Headphones and the Harman target curve

[markanini]

And the theory of gravity is just a theory...

 

[Yorkshire Mouth]

And the theory of gravity is just a theory...

There is absolutely no comparison whatsoever.

 

[raistlin65]

I think we can put to bed the ‘science’ bit.

Preference isn’t science.

Measurements are science.

You can measure preference scientifically.

That doesn’t make the preferences scientific.

Is headphone sound signature preference nature? Or nurture? Or both?

If some component of it is nature, that component most decidedly is scientific.

If it is nurture, but the brain goes through some semi- or fully permanent changes that result in that preference, I suspect a neurologist would argue that preference effect is scientific as well.

 

[Yorkshire Mouth]

Is headphone sound signature preference nature? Or nurture? Or both?

If some component of it is nature, that component most decidedly is scientific.

If it is nurture, but the brain goes through some semi- or fully permanent changes that result in that preference, I suspect a neurologist would argue that preference effect is scientific as well.

Your question isn’t specific. It’s very unclear. What does “Is headphone sound signature preference nature? Or nurture? Or both?” mean?

Preference is preference. Some people prefer more bass, or brighter. That’s up to them, and it’s not scientific.

Measuring whether a headphone’s sound is close to what you’d hear on GSIAGR is science. Measuring whether people prefer that isn’t, no matter how accurately and scientifically your measurement of their preference is.

 

[Pretorious]

But it is my understanding that Harman's research was conducted cross-culturally and that the preferences remained the same. That most certainly is scientific. Some deviation along a baseline that accounts for over 60% of people's preference still would be science.

I think Amir demonstrates this logically in his headphone measurements, where he has mentioned before that averages will only provide one with an average of that model, but tells you nothing on how your individual headphone sounds, nor how your ears and HRTF affect the frequency. It's all still measurements; it's all still science.

 

[raistlin65]

Your question isn’t specific. It’s very unclear. What does “Is headphone sound signature preference nature? Or nurture? Or both?” mean?

Actually, I'm fairly certain that there's enough shared common understanding of what nature and nurture means among most ASR members that additional specificity is unnecessary.

So what you seem to be describing is an individual need for additional specificity.

 

[Yorkshire Mouth]

But it is my understanding that Harman's research was conducted cross-culturally and that the preferences remained the same. That most certainly is scientific. Some deviation along a baseline that accounts for over 60% of people's preference still would be science.

I think Amir demonstrates this logically in his headphone measurements, where he has mentioned before that averages will only provide one with an average of that model, but tells you nothing on how your individual headphone sounds, nor how your ears and HRTF affect the frequency. It's all still measurements; it's all still science.

You’re mixing up different things, and that’s the very reason it’s a problem.

To be clear, Amir pass/fails (recommends or not) headphones based on how closely they follow an average preference.

Let’s switch that to...let’s say distortion in DACs, or even frequency response in DACs.

If people were ‘scientifically measured and averaged’ to prefer DACs with a bit of distortion, should Amir recommend them, and not recommend the transparent?

If people prefer 4dB more bass (more or less what we’re discussing here), would Amir be correct in recommending DACs with a tonal imbalance?

What people prefer is surely not what this site is about. Even if you measure that preference scientifically.

 

[Yorkshire Mouth]

Actually, I'm fairly certain that there's enough shared common understanding of what nature and nurture means among most ASR members that additional specificity is unnecessary.

So what you seem to be describing is an individual need for additional specificity.

Nature and nurture are nothing to do with what we’re discussing.

The goal MUST be audio transparency. NOT what people prefer, be that through nature or nurture.

Back to basics. In as far as possible, in as far as it can be measured, the goal needs to be us hearing what the person mixing the track heard. You take that digital master. You play it through transparent electronics. You produce a sound through the headphones which captures what that mixer heard through their speakers, in as far as that’s possible.

The Harman curve, by definition, says we’ll play music back to people and ask them if they prefer it with a bit more bass or treble, then create a curve based on the average of that preference.

That’s wrong. I mean, don’t get me wrong, if you want to EQ it it to your preference, that’s down to you. But goal number 1 is transparency, whether people prefer that or not.

 

[Robbo99999]

Nature and nurture are nothing to do with what we’re discussing.

The goal MUST be audio transparency. NOT what people prefer, be that through nature or nurture.

Back to basics. In as far as possible, in as far as it can be measured, the goal needs to be us hearing what the person mixing the track heard. You take that digital master. You play it through transparent electronics. You produce a sound through the headphones which captures what that mixer heard through their speakers, in as far as that’s possible.

The Harman curve, by definition, says we’ll play music back to people and ask them if they prefer it with a bit more bass or treble, then create a curve based on the average of that preference.

That’s wrong. I mean, don’t get me wrong, if you want to EQ it it to your preference, that’s down to you. But goal number 1 is transparency, whether people prefer that or not.

The problem is headphones are a different experience to speakers, you don't feel the bass through your body like you would with speakers, so people preferred turning up the bass in headphones - probably to better make the experience of bass closer to that of speakers, to make up for the loss of feel of the bass. In my eyes that's a valid thing to do and can still come under "transparency" and "as the artist intended" in as much as the headphones are trying their best to replicate a reference speaker experience. I think headphones are less "transparent" if the bass didn't include that boost for those reasons.

Also, unfortunately headphones work differently for different people in terms of both the actual frequency response received at the eardrum and additionally what that ideal frequency response would be at the eardrum for that individual, so if you have to choose one target curve for all people then it stands to right that it makes sense to choose one that is least offensive to the most people, so the preference part of the study acknowledges that element. I tend to think that the only way you can get true transparency with a headphone is the Smyth Realizer solution and to base the setup of that system with some proper reference speakers. In the meantime we have the Headphone Harman Curve which suits some people better than others based on the anatomical variation of the population, in my experience the Headphone Harman Curve can actually be very close to "transparent" for some people....in my testing comparing my Harman EQ'd headphones to my reference speaker system I think it's quite close, it's not perfect, but it's close for me.....so for me I'm a fan of the Headphone Harman Curve, but I acknowledge that some people will like it more than others, and by extension that for some people it will be more "accurate & transparent" than others. I haven't read all of the back & forths in this thread but figured I'd add my viewpoint.

 

[Sean Olive]

Okay, a few issues with and questions about the methodology. Please feel free to chip in.

1 - ‘Good speakers in a good room’. That’s okay, I can see the point. But why not go the whole hog? Why not go to Abbey Road and use the mixing room there? Whatever the faults and flaws, our target must surely be to hear the music exactly as it was heard when mixed.

2 - Play this back into the same type of measuring equipment used to measure the headphones. Great idea, can’t argue with that.

3 - Okay, but why not just stop there. Take that measurement, and that’s your curve. By the way, have Harman ever released this curve? I’d have thought it should be in the data somewhere. Anyone?

4 - Get trained listeners to listen to that (and other curves), and come up with an average of what they prefer. What? Why bring preference into it at all. Surely we should be asking for accuracy. The legitimate next scientific step here is to sit those trained listeners in the ‘good room with good speakers’ (or Abbey Road, see above), and with headphones tuned to the curve measured earlier, then play a variety of pieces of music both through the speakers and headphones. And ask the trained listeners if they sound the same. Not if they prefer that, or another sound. Just check that the measuring equipment is doing what it’s supposed to do. If not, tweak the curve.

5 - A few years later get untrained listeners involved. At this point, I’m out. Why on Earth is this either necessary or desirable, in regards to what we’re discussing?

There is some validity in what Harman have done, to a point. But this is supposed to be science. The kit should be there for one reason and one reason only, to faithfully and accurately reproduce the music, in as far as that’s possible. And that’s what the curve should be.

The annoying thing is, we have the ability to do just this. Harman have come pretty close to doing just this. The whole ‘preference’ question either shouldn’t be there at all, or just a minor footnote.

By the way, I suspect the end result wouldn’t be a million miles away from Harman.

1 - Why Abbey Road? What is the criteria? Are the majority of recordings sold made there? Does their listening room, loudspeakers and their calibration adhere to some recognized technical standard? My point, which is made throughout our published research, is no meaningful standard exists that is uniformly practiced within the recording industry. One of the consequences is that recordings are highly variable in sound quality, and there is no perfect headphone or loudspeaker target that can accommodate for all this variability. The best you can do is come up with some statistical average and define a target that makes most recordings sound acceptable. We chose to base it on anechoically flat loudspeakers that were calibrated in an acoustically well-controlled Reference Room. All measured and documented in our papers.

3. Yes the curve is published in our AES papers.

4. How do you define accuracy in subjective terms without a Reference? Why Abbey Road? Have you seen measurements of B&W 800D loudspeakers that they used at one time? They are not accurate based on their comprehensive anechoic measurements like the ones we do and Amir does on the Klippel scanner.

5. We have preferences for trained and untrained listeners. They mostly agree. A small percentage of the untrained like a little more bass. If you want the trained listener Target Curve turn down the bass 2-3 dB.

When there is an audio standard that defines and quantifies what is considered "accurate" in the playback chain (loudspeaker, listening room, in-room target curve) and it uniformly applied throughout the recording industry then we can use that to define what an "accurate" headphone target should be. Until then, the best we can do is to define a target response that approximates what we think simulates the in-room response of an accurate loudspeaker calibrated in a well-controlled listening room. Our listening room/loudspeaker setup is well-documented and I'm sure is not too far removed from what a good recording control room might be.

 

[Sean Olive]

Preload, I’ve read your comments on ‘preference’, and I don’t see your definition as being something completely different.

The point remains, however you dress it up. Harman start with one curve (measurement in the room), then ask people if they like it. For me (and I understand not everyone agrees), I’d either just stick with the original curve, or if I asked people to comment at all, it’d be to check if the curve was accurate to the room*. And yes, I’d maybe consider the room.

But what might get lost is that I’m agreeing with the need for a curve, and that Harman’s approach has got a lot right.

* Actually, fair enough, ask them. Then create a number of curves. One ‘accurate’ for people wanting accuracy (I presume us), then maybe one ‘preferred’ for marketing purposes. Or 2 or 3; the results suggest people can be split into 3 categories.

One of the things you are missing in your argument is "what is considered an accurate in-room loudspeaker target curve? "You have to first define that before you can apply it to the headphone target.

Here is the current ITU-R standard that defines what they consider an "accurate" in-room operational response based the steady-state response measured at the listening seat Two things come to mind:
1. The curve has very generous tolerances but no real target or curve is specified. It allows the possibility of an infinite number of different shapes and slopes of curves within those tolerances. In other words, the standard is useless,
2. However, most people interpret the target as a flat line between those tolerances.

We did some experiments several years ago on room correction products and found products that implemented a flat in-room response were rated dead last in sound quality by trained listeners. Most recordings sounded too bright and thin with a flat target curve. Were listeners wrong even though the target was flat or accurate, and met the ITU-R standard?

Instead listeners preferred a target curve that had a 4-5 dB bass shelf, which was applied to the speakers upon which our headphone target is based. The headphone target has slightly more bass, probably because the whole-body vibration/tactile experience of loudspeakers in a room is missing from the headphone experience. We also found that the amount of bass preferred depends on the program: Circle of Confusion issues again.

Hardly no one follows the ITU-R standard because it makes most recordings sound thin and bright. Genelec room correction used a flat target response at one point to meet the ITU-R standard, but I believe they have changed it based on our research and customer feedback. The people who wrote the standard have admitted they didn't fully understand all the issues.

 

[Yorkshire Mouth]

1 - Why Abbey Road? What is the criteria? Are the majority of recordings sold made there? Does their listening room, loudspeakers and their calibration adhere to some recognized technical standard? My point, which is made throughout our published research, is no meaningful standard exists that is uniformly practiced within the recording industry. One of the consequences is that recordings are highly variable in sound quality, and there is no perfect headphone or loudspeaker target that can accommodate for all this variability. The best you can do is come up with some statistical average and define a target that makes most recordings sound acceptable. We chose to base it on anechoically flat loudspeakers that were calibrated in an acoustically well-controlled Reference Room. All measured and documented in our papers.

3. Yes the curve is published in our AES papers.

4. How do you define accuracy in subjective terms without a Reference? Why Abbey Road? Have you seen measurements of B&W 802N loudspeakers that they used at one time? They are not accurate. Next question.

5. We have preferences for trained and untrained listeners. They mostly agree. A small percentage of the untrained like a little more bass. If you want the trained listener Target Curve turn down the bass 2-3 dB.

When there is an audio standard that defines and quantifies what is considered "accurate" in the playback chain (loudspeaker, listening room, in-room target curve) and it uniformly applied throughout the recording industry then we can use that to define what an "accurate" headphone target should be. Until then, the best we can do is to define a target response that approximates what we think simulates the in-room response of an accurate loudspeaker calibrated in a well-controlled listening room. Our listening room/loudspeaker setup is well-documented and I'm sure is not too far removed from what a good recording control room might be.

Before we go any further, a massive thanks for taking the time and trouble to share your extensive knowledge and expertise.

Your main question boils down to 'why Abbey Road', and answering that question clears up most of the other issues.

A person sits at a mixing desk and mixes the music. They hear what they hear, and that's how the tape/digital file is mixed. If a pair of headphones had a curve which matched what the mixer heard, then you'd be hearing the same thing (within the limitations of the technology). So if Abbey Road sounded bright, the 'phones would sound equally bright, and so on. If Abbey Road made a recording sound too bassy, the mixer would mix down the bass, and if you wore those 'phones, you'd hear the same mixed-down bass.

Why Abbey Road. A few reasons:

1. It's generally seen as the premiere recording and mixing studio on the planet
2. Many of the great recordings were originally recorded and mixed there
3. A disproportionately large number of re-issues are mixed there, even the recording wasn't originally
4. A disproportionately large number of mixing studios use these B&Ws, and other models which are commonly used have a similar sound signature. 75% of the biggest albums you hear are mixed at a relatively small number of studios, and you'll find the B&Ws in there more often than not

The point isn't whether those speakers or Abbey Road is 'perfect'. It's that what the sound engineer hears there (warts and all) is what we hear at home.

And for that we need electronics which faithfully and accurately reproduce the recording, and headphones which tonally match what the mixer was hearing.

However, having said all of that, I'm more than open to use some other studio, or just great speakers in great room as a reference. What I'm not so sure about is taking that AND THEN AFTERWARDS playing that curve back to people, getting them to tweak it to their preference, and then producing an average preference, and using that as something to which all headphones should adhere.

It's the addition of that last part with which I have an issue.

When you say:

We have preferences for trained and untrained listeners. They mostly agree. A small percentage of the untrained like a little more bass. If you want the trained listener Target Curve turn down the bass 2-3 dB.

For me that's completely the wrong way round.

Let the untrained who like more bass turn their bass up a bit. Don't expect those of us who want accurate reproduction to turn our bass down to fit in with them. Does that make sense?

I have no issue whatsoever with the current Harman Curve being there to satisfy the public's wants - manufacturers are, after all, there to make money. But surely a headphone based on your pre-preference research curve would be a viable extra option to throw into the mix (no pun intended).

Having said all that, quite aside from you taking the trouble to reply here, I'd like to thank you for at least doing the research. Because I don't see anyone else doing it.

One last question. Would you please publish for us here the unadulterated curve from your AES paper?

Thanks again.

 

[Sean Olive]

Before we go any further, a massive thanks for taking the time and trouble to share your extensive knowledge and expertise.

Your main question boils down to 'why Abbey Road', and answering that question clears up most of the other issues.

A person sits at a mixing desk and mixes the music. They hear what they hear, and that's how the tape/digital file is mixed. If a pair of headphones had a curve which matched what the mixer heard, then you'd be hearing the same thing (within the limitations of the technology). So if Abbey Road sounded bright, the 'phones would sound equally bright, and so on. If Abbey Road made a recording sound too bassy, the mixer would mix down the bass, and if you wore those 'phones, you'd hear the same mixed-down bass.

Why Abbey Road. A few reasons:

1. It's generally seen as the premiere recording and mixing studio on the planet
2. Many of the great recordings were originally recorded and mixed there
3. A disproportionately large number of re-issues are mixed there, even the recording wasn't originally
4. A disproportionately large number of mixing studios use these B&Ws, and other models which are commonly used have a similar sound signature. 75% of the biggest albums you hear are mixed at a relatively small number of studios, and you'll find the B&Ws in there more often than not

The point isn't whether those speakers or Abbey Road is 'perfect'. It's that what the sound engineer hears there (warts and all) is what we hear at home.

And for that we need electronics which faithfully and accurately reproduce the recording, and headphones which tonally match what the mixer was hearing.

However, having said all of that, I'm more than open to use some other studio, or just great speakers in great room as a reference. What I'm not so sure about is taking that AND THEN AFTERWARDS playing that curve back to people, getting them to tweak it to their preference, and then producing an average preference, and using that as something to which all headphones should adhere.

It's the addition of that last part with which I have an issue.

When you say:



For me that's completely the wrong way round.

Let the untrained who like more bass turn their bass up a bit. Don't expect those of us who want accurate reproduction to turn our bass down to fit in with them. Does that make sense?

I have no issue whatsoever with the current Harman Curve being there to satisfy the public's wants - manufacturers are, after all, there to make money. But surely a headphone based on your pre-preference research curve would be a viable extra option to throw into the mix (no pun intended).

Having said all that, quite aside from you taking the trouble to reply here, I'd like to thank you for at least doing the research. Because I don't see anyone else doing it.

One last question. Would you please publish for us here the unadulterated curve from your AES paper?

Thanks again.

I understand the underlying reason/concept for measuring in a recording studio: "to measure what the recording engineer/producer heard". We own a pair of 800D loudspeakers and I have measured them and put them in many double-blind listening tests so I have quite a lot of subjective/objective data on them.

If we assume they are relatively flat on-axis then the direct sound at Abbey Road should be similar to the direct sound measured in our Reference Room. The loudspeaker we used has better off-axis and directivity than the 800D so the reflected sounds that contributed to the headphone target curve should be more accurate. I suspect that above 4-5 kHz the direct sound makes up most of the steady-state response due to the increased directivity of the loudspeaker.

I would argue that most recording studios in the world -- do not use B&W'800Ds. They are primarily used for mixing classical music. I would guess that Genelecs are more common and JBL is popular here in the USA. Both are essentially designed to similar targets: flat on-axis, smooth off-axis and directivity. Most studio monitors aim for flat on-axis, so the big differences occur below 300 Hz due to interactions between the speaker and room acoustics, which can cause 30dB variations. I didn't include those in our target curve. If Abbey Road has those variations should they be included in a headphone target curve? Does Abbey Road publish measurements of their in-room response?

Amplifier? Not really an issue. Most professional amps are designed to be flat. Even $100 amps can achieve that performance.

The recording mixer: that is a bigger variable in this process. Hearing loss is an occupational hazard, and the recordings they make may never sound correct for someone with normal hearing even if we perfectly replicate the Abbey Road sound field through the headphone. Note: Our trained listeners are tested for normal hearing.


For an active headphone is easy to program presets that includes "accurate" (whatever that means), trained listener, untrained listener and even hearing impaired. We could have presets called Abbey Road, Capitol Records, Carnegie Hall, or whatever marketing folks think will sell.. Y Or I could just provide a bass/treble knob that allows you to adjust to taste and/or compensate for bad recordings and hearing loss. You are giving me lots of good ideas..

So in conclusion, I would argue that our setup would likely produce a very similar Headphone Target Curve of any studio that uses monitors that have flat on-axis response including Abbey Road. Below 300 Hz, our setup would also produce similar results assuming the studio did proper speaker calibration, bass management and/or acoustic treatment to minimize speaker/room acoustic interactions. I think you said before that the differences are probably quite small.

 

[Yorkshire Mouth]

I understand the underlying reason/concept for measuring in a recording studio: "to measure what the recording engineer/producer heard". We own a pair of 800D loudspeakers and I have measured them and put them in many double-blind listening tests so I have quite a lot of subjective/objective data on them.

If we assume they are relatively flat on-axis then the direct sound at Abbey Road should be similar to the direct sound measured in our Reference Room. The loudspeaker we used has better off-axis and directivity than the 800D so the reflected sounds that contributed to the headphone target curve should be more accurate. I suspect that above 4-5 kHz the direct sound makes up most of the steady-state response due to the increased directivity of the loudspeaker.

I would argue that most recording studios in the world -- do not use B&W'800Ds. They are primarily used for mixing classical music. I would guess that Genelecs are more common and JBL is popular here in the USA. Both are essentially designed to similar targets: flat on-axis, smooth off-axis and directivity. Most studio monitors aim for flat on-axis, so the big differences occur below 300 Hz due to interactions between the speaker and room acoustics, which can cause 30dB variations. I didn't include those in our target curve. If Abbey Road has those variations should they be included in a headphone target curve? Does Abbey Road publish measurements of their in-room response?

Amplifier? Not really an issue. Most professional amps are designed to be flat. Even $100 amps can achieve that performance.

The recording mixer: that is a bigger variable in this process. Hearing loss is an occupational hazard, and the recordings they make may never sound correct for someone with normal hearing even if we perfectly replicate the Abbey Road sound field through the headphone. Note: Our trained listeners are tested for normal hearing.


For an active headphone is easy to program presets that includes "accurate" (whatever that means), trained listener, untrained listener and even hearing impaired. We could have presets called Abbey Road, Capitol Records, Carnegie Hall, or whatever marketing folks think will sell.. Y Or I could just provide a bass/treble knob that allows you to adjust to taste and/or compensate for bad recordings and hearing loss. You are giving me lots of good ideas..

So in conclusion, I would argue that our setup would likely produce a very similar Headphone Target Curve of any studio that uses monitors that have flat on-axis response including Abbey Road. Below 300 Hz, our setup would also produce similar results assuming the studio did proper speaker calibration, bass management and/or acoustic treatment to minimize speaker/room acoustic interactions. I think you said before that the differences are probably quite small.

Thanks again.

Yes, I think that's almost certainly correct.

So I come back to this. Would it be possible for you to let us have the pre-preference curve?

I appreciate there may be some copywrite issues.

Cheers.

 

[markanini]

So I come back to this. Would it be possible for you to let us have the pre-preference curve?

Someone will correct me if I'm wrong. That's the green dotted line in the pdf you linked to before. Essentially simulating and anechoic response in a live room. I don't know if that's a common speaker response in recording studios.

 

[Yorkshire Mouth]

Someone will correct me if I'm wrong. That's the green dotted line in the pdf you linked to before. Essentially simulating and anechoic response in a live room. I don't know if that's a common speaker response in recording studios.

We need to be careful and clear with our terms.

My understanding was it was a speaker which was near flat, EQd so it’d sound flat in an anechoic chamber, then placed in a ‘great room’. So it’s not flat anechoic. It’s that + the colouration of the great room.

A good mixing room should sound very close to Harman’s ‘great room’.

 

[raistlin65]

Nature and nurture are nothing to do with what we’re discussing.

The goal MUST be audio transparency. NOT what people prefer, be that through nature or nurture.

Your statement, which I replied to, was, "Preference isn’t science."

Sound signature preference could be biologically derived and/or culturally derived. If purely the latter, then it would seem to be social science. Otherwise, saying it's not science seems silly.

The goal MUST be audio transparency. NOT what people prefer, be that through nature or nurture.

Back to basics. In as far as possible, in as far as it can be measured, the goal needs to be us hearing what the person mixing the track heard. You take that digital master. You play it through transparent electronics. You produce a sound through the headphones which captures what that mixer heard through their speakers, in as far as that’s possible.

The Harman curve, by definition, says we’ll play music back to people and ask them if they prefer it with a bit more bass or treble, then create a curve based on the average of that preference.

That’s wrong. I mean, don’t get me wrong, if you want to EQ it it to your preference, that’s down to you. But goal number 1 is transparency, whether people prefer that or not.

Ok. Let's say I agree with your premise of what the goal is (which I do not).

1) Your assumption is that there is some, singular accurate reproduction of the sound that is definitive. Rather than the idea that someone might have mixed the sound for listening to on various speakers and/or headphones.

2) Assuming the person mixing had specific sound in mind, how does a user know if what they are perceiving has achieved that? Seems to me the only way to do that would be to set up the exact same equipmen,t with the same room acoustics, as the person mixing was using.

 

[Sean Olive]

Before we go any further, a massive thanks for taking the time and trouble to share your extensive knowledge and expertise.

Your main question boils down to 'why Abbey Road', and answering that question clears up most of the other issues.

A person sits at a mixing desk and mixes the music. They hear what they hear, and that's how the tape/digital file is mixed. If a pair of headphones had a curve which matched what the mixer heard, then you'd be hearing the same thing (within the limitations of the technology). So if Abbey Road sounded bright, the 'phones would sound equally bright, and so on. If Abbey Road made a recording sound too bassy, the mixer would mix down the bass, and if you wore those 'phones, you'd hear the same mixed-down bass.

Why Abbey Road. A few reasons:

1. It's generally seen as the premiere recording and mixing studio on the planet
2. Many of the great recordings were originally recorded and mixed there
3. A disproportionately large number of re-issues are mixed there, even the recording wasn't originally
4. A disproportionately large number of mixing studios use these B&Ws, and other models which are commonly used have a similar sound signature. 75% of the biggest albums you hear are mixed at a relatively small number of studios, and you'll find the B&Ws in there more often than not

The point isn't whether those speakers or Abbey Road is 'perfect'. It's that what the sound engineer hears there (warts and all) is what we hear at home.

And for that we need electronics which faithfully and accurately reproduce the recording, and headphones which tonally match what the mixer was hearing.

However, having said all of that, I'm more than open to use some other studio, or just great speakers in great room as a reference. What I'm not so sure about is taking that AND THEN AFTERWARDS playing that curve back to people, getting them to tweak it to their preference, and then producing an average preference, and using that as something to which all headphones should adhere.

It's the addition of that last part with which I have an issue.

When you say:



For me that's completely the wrong way round.

Let the untrained who like more bass turn their bass up a bit. Don't expect those of us who want accurate reproduction to turn our bass down to fit in with them. Does that make sense?

I have no issue whatsoever with the current Harman Curve being there to satisfy the public's wants - manufacturers are, after all, there to make money. But surely a headphone based on your pre-preference research curve would be a viable extra option to throw into the mix (no pun intended).

Having said all that, quite aside from you taking the trouble to reply here, I'd like to thank you for at least doing the research. Because I don't see anyone else doing it.

One last question. Would you please publish for us here the unadulterated curve from your AES paper?

Thanks again.

You can find these curves in the AES papers, presentations and websites throughout the internet.

This graphic shows the Harman AE/OE Target from 2013 (black) and 2015 (blue) from my presentation. 2013 curve is before we added listener preferred bass and treble adjustments, what you call the "unadulterated curve". It represents the accurate speaker in the room equalized to the in-room target curve. You can see there is little difference between 2013 and 2015 curve: about 2 dB more bass and 1 dB less treble.

The green curve is the in-room response of the loudspeaker after it has been equalized to a flat in-room target response, which we did as the baseline curve for the method of adjustment study where people could adjust a bass and treble shelving filter. I already pointed out that a flat in-room loudspeaker target is not preferred based on a previous room correction study. This has been again confirmed in the 2015 study where listeners adjusted the flat bass up 6 dB and reduced the treble to something that resembles the 2013 curve based on a measurement of an accurate speaker in the Harman Reference Room.

 

[Robbo99999]

@Yorkshire Mouth , seeing as Sean above is intimating your could try the 2013 Harman Curve I made some EqualiserAPO EQ config files months ago for turning a 2018 Harman EQ'd headphone into a 2013 Harman EQ'd Headphone, and I also did one to turn a 2018 Harman EQ'd heapdhone into 2015 Harman EQ'd headphone. I did this by tracing the different curves in VirtuixCAD and then using REW to manually create a few parametric filters that describe the differences between the different versions of the Harman Curves. I'll attach them to the end of this post so people can add them onto their EQ's (you could try them too if you currently use an Oratory 2018 Harman EQ for instance, you'd just activate my config files on top of the Oratory EQ).

Note: these contain a few Low Shelf Filters, and I've noticed importing config files into EqualiserAPO can sometimes result in the Low Shelf filter not activating...it'll normally be seen by the Low Shelf filter listing in EqualiserAPO looking different to the other filters (font & layout, etc)....in which case then manually input the bugged Low Shelf Filters into EqualiserAPO and delete the bugged Low Shelf Filters that were imported earlier.

 

[Magma]

Especially with the Ananda it's like throwing a thick blanket over the sound.

Interesting. I tried Oratory's EQ with both AKG K371 and Focal Clears and with both, I had the same experience. Using EQ made them sound like there is a blanket on top of them. K371 is close to the Harman curve already though.