and that it could improve from a bass and upper treble enhancement and that they could read a frequency response graph and tell you wether some cans had soundstage. A lot of users disagreed. It seemed to me this frequency response explanation was a bit over-the-top and not necessarily true, however I stumbled upon a lot of threads where people claim headphones having a soundstage are some that have some more present bass and certain highs.
Some say that headphones with a huge soundstage and a very pleasing listening experience with “wow” effect can even be bad for mixing (good would be linearity and precise imaging). I guess so, If “pleasing” means delivering more bass or treble than a track might be mixed for on “flat” loudspeakers in a well treated room. But then again, what if soundstage is not about frequency response but about showing the results of clever mixing (when for example psychoacoustic tricks were used). What if I’m missing these details with headphones that have less soundstage? After all people listen to mixes on headphones, and as a mixer I would take soundstage into consideration. So I should hear it. Or, to dive further, I could ask the question, how mixing engineers produce soundstage. Is it just about reverb, early reflections and pre-delay (in the digital/artifical reverb domain) or very good stereo recordings in the real-world-recording-domain?
I’m asking these questions obviously to understand if soundstage is even an important aspect to consider when choosing for a pair of mixing headphones, and I am starting to suspect that cans with an amazing soundstage are not necessarily linear anymore. Or that soundstage was a tradeoff with imaging/precision on the other side, or with linearity. But is that true?
The Neumann NDH 30 don’t seem to have a very deep or wide soundstage, LCD-X often seemed to have more soundstage, but that was more of a feeling. But, like I said in the notes about NDH 30 and LCD-X above, soundstage also differed from track to track. In some good binaural recordings and some stereo church organ recordings (Toccata and fugue in D Minor BWV 565 Fugue, Kay Johansson) NDH 30 sounded more natural to me and I good hear a deeper reverberation in the room and had a feeling of being more immersed, a stronger feeling of a deep and wide soundstage. Or in Buena Vista social clubs number-one-loudspeaker-and-headphones-test-track “Chan Chan”, where I could hear a very spatial reverb in the first minute, caused by a percussive instrument that was sitting to the mid-left with beautiful echoes/reverb in the mids and highs, reverberating in the right and mid-right position that I was missing in this precision with the LCD-X. But then, for other tracks like Jazz or RnB, the LCD-X sounded more realistic/externalizing. In a lot of pop mixes, like D’Angelo’s “Send it on” for example, LCD-X’s soundstage seemed better (background voices were further away, the horns to the sides left and right seem more exterior and further away from the ears). Some people say the “further away” externalisation also depends on how close the drivers are to the ears. Is that so?
What about reverb and reverb details? Shouldn’t cans that show you all the details of room reverberation also add to the most realistic room feeling? In recordings like “Meus 26” by Tim Bernandes, for example. Here, by the way, I heard the reverb on the vocals mainly to the right - what a weird decision in mixing! Flipping my NDH 30 and wearing their right pad on my left hear to prove it's not my hearing didn't work, they just sounded weird (angled drivers etc.). Anyway, I could hear more details of this reverb on the NDH 30 than on the LCD-X, though, but also it sounded a little bit more isolated, taking a bit of the soundstage immersion, and the LCD-X sounded like having more soundstage, also because the bass and lower mids were more smooth and organically blending into the room reverb, making it more of a round thing and adding to an impression of soundstage and immersion more.
Also quite a nice track for room reverb and depth: Papajazz - “Money Speaks”. Both LCD-X and NDH 30 sound amazing on this, with LCD-X sounding more deep and round and organic and warm and effortless, while NDH 30 sounding a bit more precise and revealing the room differently. Is any of these better? It’s so hard to tell.
I read this article of Diego Estan (https://www.soundstageaccess.com/in...ic-i-use-to-evaluate-soundstaging-and-imaging), which is about speakers, not about headphones! But he listed a lot of songs that would have extraordinary good imaging and soundstage. I listened back on my Neumann KH system, and I could hear what he would mean, but I was not impressed as he was - probably because my room is not treated as well as his was (also a reason why I need open back headphones for checking for some mix anomalies due to room treatment problems).I wondered if these would translate to headphones, too. But I was really missing the deep externalisation he was talking about, both on my speaker system and on my headphones. When he talks about elements sounding like coming from behind the walls of his room - no, I can’t hear that. I remember listening to a recording of percussion in a medium reverberating room on medium quality studio speakers in an acoustically treated room (but not really with bass traps), that was a lot of years ago, and I was amazed on how 3D I could hear the percussive elements - some pianissimo to piano sounds felt like they were actually played super close to my nose (!) - that was really holographic! Unfortunately, I never had that impression again and when some years later I listened to a lot of studio monitors and found the Neumann KH 150 perfectly suited for me, did quite some room treatment (but for mainly mids and treble, no huge bass traps) and used the MA-1 room correction DSP system, I still couldn’t get anywhere as close to what I heard years back then. The only time when I heard a very precise imaging or something close to a very deep soundstage was, if I remember correctly, on coaxial speakers (Geithain and KS Audio), something I couldn’t hear on non-coaxial. Friends who listen on my system also seem to hear isolated mono sources centred better than I do. Then again, they don't always have trained ears. I am saying all this because I am kind of speculating on the fact that...
...my hearing may impaired:
and I may not be able to actually hear (anymore) what good stereo imaging is. Let me quickly summarise:
When listening to speakers in a well treated room, a mono source appears sharp and in the center only on a few rare occasions. It mostly moves a little to the left and the right jumping around a slight-left center, but most importantly: It mainly sits somewhere to the slight left. This may be bad loudspeaker correction or bad room acoustics, so listening with headphones should do different, no? Well, only sometimes. Sometimes I hear a voice in the center, but mostly just slightly to the left, regardless of the headphone orientation. I am not talking 9 o clock, I am more talking 11h45. In fact, in most of the mixes that I am listening to, the majority of content seems to be heard very slightly to the left, they don’t seem completely balanced (there is a thread somewhere where users discuss that music is actually mixed more to the left - is it music or is it my ears?). Listening to some good stereo recordings of pianos that seem to make use of stereo in the sense that lower register is panned more to the left and higher register to the right, a weird thing happens sometimes in the mids: they seem to jump around, and even if a player played a chromatic or diatonal series of notes going up in pitch in middle register, they wouldn’t move from left to right but jump around from centre-slight-left to left and to center (I don’t know specifically how most of these were recorded and unfortunately I can’t do my own piano recordings to further investigate this right now). What did the doctors say? Well, they said that 1. my hearing is still very good and my ear canals are quite clean and there is no earwax on my eardrums and 2. that there are some dips at the 4k octave (typical for some damages of loud music in concerts/clubs etc) and 3. that my left ear is a bit worse than my right ear (if I remember correctly, I hear just slightly less on the left ear and maybe here the dip around 4k was a bit deeper, probably due to 11 years of violin playing). But shouldn’t that result in me hearing sources more to the right because my right ear hears better? Or could it be that my location cues are actually affected and my brain makes up for the differences and unfortunately does the job too well, shifting things to the left? My right ear seems to be slightly different to my left ear also in physiognomy, as sometimes in-ear headphones will hurt a bit more in the right ear and with the NDH 30, the helix of my right ear always touches the fabric on the driver, when the left ear's helix doesn't. And why is it, that sometimes I actually do hear things very precisely in the center, but I can’t seem to find a specific reason that would be reproducible? Is it micro-details in the 2k and 4k octave for location cues? Can I re-train my brain? I've asked ear specialists about this specific problem, but they said they can't find a reason and don't know the answer and recommended me to google it. Well... haven't been successful so far
Ear canals:
One thing to throw in to complicate this even more:
I found this video link in some headphone forum:
Speaking of "linearising" headphones to your "preference": Does anybody know how to get a grip on something like his headphone app, only not for Windows but either for MacOS or platform-independent, or the basics of what the software is doing to write a little program (for example in Cycling 74's Max) that does the same as what his headphone app is doing? He says what his app does differently than others is it tests frequency bands while always interrupting the test band with a 500 Hz tone, because a) the ear canal has least resonances at 500 Hz and b) the human hearing always adapts and needs to be "recalibrated" all the time by a reference tone. I would like to try to take this into consideration and build a little program to calibrate my headphones, but I neither know how this app really works nor do I know if that would even make sense for listening/mixing on headphones. Here is his website: http://www.davidgriesinger.com
Here are some of the graphs from his video, he says that different persons would hear a very different frequency content using the same (HD600 in that case) headphones, you can see that one would hear almost linearly while another not at all:
EQing:
If taking your personal hearing into consideration (and knowing it well, I guess), you could of course always EQ your headphones to your personal liking. But this strikes me as a problem, not having listened to very linear headphones a lot and not really KNOWING exactly what flat would be. I could of course always rely on measurement curves from autoeq.app (oratory1990’s curves for example), but these are just curves of his measurement equipment with a specific sample of a headphone model. How can I know that EQ-ing my headphone is actually not making it worse somewhere? I think it’s impossible to say, no?
There was a reddit of a user who had the LCD-X for 6 months and only then could adjust EQ to preference, the online frequency responses were not good enough. This is the thread: But what to do if you can only listen to the headphones for some weeks and then have to decide?
Speaking of EQ: I tried sonarworks (didn’t sound nice in no single case, always weirdly changed the sound),
I tried equalising with Fab Filter using curves from for example oratory1990 on autoeq.app. (how long it took to find out the Q values are calculated differently… ugh… I can try to find the source if someone needs this, also went crazy about Ableton Live’s Eight band EQ, again different Q values than the standard…)
And I tried Realphones 2. Realphones sounded best in some cases, Fabfilter EQ sounded always too strong, but when I dialed down the Fabfilter correction curve in amplitude, it sounded always quite good. Dialing down the sonarworks dry/wet: weird. Dialing down realphones: okay-ish. Applying Fabfilter EQ seemed the most reasonable thing to do because you would really know what you were doing as opposed to some algorithms you don’t really know about (now, is there a "room simulation" curve applied or not, and am I sure I really know what the plugins are doing?). EQing the NDH 30 with oratory1990’s curve (bass boost, treble boost) made them sound more like LCD-X, but more like LCD-X un-EQed! Interesting. Now - is that getting closer to linearity or closer to hyped sound? I guess in the treble a little boosting is a good idea, but then again this a very unpredictable area. And as I will be using the NDH30 also in situations where I am listening directly without any EQ/DSP, I still should be comfortable with how my headphones sound as-are.
Speaking of EQ/frequency response curves you find online: are they really useable? They are so different! You'll find different measurements with different data for treble above 2…3kHz and you simply won’t know which to choose and if to use the treble part of the curve or not, right? Some say there's some measurement influences around (if I remember correctly) 10kHz you should ignore and above a certain frequency the data is unreliable, but how do I know if the correction curves on autoeq.app or elsewhere took that into consideration or not? and not EQing NDH30 in the upper treble - maybe I should do it just a bit, its quite warm.
Even more so: autoEQ.app seems to do weird calculation stuff: I downloaded the EQ settings for the LCD-X three times on three different days and I got three slightly different curves (frequencies would differ by up to 10 Hz and dB values in some cases by up to something like 3...6 dB).
Also: I got the EQ curve for the specific LCD-X pair I had from Audeze’s quality control. Tried to use it for EQ, I don’t know, maybe helped a bit? But it sure was quite different from oratory1990’s on autoeq.app. So which one to use?
How can I know if the curves I found were actually the right ones? I talked about it earlier in this post, where I also put the frequency response curves. Correction, yes please, but based on what exactly? There's so many different curves. And even if I put a curve with an EQ, switch it on, compare it A/B back and forth, find that "wow this curve does the trick", go for a coffee, come back, only to find that "what? Yeah maybe the curve added some missing sparkle and detail in the upper treble region, but where are my nice natural mids? They're all gone! Back to no EQ, it was better!" etc etc... My point/question here is: We don't know how Sonarworks or other companies measured, we don't always know what additional algorithms or curve they add, and most importantly we don't know the frequency response of our specific set of headphones. Do people really install Sonarworks or download the autoeq.app curve, and that's it, "boom. Perfection." (this is actually a quote from a NDH 30 review)? I mean I'd love that, but I can't say that it worked for me in any case...
I also called a person at Neumann, asking them for a quality control frequency curve of my specific NDH30 pair. Surprise: two employees independently both said no, both explained it similarly: handing out frequency response curves wouldn’t make any sense. They would be highly dependent on measurement equipment and the measurement adaptors that were used. Every company had different techniques/adaptors. And it would only make sense to compare different headphones from the same company using the same measurement technique. I didn’t fully understand though how that was linked to completely unusable curves, I mean the curves should show at least something, no? They are saying: no, we won’t give these away, they will show up in forums and people will discuss them like crazy, but they won’t be representative - does that maybe also show some kind of honesty/responsibility/more scientific-based than marketing oriented positions? Or I can ask the question differently: Why are Ollo Audio (in their case they send you a measurement of your sample, it's part of their marketing even!) and Audeze (I'm not the only one who got a sample-specific measured frequency curve from their quality control department) handing out their frequency response measurements for the individual headphones, whereas Neumann don't, arguing that it wouldn't make any sense. Neumann are almost sounding like in an ideal world they would love to not give any frequency response for their headphones at all?
(The person at Neumann said I should try the HD600 if the NDH30 was not comfortable enough, by the way. Haven’t tried them yet. I am afraid of missing the bass, as some people claim.)
so...???
To be honest - I would really like to keep the NDH30 (for different reasons: portability, design/looks, simlilarity to KH120/KH150, …), just get used to them and get to love them as a precise and good working tool. Change the headband using a HD5xx or HD6xx headband for more comfort. And try to get to know exactly how they sound and apply some EQ for some work situations maybe… But I have been reading so much online and made so many listening tests that I don’t really know anymore what I really hear and what to think and some things really irritate me. Maybe this post can give some insights on the complexity of the topic (for an audio person/professional, but a headphone beginner like me), and maybe you are having some ideas concerning my irritations, misunderstandings and maybe even some theories on my hearing irritations in stereo/depth/stable phantom center - I would like to find out whether I should trust my ears and think of them as very precise and able to hear imprecise imaging, or if I should mistrust them and think of my hearing as being slightly impaired and my brain of trying to make up for it a bit too much - which would again really complicate the journey on finding a matching pair of headphones. What do you think?
Thanks a lot for your patience and time!
