• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

ZMF Caldera Headphone Review

Rate this headphone:

  • 1. Poor (headless panther)

    Votes: 47 26.0%
  • 2. Not terrible (postman panther)

    Votes: 85 47.0%
  • 3. Fine (happy panther)

    Votes: 31 17.1%
  • 4. Great (golfing panther)

    Votes: 18 9.9%

  • Total voters
    181
serious question, if all headphone mfr were to adhere to a singular target and all music production going forward as well, is the idea that all past music previously recorded would be remastered to the Harman curve in some capacity? I'm having trouble wrapping my head around the totality of your expectations here and would like to understand them better. What would all music and reproduction look like in a perfect world and how would you expect that that task is carried out?
Ups, and here lies the problem, I admire your expertise in designing headphones, they are awesome and capable, but you need to familiarize more with standardization, and how is so helpful in all scientific fields. When Amir recommends a HP is because data supports that it will please the majority of listeners (not all of course), Amir can not recommend a product that the data said will only please a small minority. If you think your Standard is better then you have to put the data to support that, If you design the HP to a special minority of gifted listeners, still you have to show the data to support that too, you see facts, no opinions.
 
Last edited:
Amir can to recommend a product that the data said will only please a small minority. If you think your Standard is better
I'm making no such argument! Standardization and the understanding of research is important of course.

Everyone should like what they like, believe whatever data that they like to meet their personal needs.

I am honestly interested in how the interplay between the audio recording industry and audio reproduction industry would work in a perfect world according to Amir. I do hope Amir thoughtfully answers my question.
 
Ups, and here lies the problem, I admire your expertise in designing headphones, they are awesome and capable, but you need to familiarize more with standardization, and how is so helpful in all scientific fields. When Amir recommends a HP is because data supports that it will please the majority of listeners (not all of course), Amir can to recommend a product that the data said will only please a small minority. If you think your Standard is better then you have to put the data to support that, If you design the HP to a special minority of gifted listeners, still you have to show the data to support that too, you see facts, no opinions.
The arrogance on display here is amazing. ZMF is a very successful business with thousands of delighted customers, many of whom apparently do not share the tastes of the almighty "majority" of Harman adherents. You may not approve of ZMF headphones or the sonic preferences of their clientele, but I don't see how that gives you license to demand that others bend to your notion of what is proper. I can see the benefit of standards on the production side of music, but some people simply like various colorations of sound, and thus there will always be a market for consumer equipment that caters to such preferences.
 
Ah I did some exploring and I think all major headphone papers can be fine on AES. They require paying for the papers but I think that is reasonable enough. I'm going to have to go read them to better understand and participate in discussions!
 
  • Like
Reactions: GDK
Ah I did some exploring and I think all major headphone papers can be fine on AES. They require paying for the papers but I think that is reasonable enough. I'm going to have to go read them to better understand and participate in discussions!
I would strongly suggest joining the AES - it will cost less to have an associate membership (which gets you all the papers for free) than it will to buy the full Harman papers alone.
 
some people simply like various colorations of sound,
The Harman target is a coloration that 64 percent prefer per the research. And I forget but I believe this is based on a wide array of listeners, not just online forum headphonophiles.

We as headphone audiophiles are a very particular group, and what I've found is that many headphone audiophiles are not part of that 64 percent, and ZMF certainly wouldn't exist of those 64 percent were such a majority.

And this is part of why the Harman target is so great, it's a conversation and gateway to further and other great research, not closed "right" or "wrong" judgment, which is very exciting in the days/years to come.
 
Amir - serious question, if all headphone mfr were to adhere to a singular target and all music production going forward as well, is the idea that all past music previously recorded would be remastered to the Harman curve in some capacity? I'm having trouble wrapping my head around the totality of your expectations here and would like to understand them better. What would all music and reproduction look like in a perfect world and how would you expect that that task is carried out?
I cannot answer for Amir, but I believe I can answer for Floyd and Sean, at least as I understand their positions:

  • Present research on preferences for headphones and speakers is done with extant recordings
  • Ergo, the currently-most-preferred equipment is most preferred with recordings that predate any standardization, on average
  • Standardizing playback equipment means that this relatively backward-compatible standard then also becomes forward-compatible in perpetuity
  • Both Sean and Floyd have consistently advocated for easily user-accessible tone controls to adjust for per-recording and other individual preference variation.
I'm not a "reproductionist", so I personally don't care much one way or another about the circle of confusion - I have a much more "mort de l'auteur" view on the matter that whatever is preferred by end users is preferable, end of story - but that's my understanding of the brass tacks of the circle of confusion argument.
 
The arrogance on display here is amazing. ZMF is a very successful business with thousands of delighted customers, many of whom apparently do not share the tastes of the almighty "majority" of Harman adherents. You may not approve of ZMF headphones or the sonic preferences of their clientele, but I don't see how that gives you license to demand that others bend to your notion of what is proper. I can see the benefit of standards on the production side of music, but some people simply like various colorations of sound, and thus there will always be a market for consumer equipment that caters to such preferences.
Well, sorry if I sounded arrogant, it is actually in the realm of possibilities the FR showed by this HP plus the still unknown properties we do not know about HPs makes a combination that is superior than the standard?, of course it is possible, as a man that believe in science can not deny that posibility. But without data, the total opposite is also as possible. you see the my problem here, I am not demandinding anything, just trying to explain why Amir can not recommend this HP without EQ
 
  • Standardizing playback equipment means that this relatively backward-compatible standard then also becomes forward-compatible in perpetuity
  • Both Sean and Floyd have consistently advocated for easily user-accessible tone controls to adjust for per-recording and other individual preference variation.
So unifying all headphones response and then having all amps/dacs have tone controls that adhere to adjustments in the areas that the Harman target has identified to account for recordings (and user preference) that predate this total unification?

Am I getting that right?

Amir is this how you would do it?
 
Last edited:
I would strongly suggest joining the AES - it will cost less to have an associate membership (which gets you all the papers for free) than it will to buy the full Harman papers alone.

I might. It's going to really cut into my time for reading papers in my actual field
 
So unifying all headphones response and then having all amps/dacs have tone controls that adhere to adjustments in the areas that the Harman target has identified to account for recordings (and user preference) that predate this total unification?

Am I getting that right?
I don't think Sean has specifically advocated having all headphones have the exact same response - certainly, Floyd hasn't advocated having all speakers have the exact same directivity. That level of...specificity in prescription is something I don't see in their work.

If I were forced at gunpoint to stylize the solution to the circle of confusion per what I understand Sean and Floyd to believe, it would be that all speakers should be designed with reasonably constant directivity and axial responses free of significant resonances, and that all headphone responses should be "Harman-like" in general insofar as having a similar shape for ear gain, and either defaulting to or at least having the option to have something close to the Harman target's overall tilt.

Like, a lot of Sean's commentary is mostly about headphones that...like, most people can buy, which are usually active DSP designs these days. Those definitely should have a "Harman preset", I think we can all agree that's a good idea. For passive headphones and audiophile DAC-amp stuff, having a robust DSP and measurements that would allow making profiles to EQ a headphone to be close to Harman sounds like a good idea, and I know Sean sees value in ex. @oratory1990's database.

As far as how tightly grouped things need to be, at the very least when I've talked to Sean (both publicly and privately), he has seemed supportive of the idea of having options for different tastes, so long as they're like...real, identifiable-in-blind-listening-tests tastes, not just something somebody is asserting based on a sighted preference. I wouldn't really feel comfortable drawing boundaries for precisely what Sean or Floyd would consider "good enough", I guess that's something you'd have to ask them directly.

Re:
tone controls that adhere to adjustments in the areas that the Harman target has identified
I would not say, and I don't think that Sean would say, that the publicly available Harman work has identified where ideal tone control adjustments would happen. Like, that just plain isn't something that was tested in the research, with the exception of allowing frequency adjustments on shelf filters. They may have done this internally, but in terms of what's published, we have results from a number of methodologies that all point in the same direction as far as what people prefer for the end result headphone frequency response, but not a lot of data to say which methodology is the best one for EQing.

It'd be a very interesting thing to do a paper on, in fact. Anyone got some spare budget?
 
I don't think Sean has specifically advocated having all headphones have the exact same response - certainly, Floyd hasn't advocated having all speakers have the exact same directivity. That level of...specificity in prescription is something I don't see in their work.

If I were forced at gunpoint to stylize the solution to the circle of confusion per what I understand Sean and Floyd to believe, it would be that all speakers should be designed with reasonably constant directivity and axial responses free of significant resonances, and that all headphone responses should be "Harman-like" in general insofar as having a similar shape for ear gain, and either defaulting to or at least having the option to have something close to the Harman target's overall tilt.

Like, a lot of Sean's commentary is mostly about headphones that...like, most people can buy, which are usually active DSP designs these days. Those definitely should have a "Harman preset", I think we can all agree that's a good idea. For passive headphones and audiophile DAC-amp stuff, having a robust DSP and measurements that would allow making profiles to EQ a headphone to be close to Harman sounds like a good idea, and I know Sean sees value in ex. @oratory1990's database.

As far as how tightly grouped things need to be, at the very least when I've talked to Sean (both publicly and privately), he has seemed supportive of the idea of having options for different tastes, so long as they're like...real, identifiable-in-blind-listening-tests tastes, not just something somebody is asserting based on a sighted preference. I wouldn't really feel comfortable drawing boundaries for precisely what Sean or Floyd would consider "good enough", I guess that's something you'd have to ask them directly.

Re:

I would not say, and I don't think that Sean would say, that the publicly available Harman work has identified where ideal tone control adjustments would happen. Like, that just plain isn't something that was tested in the research, with the exception of allowing frequency adjustments on shelf filters. They may have done this internally, but in terms of what's published, we have results from a number of methodologies that all point in the same direction as far as what people prefer for the end result headphone frequency response, but not a lot of data to say which methodology is the best one for EQing.

It'd be a very interesting thing to do a paper on, in fact. Anyone got some spare budget?
Hey thanks for the thorough and thoughtful response! So very interesting and I do look forward to future work on this stuff. Thanks and appreciate you and your time.
 
I am honestly interested in how the interplay between the audio recording industry and audio reproduction industry would work in a perfect world according to Amir. I do hope Amir thoughtfully answers my question.
I think Amir's ideal outcome would be an end to the "circle of confusion" in which studios mix for home listeners that have an unknown / random response, and (worst case) studios are also mixing on monitors with an unknown / random response, whose recordings are in turn used to evaluate the home speakers, (even by speaker builders)... and around and around.

In the ideal situation, studios would have a good idea of how recordings would sound on the listener's end, because everyone would have more or less the same frequency response in their home systems. Listeners wouldn't have to wonder if their speakers sound right because the home speakers would at least be targeting the same response as the studios.

An aside: This may sound dreary compared to the "wealth" of options that exist today. On the other hand, TVs have generally solved this problem by calibrating to standard color spaces both at the studio and homes, and if you don't like the standard, there are always "EQ" controls to change color balance if you want to. Everyone seems fine with this situation.

Not speaking for Amir here, just my understanding of the logic behind lobbying for a standard response.

However, it's worth noting that Amir has both a positive and normative outlook on audio. On the former count, he publishes objective data on everything he reviews, which I like because it lets us make up our own minds armed with good info.

On the latter count, he believes that the Harman response is the right one, other ones are to at least some extent wrong. While you and Amir disagree on this, I think that's fine - it's a matter of opinion if standardization is really desirable or not.

Olive's article on Toole's concept of circle of confusion: https://seanolive.blogspot.com/2009/10/audios-circle-of-confusion.html

Breaking the Circle of Circle of Confusion



As Toole points out in [1], the key in breaking the circle of confusion lies in the hands of the professional audio industry where the art is created. A meaningful standard that defined the quality and calibration of the loudspeaker and room would improve the quality and consistency of recordings. The same standard could then be applied to the playback of the recording in the consumer’s home or automobile. Finally, consumers would be able to hear the music as the artist intended.

There is precedent for this. THX is one we all know. But there would be requirements for linearity, dispersion, etc. for speakers and likewise for headphones.
 
Last edited:
I think Amir's ideal outcome would be an end to the "circle of confusion" in which studios mix for home listeners that have an unknown / random response, and (worst case) studios are also mixing on monitors with an unknown / random response, whose recordings are in turn used to evaluate the home speakers, (even by speaker builders)... and around and around.

In the ideal situation, studios would have a good idea of how recordings would sound on the listener's end, because everyone would have more or less the same frequency response in their home systems. Listeners wouldn't have to wonder if their speakers sound right because the home speakers would at least be targeting the same response as the studios.

An aside: This may sound dreary compared to the "wealth" of options that exist today. On the other hand, TVs have generally solved this problem by calibrating to standard color spaces both at the studio and homes, and if you don't like the standard, there are always "EQ" controls to change color balance if you want to. Everyone seems fine with this situation.

Not speaking for Amir here, just my understanding of the logic behind lobbying for a standard response.
Man, wait till Amir finds out about SoundCloud
 
I don't think Sean has specifically advocated having all headphones have the exact same response - certainly, Floyd hasn't advocated having all speakers have the exact same directivity. That level of...specificity in prescription is something I don't see in their work.

If I were forced at gunpoint to stylize the solution to the circle of confusion per what I understand Sean and Floyd to believe, it would be that all speakers should be designed with reasonably constant directivity and axial responses free of significant resonances, and that all headphone responses should be "Harman-like" in general insofar as having a similar shape for ear gain, and either defaulting to or at least having the option to have something close to the Harman target's overall tilt.

Like, a lot of Sean's commentary is mostly about headphones that...like, most people can buy, which are usually active DSP designs these days. Those definitely should have a "Harman preset", I think we can all agree that's a good idea. For passive headphones and audiophile DAC-amp stuff, having a robust DSP and measurements that would allow making profiles to EQ a headphone to be close to Harman sounds like a good idea, and I know Sean sees value in ex. @oratory1990's database.

As far as how tightly grouped things need to be, at the very least when I've talked to Sean (both publicly and privately), he has seemed supportive of the idea of having options for different tastes, so long as they're like...real, identifiable-in-blind-listening-tests tastes, not just something somebody is asserting based on a sighted preference. I wouldn't really feel comfortable drawing boundaries for precisely what Sean or Floyd would consider "good enough", I guess that's something you'd have to ask them directly.

Re:

I would not say, and I don't think that Sean would say, that the publicly available Harman work has identified where ideal tone control adjustments would happen. Like, that just plain isn't something that was tested in the research, with the exception of allowing frequency adjustments on shelf filters. They may have done this internally, but in terms of what's published, we have results from a number of methodologies that all point in the same direction as far as what people prefer for the end result headphone frequency response, but not a lot of data to say which methodology is the best one for EQing.

It'd be a very interesting thing to do a paper on, in fact. Anyone got some spare budget?

This is perhaps the most interesting aspect of all this to me.

Even going back to how the curve was made. How many deviations from the curve were tested and how granular were the changes. In the context of this forum was the Harman curve tested against a curve that looked like the Caldera. This is not to make a point to support the Calderas just illustrating over of the thoughts I have regarding the matter in general.

Also brain EQ is very real. Did the studies look at what the effects were if you kept listening to an inferior curve and then went back to a better curve? Do we even really need a target curve or can our brain do with EQ that unless things are truly horribly of we can compensate.
 
Even going back to how the curve was made. How many deviations from the curve were tested and how granular were the changes.
In short, not very granular. I think this is my biggest contention with the popular conception of the Harman research: it is an extremely well evidenced general picture of what people prefer, but it is not hyper-specific regarding small features. The most specific adjustment I'm aware of is the change in the ear gain between 2015 and 2018, and, to my eternal sorrow, the methodology and reasoning behind that change isn't in any of the published work. Harman gives us an extremely solid general idea of the preference space for headphones (in the same way that ex. @Floyd Toole's work does for speakers), but I think people read it a bit...pedantically in terms of small differences sometimes.

In the context of this forum was the Harman curve tested against a curve that looked like the Caldera.
This said, given what we know from the comparisons in the Harman research - I'm thinking in particular about Listener Preferences for Different Headphone Target Response Curves, A Study of Listener Bass and Loudness Preferences over Loudspeakers and Headphones, and A Statistical Model that Predicts Listeners' Preference Ratings of Around-Ear and On-Ear Headphones - it is very probable that a large majority of listeners would prefer a Harman curve matching headphone to a headphone with the Caldera's unequalized frequency response. Acoustically flat bass is very seldom preferred when listeners have options, and I'm dubious that its treble would be preferred as well based on how Lorho's target, which is somewhat similar, compared to the "RR1_G" curve.

This doesn't mean that 100% of listeners would prefer the Harman headphone, of course, but in my opinion we can be pretty confident that the majority would.
Also brain EQ is very real. Did the studies look at what the effects were if you kept listening to an inferior curve and then went back to a better curve? Do we even really need a target curve or can our brain do with EQ that unless things are truly horribly of we can compensate.
In general, humans tend to be comparative in our evaluations - it's part of why studies with references (even hidden ones, as Sean favours) tend to produce more consistent results. People do seem to "adapt" to an extent to equipment with familiarity (you'll have to pardon my forgetting the title, but Sean has linked a paper before regarding subjective evaluations of speakers before and after long exposures), but people still tend to prefer "on average better" sound, even when it's very different from their references.
 
In short, not very granular. I think this is my biggest contention with the popular conception of the Harman research: it is an extremely well evidenced general picture of what people prefer, but it is not hyper-specific regarding small features. The most specific adjustment I'm aware of is the change in the ear gain between 2015 and 2018, and, to my eternal sorrow, the methodology and reasoning behind that change isn't in any of the published work. Harman gives us an extremely solid general idea of the preference space for headphones (in the same way that ex. @Floyd Toole's work does for speakers), but I think people read it a bit...pedantically in terms of small differences sometimes.


This said, given what we know from the comparisons in the Harman research - I'm thinking in particular about Listener Preferences for Different Headphone Target Response Curves, A Study of Listener Bass and Loudness Preferences over Loudspeakers and Headphones, and A Statistical Model that Predicts Listeners' Preference Ratings of Around-Ear and On-Ear Headphones - it is very probable that a large majority of listeners would prefer a Harman curve matching headphone to a headphone with the Caldera's unequalized frequency response. Acoustically flat bass is very seldom preferred when listeners have options, and I'm dubious that its treble would be preferred as well based on how Lorho's target, which is somewhat similar, compared to the "RR1_G" curve.

This doesn't mean that 100% of listeners would prefer the Harman headphone, of course, but in my opinion we can be pretty confident that the majority would.

In general, humans tend to be comparative in our evaluations - it's part of why studies with references (even hidden ones, as Sean favours) tend to produce more consistent results. People do seem to "adapt" to an extent to equipment with familiarity (you'll have to pardon my forgetting the title, but Sean has linked a paper before regarding subjective evaluations of speakers before and after long exposures), but people still tend to prefer "on average better" sound, even when it's very different from their references.

Thank you! That's all very good to know. I think people seeing and appreciating these studies and the details would help demystify things a lot. In the end they might still like X, Y, Z headphone not but at least they could appreciate the value and insights of the work.

And yeah I think it's fair to say that most listeners would probably enjoy Caldera "harmonized " vs not. at least in the basis of research. and some people may genuinely fall into the category of being in the few exceptions that might not. The obvious issue being many other biases driving people to that decision.

personally I need to do more side by side comparisons of Harmon tuned headphones to deviant headphones and where those deviations are and my preferences regarding the whole thing. I actually ended up ordering a pair of the DCA expanse so I can side by side compare them to some other headphones I own over a longer period of time. I got to try them initially and while they were good I actually really enjoyed the Sennheiser 800 s and the susvara more which are both pretty bass deficient headphones.
 
Back
Top Bottom