[Rob80b]

Yes but, I wonder how such a score can be adjusted to your personnel preferences ?

Maybe not initially but if this type of measuring becomes more common it would IMHO be akin to following certain reviewers who appear to have similar preferences, or not, to your own.

So the detailed frequency response of a device is probably still more interesting for a buyer t

That really depends on where one places their priorites. Today I believe most audio transducers do, more than less, get the frequencies correct....timbre on the other hand is what gives, in this case headphones, their overall character. "Timbre distinguishes different types of sound production, such as choir voices and musical instruments. It also enables listeners to distinguish different instruments in the same category"... for for me is of utmost importance and "frequency response" charts do not show us this at all, spectral analysis does a better job.

https://en.wikipedia.org/wiki/Timbre
https://www.jsr.org/hs/index.php/path/article/view/1292

 

[Beagle]

Maybe not initially but if this type of measuring becomes more common it would IMHO be akin to following certain reviewers who appear to have similar preferences, or not, to your own.


That really depends on where one places their priorrites. Today I believe most audio transducers do, more than less, get the frequencies correct....timbre on the other hand is what gives, in this case headphones, their overall character. "Timbre distinguishes different types of sound production, such as choir voices and musical instruments. It also enables listeners to distinguish different instruments in the same category"... for for me is of utmost importance and "frequency response" charts do not show us this at all, spectral analysis does a better job.

https://en.wikipedia.org/wiki/Timbre
https://www.jsr.org/hs/index.php/path/article/view/1292

I agree. Materials used in the construction are also very important because everything eventually sounds like what it is made of. You can hear this by just tapping on the cups/headband/frame with no music playing.

 

[cementnothing]

Super interested in this! I think the FR measurement is non-trivial in what it can tell someone, but I do think that there are such differences in timbre, materials used, shape of earcup, etc., that FR alone can't necessarily show exactly whether one will like it. I think the caveat is that a FR can be a pretty good indicator of whether one will like a headphone, but I don't think that's everything.

For example, I have the Meze 99 Classics, and they definitely tilt pretty dark. I figured I'd like them because I prefer a warmer, even leaning towards dark, sound signature, but there's something with the timbre that causes pause. I don't know what it is, but the airiness of it feels off in some way. Maybe it's because I'm used to open-backs and not closed-backs.

At any rate, it seems that FR can tell us whether it's probable we'll like something, but it's not the full story!

 

[Rob80b]

At any rate, it seems that FR can tell us whether it's probable we'll like something, but it's not the full story!

Case in point..the previously mentioned Sennheiser HD820....the FR definitely looked wonky and the recessed mid-range was not to many folks liking but upon direct on ears comparisons with a few contenders, the Focal Stellia, Meze Liric and the Dan Clarke Aeon, the Sennheiser ticked my boxes for timbre, distortion and immersiveness ....and while being a closed back the frequency response was fine in my books.
Just listening to some BBC 3 live opera and the main singer was strutting back and forth across the stage with the chorus was set in the background while in the foreground the conductors score was easily heard when the pages were turned.
Still that doesn't make it a great headphone, but for me the timbre, distortion (or lack of) and immersiveness is high up there, one just needs to know what they like and if the "innovative MDAQS algorithm" can supply more information than just how prominent, recessed or exaggerated certain frequencies are, I believe it could be a valid and innovative tool to pursue.

 

[cementnothing]

Case in point..the previously mentioned Sennheiser HD820....the FR definitely looked wonky and the recessed mid-range was not to many folks liking but upon direct on ears comparisons with a few contenders, the Focal Stellia, Meze Liric and the Dan Clarke Aeon, the Sennheiser ticked my boxes for timbre, distortion and immersiveness ....and while being a closed back the frequency response was fine in my books.
Just listening to some BBC 3 live opera and the main singer was strutting back and forth across the stage with the chorus was set in the background while in the foreground the conductors score was easily heard when the pages were turned.
Still that doesn't make it a great headphone, but for me the timbre, distortion (or lack of) and immersiveness is high up there, one just needs to know what they like and if the "innovative MDAQS algorithm" can supply more information than just how prominent, recessed or exaggerated certain frequencies are, I believe it could be a valid and innovative tool to pursue.

That's super interesting to know. I think another benefit of not thinking that only FR matters is that you can miss out on HP's that you might actually really enjoy despite the FR. As you said, the FR of the HD820 looked wonky, but you genuinely enjoy it. If someone were to only look at FR and decide based on that, they might miss out on a surprising listen.

 

[Beagle]

Case in point..the previously mentioned Sennheiser HD820....the FR definitely looked wonky and the recessed mid-range was not to many folks liking but upon direct on ears comparisons with a few contenders, the Focal Stellia, Meze Liric and the Dan Clarke Aeon, the Sennheiser ticked my boxes for timbre, distortion and immersiveness ....and while being a closed back the frequency response was fine in my books.
Just listening to some BBC 3 live opera and the main singer was strutting back and forth across the stage with the chorus was set in the background while in the foreground the conductors score was easily heard when the pages were turned.
Still that doesn't make it a great headphone, but for me the timbre, distortion (or lack of) and immersiveness is high up there, one just needs to know what they like and if the "innovative MDAQS algorithm" can supply more information than just how prominent, recessed or exaggerated certain frequencies are, I believe it could be a valid and innovative tool to pursue.

I’ve also had experiences where similarly skewed FR responses (on two or more headphones) sounded different to me.

One thing that is consistent for me is a dip in upper mid/lower treble is going to be irritating for me. I have found this in pretty much every LCD and Focal (except the Spirits) headphone I have owned.

 

[castleofargh]

FR is accepted as a main variable for preferences. It has been established to be by some of the people doing all the research/books I read(and is also mentioned in the videos of this thread), while I have never seen evidence that it wasn't.
On the other hand, I don't think anybody ever claimed FR graphs to be everything. So I'm always confused when I read people reacting as if it is a common position that needs debunking.

IMO, FR graphs mostly get bad reputation from people who overestimate what you can deduce from them, and overestimate their own ability to do so. Without psychoacoustic knowledge or at least some serious personal experience in fooling around with EQ, headphones, and some basic knowledge about measurements (in general, and how those particular graphs were made), people are bound to make interpretation mistakes. That's not the measurement's fault.
I've been surprised a number of times by devices I didn't expect to really enjoy based on a FR graph. But pretty much every time I thought something was going to sound horrible based on FR, it did just that for me. Speakers, headphones, IEMs, when the FR screams failure, it's usually a waste of time to go audition them. That's my experience.

Anyway, better analysis models are always great news, and if they satisfy a big percentage of the population with a result that's harder to misinterpret, hooray!!



For the HEAD acoustic guys.
In the second or third video, we see results for, I think 3 products with... was it speakers, a pair of headphones, and maybe something in a car? Maybe I'm mixing things up in my head, I watched quite a few sped up videos in one sitting and somehow ended up needing Till Papenfus to tell me all about hammers and acoustic analysis. It was just one example anyway, but I'm curious. Were there actual trials with listeners having to A/B and rate headphones against speakers(or IEM vs car stereo, or...), and if so, what were the trends for global ratings? Or did the trials remain between same playback systems types, and it's your own SkynetGPT that decided it could use the rating basis for all sound systems?

 

[o0genesis0o]

IMO, FR graphs mostly get bad reputation from people who overestimate what you can deduce from them, and overestimate their own ability to do so. Without psychoacoustic knowledge or at least some serious personal experience in fooling around with EQ, headphones, and some basic knowledge about measurements (in general, and how those particular graphs were made), people are bound to make interpretation mistakes. That's not the measurement's fault.
I've been surprised a number of times by devices I didn't expect to really enjoy based on a FR graph. But pretty much every time I thought something was going to sound horrible based on FR, it did just that for me. Speakers, headphones, IEMs, when the FR screams failure, it's usually a waste of time to go audition them. That's my experience.

Couldn't agree more with these points.

 

[Rob80b]

FR is accepted as a main variable for preferences. It has been established to be by some of the people doing all the research/books I read(and is also mentioned in the videos of this thread), while I have never seen evidence that it wasn't.

Don't believe anyone would question the validity of FR.....I've relied on them or at least made reference to them for over 50 years with regards to audio playback and recording equipment, although back then we had to use oscilloscopes.
What has been presented here if I'm understanding correctly is the possibility to define other attributes such as timbre, distortion and immersiveness... giving us a more holistic interpretation of the headphones ability in the reproduction of an audio recording.

 

[Redcarmoose]

This is what I get out of the talk. IMHO
The MDAQS algorithm is attempting a more comprehensive way of understanding headphone measurements:
Outside of FR graphs there are 3 dimensional components MDAQS algorithm concerns itself with.........

1. Immersive Quality
2. Timbre X5 perimeter aspects
3. Distortion

Each of the above is interrelated to FR and inescapable to FR still none-the-less. As such a suggested target FR was developed in relation to the 6 music genre examples through human listening. Thus higher tones have a (more dramatic) soundstage/immersive affecting factor than lower tones, and are related also to FR. Timbre has maybe a realism factor which I’m surprised was not a term used in this talk? Maybe the word realism is not needed?

Cross correlation bands showing a graph, as one graphical example of MDAQS technology.


MDAQS
Measuring beyond the eardrum. Beyond the eardrum physical FR measurement, to a perceived measurement. Thus the binaural nerve information processed independently inside the brain, regardless of sound waves hitting the eardrums. Thus only the brain establishes what level immersive quality is. Perception of Timbre is only subjectively. (From both learning and memory?) So they need to model the timbre by using the 5 perimeter aspects. Using AI instead of the standard testing methods, to discover the nerve-networks, or to at least get an idea as to the subjective response.

Distortion:
Quite possibly the least important of the three listed. In headphones it’s basically under control, thus more relevant to this system of measurements other than headphone music reproduction.

Here two different FR reponses of two IEMs still rate equally (in their MDAQS) rating system. As such showing that things like timbre can result in equally beneficial results. Yet timbre is impossible to measure, but in a series of computations with this MDAQS, can still be measured with a perceptional based methodology.

I have often wondered how one area in an IEM response may lead way as a departure from (standards of accepted personal FR) curves gaining profound realism, and bypassing part of what is considered a great FR (whatever that means). That through contemplation of timbral accuracy and soundstage/immersion we go beyond simple FR graphs..........into which may be perceived as greater accessibility and music interaction, beyond just the measured FR.

 

[pail]

Timbre, distortion and immersiveness ... not just frequency response graphs...

Immersiveness is subjective (as long as we don't talk about an Immersive electronic system and we measure data in it), Timbre and Distortion we can say are less subjective concepts or not at all and purely objective.

 

[DarginMahkum]

I wonder if the frequency response haters really listened to the video: He says "if you have a non-ideal frequency response no matter what the target is, you will fail, because human perception will tell you it is lousy". So Frequency response is not everything, but it is the ground work. FR can reflect a lot of things, but of course it is not a model of perception, it is how a device under test behaves under a particular scenario. Plus FR we see is smoothed out, an includes the measurement system character.

 

[Rob80b]

Immersiveness is subjective (as long as we don't talk about an Immersive electronic system and we measure data in it), Timbre and Distortion we can say are less subjective concepts or not at all and purely objective.

At the end of the day it is all subjective....real measured distortion is actually preferred in some circles as they say it adds character and a measure of warmth.
Immersiveness could be construed as the ability to grab ones emotions and get up and dance. I am thinking more in terms of recreating to the illusion of depth, layering, width, height etc...some headphones having the ability to make you imagine you are in the audience, while another puts you in the recording booth or on stage...taking you to the venue or bringing the performers to you. It's what this hobby is all about.
Obviouly the phones that exhibit less distortion and recreate the delicate overtones and harmonics of the fundamentals will more clearly define timbre.... and immersiveness may be solely reliant on baffle and cup design and how the the audio vibrations interact with the individuals ear... and that is totally subjective.
But...to maybe, possibly have a measuring system that can more clearly define those attributes...that would be interesting.

 

[Beagle]

I think I understand the potential need for measurements but at the same it is kind of sad (I’ll resist using “pathetic”) that the magic and emotion we felt when we were kids listening to a transistor radio or Walkman has transformed into this type of overthinking/over analyzing. And I’ve become immersed in it, like someone who semi-voluntarily left the wonderful atmosphere of the beach and dived headfirst into a cesspool. I lost my sense of original purpose.

Like instead of enjoying grandma’s baking/cooking, we are trying to decipher her recipes to determine what it tastes like. It’s literally the same thing.

They say ignorance is bliss. I’m slowly beginning to realize that they were probably right.

 

[DarginMahkum]

I think I understand the potential need for measurements but at the same it is kind of sad (I’ll resist using “pathetic”) that the magic and emotion we felt when we were kids listening to a transistor radio or Walkman has transformed into this type of overthinking/over analyzing. And I’ve become immersed in it, like someone who semi-voluntarily left the wonderful atmosphere of the beach and dived headfirst into a cesspool. I lost my sense of original purpose.

Like instead of enjoying grandma’s baking/cooking, we are trying to decipher her recipes to determine what it tastes like. It’s literally the same thing.

They say ignorance is bliss. I’m slowly beginning to realize that they were probably right.

The more a hobby gets expensive, the more people will become concerned about what they are buying and try to be informed about the background. The amount of BS that is constantly being recreated and repackaged and resold with ever increasing prices is for me a concerning issue, as it diverts our attention from enjoying the music to a wild consumerism that never really satisfies us. We are often not satisfied with what we are listening to not because we over-analyse what we have, but because we are often pushed to follow a never ending chain of trades that doesn't leave any room for proper analysis to do informed shopping.

When I was a kid, the elders at home had something to listen to music but they never needed to analyse things, as it was not a hobby for them in which they were constantly buying and selling equipment every two months. I didn't analyse what my grandmother cooked as eating her food was not a hobby for me, nor she was trying to sell me her cookies with ever increasing prices with a new name and packaging every time I see her.

At the end of the day it is all subjective....real measured distortion is actually preferred in some circles as they say it adds character and a measure of warmth.
Immersiveness could be construed as the ability to grab ones emotions and get up and dance. I am thinking more in terms of recreating to the illusion of depth, layering, width, height etc...some headphones having the ability to make you imagine you are in the audience, while another puts you in the recording booth or on stage...taking you to the venue or bringing the performers to you. It's what this hobby is all about.
Obviouly the phones that exhibit less distortion and recreate the delicate overtones and harmonics of the fundamentals will more clearly define timbre.... and immersiveness may be solely reliant on baffle and cup design and how the the audio vibrations interact with the individuals ear... and that is totally subjective.
But...to maybe, possibly have a measuring system that can more clearly define those attributes...that would be interesting.

A live recording in a hall will sound different from a studio recording done in an anechoic chamber. So what is the right reproduction for these two extreme cases? What about if some 3D effect plug-in was applied during production? What about a jazz club without any acoustic treatment? Can we or shall we even evaluate the immersiveness for a transducer, a feature which should be mostly inherent in the recording? If yes, to what degree is it a component of reproduction quality? All very subjective and very much recording, music and person dependent. As you said, the "immersiveness" not coming from the recording should be the interaction of the sound waves with the environment (cups or the room) and the ears. There are not so many parameters for the audio waves: Amplitude, frequency and phase / timing. Rest is up to the interpretation of our brain - which is very adaptive. So, joining all this information to create a single note like "3.9", doesn't make sense for me, or for anyone that has specific expectations for an informed listening.

Also, let's not forget the definition: Frequency response of a system (HP etc) gives us the frequency domain information about how it transforms the input signal. Obviously something that defining cannot be ignored or taken as unimportant. I would encourage anyone that sees what we are doing here in these forums as a hobby, to at least be aware of it and how it relates subjectively to his hearing for the music he likes. Otherwise it is not a hobby, but a mere shopping therapy.