• 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!

The Etymotic Target (R.I.P. Harman)

bluefuzz

Major Contributor
Joined
Jan 17, 2020
Messages
1,043
Likes
1,772
A sine sweep is a bogus test. If the tones are supposed to be equally loud the headphone or loudspeaker will look like an equal loudness contour which will vary depending on the SPL

Doesn't that somewhat contradict your statement:

As I said earlier the Harman Target has nothing to do with loudness contours as all the listening tests were done at average sound pressure levels well above where large growths in loudness in bass and treble occur.

Surely if the sine sweep is done at the same level as the listening level then they should sound perceptually 'flat'. Or are you saying that a measured 'flat' or slightly downward sloping frequency response is not (or should not be) perceptually 'flat', i.e. all frequencies are of the same subjective level?
 

solderdude

Grand Contributor
Joined
Jul 21, 2018
Messages
15,891
Likes
35,912
Location
The Neitherlands
I think S. Olive means listening to a sine sweep is a poor test. It isn't a poor test when measuring though.
He is right as well.

Better to switch between say 440Hz and other frequencies and try to detect level differences. That too demands listening skills though.
What S. Olive states is quite true. When one tries to audibly get all frequencies equally loud you get different 'response' depending on the used level.

The hearing is not linear in amplitude at different frequencies and also is level dependent.
So hearing/perceiving all frequencies equally loud is something very different than equal SPL (measured).
 
OP
Sharur

Sharur

Senior Member
Joined
Apr 10, 2021
Messages
476
Likes
213
I think S. Olive means listening to a sine sweep is a poor test.
Using a tone generator is an incredibly good way to EQ a headphone to your ear's peaks and dips because we all know EQ presets are more or useless for that. Dr. Olive's statement about equal loudness contours definitely makes sense though. Which is why having a bass shelf at 200 Hz doesn't really make any sense. It just presents itself as a mud cut.
 
Last edited:

solderdude

Grand Contributor
Joined
Jul 21, 2018
Messages
15,891
Likes
35,912
Location
The Neitherlands
using a tone generator or sweep to determine tonal balance is what S. Olive is talking about.
Looking for sharp dips and peaks a tone, warble or narrow noise band is handy.
Of course it all depends on what the goal is.

A bass shelf (upwards or downwards) can make a lot of sense. How much depends on the frequency response.
 

maverickronin

Major Contributor
Forum Donor
Joined
Jul 19, 2018
Messages
2,527
Likes
3,308
Location
Midwest, USA
I am quite careful to qualify that the Harman Target is optimized for stereo recordings which are made for speakers in rooms.

For binaural recordings and spatial rendering of HRTF sources a different target is likely needed. Dr Mead Killion at etymotic liked to make binaural recordings of jazz performances that were used to validate the headphones. I have often told him that while they may work for those recordings a different tuning is needed for stereo ones.
Obviously you want to ensure the ear canal resonance is not duplicated twice in the playback and manikin recording. Secondly the recording depending on how it was made may have sufficient bass from the room.
In fact this was confirmed by Facebook Reality Labs who recently published an AES paper showing that while the Harman Target was preferred for stereo recordings over DF, for binaural rendering of VR/AR material a different target is more optimal.

Have you ever done any research into basic crossfeed or more advanced HRTF virtualization techniques?

To me, with most full size headphones, stereo mixes are essentially unlistenable without crossfeed. Anything with much stereo separation at all will cause headaches after a period of time so I always use at least a basic Meier/CMoy style crossfeed.

That may be why I prefer a different target curve, something close to Etymotic, but with a mild bass shelf below 100Hz.
 
U

UKPI

Guest
Surely if the sine sweep is done at the same level as the listening level then they should sound perceptually 'flat'. Or are you saying that a measured 'flat' or slightly downward sloping frequency response is not (or should not be) perceptually 'flat', i.e. all frequencies are of the same subjective level?

Using a tone generator is an incredibly good way to EQ a headphone to your ear's peaks and dips because we all know EQ presets are more or useless for that.

https://www.audiosciencereview.com/...ngful-headphone-measurement.7919/#post-191918
The paper linked at the end of this post should be pertinent to this part of the discussion.

Edit: Fixed the link.
 

JohnYang1997

Master Contributor
Technical Expert
Audio Company
Joined
Dec 28, 2018
Messages
7,175
Likes
18,292
Location
China
A sine sweep is a bogus test. If the tones are supposed to be equally loud the headphone or loudspeaker will look like an equal loudness contour which will vary depending on the SPL
A better test for judging spectral balance is wideband noise and music. We used music.
Sine sweep (manually) is exceptionally good for identifying peaks and dips in the frequency response. And it's really a great tool to have for the valleys and bump in the bass(<500hz) and highs (>6khz). Usually it's difficult to use music to identify issues at those regions. 10khz-14khz range is still quite important especially when the sound balance is already good.
It's quite similar to the technique used in mixing that using a high Q peak filter to find annoying peaks to notch out.
With just listening to music normally one cannot identify how many and exactly where the peaks are in the evaluated system (speakers or headphones).
 

JohnYang1997

Master Contributor
Technical Expert
Audio Company
Joined
Dec 28, 2018
Messages
7,175
Likes
18,292
Location
China
This is not a Harman in room target.



Neutral loudspeakers aren't flat in a well treated room. Its a downward slope. If you get a flat (horizontal) response in-room you started out with speakers with a big boost in the treble.
Neutral is flat speakers in well treated room.
Read that again. What it means is that a pair of speakers with flat anechoic response put in a well treated room. Not in room response being flat.
 

ctrl

Major Contributor
Forum Donor
Joined
Jan 24, 2020
Messages
1,616
Likes
6,086
Location
.de, DE, DEU
Here comes my attempt, in a simple way, to explain why the approach with a flat HRTF in the bass sounds less natural.

Let's start from the situation that is indisputable and that everyone can agree with. What sound pressure curve arrives at the eardrum with anechoic flat measuring speaker in an anechoic chamber (in free field)?

1631639757146.png

Source: Audio precision - Application Note: Headphone Electroacoustic Measurements

To get from free-field equalization to diffuse-field equalization, one needs further measurements from different directions or a measurement in the reverberation chamber. But for the sake of simplicity, let's ignore this (the changes are not dramatic).

Now we take the flat-measuring loudspeaker from the anechoic chamber and place it in a "normal, average" listening room.
What happens in the bass range? Depending on the placement in the listening room, there is boundary reinforcement in the bass frequency range.

Under ideal conditions, it could look like this:
1631641498919.png
I am not using the optimal tool here, the green curve corresponds to our anechoic flat measuring loudspeaker (here the curve is not flat, you have to imagine). The green curve corresponds to the speaker in an anechoic chamber.
If now two boundary surfaces are added (in the sketch on the left, these are the two green lines at the bottom and right of the speaker), the sound pressure reproduction in the bass range increases. The blue curve corresponds to a measurement of the speaker in our "normal" listening room (with ideal walls) - ignore the numerical values (and ignore room modes, ...), it's all about the principle here.

Our measurement microphone no longer measures only direct sound in our normal listening room, as in an anechoic chamber, but a mixture of direct and diffuse sound.
This leads to the well-known frequency response that is tilted towards high frequencies. In the following sketch, this now corresponds to the frequency response curve in cyan. This is the in-room measurement of our loudspeaker in our normal listening room at the listening position.
1631642942799.png


The frequency response curve at the eardrum changes accordingly and is shown in purple in the sketch.
Thus, for our "ideal" HRTF, which is supposed to reflect a neutral loudspeaker in a normal room, we get a bass boost and treble rolloff, compared to the pure diffuse field equalization of the loudspeaker in an anechoic chamber.
 
Last edited:

Sean Olive

Senior Member
Audio Luminary
Technical Expert
Joined
Jul 31, 2019
Messages
329
Likes
3,004
Was this test conducted with a tone generator? Etymotic target is perceived as flat to me when performing a sine sweep. I don't all of a sudden feel a chunk of SPL dropping at 200 Hz.
Sine sweep (manually) is exceptionally good for identifying peaks and dips in the frequency response. And it's really a great tool to have for the valleys and bump in the bass(<500hz) and highs (>6khz). Usually it's difficult to use music to identify issues at those regions. 10khz-14khz range is still quite important especially when the sound balance is already good.
It's quite similar to the technique used in mixing that using a high Q peak filter to find annoying peaks to notch out.
With just listening to music normally one cannot identify how many and exactly where the peaks are in the evaluated system (speakers or headphones).
Sine tones are good for identifying buzzes and rattles... and perhaps peaks/dips within a local frequency region where loudness sensitivity is constant. But there are more efficient ways to determine if the headphone has a smooth response/accurate spectral balance.In order of efficiency:
1) Measure it
2) listen to broadband pink noise -- the most sensitive signal for detecting low/medium/high Q resonances.

If you are trained you can identify the frequencies at which the resonances/anti-resonances occur. The more it rings or whistles the higher the Q
 

Sean Olive

Senior Member
Audio Luminary
Technical Expert
Joined
Jul 31, 2019
Messages
329
Likes
3,004
Due to the missing component of tactile feel (when reproduced at the same SPL at eardrum level) the X perceived sensation will still be that the headphones are missing bass because the tactile feel input to the brain is missing.

Solution is the subwoofer backpack or adding some extra lows at the eardrum level that kind of compensates for that loss.

Now the thing is the latter works well for the general public but is not really needed for more experienced listeners and a lot of recording professionals.
The bass boosted headphones are intended for the general public. For them Harman curve is fine. For real bassheads and DJ's in live venues Harman bass is 'bass-shy' to them.

A single target thus will NEVER satisfy anyone. One will favor this and someone else will favor something else for different reason.
There is no single correct curve. Just look at Oratory who started out with Harman but developed his own (less bassy and different in clarity) target curve. Some prefer the Etymotic target, some really like bass shy headphones (HD600 or K5xx to K702) or prefer overly bassy ones.

Comfort, ease of use, looks, functionality, price, use case etc. all create a more diverse market.
When audiophiles (mostly folks with cash to spend) would all prefer the same 'target' only very few models would sell well and others would not.
Here too comfort plays a role widening the market but the reality is many models are sold with widely varying tonal balances.

Our data indicates at least 64% of listeners are happy with the Target. The others want less bass (21%) or more bass (15%).

So give them a tone controls or an app to customize the bass/treble balance of the headphone to compensate for their training/age/ level of testosterone/estrogen and hearing loss.

It's not rocket science to personalize the headphone to satisfy everyone's taste. But you have to start with some baseline or reference that most people agree is relatively neutral and go from there.
 

solderdude

Grand Contributor
Joined
Jul 21, 2018
Messages
15,891
Likes
35,912
Location
The Neitherlands
Our data indicates at least 64% of listeners are happy with the Target. The others want less bass (21%) or more bass (15%).

2/3 so the majority of music consumers which are mostly young folks on phones with BT headphones listening to popular music which is compressed like hell, will certainly prefer a little bass boost and some top end lowered. There is no doubt about that. Also people in transit will almost all prefer some bass boost.

The 'less bass' folks can choose other headphones (or use a tone control). Bass heads (a smaller group) will want humongous bass and nothing else.
For them too there are products they like and can buy. And they can always use bass boost switches or settings.

This has always been the case even before the Harman target was there. Harman research has simply shown that the majority simply prefers some extra bass for playback. There can be several reasons for this but regardless the facts are there.

So give them a tone controls or an app to customize the bass/treble balance of the headphone to compensate for their training/age/ level of testosterone/estrogen and hearing loss.

Yes, and it has always existed. Tone controls... Some buy headphones or speakers with 'permanent' tone controls in it as well. This too has always been the case. Only purists hated tone controls and wanted defeat buttons so the market gave it to them.

It's not rocket science to personalize the headphone to satisfy everyone's taste.

certainly today it isn't with the huge amount of headphones available and smartphones having all kinds of EQ possibilities.
Still... a lot of folks still believe it is rocket science and want others to do it for them.
They are so afraid to do it 'wrong' or have an incorrect EQ based on this or that.
For them a baseline or reference is handy.

I think the whole discussion is the 1/3 of folks that want or feel the prescribed amount and spectrum of the bass lift is not to their liking.
It's those folks that feel the baseline is incorrect. They will always moan. What most of them don't seem to get is that it is not obligated to listen to headphones with that exact tonal balance. Besides..,. I only know of a handful of headphones that actually are close to that target.

Headphone owners also do not have to EQ to that baseline nor are they forced to.
One has a choice not to boost bass, only boost a little or a bit more.

IMO headphones should have a smooth response, sound tonally correct (as real as sounds around you) and if the recordings ask for it adjust the bass or other parts using simple tone controls.

One thing is for sure there is no target curve that can satisfy all people. Harman curve does this for the majority of people (2/3) so the other third will always question/debate or hate it. This can't be helped. Combine this with strong opinionated people and discussions are there.

R.I.P. Harman target ? No of course not. Will some folks prefer another tonal balance ? Yes, always for whatever reason they have.
Is one tonal balance THE correct one ... in all cases... for all people and is something to strive for ? No.
But is it a good baseline from where one can adjust tonal balance ? Yes IMO it is, My personal one is close but not the same though.

To me an even response (no large dips and peaks) low enough distortion with a few dB of bass boost sounds great.
 
Last edited:

Sean Olive

Senior Member
Audio Luminary
Technical Expert
Joined
Jul 31, 2019
Messages
329
Likes
3,004
2/3 so the majority of music consumers which are mostly young folks on phones with BT headphones listening to popular music which is compressed like hell, will certainly prefer a little bass boost and some top end lowered. There is no doubt about that. Also people in transit will almost all prefer some bass boost.

The 'less bass' folks can choose other headphones (or use a tone control). Bass heads (a smaller group) will want humongous bass and nothing else.
For them too there are products they like and can buy. And they can always use bass boost switches or settings.

This has always been the case even before the Harman target was there. Harman research has simply shown that the majority simply prefers some extra bass for playback. There can be several reasons for this but regardless the facts are there.



Yes, and it has always existed. Tone controls... Some buy headphones or speakers with 'permanent' tone controls in it as well. This too has always been the case. Only purists hated tone controls and wanted defeat buttons so the market gave it to them.



certainly today it isn't with the huge amount of headphones available and smartphones having all kinds of EQ possibilities.
Still... a lot of folks still believe it is rocket science and want others to do it for them.
They are so afraid to do it 'wrong' or have an incorrect EQ based on this or that.
For them a baseline or reference is handy.

I think the whole discussion is the 1/3 of folks that want or feel the prescribed amount and spectrum of the bass lift is not to their liking.
It's those folks that feel the baseline is incorrect. They will always moan. What most of them don't seem to get is that it is not obligated to listen to headphones with that exact tonal balance. Besides..,. I only know of a handful of headphones that actually are close to that target.

Headphone owners also do not have to EQ to that baseline nor are they forced to.
One has a choice not to boost bass, only boost a little or a bit more.

IMO headphones should have a smooth response, sound tonally correct (as real as sounds around you) and if the recordings ask for it adjust the bass or other parts using simple tone controls.

One thing is for sure there is no target curve that can satisfy all people. Harman curve does this for the majority of people (2/3) so the other third will always question/debate or hate it. This can't be helped. Combine this with strong opinionated people and discussions are there.

R.I.P. Harman target ? No of course not. Will some folks prefer another tonal balance ? Yes, always for whatever reason they have.
Is one tonal balance THE correct one ... in all cases... for all people and is something to strive for ? No.
But is it a good baseline from where one can adjust tonal balance ? Yes IMO it is, My personal one is close but not the same though.

To me an even response (no large dips and peaks) low enough distortion with a few dB of bass boost sounds great.
Add to those arguments the fact that I can choose a group of recordings that will make you love or hate any one of those three alternative target functions based on how much bass or treble they contain. Circle of Confusion issues will always necessitate the need for adjustment no matter what Target you prefer.
 
OP
Sharur

Sharur

Senior Member
Joined
Apr 10, 2021
Messages
476
Likes
213
If outside noise decreases perceived bass, 35-42 dB of isolation would increase perceived bass when compared to a headphone with the same frequency response in a room with average loudness, no?
 

Sean Olive

Senior Member
Audio Luminary
Technical Expert
Joined
Jul 31, 2019
Messages
329
Likes
3,004
The Target curve research was done indoors in relatively quiet listening rooms so masking from background noise was not a factor.

If you listen in noisy environments and don’t have good passive or active attenuation of the background noise , that can affect your perception of the sound
 

Mauro

Member
Joined
Nov 23, 2019
Messages
92
Likes
81
Sine tones are good for identifying buzzes and rattles... and perhaps peaks/dips within a local frequency region where loudness sensitivity is constant. But there are more efficient ways to determine if the headphone has a smooth response/accurate spectral balance.In order of efficiency:
1) Measure it
2) listen to broadband pink noise -- the most sensitive signal for detecting low/medium/high Q resonances.

If you are trained you can identify the frequencies at which the resonances/anti-resonances occur. The more it rings or whistles the higher the Q
I do agree. My experience is this:
Not having the chance to measure I use a sine generator to jump of a few Hz to identify resonances in localized areas of the spectrum.
it is also very useful using a sine generator to identify which are the frequencies where the two drivers of the headphone mismatch and corrupt positioning of apparent sound sources.

But I think the best training is using the software by Harman which is linked by sean olive’s blog.
you can even use/add pink noise and try what he suggests. With pink noise I was more able to discriminate peaks and valleys in headphones compared to music (12 level vs 8/9).

since then I use pink noise for speakers and headphones.
Give it a check. You will be repaid!
http://harmanhowtolisten.blogspot.com/2011/01/welcome-to-how-to-listen.html?m=1
 
OP
Sharur

Sharur

Senior Member
Joined
Apr 10, 2021
Messages
476
Likes
213
I've received some new information which I wanted to discuss regarding high frequencies. Somewhere along the line I mixed up how diffuse field and free field are derived. If free field is no reflection, diffuse field is highly reflective, and harman target is somewhere in between, why is the ear canal gain pushed to a higher frequency, why is there so much energy from 5-10 kHz, and why does it drop like a rock after 10 kHz?
graph - 2021-09-14T193648.103.png

graph - 2021-09-14T193723.451.png

graph - 2021-09-14T193654.290.png

graph - 2021-09-14T193650.776.png
 

Sean Olive

Senior Member
Audio Luminary
Technical Expert
Joined
Jul 31, 2019
Messages
329
Likes
3,004
I’m pretty sure this is not a scientific poll. Even the way in which the question posed in biased.
I've received some new information which I wanted to discuss regarding high frequencies. Somewhere along the line I mixed up how diffuse field and free field are derived. If free field is no reflection, diffuse field is highly reflective, and harman target is somewhere in between, why is the ear canal gain pushed to a higher frequency, why is there so much energy from 5-10 kHz, and why does it drop like a rock after 10 kHz?
View attachment 153405
View attachment 153406
View attachment 153407
View attachment 153408
The Harman Target is based on a stereo pair of anechoically flat loudspeakers measured at the DFP in a semi-reflective room, which is neither free-field nor diffuse field. The DRP was measured in the stereo seat using a +- 30 degree spatial average.

The differences you see up to 10 kHz between ours and the others can be explained by those differences. If you look at the predicted-in room curve (PIR) of the loudspeaker which is a weighted sum of the direct sound, early-reflected and sound power it pretty well lines up with the Harman Target Curve. No one knows what the exact shape of the curve should be above 10 kHz because there are so many measurement errors in the IEC couplers, and the subjective and objective measurements themselves.

The B&K 5128 and the latest GRAS RA0045 (?) couplers have lower measurement errors above 10 kHz (the resonance at 13.5 kHz in the GRAS RA0045 has been damped making measurements more repeatable particularly distortion) and the 5128 is based on a large study that scanned human ear canals -- so there is promise that the measurement device better represents average humans.


For those reasons mentioned above, we don't even calculate predicted preference scores of headphones using data beyond 10 kHz. Add to that, the fact that most adults have greatly elevated hearing thresholds above 10 kHz makes the task of defining what is the ideal HF target a fuzzy, grey area that is very dependent on the individual and the music signal. This is topic that needs more research.
 
Last edited:
Top Bottom