• 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!

IEMs equalized to Harman IE target sound total different from AKG K371 ,or any other over the ears Headphones EQ to Harman OE target.

ObjectAudio

Member
Joined
Jan 22, 2021
Messages
50
Likes
70
I equalizing few IEMs that I have to Harman IE target , and they always sound total different from the Harman target poster boy headphone the AKG K371, or any other over the ears headphones I have that was equalized to Harman OE target.
With Sean Olive information that I was looking, I didn't see any mention that it was verified that that IEMs equalized to IE Harman target sound similar to over the ears headphones equalized to OE Harman target. If it was verified I will be happy to get this information, as to my ears they sound total different.
If any one tried it please share your experience.
 

Exprymer

Member
Joined
Apr 26, 2021
Messages
33
Likes
47
Yes, I have had this experience too. I KZ ZS10 I own when equalized to the Harman Target via the Wavelet app sounds a lot brighter than the k371. Could be a lot of factors, from QC to how our ears behave.
 

tomtoo

Major Contributor
Joined
Nov 20, 2019
Messages
3,607
Likes
4,514
Location
Germany
Thats very interesting. Is there a direction where they differ?
It would be very interesting for me to hear more about your impressions?
 

tomtoo

Major Contributor
Joined
Nov 20, 2019
Messages
3,607
Likes
4,514
Location
Germany
Yes, I have had this experience too. I KZ ZS10 I own when equalized to the Harman Target via the Wavelet app sounds a lot brighter than the k371. Could be a lot of factors, from QC to how our ears behave.

I eqed that pana iems to harman target, they did sound bright to me.

Maybe iam not alone?
 

DVDdoug

Major Contributor
Joined
May 27, 2021
Messages
2,920
Likes
3,831
The K371 doesn't perfectly match the target and EQ'd headphones IEMs probably aren't perfect either.

It also depends on how they fit in/on your ears. And your perception changes with volume changes and it's probably impossible to level-match headphones with IEMs.
 

bigjacko

Addicted to Fun and Learning
Joined
Sep 18, 2019
Messages
721
Likes
359
I feel the same. My ER4SR not eqed sound the same as over ear headphones eqed. After eqing ER4SR the sound is totally different.
 

julian_hughes

Addicted to Fun and Learning
Joined
Aug 23, 2020
Messages
657
Likes
901
There's always the distinct possibility that applying an eq curve to any headphone or IEM doesn't actually result in the expected curve. Everyone measures headphones, then calculates a curve. But who then measures again to see how that combination now measures? I've never, ever seen anyone publish this type of verification. Also when I've used those apps/tools which supposedly make one headphone's FR response the same as another, they still sound very different indeed. Another thing to consider is that many headphones and IEMs have impedance which varies, some extremely, so will never sound how they measure unless you're using a playback device with similar output impedance as the tester. Really this whole culture of trying to manipulate FR curves is in large part a crock of ess aich one tee. You can sometimes correct some faults but you can't in fact turn a KZ boom-ting-boom-fart-&-scream IEM into a Campfire Andromeda or a Sennheiser HD800.
 

Exprymer

Member
Joined
Apr 26, 2021
Messages
33
Likes
47
Yes, that's the QC part of the issue. When I bought my akg k371 I bought them expecting they sound as close as possible to the sample amir reviewed or as the sample people used to measure and generate a correction IR. In my case, the IR does really seem to balance it out, but how is the response after the correction is an incognita.
With the KZ ZS10, which I bought because it was the cheapest option ranking high on the auto eq GitHub repository, the correction seems to overcompensate the lack of highs, making me consider the possibility that the response before IR is not too close to the sample used for the correction. I have to test the correction with other IEMs to see a trend of brighter characteristics of the Harman Target in IEMs. ( Which I remember is a matter of taste).
Could be how our ears behave at different pressures, once the IEMs basically seals the ear canal.
 

solderdude

Grand Contributor
Joined
Jul 21, 2018
Messages
15,891
Likes
35,912
Location
The Neitherlands
But who then measures again to see how that combination now measures? I've never, ever seen anyone publish this type of verification.

Oratory 1990 does exactly this in all his pdf's.

as to my ears they sound total different.
If any one tried it please share your experience.

It's just more obvious between IEMs and over-ears.
This is also the case between 2 headphones both EQ'ed to the same target on 1 specific test fixture that kind-of resembles an 'average' ear.
An average ear does not exist.
Even 2 B&K test fixtures that both adhere to certain standards can give substantially different measurement results on the exact same headphone and will differ in a different way when another headphone is measured on both test fixtures.

The conclusion you can walk away with is that a measurement gives 'a test result' and that test result is obtained in a specific way that is supposed to be very close to your and my hearing but secretly isn't. And that's what certain people EQ on believing it is based on an exact measurement.
The measurement is exact but only to that specific test fixture in 1 or ' averaged' conditions.
 
Last edited:

julian_hughes

Addicted to Fun and Learning
Joined
Aug 23, 2020
Messages
657
Likes
901
Oratory 1990 does exactly this in all his pdf's.
...
Thank you for the info, I'm having another look at his stuff. Previously I have used Oratory's plots via AutoEQ and in the end thought that, for example, my Massdrop Plus Universals (from a source with output impedance of aprox 1 or lower) sound much, much better just as they are than when "corrected". I can't help thinking that if your headphones need 5 or 7 or 10 or 15 dB correction at different frequencies then either you bought some really lousy headphones or there is some serious flaw in the assumption that FR exists by itself as an indication of sound quality and can be predictably manipulated without regard to other factors. This stuff like AutoEQ and Morphit is great and very interesting but adjusting the FR simply does not do what people think it does. You cannot make a Shure SE215 sound like a Koss KSC75 and you cannot make a Pinncale P1 sound like a Sennheiser HD*whatever*. Maybe the FR curves look similar but the sound is still very, very different and usually much less good than the original item unfiltered.

I have a cunning trick. Never fails. It's great. I use headphones and IEMs I like the sound of. Yes, I got some of that Einstein DNA in me.
 

solderdude

Grand Contributor
Joined
Jul 21, 2018
Messages
15,891
Likes
35,912
Location
The Neitherlands
EQ'ing to a target measured on a specific rig that may come close to reality with certain headphones will give an improvement most certainly when the actual response is deviating a lot from the target.
For that reason in most cases there will be an audible improvement for headphones compared to no EQ.
It just isn't a guarantee that the used target is similar to your personal target or someone else's target nor reality.
The science isn't as exact as what HATS manufacturers and owners that spend a small fortune on them would care to admit.

Still I believe it is best to look for a headphone that already sounds good and then look for as many different measurements you can find on different fixtures, look for common deviations in all these measurements and EQ only the biggest deviations they have in common.
Chances are that EQ is more accurate than just relying on one type of test fixture.
That's why generated EQ by different masters of EQ is never the same but do have some recurring deviations in all of them.
 
OP
O

ObjectAudio

Member
Joined
Jan 22, 2021
Messages
50
Likes
70
I will add that the EQ of the IEMs is done on IEC711 rig, and I don't believe the issue is some headphones variation in production. The differences is too significant to my ears especially in the 1-3Khz region.
It also looks like that companies like Sennheiser, Westone , Shure , Ultimate ears all of them usually tuned their IEMs with less pinna gain in the 1-3Khz are match closer in sound to over the ears headphones targeted to Harman eq.
 
OP
O

ObjectAudio

Member
Joined
Jan 22, 2021
Messages
50
Likes
70
Reading Harman information slide "Relationship between headphones reference and amplitude response"
There are three groups of preference ratings:
1.6 to 9 - This group with the lowest preference', all the IEMs tested have big bass boost and most of them around 10dB Pinna gain.
30 to 49 - This group the mid one most of them have too much bass boost or they don't have any bass boost.
50 to 70- The highest rating group. The most balanced group ; all the IEMS have bass boost around 3-7 dB that is still less than the suggested Harman target of 10dB, and all of them have pinna gain around 3 to 7 dB that is less than the Harman suggested 10dB boost.

The higher preference IEMS in this test are agreeing with my preference and what to my ears sound similar to over the head headphones EQ to Harman target ; less bass boost than what Harman IE suggesting, and less pinna gain in the 1-3Khz than what Harman target is suggesting.



1633527210721.png
 

solderdude

Grand Contributor
Joined
Jul 21, 2018
Messages
15,891
Likes
35,912
Location
The Neitherlands
all the IEMs tested have big bass boost and most of them around 10dB Pinna gain

IEM's have no pinna gain, there is no pinna in play.
Only (partially) an ear canal.

The plots you posted are over-ear headphones and only shows most people prefer headphones close to the Harman target.
 
OP
O

ObjectAudio

Member
Joined
Jan 22, 2021
Messages
50
Likes
70
IEM's have no pinna gain, there is no pinna in play.
Only (partially) an ear canal.

The plots you posted are over-ear headphones and only shows most people prefer headphones close to the Harman target.
Both of your statements are incorrect.
The graph is in the IEMs presentation section (slide number 107) starting in slide number 77 and ending in slide 116 on the Harman presentation PDF.
The pinna gain region in IEM (1-3Khz) is trying to compensate for the non existence of pinna and the gain the it present to speakers and over the ears headphones.
 

solderdude

Grand Contributor
Joined
Jul 21, 2018
Messages
15,891
Likes
35,912
Location
The Neitherlands
You mean compensation for the lack of concha and part of the ear canal ?
Usually people refer to pinna gain as the acoustical gain the pinna has for incoming signals.
Of course both pinna and concha gain as well as part of the ear canal has to be compensated for in order for IEM's to sound realistic.

In other words my pinna gain is not your pinna gain.;)

The plots are indeed for IEM's (my bad) as the concha and ear-canal gain are visible (the pinna influence itself is marginal) simulating an angle speakers would have.
 
Last edited:

Lunafag

Member
Joined
Apr 14, 2021
Messages
89
Likes
127
The pinna gain region in IEM (1-3Khz)
It's a common misnomer. Pinna makes a dip at 8kHz and a boost around 11kHz, the boost at 2.5k comes from the ear canal itself.
As for the topic, Harman target calculates preference rating. It's made for a particular rig and when headphones are measured on that rig it tells which headphones people will prefer more. It doesn't even reflect what I'm hearing as peaks above 3k on all the headphones I've tried never seemed to line up with what what was measured. And even if you measure IEMs on the same rig there are still psychological factors that will make them sound different from headphones, and even open-back IEMs from regular IEMs(I had to shift the bass shelf way back to 50Hz on my gl20 for them to sound natural even though 3 sources confirmed it's ruler-flat in bass). Those factors are still up to debate but Harman IEM and headphone targets deviate from each other so people do seem to prefer different signatures in different form factors. Will we ever be able to make headphones and IEMs sound the same? I doubt it will happen any time soon.
 
OP
O

ObjectAudio

Member
Joined
Jan 22, 2021
Messages
50
Likes
70
For simplicity lets call the 1-5Khz region the presence region.
My point is that very respectable IEM companies like Sennheiser, Westone , Shure and Ultimate ear have different IEM tuning target than the Harman IE in mind, and I think their target is closer in sound to over the head headphones Eq to Harman target.
I am sure they doing also their homework about what is the preferred IEM sound , and there is some similarity between them for the tuning especially in the presence region.
For reference some of these companies top of the line IEMs compared to Harman IE target:

Sennheiser top of the line IE900:

1633537028066.png

Ultimate Ears 18 PRo
1633537207503.png


Shure 846

1633537319998.png
 

charleski

Major Contributor
Joined
Dec 15, 2019
Messages
1,098
Likes
2,239
Location
Manchester UK
less bass boost than what Harman IE suggesting, and less pinna gain in the 1-3Khz than what Harman target is suggesting.
The one thing that really stands out from those graphs is that they all have a downward slope of 5dB or more from 200-1000Hz, whereas the diffuse-field measure on which the Harman target is based has an upward slope. Given that this region is where the fundamentals of almost all musical tones lie, I think it makes more sense to worry about getting that right first before we obsess over the bass and overtone regions. This is also the part of the spectrum where there's the least variability in measurements at the eardrum:
Headphone simulation of free-field listening. I- Stimulus synthesis - J Acoust Soc Am 1989 - W...png

[Wightman & Kistler 1988]

Unfortunately this is a common failing of IEMs, thought the graph of the Shure you posted shows far better tracking in this region.

Frankly, though, I'm coming to the opinion that both headphone and earphone measurement techniques are still a bit of a mess, with significant non-linear variations between different types of fixture. Blindly EQing to match a curve is never going to produce an optimal result, and people need to be prepared to alter the EQ as needed.
 

MayaTlab

Addicted to Fun and Learning
Joined
Aug 15, 2020
Messages
924
Likes
1,512
significant non-linear variations between different types of fixture.

I agree to a degree, but I wouldn't put all headphones types in the same basket in regards to that degree for various ranges of the FR spectrum. No idea about IEMs, but I suspect that there is quite a significant difference between, let's say, a pair of HD650, K371 and Bose QC35II in regards to how they vary across fixtures / individuals for different parts of the FR spectrum. I would expect, for example, the K371 to show a lot more variation at lower frequencies than the other two because of sealing issues mostly, or the QC35II to show a lot more variation in the 1-5kHz range (and nearly none below thanks to its ANC feedback system and good design).

Harman's research cleverly tested their over-ears targets on several open over-ears known / presumed for decently low individual to individual variation and decent human to test fixture translation to mitigate the question of HPTF variation and on all of these various headphones they tended to be more or less significantly preferred over other targets, so in the case of these types of headphones the variation wasn't too high to invalidate the trends.
 
Last edited:
Top Bottom