ZMF Caldera Headphone Review

[Robbo99999]

Science needs to applied and paid attention to, but the practice of a scientific study in application across all platforms cannot be adhered to as a gospel as there are many cases where users do not prefer harman tuned headphones. It's the same reason why a group gathers here who likes it, and groups gather on other internet forums of those who do not.

Finite statements like this are dangerous, because it forwards a group think aspect without the applied practice and listening. I understand there is great research done for harman, but it's a starting point not an end-point. We are all individuals and not a group and see the world from one lens, so closing yourself off to a very general mean of FR will very likely not lead you to your own audio preferences and favorite sound.

I would urge anyone regardless of what side you stand on, like anything in life, to stay open minded. Go on a journey to find and understand what sound you like, and why or why it doesn't adhere to harman. Which yes, would mean listening to the Caldera in person utilizing the tuning system that is innate to ZMF's instead of just making a statement here to follow your peers.

The Harman Research is the best research out there in terms of a valid target for headphones. In this thread you seem to be advertising the Caldera as the answer to personalised HRTF with the various pads, but I fail to see how a "random frequency response" combined with random different pads will get people closer to their HRTF - it all seems like a stab in the dark for both user & the designer.

 

[jhwalker]

@amirm - If you are going to be continuing to ruin this forum responsibly, you really need to get the rig that most/many us manufacturers are using which is the 5128 from B/K. It will accurately give you results above 8K and more properly show the rest. I know the 711 coupler is great since it will reflect the Harman research, but the industry is moving past that as the 711 coupler is dated at this point. Maybe you can develop some findings with your own research more based on what your personal HRTF is as we all have large variances in certain areas.

Further - and as any ZMF owner will tell you, there's 6 sets of different pads for the Caldera, and a titanium mesh for the front that helps tune the headphone to your preferences/HRTF and without testing all of that the way an actual owner is and finding what works best for your own personal HRTF BEFORE doing EQ you are also doing a dis-service to anyone who reads this by stating "facts" without utilizing the headphone in it's designed sonic modular fashion. For example there's three different depths of Caldera pads, all which adjust the distance of the driver to the ear to account for different anatomies, as well as a fully perforated pad that more follows harman.

Please be more responsible and use the headphone the way an owner would, so you can provide accurate information since you have so many followers.

-----------

@solderdude and everyone else who has commented on the tuning of ZMF's. I work with, and know most manufacturers, and have talked at length with all of them about the Harman curve and ofcourse read all the research and as you know designed and tuned many headphones. And for anyone who does this laborious practice of spending years at a time tuning a single headphone, you try many things. I use and own flat panel, DIY, 711 coupled Larson Davis AEC206, and two 5128's from B/K. I measure every headphone and driver I design in various stages and always tune much closer to harman just to try what each headphone will sound like, as I do want to test and utilize the great research that Sean Olive has done because I am not a denier of it and find it useful.

But what Sean Olive will tell you, as any other responsible headphone MFR that actually makes headphones, that the harman curve was meant as a starting point not as a defined endpoint that all headphones need to meet to be enjoyable.

Each system, driver/cups etc has many innate qualities of airflow that are measured in other ways, some that are worth exploring via measurements and some that are experienced more subjectively that are the culmination of many measurements like THD, FR, impulse, CSD, bursts etc. Further, open vs closed full sized headphones vary quite differently in how we each experience them when tuned the same. I know this because when I'm working on the closed version (and vice-versa) of each of our open headphones I tune them to match, and also tune them to harman and then EQ them to harman, and every single time the closed version or open version tuned like the other sound way different. There's just so much more that goes into how we experience sound aside from FR.

I'm pretty sure there's many of you on this forum who enjoy a headphone that isn't tuned exactly to @amirm 's cult of Harman tuning, but are afraid to speak up because of the insane groupthink that happens here. I'm here to tell you that it's OK, because all of us on the side of things that are actually making headphones, know that there are many variations, including liking the Harman curve that can be enjoyable for many different reasons. I implore you to like what you like and stand up for yourself, because there's a reason you like what you like, and you're not alone!!!!

Many people seem to think our headphones are just me sitting in a basement tuning the headphones and making sure they look some special woody way, this is just not true. I do tons of acoustic research and testing on each headphone and they are tuned in very specific ways based on the driver and platform (open/closed/semi-open) that they exist within. I follow advice, thoughts and measurements from many sources before they are released and you'll find that none-of our headphones have .1 // .2 // .3 versions after they are released. If you don't like them because they are not tuned to harman that is totally awesome and cool! But I would implore more of you to actually listen to the Caldera before posting here, and do more work to find what your actual HRTF is because that's what my headphone designing is all about, making a platform that is semi-modular within the pads and damping mesh where a gamut of owners can end up with something they really like.

BONUS COMMENT: If you want a ZMF that is tuned to closer harman, the Atrium Closed with suede pads is quite close. I don't particularly love the way it sounds due to my HRTF, I like the lambskin version that has quite a bit more mid-bass and is below harman in the 4-6 khz area.

BONUS II: If you want a Caldera that conforms closest to harman, use the ultra perf pads, gently EQ to the curve in the area (less has to be done) and then EQ the bass up as the ultra perf will cause a roll-off below 70 HZ. I know about 2 out of hundreds of owners who prefer this, but if that's what you're into, it's the easiest way to get there.

While I appreciate your participation, the caustic tone is a bit over the top. "Cult". "Insane groupthink". "Be more responsible".

If you want more people to respect you and your products, I'd respectfully suggest you show more respect in return.

 

[zach915m]

Yes, I agree, but science is refuted with more science, not with less of it.

Which is why if you want to make finite statements here, you should buy a measurement rig that adheres to the common standards as I have done, and find what works for your preferences before making statements based on no actual experience about a headphone you have never heard.

Also with @amirm calling things "errors" vs "deviation" it furthers the ideas that there is a "right" and a "wrong" vs a preference as a guideline used for us all to understand our own audio journey better. I get that it leads him gathering the type of followers he is after (edit: it promotes controversy instead of learning, just look at the eyes/viewers this site gains because of it and across the web forums), but it doesn't further our hobby, which most of us are here to enjoy (and do!).

The aim of science is to delineate areas, fostering a collective understanding and advancing shared knowledge through impartial language and objectivity. It involves consistently employing variables, such as products or headphones, in a standardized manner (as seen in amir's reviews; note that practices vary). It encourages audiences to embark on their individual journeys, recognizing the diverse range of variables at play, as each person has unique Head-Related Transfer Function (HRTF) preferences.

 

[cheapmessiah]

I get that it leads him gathering the type of followers he is after

I dont share that statement, as I dont think there is a ploy for some sort of ultimate malicious goal, but I agree with the rest of the post.

 

[solderdude]

The aim of science is to delineate areas, fostering a collective understanding and advancing shared knowledge through impartial language and objectivity.

Isn't that what the Harman research was all about ?
They ended up with an averaged target that seemed to satisfy the majority of people.
Of course there is a minority that is more likely to prefer the 'optimum hifi' or Crinacle, or another target and don't want the typical bass boost... however impressive that boost may sound.

but it doesn't further our hobby

Furthering the hobby isn't done with just reporting but with experimentation. That's not what ASR is about. You need DIY forums for that. Sometimes some of it is discussed here too but checking technical performance is the main goal.

 

[zach915m]

Isn't that what the Harman research was all about ?

My post is about the use of harman research by this website, not about the research itself.

 

[Soria Moria]

Also with @amirm calling things "errors" vs "deviation" it furthers the ideas that there is a "right" and a "wrong"

I think most people would agree that these are errors (and keep in mind it's $3 500). But it seems to be mostly down to semantics.

 

[solderdude]

My post is about the use of harman research by this website, not about the research itself.

Amir uses the Harman 'standard' because it is the only properly researched one
For speakers the Klippel is the standard.
For audio gear he also uses standards so gear performance can be compared.

It would not be wise to use less defined standards instead. ASR members usually agree with this approach.

 

[zach915m]

Amir uses the Harman 'standard' because it is the only properly researched one
For speakers the Klippel is the standard.
For audio gear he also uses standards so gear performance can be compared.

It would not be wise to use less defined standards instead. ASR members usually agree with this approach.

Its not about amir using the Harman standard, it's about his practice in doing so. The best researchers fully understand the variable they are using and are able to break down the science they use, within the product, in this case a headphone, so that the end user can understand the "practice" of using said product/headphone.

There's a reason ASR is so controversial, and it's not because the info presented is blanketly true or false. As with most of life the answers are in between. I'll leave you all to further discuss your own truths.

 

[Mad_Economist]

@amirm - If you are going to be continuing to ruin this forum responsibly, you really need to get the rig that most/many us manufacturers are using which is the 5128 from B/K. It will accurately give you results above 8K and more properly show the rest. I know the 711 coupler is great since it will reflect the Harman research, but the industry is moving past that as the 711 coupler is dated at this point. Maybe you can develop some findings with your own research more based on what your personal HRTF is as we all have large variances in certain areas.

I really profoundly don't agree with this, speaking as a strong advocate of the 5128/ITU-T P57 4.3 standard. It has a lot of merits, including:

• Wider band accurate impedance
• Accurate impedance at low frequencies relative to the 60318-4
• Accurate canal geometry (for if you really want a truly deep insertion, I suppose)

But the 60318-4 and 60318-7 (and corresponding ITU-T P57 3.3) ear simulators and HATS are absolutely usable. Indeed, more than usable, there are cases where the 5128 has been demonstrated to have worse correlation with in situ measurements of real human ears (admittedly, this is typically either leakage response stuff or pinna flexibility, but while controllable, these are still real merits of the KB500x).

Sean has done the lion's share of his work on 60318-4/711 couplers, but more than that, an even larger majority of all work on headphone acoustics has used these couplers, and it has consistently yielded results that correlate strongly with human experience. This does not mean that it is perfect and cannot be improved (I've said here and elsewhere how I see the 5128 as better), but it absolutely does mean that it isn't necessary for Amir (or any reviewer) to have a 5128 in order to make reviews of products with measurements. There are cases where differences between 5128 and other HATS/ear sim measurements may favour the 5128, it's true - but "there is some chance that a small difference, likely at very low or very high frequencies, may be more realistically reflected on this ear sim" is not a valid disqualifier for other ear sims.

Each system, driver/cups etc has many innate qualities of airflow that are measured in other ways, some that are worth exploring via measurements and some that are experienced more subjectively that are the culmination of many measurements like THD, FR, impulse, CSD, bursts etc.

I had a presentation at Canjam SoCal about this, which hopefully at some point I'll be turning into a video or web presentation, but it should be noted that THD has a pretty poor correlation with our perceptions of headphone differences, and the "time" domain things (burst respond, CSD, impulse) are only providing new information when the headphones cease to behave as minimum phase devices. Which is pretty damn rare. More often than not, these measurements (particularly views like CSD and time views of bursts) mislead the viewer into conflating perfectly normal min phase behavior which is visible in frequency response with "time domain problems".

Further, open vs closed full sized headphones vary quite differently in how we each experience them when tuned the same. I know this because when I'm working on the closed version (and vice-versa) of each of our open headphones I tune them to match, and also tune them to harman and then EQ them to harman, and every single time the closed version or open version tuned like the other sound way different. There's just so much more that goes into how we experience sound aside from FR.

I'm not sure if you've seen Axel Grell's presentation regarding acoustic impedance, but he has a theory that this is the main factor which influences different perception when the mannequin FR is the same. He's apparently got a German university working on a study on it, so hopefully we may see some data in the next couple years. I will say, my personal two bits is that, since acoustic source impedance changes transfer level with different load impedance, it will likely all lead back to the FR at the eardrum, but that's conjecture.

--

If people don't mind me briefly getting on a soapbox, there are some interesting technical and philosophical questions in play here: How much interpersonal variation in preference should manufacturers account for; how do different measurement systems differ and which best reflects human experience; whether targeting the 1stdev population segment should be the goal for 100% of headphones; etc. It would be a shame to lose the opportunity to discuss them in acrimony, in my opinion.

 

[cheapmessiah]

I cant help but think that the controversy of ASR is that is not a relay station for marketing, and that besides manufacturers and people dealing in esoterics, most people find the general guidelines quite agreeable, much like the harman target, go figure.

 

[solderdude]

Its not about amir using the Harman standard, it's about his practice in doing so. The best researchers fully understand the variable they are using and are able to break down the science they use, within the product, in this case a headphone, so that the end user can understand the "practice" of using said product/headphone.

Isn't that what Amir is doing ? I mean... he uses the same fixture. Uses the same target (based on research), reports the results and ways to get the tonal balance up to said standard so that the end user can 'correct' that headphone towards that standard.

The fact that not everyone agrees about that standard and that many, many people just use a headphone 'as is' is besides the point. No one is forcing anyone to only like the Harman target. People get informed and ASR members vent their opinions.

There might be a few people that heard this headphone. ZMF headphones have been measured and discussed by others and on other fixtures. People can go there. They come here for ASR measurements.

Aside from just a few headphones out there most of them do not adhere anywhere close to the Harman target yet many people love those headphones to bits. This one just happens to be one of many that don't follow the target. Intentionally even as it seems. That doesn't make it a bad headphone. Most objections appear to be the price/Harman target ratio ... if something like that exists.

Most people buy headphones because they like the looks or price, read reviews/opinions, look at what trendsetters are wearing or like the brand.
Some people buy headphones for how the sound and others like to modify, use EQ or tone controls.
There is no 'best way' to do this but you can EQ towards a standard in which case ASR is one of the websites that can help achieve that.

 

[zach915m]

I really profoundly don't agree with this, speaking as a strong advocate of the 5128/ITU-T P57 4.3 standard. It has a lot of merits, including:

• Wider band accurate impedance
• Accurate impedance at low frequencies relative to the 60318-4
• Accurate canal geometry (for if you really want a truly deep insertion, I suppose)

But the 60318-4 and 60318-7 (and corresponding ITU-T P57 3.3) ear simulators and HATS are absolutely usable. Indeed, more than usable, there are cases where the 5128 has been demonstrated to have worse correlation with in situ measurements of real human ears (admittedly, this is typically either leakage response stuff or pinna flexibility, but while controllable, these are still real merits of the KB500x).

Sean has done the lion's share of his work on 60318-4/711 couplers, but more than that, an even larger majority of all work on headphone acoustics has used these couplers, and it has consistently yielded results that correlate strongly with human experience. This does not mean that it is perfect and cannot be improved (I've said here and elsewhere how I see the 5128 as better), but it absolutely does mean that it isn't necessary for Amir (or any reviewer) to have a 5128 in order to make reviews of products with measurements. There are cases where differences between 5128 and other HATS/ear sim measurements may favour the 5128, it's true - but "there is some chance that a small difference, likely at very low or very high frequencies, may be more realistically reflected on this ear sim" is not a valid disqualifier for other ear sims.


I had a presentation at Canjam SoCal about this, which hopefully at some point I'll be turning into a video or web presentation, but it should be noted that THD has a pretty poor correlation with our perceptions of headphone differences, and the "time" domain things (burst respond, CSD, impulse) are only providing new information when the headphones cease to behave as minimum phase devices. Which is pretty damn rare. More often than not, these measurements (particularly views like CSD and time views of bursts) mislead the viewer into conflating perfectly normal min phase behavior which is visible in frequency response with "time domain problems".


I'm not sure if you've seen Axel Grell's presentation regarding acoustic impedance, but he has a theory that this is the main factor which influences different perception when the mannequin FR is the same. He's apparently got a German university working on a study on it, so hopefully we may see some data in the next couple years. I will say, my personal two bits is that, since acoustic source impedance changes transfer level with different load impedance, it will likely all lead back to the FR at the eardrum, but that's conjecture.

--

If people don't mind me briefly getting on a soapbox, there are some interesting technical and philosophical questions in play here: How much interpersonal variation in preference should manufacturers account for; how do different measurement systems differ and which best reflects human experience; whether targeting the 1stdev population segment should be the goal for 100% of headphones; etc. It would be a shame to lose the opportunity to discuss them in acrimony, in my opinion.

I agree, and I think maybe I'm being taken out of context. Both rigs should be used to get the best picture of a headphone along with other factors. All these topics are pertinent and ofcourse the true answers are hard to define.

 

[kemmler3D]

I'm pretty sure there's many of you on this forum who enjoy a headphone that isn't tuned exactly to @amirm 's cult of Harman tuning, but are afraid to speak up because of the insane groupthink that happens here. I'm here to tell you that it's OK, because all of us on the side of things that are actually making headphones, know that there are many variations, including liking the Harman curve that can be enjoyable for many different reasons. I implore you to like what you like and stand up for yourself, because there's a reason you like what you like, and you're not alone!!!!

Thanks for coming by and the informative post Zach! I've spent a good bit of time here and what I would say about this - the cultishness and groupthink is not as intense as you might expect. In general the most common advice in headphone threads (outside of review threads) is to go and listen for yourself because no two HRTFs are quite the same.

Speaking for myself, but also what I've seen from a lot of members - the data is a very useful guide and that's why most of us are here. Amir's opinions and preferences are his own, and most ASR users don't forget to have some of their own. Harman being a starting point or general guideline is generally understood around here from what I've seen.

 

[cheapmessiah]

If people don't mind me briefly getting on a soapbox, there are some interesting technical and philosophical questions in play here: How much interpersonal variation in preference should manufacturers account for; how do different measurement systems differ and which best reflects human experience; whether targeting the 1stdev population segment should be the goal for 100% of headphones; etc. It would be a shame to lose the opportunity to discuss them in acrimony, in my opinion.

There is one huge advantage to a singular standardized tunning, it would open the gates to the possibility of widespread use and understanding of PEQ, be it analog or digital, as a tool to dial preferences. It would also mean that the "kwnoledge" of a tunning becomes instantaniously available for everyone in the world the moment it is put on the internet. So the question of what tunning to target becomes less trivial the moment proper and convenient customization becomes available with it.

 

[kemmler3D]

There is one huge advantage to a singular standardized tunning, it would open the gates to the possibility of widespread use and understanding of PEQ, be it analog or digital, as a tool to dial preferences. It would also mean that the "kwnoledge" of a tunning becomes instantaniously available for everyone in the world the moment it is put on the internet. So the question of what tunning to target becomes less trivial the moment proper and convenient customization becomes available with it.

I think there's a lot to be said for this idea, but in a world where PEQ is only available on a minority of playback devices aside from computers, there's still a need for a wide variety of tunings.

 

[cheapmessiah]

I think there's a lot to be said for this idea, but in a world where PEQ is only available on a minority of playback devices aside from computers, there's still a need for a wide variety of tunings.

Indeed, I hope it wont be long until chinese manufacturers start implementing 10band PEQ on DACs like RME, but priced for the masses. I mean, at some point they have to stop chasing an extra DB of SINAD per year dont they? XD

/dreams

 

[Mad_Economist]

I agree, and I think maybe I'm being taken out of context. Both rigs should be used to get the best picture of a headphone along with other factors. All these topics are pertinent and ofcourse the true answers are hard to define.

Perhaps I'm misreading you, what I was responding to was specifically the sentiment I perceived that 60318-4-only reviews were unsuitable. Like, ideally, yes, both. But an absence of 5128 does not make a review insufficient.

 

[SuicideSquid]

You could easily extend that to all luxury products that are floating around on the planet.
Is a $ 100k watch really worth the money when the goal is to keep time ?
Yet... plenty are sold to people that like to own it even though a $ 10.- watch may be more accurate when it comes to its primary task (telling time).

If headphones are sold that cost over $ 1k then that obviously is worth it to the one buying it. If not they will return it or simply buy another one.
That's the beauty of the market.
That does not mean a $ 10.- IEM can not sound more than good enough to someone else.
That person is correct but so are people spending thousands on it ... when they feel it is worth it.

It also does not mean you need to spend more than a few hundred to get top notch sound either.
It just means that some products are worth more to some people than to others.
Different goals.

Two things:

1. Spending $100,000 on a watch is done purely to be ostentatious and wasteful. It's shameful behaviour and should be criticized, in the same way spending tens of thousands of dollars on high-end audio "tweaks" is purely an ostentatious display of wasted wealth, is shameful, and ought to be criticized.

2. No one who spends $100,000 on a watch pretends it keeps better time than a $20 Casio. Plenty of people drop $3,500 on headphones thinking it will get them better performance than a $350 set of headphones. Those people are mistaken, and people are right to call out those errors in the comments.

So yeah, there's no good reason to spend $3,500 on a product that doesn't give you better performance, aesthetics, or comfort compared to a product 1/10th the price, and to the statement, "well it's just an ostentatious display of wealth like other luxury products", I say "yes, and that's gross".

 

[usern]

@amirm - If you are going to be continuing to ruin this forum responsibly, you really need to get the rig that most/many us manufacturers are using which is the 5128 from B/K.

Where is easily accessible and searchable database of B/K measurements for frequency response and distortion?