• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required as is 20 years of participation in forums (not all true). Come here to have fun, be ready to be teased and not take online life too seriously. We now measure and review equipment for free! Click here for details.

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

Pretorious

Member
Joined
Nov 20, 2020
Messages
51
Likes
62
Location
Illinois
The issue is that the Harman target does not reproduce the sound of flat speakers in a treated room at the eardrum. Only the alleged perceived tonal balance according to a test with some people that funnily enough includes no one in this thread.

Which however can be tested against to get someone’s preference for a particular frequency response. Hence the whole point of the study by Harman. The majority preferred that curve over DF and FF curves, a point that has been belabored and emphasized by @Sean Olive in many threads many times on this forum alone. Any of us here can take flat speakers in a well-treated room, listen to them, and then compare a similar FR on headphones and see if we prefer that over other curves, just like the study. If Harman is right, that FR preference should more closely match their curve than any other.

Keep in mind, I’m playing a bit of devil’s advocate here. I don’t EQ to Harman, and listen to stock HD600s. We should support the scientific method on principle, which is why I make the argument here. The results from the Harman study are reproducible and verifiable. At the same time a preference is a preference, and I have no problem with that. Listen to what you like best and gives you the most pleasure. This isn’t a religion. But we should argue on principle, too, and apply the same scientific standard to everything we do.
 
OP
Sharur

Sharur

Senior Member
Joined
Apr 10, 2021
Messages
447
Likes
150
The majority preferred that curve over DF and FF curves
Raw DF and FF are unlistenable. Where was the comparison to Etymotic target? ER-4P does not count.
 
Last edited:

Jimbob54

Master Contributor
Forum Donor
Joined
Oct 25, 2019
Messages
6,954
Likes
8,148
I honestly have no clue how someone at a comfortably loud level and good seal will dial up the bass on ER2SE or ER4SR by 10 dB
I agree there, a 3-5 db low shelf from 100hz is ample . Any more than that and the preamp reduction I have to make means some portables really struggle to drive the 2SE. I had to get a balanced cable for the Qudelix 5k to avoid maxing it. Harman bass levels are generally not a good sound for me on anything. I agree with @restorer-john on that- and I rarely adjust the mids and treble to the same extent Harman based EQs require.

Doesnt mean I cant "use" the Harman curve as a consistent bassline to judge broadly how a headphone might present
 

Jimbob54

Master Contributor
Forum Donor
Joined
Oct 25, 2019
Messages
6,954
Likes
8,148
To be honest, I think the President of the United States is a greater authority figure than Harman so I'm going to go with him :)
View attachment 152606

Not really, he's got it pointing in wrong. Terrible insertion and will definitely need a huge bass boost to sound normal ;-)
 

Thomas_A

Major Contributor
Forum Donor
Joined
Jun 20, 2019
Messages
1,287
Likes
827
Location
Sweden
I mean, the curve is just Diffuse field with small-room compensation

What is the small room compensation? The x-curve? This is a large reflective concert hall compensation if anything. Not even a cinema.
 
Last edited:

jae

Addicted to Fun and Learning
Joined
Dec 2, 2019
Messages
539
Likes
563
End consumers are not getting a better, more accurate product from all this. They never were. The curve was/is designed to find out what people "liked" and let's face, what people "like" and what's actually uncoloured accuracy are completely different things.

The so-called "harman" curve is no different to a 1970s loudness button on a receiver or amplifier. Sounds big, fat and fun at low levels, but absolutely horrid at anything approaching medium levels. There's a reason loudness contours fell out of fashion, and so will this slavish adherence to an arbitrary brand's key selling point of difference.

I would think any manufacturer that goes out of their way to use some kind of peer-reviewed literature as a basis for their design, rather than some designer's arbitrary subjective approach is likely one that is going to be taking a more methodological approach in all aspects of the design and probably make a "better" product overall regardless of their end goal.

There are modern offerings from AKG, Focal, Sennheiser, Hifiman, Moondrop, DCA, and even common lifestyle/consumer brands that are clearly created with this target in mind and adhere to it in varying degrees based on the cohort of people the respective products are catered to. These products are some of the best regarded products in their respective lineups, as far as the market is concerned.
 

jae

Addicted to Fun and Learning
Joined
Dec 2, 2019
Messages
539
Likes
563
In a thread comparing the "correctness" of one target to Harman, one would expect there to be more discussion on whether or not "good speakers in a good room" should be the convention by which earphone tonality should be tuned in the first place. If we are going to challenge this we need some sensible reasoning or psychoacoustic explanation rather than "I don't like it/some other target sounds better to me/too much bass". We all know that music is produced differently, individual preferences exist, we know what rooms do, people prefer different volumes and so on, but we are also sensible enough to know we can control for these things and find the lowest common denominator in spite of these facts. We know the Harman data is not perfect and of course are only grouped averages, and on top of this, these averages I assume only applies to one setting- one position/distance, one axis, one angle of incidence, one SPL, and the tonality of one well-regarded speaker target. Beyond this, what is common rather than different and why is this the case? Is the variability of preference only restricted to certain bands for a reason? Is anatomical variability enough to explain it if preference biases are controlled?

We know that FR, amplitude, and time-domain characteristics can affect perception of sound and these can all affect perceived loudness, pitch, and timbre/tonal balance in various non-linear ways. What is to say that some portion of a boost on a target curve is not simply a "more accurate" frequency correction that emulates or compensates for things like ITD/ILD differences, phase coherence, or reflection cues in a broad group of test subjects? How much of Harman's reference is actually true individual "preference" vs an implicit or demonstrable psychoacoustic phenomenon that could potentially be quantified, or even predicted by anatomical features? If we were able to easily take various measurements at the eardrum, how would it compare between the two sources and also how would it compare to an average response? To people who are dismissing or prefer a certain established target, how many of you are actually taking personal in-ear measurements of both earphones and speakers and comparing them in any objective way to these purported targets?
 
Last edited:

aac

Active Member
Joined
May 17, 2020
Messages
173
Likes
126
I don't think Harman target has anything to do with good speaker in a good room as it is now, at least for IEMs.
 
Last edited:
OP
Sharur

Sharur

Senior Member
Joined
Apr 10, 2021
Messages
447
Likes
150
We have no proof that President Obama does not deep insert his Etymotic ER4 when not in public but it's undeniable that 40 dB of noise attenuation is not something someone as important as the President of the United States should have in a public space.
 

Soundmixer

Active Member
Joined
Mar 8, 2021
Messages
210
Likes
109
The x curve is one of the worst thought of "standards" for room sound there is. There is a lot of work going ton to revise this old and ad-hoc response of a speaker in a room that is forced by Dolby for movie mixing and mastering.

You are so correct about this. SMPTE started taking a very good look at the X-curve a few years ago, and I am sure it will be jettisoned when there is a new curve established.
 
OP
Sharur

Sharur

Senior Member
Joined
Apr 10, 2021
Messages
447
Likes
150

solderdude

Master Contributor
Joined
Jul 21, 2018
Messages
9,748
Likes
21,081
Location
The Neverlands
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.
 

Sean Olive

Active Member
Audio Luminary
Technical Expert
Joined
Jul 31, 2019
Messages
200
Likes
1,686
Which however can be tested against to get someone’s preference for a particular frequency response. Hence the whole point of the study by Harman. The majority preferred that curve over DF and FF curves, a point that has been belabored and emphasized by @Sean Olive in many threads many times on this forum alone. Any of us here can take flat speakers in a well-treated room, listen to them, and then compare a similar FR on headphones and see if we prefer that over other curves, just like the study. If Harman is right, that FR preference should more closely match their curve than any other.

Keep in mind, I’m playing a bit of devil’s advocate here. I don’t EQ to Harman, and listen to stock HD600s. We should support the scientific method on principle, which is why I make the argument here. The results from the Harman study are reproducible and verifiable. At the same time a preference is a preference, and I have no problem with that. Listen to what you like best and gives you the most pleasure. This isn’t a religion. But we should argue on principle, too, and apply the same scientific standard to everything we do.

I don't think anyone from Harman is claiming that the IE Target Curve matches the response of an anechoically flat loudspeaker in a room. In fact, the bass response exceeds what the

was tested against 30 IE headphones including the ER4P. The listeners included
End consumers are not getting a better, more accurate product from all this. They never were. The curve was/is designed to find out what people "liked" and let's face, what people "like" and what's actually uncoloured accuracy are completely different things.

The so-called "harman" curve is no different to a 1970s loudness button on a receiver or amplifier. Sounds big, fat and fun at low levels, but absolutely horrid at anything approaching medium levels. There's a reason loudness contours fell out of fashion, and so will this slavish adherence to an arbitrary brand's key selling point of difference.

There are many false assumptions in this video:

1. "That Studio Monitors are not flat and rolled off" This is only true if are talking about the sound power response, which gently rolls off on most direct radiator speakers because of increased directivity with frequency.

Most studio monitors are flat on-axis and within the listening window, and this 100% determines the direct sound of the loudspeaker for someone sitting in that window. Speakers that have flat sound power do not generally rate well in controlled listening tests because to achieve this the direct sound has to be boosted at high frequencies to compensate for the rolled-off sound power. Also, the sound power by itself has very little predictability of the perceived sound quality of the speaker, as we have shown. The Consumer Reports loudspeaker "accuracy score" from which the Etymotic accuracy scores were derived is complete bunk to the point where they have discontinued using it ( see https://seanolive.blogspot.com/search?q=Consumer+Reports). It may be more valid for headphones but certainly not for loudspeakers.

2. The notion that all recordings are bright because the studio monitor speakers do not have flat power is pure speculation with no published scientific proof to back it up. Yes, most studio speakers do not have flat sound power but that does't not prove that studio engineers are cranking the treble on their microphones and mixes to compensate for it.. And if they do, they could be compensating for their hearing loss.

In fact, most well-made recordings are neutral when listened to through loudspeakers similar in FR to studio monitors (ie flat on-axis/listening window).

3. Amir has already pointed out that the Academy X-curve has been shown to be seriously flawed in recent publications by Drs. Floyd Toole and Linda Gedemer whose PhD research focused on that topic. To even mention it here as support for accurate sound, indicates the person has not carefully read the literature.

4. We included the Etymotic ER4P as one of the 30 headphones tested against the Target . The 70+ listeners (about 1/2 of whom are trained) on average did not rate it as high as the Harman Target (the results are below). The 2nd graph shows results of trained vs untrained listeners and we see the ranking of the headphones did not change much due to training.

A common complaint was that the Etymotic sounded thin in the bass, and/or a bit bright. I've owned many pairs of Etymotic headphones over the years, and had many discussions with Mead Killion (a great guy for whom I have a lot of respect) about the target. While these headphones are otherwise great, IMO they generally sound thin and I have always had to equalize the bass for stereo recordings. Floyd and I even participated in experiments Mead ran in which different versions of their target were rated. We were never told the results, and they were not published as far as I know. After these tests the company line became, the Etymotic target had to be modified to compensate for recordings because most are too bright due to studios using dull speakers (because they don't have flat sound power).. Consumer Reports acknowledged the fallacy of flat sound power speakers being accurate -- but apparently Etymotic stands firm in denial of this.


1631558007129.png

1631558168295.png
 
Last edited:
OP
Sharur

Sharur

Senior Member
Joined
Apr 10, 2021
Messages
447
Likes
150
I don't think anyone from Harman is claiming that the IE Target Curve matches the response of an anechoically flat loudspeaker in a room. In fact, the bass response exceeds what the

was tested against 30 IE headphones including the ER4P. The listeners included



There are many false assumptions in this video:

1. "That Studio Monitors are not flat and rolled off" This is only true if are talking about the sound power response which gently rolls off on most direct radiator speakers because of increased directivity with frequecy

Most studio monitors are flat on-axis and within the listening window, and this 100% determines the direct sound of the loudspeaker for someone sitting in that window. Speakers that have flat sound power do not generally rate well in controlled listening tests with trained/untrained listeners because to achieve this the direct sound has to be boosted at high frequencies to compensate for the rolled off. The Consumer Reports loudspeaker "accuracy score" from which the Etymotic accuracy scores were derived is completely bunk to the point where they have discontinued using it ( see https://seanolive.blogspot.com/search?q=Consumer+Reports). It may be more valid for headphones but certainly not for loudspeakers.

2. The notion that all recordings are bright because the studio monitor speakers do not have flat power is pure speculation with no published scientific proof to back it up. Yes, most studio speakers do not have flat sound power but that does't not prove that studio engineers are cranking the treble on their microphones and mixes to compensate for it.. And if they do, it could be they are compensating for their hearing loss.

In fact most well-made recordings are neutral when listened to through loudspeakers similar in FR to studio monitors (ie flat on-axis/listening window).

3. Amir has already pointed out that the Academy X-curve has been shown to be seriously flawed in recent publications by Floyd Toole and Dr. Linda Gedemer whose PhD research focused on that topic. To even mention it here as support for accurate sound, indicates the person has not carefully read the literature.

4. We included the Etymotic ER4P as one of the 30 headphones tested against the Target and the 70+ listeners (have of whom are trained on average did not rate it as high as the Harman Target (the results are below). The 2nd graph results of trained vs untrained listeners shows the ranking of the headphones did not change much due to training.

A common complaint was that they sounded thin and a bit bright. I've owned many pairs of Etymotic headphones over the years, and had many discussions with Mead Killion (a great guy for whom I have a lot of respect) about the target. While these headphones are otherwise great,IMO they generally sound thin and I have to equalize the bass for stereo recordings. Floyd and I even participated in experiments he ran in which different versions of their target were rated. We were never told the results, and they were not published as far as I know.The company line became, our headphones sound accurate but most recordings are too bright to compensate for dull speakers (because they don't have flat sound power).. Consumer Reports acknowledged the fallacy of flat sound power speakers but apparently Etymotic stands firm.


View attachment 153136
View attachment 153137
ER4P does not equal ER4S though? How was insertion depth ensured for ideal bass response?
 

Sean Olive

Active Member
Audio Luminary
Technical Expert
Joined
Jul 31, 2019
Messages
200
Likes
1,686
How many times must it be said that ER4P does not equal ER4S? How was insertion depth ensured?
Based on that question, you clearly have not read the papers. And I have tested and owned both models.
 
OP
Sharur

Sharur

Senior Member
Joined
Apr 10, 2021
Messages
447
Likes
150
Based on that question, you clearly have not read the papers. And I have tested and owned both models.
Was ER4S mentioned in the papers? What dBA were the tests conducted at?
 

Sean Olive

Active Member
Audio Luminary
Technical Expert
Joined
Jul 31, 2019
Messages
200
Likes
1,686
Was ER4S mentioned in the papers? What dBA were the tests conducted at?
It's in the paper, if you read it.
https://www.aes.org/e-lib/browse.cfm?elib=19237

"2.7 Relative and Absolute Playback Levels The relative levels of the virtual headphones were matched according to ITU-R BS 1770-4 loudness model [18]. The authors then verified the relative loudness through informal listening. The absolute equivalent diffuse field level was 82 dB (slow, C weighted) averaged over the duration of the sound loop."

This is a reasonably loud but still comfortable level that is well above the low levels where loudness contours come into play,

The FR of the 4s and 4p are pretty identical up to 1 kHz (flat) and differ slightly in gain above that. The 4S has more HF and if anything would be discerned as brighter than the 4P. The issue is mostly the lack of bass shelf.

Including the 4S in the sample of 30 headphones vs the 4P would not have significantly changed the results of the listening tests because this magnitude in FR is not significant compared to the range of FR tested.
 
OP
Sharur

Sharur

Senior Member
Joined
Apr 10, 2021
Messages
447
Likes
150
I don't think anyone from Harman is claiming that the IE Target Curve matches the response of an anechoically flat loudspeaker in a room.
Well, that is what this thread is about. Is the Harman target not supposed to reproduce the timbre of anechoically flat speakers in a treated room in headphones?
 

Sean Olive

Active Member
Audio Luminary
Technical Expert
Joined
Jul 31, 2019
Messages
200
Likes
1,686
Well, that is what this thread is about. Is the Harman target not supposed to reproduce the timbre of anechoically flat speakers in a treated room in headphones?
That is where we started. But after you adjust the speaker to an in-room target curve and match the headphone to that target they do not sound the same. And the difference is even greater between AE/OE and IE headphones.. Psychoacoustics comes into play. We found the same to be true when sitting in a car comparing the the audio system to a binaural capture/reproduction of it through headphones... The perception of bass is not the same.

And others have reported this too, including researchers from Fraunhofer. Sadly, they never published their Target Curve but the fact that they required experts to tune the headphone to match the speaker suggests that the DRP measurement of the speaker was not a good match. They also found that the DF and modified DF (Lorho) responses were not a good match. What do they have in common? No bass shelf (like the Etymotic headphones).

https://www.aes.org/e-lib/browse.cfm?elib=16482
 
Last edited:
Top Bottom