Pink Noise sounds boomy with Harman tuned headphones.

[Tangband]

For loudspeakers there isn't and can never be a single target as the response at the LP depends on the directivity of the loudspeakers, the reflectivity of the room and listening distance, the only reasonable target for loudspeakers above transition frequency is linear direct sound and smooth directivity, more about the topic here.

True.
About below the schroeder frequency:
our eyes sees the room and expect it to sound with a certain bass tilt. Ultimately, one need to listen to the cello or electric bassguitarr and judge the bass quality and maybe change it with a dsp or the placement of the two speakers. Pink noise is unfortunately a very bad tool to judge how the pitch of a cello will sound.

Also - our inner sound reference are always colored by the speakers we normally listen to - so the circle of confusion is very real, starting with the owners loudspeaker that might lack bass below 70 Hz . This is good to know when you listen to a new speaker in a store , it will take about an hour of listening in the new room with the new speaker to be somewhat free from your own ”at home loudspeaker sound ” bias.

This is something that true for all of us.

To make it even more complicated, and if we go further than Toole/Olive:s research, - two stereo loudspeakers in a room can sound very different because of the setup in the room . As an example - If you have the same room and the same loudspeakers but change the distance between the speakers with 1 meter, you gonna have a massive change in tonality in the bass region and the stereo Illusion will be very different.

 

[solderdude]

Yes, that will certainly help with speakers.
For headphones (OP) only 'bias', listening SPL, preference and proper seal would be affecting the tonal balance.... and of course the recording which in most cases has the biggest contribution to 'accepting instruments as realistic'.

 

[ZolaIII]

Come on guys read the:
Precise and Full-range Determination of
Two-dimensional Equal Loudness Contours conducted by Yôiti Suzuki and with great research team. It's time to stop with all this Harman nonsense.

 

[thewas]

True.
About below the schroeder frequency:
our eyes sees the room and expect it to sound with a certain bass tilt. Ultimately, one need to listen to the cello or electric bassguitarr and judge the bass quality and maybe change it with a dsp or the placement of the two speakers. Pink noise is unfortunately a very bad tool to judge how the pitch of a cello will sound.

It should be added also that it also depends on the recordings (see circle of confusion) and reverberation curve, for a example in a room with low bass reverberation a less tilt sound neutral.

 

[Maciek]

Amir would have to buy a mannekin (with its own issues/compensation challenges) and would have to build a 'reference reflective' room to do such measurements.
Klippel exists for a good reason. In ear measurements with a really small diameter mic that actually is truly reference is a problem just as it is to use real ears as all ears have differences in length, diameter, pinna and headsize as well. Who will be the standard .. the 'average joe', Amir's ears ?

Nah... speakers in Klippel or anechoic measured all the way around (not only on axis) can give more valuable insight in a speakers performance.

I wonder why nobody tried to compute the dofference between ear coupler used for measurement and "personal" ears. I mean, ear's transfer function im db/f is just a different way of measuring oiter and inner ear diameter, depth, height etc. So for example if spmeone has ears significantly larger, or an ear canal sognificantly wider than the one used for measurements, should for headphpnes that fall below/above the target in specific areas. This sort of understanding (translating physical dimensions of the ear to specific frequency ranges) would make it so much easier to evaluate headphones through graphs, with as much confidence as speakers...

 

[antoninomubarik]

I was wondering whether it was normal to hear boomy-ness when listening to pink noise with Harman tuned headphones?

Are they supposed to sound boomy because they’re compensating for lack of physical impact?

It is impressive to listen to and sounds great with a lot of pop recordings but too 'fat' for me with well made recordings.

 

[odyo]

Afaik, harman bass is elevated to compensate lack of bass impact on headphones. With speakers, you feel the bass with your whole body. Headphones are limited to your ear/head. So harman bass is generous to overcome this.
I like harman bass with music types that rely on sub bass like Lorn's albums. But in general i don't like harman bass as it kinda masks the agility of mid bass/punch. My current headphone does the sub bass well so when i feel like it, i boost the bass a lot.

 

[ADU]

Afaik, harman bass is elevated to compensate lack of bass impact on headphones. With speakers, you feel the bass with your whole body. Headphones are limited to your ear/head. So harman bass is generous to overcome this.

It's an interesting idea that has been around for quite some time. But the science to support it seems weak, at best imo.

For starters, the 711 measurement systems used by Harman to develop the target are not that reliable outside of the midrange frequencies. That alone should tell you to take the levels in the sub-bass and treble with a grain of salt.

Harman's 2018 over-ear target appears to be more or less consistent though with the measured response of neutral loudspeakers that are well-extended into the sub-bass, or accompanied with a sub-woofer, based on recently acquired spinorama data.

This is also something that I pointed out to Dr. Olive in another recent discussion about a year ago, and here was his response...

<<<I see no appreciable differences between your preferred in-room loudspeaker curve or PIRL (and the current Harman over-ear headphone targets based to some extent on this), and the estimated in-room responses of some of the best extended speakers in the sub-bass, which also have a reasonably flat/neutral on-axis response...>>

I agree. And that was the point of showing the anechoic spins of the Revel vs the in-room and headphone target curves that you show in the next post. The bass roll-off of the speaker only means that a subwoofer or EQ needs to be applied.

And here is another reply where he addresses the question of headphones with no elevation in the bass (or a flat in-ear response below about 100 Hz, or 1 kHz), which seems to be the basis for some other recently proposed "hifi" in-ear curves...

We have published several papers on headphone and loudspeakers that show that in-room loudspeaker and headphone target curves with flat bass contours below 1kHz are not preferred by most listeners whether they are Harman employees or not, trained or untrained( see https://www.aes.org/e-lib/online/search.cfm?type=elib)

The graph below taken from (https://www.aes.org/e-lib/browse.cfm?elib=17500) shows the results from a study where 238 listeners from four countries rated 4 headphones. HP1 is the Harman Target (2015) and HP2 is a headphone with flat bass. HP3 also has flat bass but also less energy at the ear canal region @ 3 kHz. Three of the listening groups included external listeners-- recording engineering/mastering students from Canada Harris College, USA Citrus College and Loyola Marymount University-- all preferred the HP1 (Harman Target ) over that flat bass headphones HP2 and HP3. These recording engineering students tastes were no different than those of the untrained Harman listeners.

I am a trained classical musician and formally trained Tonmeister and I do not prefer loudspeakers with a flat in-room response or headphones with flat bass below 100 Hz. The people who tend to like flat bass headphones are disproportionately represented by females and older listeners who are more likely to have hearing loss. Less bass and more treble tends to compensate for hearing loss making voices more intelligible

Whether listeners are from Harman are not is really irrelevant because all of the listening tests are conducted double-blind and listeners are unaware of the products brands and models being tested. Also, there is 25+ years of data showing Harman listeners' sound quality tastes are representative of external listeners tested outside Harman. The main factor influencing how they rate loudspeakers and headphones is whether they are trained or not. While both trained and untrained listeners tend to have similar tastes, the trained listeners are more discriminating and repeatable as the graph below indicates.

Contrary to what you say, many "saine" (sic ) people accept the results of studies done using Harman Listeners including audio scientists and engineers who peer review our papers for publication in scientific journals.


View attachment 170730

As markanini said above, Harman is not targeting the sound of bookshelf speakers with these curves. They appear to be targeting a speaker that's fairly neutral, but also very well extended into the sub-bass frequencies.

Some of the details and caveats in their studies, and the equipment used for them leaves the door open to question the accuracy and precision of some of their results in both the lower and higher frequencies though, imo.

 

[markanini]

For starters, the 711 measurement systems used by Harman to develop the target are not that reliable outside of the midrange frequencies.

The Harman curve is developed around GRAS 43AG and 45CA systems.

 

[odyo]

It's an interesting idea that has been around for quite some time. But the science to support it seems weak, at best imo.

For starters, the 711 measurement systems used by Harman to develop the target are not that reliable outside of the midrange frequencies. That alone should tell you to take the levels in the sub-bass and treble with a grain of salt.

Harman's 2018 over-ear target appears to be more or less consistent though with the measured response of neutral loudspeakers that are well-extended into the sub-bass, or accompanied with a sub-woofer, based on recently acquired spinorama data.

This is also something that I pointed out to Dr. Olive in another recent discussion about a year ago, and here was his response...



And here is another reply where he addresses the question of headphones with no elevation in the bass (or a flat in-ear response below about 100 Hz, or 1 kHz), which seems to be the basis for some other recently proposed "hifi" in-ear curves...



As markanini said above, Harman is not targeting the sound of bookshelf speakers with these curves. They appear to be targeting a speaker that's fairly neutral, but also very well extended into the sub-bass frequencies.

Some of the details and caveats in their studies, and the equipment used for them leaves the door open to question the accuracy and precision of some of their results in both the lower and higher frequencies though, imo.

I hope we will see newer harman standard with better measurement gear and maybe less smoothed target if possible.

 

[ADU]

The Harman curve is developed around GRAS 43AG and 45CA systems.

I believe these are both based on the 711 standard, as explained here by Jude...

 

[beagleman]

Not sure if it is just me, but the Harman Curve Target almost sounds like the sound I got when I first got a new subwoofer and had it dialed up too much, in my excitement to really hear loud deep bass.

 

[Tangband]

Not sure if it is just me, but the Harman Curve Target almost sounds like the sound I got when I first got a new subwoofer and had it dialed up too much, in my excitement to really hear loud deep bass.

Maybe you are used to listen to headphones/ speakers that have to little bass ? Youre not alone in this.
Some other people also complain about the Harman curve , that its sounding dull with to little treble. They are often used to listening with speakers that have +5 dB at 10 KHz ( very common with newer B/W speakers and others) .
Many microphones used in a studio has a slight rise in the frequency response from 4-10 Khz, sometimes more than 5 dB ….

The compression effects in the studio often amplify 2-10 KHz and this makes everything worse.

So , shortly - most of us are used to listen to a thin, treble forward sound that dont sound at all as the real thing.
We all need to sober up and use frequency linear gear.

 

[beagleman]

Maybe you are used to listen to headphones/ speakers that have to little bass ? Youre not alone in this.
Some other people also complain about the Harman curve , that its sounding dull with to little treble. They are often used to listening with speakers that have +5 dB at 10 KHz ( very common with newer B/W speakers and others) .
Many microphones used in a studio has a slight rise in the frequency response from 4-10 Khz, sometimes more than 5 dB ….

The compression effects in the studio often amplify 2-10 KHz and this makes everything worse.

So , shortly - most of us are used to listen to a thin, treble forward sound that dont sound at all as the real thing.
We all need to sober up and use frequency linear gear.

I wish that were the case, would make it a simple solution.

No, my set up is not tilted up treble at all.
I DO have other speakers in rotation that have that sound, so I know the difference for sure.
I have a Sub now, that is decent down to about 25-28 Hz and maybe just not used to strong low bass in a headphone. Not sure really.

 

[dasdoing]

bought IEMs, EQed to Harman-wo-bass-boost and was shocked initially (no bass). I think the boost makes sense for most because of equal loudness contours. we tend to listen quietly on headphones for two reasons: less background noise because of noise rejection and because we tend to think that high volume close to the ears is worse than in a distance. As soon as I turned the volume up the bass was there. As with loudspeakers, loudness compensation is the better solution.

 

[markanini]

bought IEMs, EQed to Harman-wo-bass-boost and was shocked initially (no bass). I think the boost makes sense for most because of equal loudness contours.

"IIt's not the room, you just need to turn it up"

 

[abdo123]

bought IEMs, EQed to Harman-wo-bass-boost and was shocked initially (no bass). I think the boost makes sense for most because of equal loudness contours. we tend to listen quietly on headphones for two reasons: less background noise because of noise rejection and because we tend to think that high volume close to the ears is worse than in a distance. As soon as I turned the volume up the bass was there. As with loudspeakers, loudness compensation is the better solution.

the harman curve was studied at 85dBSPL.

 

[trevintheaudioboy]

Tuning bass way up like Harman suggests just results in bloatedness. With drivers that close to your ear and not making contact with other body parts, there’s no way you can replicate good loadspeakers experience in headphone enclosure. Why do we feel the need to try to make headphones sound like speakers anyway? Why compromise bass accuracy and tightness to attempt to compensate for something headphones cant provide? I think in a few years Harman headphone target will be obsolete because we will accept that headphones are playing our ears and speakers are playing a room. Celebrate our differences, ya know? The thread title is a statement, and measurements (on this website and elsewhere) back it up. It sounds boomy because harman is boomy.

 

[MajoloBama]

To my Harman sounds like good speakers in a good room. The room being very important here. Of course speakers are better but the timbre, or character of speakers. That's what harman is about. Flat earphones sound really, really boring and weird.

 

[markanini]

Why compromise bass accuracy and tightness to attempt to compensate for something headphones cant provide? I think in a few years Harman headphone target will be obsolete because we will accept that headphones are playing our ears and speakers are playing a room. Celebrate our differences, ya know?

There are many inconvenient facts about headphones to get you red pilled still. Have you considered you might be biased by bass light headphones? Does an orchestra bass drum sound less true to life on a Harman tuned headphone or more true to life when doing an A-B test? Do you know about the many classic and modern headphones developed independently of Harman that happen to line up with the target for most of the frequency range? https://github.com/jaakkopasanen/AutoEq/blob/master/results/RANKING.md See the entries with a score of 85 and above.