[castleofargh]

I am very puzzled about "Harman target curve" - to me that sounds really bad and I am not sure why is so popular.
I have applied compensation curves with AutoEQ and various results from here: https://github.com/jaakkopasanen/AutoEq/tree/master/results
I have used Grado SR60i, AKG K701, HIFIMAN Deva Pro for those tests. The results were strange with all of them, for me mids were too strong and I was bothered especially by the high-end loss.
I know about the history, but I am not convinced of the "science" behind it. Mainly because is based on "a panel of 10 trained listeners" that were not specified.
Education and genre preference has a lot to do with how music is perceived by the brain. All those "listeners" seemed to prefer the female voice and really hated the cymbals/high hat.
To me that's not good at all...

Am I alone in this?

1/ The Harman curve for headphones was the result of a dozen different papers over several years, not 10 trained listeners(which were their own people). the link to his blog shows the beginning of their work on headphones, just looking at more recent articles on the same blog would clearly show that. They tested for ages, on different continents, they gave people some simplified bass and treble EQ to play with, they tested some objective hypothesis, they tried to simulate one headphone on another one to remove the risk of people recognizing the headphone or being biased in some way by the comfort or weight... I think the first target came out from 250 listeners, and the all series of experiments probably goes beyond a thousand listeners.
It's by far the most involved and solid series of studies on a preferred headphone FR that were publicly(at least for a time) available.

2/ The result is a statistical one. Nowhere did they claim that it would or should please you or me in particular. When cool dude Olive discussed one of their headphone following the target pretty tightly, he had stats on how many would prefer it over something else, and that was 64%. It leaves plenty of room for you and your personal taste.

3/ The curve is from a specific measurement rig(that basically nobody posting graphs on this forum is using), and Autoeq is an aggregate of graphs that Jaako collected online. So it wouldn't be at all surprising to find out deviations from target of varying amplitudes on most compensations. It is also going to be the case for most reviewers showing the Harman curve or using it to compensate their graphs. They may, or may not use the right curve for their system. Even then, the pairs of headphones you're using probably has a few dB here and there that differ from the pair measured(+pad wear+...).

4/ Listening level(cf. equal loudness contour), age and hearing damage.

5/ Just personal taste from a lifetime of more or less conventional experiences.

I am curious... Would it be possible to know what's your age (as a range if you want)? Maybe that makes the difference?

I'm 45 and I approve this message.. I mean, my preferred response is pretty much right between DF and Harman, so both were a very good starting target for me that I would just fine-tune with EQ. The amount of bass I like, changes greatly depending on the type of headphone and I'm guessing the quality of the seal on my head(Something clearly mentioned by Olive). I have some intuition that the wider the driver, the less I feel the urge to get boosted low freqs(for whatever reason).
All in all, I guess you can interpret my case as validation that the Harman curve is pretty good for me, and also that it isn't. I'm not difficult.

 

[SoNic67]

I am of the same age and like the highs/treble un-corrected. I always liked the treble either slightly boosted or on zero, since I was a teenager listening to LPs and reel-to-reel.
I can't stand the "Harman" curve because it takes a dive after 10-13 kHz, especially when compared with the boosted mids.
If the recording engineers EQ that program in one way, why do we find that "people" prefer a different EQ?
I get the bass-boost, if the headphones have an attenuation in that region, especially open-back versus closed-back. But not liking the treble?
Is that because they listen mostly to compressed streaming formats (that usually cut that region) and when they hear those treble, they find them "un-natural"?

It's kind of mind boggling to me.

 

[sander99]

If the recording engineers EQ that program in one way, why do we find that "people" prefer a different EQ?

Recording engineers mix music for speakers. Because there is no one unique objective flat headphone curve for everyone they can not really do anything else.

Headphones skip a part of your HRTF, loudspeakers don't, and sounds in the real world coming from a distance don't. That's why there exists no objective neutral for headphones, but objective neutral for loudspeakers does exist (although the latter depends on the room as well as the loudspeaker).

If you were talking about any other audio component, even loudspeakers, but not headphones then everything you say sounds reasonable.
Only with headphones there is one complication: there is no objective flat or neutral for everyone.
If you are listening to sounds coming from your environment, from a certain distance, for example to the drum you mentioned then the sound is subjected to your personal head related transfer function (hrtf) filtering. (The sound is bouncing off your torso, bending round your head, bouncing of your pinnae, bending into your ear canal, etc.)
The resulting filtering of the audio differs per person. If you use headphones a part of this filtering is skipped. Because the part that is skipped differs per person one and the same headphone will sound objectively different to different people.

 

[SoNic67]

Recording engineers mix music for speakers.

So my Klipsch reference speakers have less treble than my headphones? Or more mids? That's why I have to apply that strange 20dB curve on headphones (that accentuates mids and lowers the highs)?
Also HTRF is the same for over-ear headphones and speakers, because it's the same ear (not IEM's). As for bouncing on walls and body... that is not considered while mastering, don't they use they use near-field monitors?

Makes no sense to me. Anyway, what do I know?
I just tested the Harman target curves and, to me, they sound bad.

 

[sander99]

Also HTRF is the same for over-ear headphones and speakers

HRTF is different for different directions. And that by the way gives the brain important information about the direction of sound.
For most people blowing highs straight into the ears from the side will result in more highs reaching the eardrums compared to highs coming from 30 degrees left or right from center (dead ahead).
Sound bouncing of the walls is part of room acoustics, not HRTF.
Sound bouncing of your own head is part of your HRTF.
Sound engineers by the way anticipate a bass bump in the consumer's playback system that often is there due to room acoustics, they use a monitor system with a bass bump.

 

[hakunamakaka]

Curious how they come up with 64% of people preferring harman curve. It can be very dependant on the music genre that person listens. Matching different set of headphones is very limited and in my view is only possible with the ones that sounds similar to begin with

 

[ST33L]

Great thread! Thanks

 

[SoNic67]

I am listening to more of classic/prog rock and pop music. I can't imagine how people can enjoy the cymbals muted, it's like something is missing.
The HRTF argument, IMHO, is just that - an argument.
I don't hear more highs from my headphones than from my Klipsch R-51M speakers. Or should those speakers be muted on treble part too?
My theory is that people's brains get used to certain equipment response curve and then, when presented with something better, they feel that they need to bring it down to their brain comfort level, where they are familiar with. I guess for me, that "average" of opinions doesn't cut it. It would be interesting to know what was their home equipment and music sources.

PS: My 2 cents, not a scientist. Listening to Deep Purple's "Smoke On The Water" right now. "Child In Time" is next". FLAC.

 

[bigshot]

I am curious... Would it be possible to know what's your age (as a range if you want)? Maybe that makes the difference?

I'm 62, but that wouldn't explain your description of what Harman sounded like to you. I can hear mids as well as you can, and the balance in the mids is what makes Harman so good for me. That's the most important part of the frequency range to get right.

It could be that the music you're listening to has recessed high end and needs boosting. Analog tape back in the 70s rolled off at the top end and LPs rolled it off further. Try listening to a few good DDD recordings and see if it is still a problem.

 

[Lunatique]

Personally, I find the Harman Curve a bit too bright and can get uncomfortable in the upper-mids. This is why I only use it as a basic starting point, but then EQ according to how loud each frequency band sounds, which then tailors the curve to my physiology and hearing. This is the best approach IMO, because we all have different hearing. For example, you might be more sensitive to, let's say, 8 KHz, while I might have a bit of hearing loss in that range, so we should EQ that region according to how loud we hear in that frequency band, and we'll both end up with the most accurate/neutral sounding result for us individually, although they would be different EQ curves.

 

[SoNic67]

I can hear mids as well as you can, and the balance in the mids is what makes Harman so good for me.

I get the correction of mids, if the headphones are not that linear in that portion. But the deliberate boost of mids is not so good. I can hear well enough without that hump.

Try listening to a few good DDD recordings and see if it is still a problem.

It's not that, and even if it was, why is the Harman attenuates that portion even more, at -10dB???
I am using Foobar2000 and it's spectrum shows the bandwidth. Hard to screen shoot... but here are a couple:

 

[castleofargh]

I am of the same age and like the highs/treble un-corrected. I always liked the treble either slightly boosted or on zero, since I was a teenager listening to LPs and reel-to-reel.
I can't stand the "Harman" curve because it takes a dive after 10-13 kHz, especially when compared with the boosted mids.
If the recording engineers EQ that program in one way, why do we find that "people" prefer a different EQ?
I get the bass-boost, if the headphones have an attenuation in that region, especially open-back versus closed-back. But not liking the treble?
Is that because they listen mostly to compressed streaming formats (that usually cut that region) and when they hear those treble, they find them "un-natural"?

It's kind of mind boggling to me.

I believe the state of things above 10kHz is mostly "mehhh, whatever". First, the body of research has mostly been done on a standard that wasn't accurately simulating human ear above that frequency(not even a statistical averaging of one). Another problem is how high frequencies(short little things), are easily affected by the smallest physical change, from placement of the headphone to ear shape.
Second problem: What is your idea of neutral sound on those graphs? spoiler, a flat line is not it.
For us random consumers, to get around those problems and stop going for the false intuitive answers that cripple the audiophile hobby, we need to go read a all lot, or to see our headphone(ideally, our own pair) measured on that rig. Then it's much easier to at least correlate the graph and how we feel about it. But that brings me back to my previous post and the variables we don't control or don't know where to find the references used.

The good news is that if you enjoy more treble, you're 100% free to add more.
Just to be sure, @jaakkopasanen did you strictly apply the delta between a graph and the Harman target at all frequencies?

Curious how they come up with 64% of people preferring harman curve. It can be very dependant on the music genre that person listens. Matching different set of headphones is very limited and in my view is only possible with the ones that sounds similar to begin with

It's from this paper I think:

Segmentation of Listeners Based on Their Preferred Headphone Sound Quality Profiles​

From memory they end up with 64% out of 130 listeners being happy with the target(because it's the one they picked as best within X headphones?), and like 20% picked others but are grouped as basically wanting less bass because that's what the trend shows for the FR of the headphones they preferred. while the rest is said to want more bass(the smaller group but for similar reasons).
I wouldn't bet money on me and that summary being entirely right, but that's how I seem to remember it.

 

[SoNic67]

What is your idea of neutral sound on those graphs? spoiler, a flat line is not it.

Why not? I exposed above my considerations: Flat on headphones sounds closer to my speakers than Harman target" corrected.
I don't see the point. So some random people "voted" that this "sounds better" and no we are taking it as gospel? People that for all purposes might have been used to listen to Spotify free, with iPods or Beats?

 

[sander99]

I don't hear more highs from my headphones than from my Klipsch R-51M speakers. Or should those speakers be muted on treble part too?

You seem to have missed the point: No, those speakers should not be muted in the treble part, the treble is attenuated by bending around your head into your ears (if the speakers are positioned at about 30 degrees or less left and right of center). The headphones blow the highs straight into your ears from the sides, so to compensate the headphone's frequency response rolls off in the highs. At least that is on average and approximately what happens. It is entirely possible that for you as an individual something else is happening, and that the harman curve is not suitable for you.

I don't hear more highs from my headphones than from my Klipsch R-51M speakers.

Ideally you should hear the same amount. But not everyone can achieve that with the same headphones (not without additional EQ).

 

[bigshot]

I have a really good speaker system, and Oppo PM-1s, which are very close to Harman. The Oppos don't sound like my speaker system (no headphones do) but they have the same general response balance. My only deviation from Harman is a -2dB cut from 2-3kHz. That frequency range is where human hearing is the most sensitive. It also might vary a lot from person to person, I don't know. But try doing a small cut right in there and see if the flinch response goes away.

Klipsch speakers are horn loaded and have a tendency to project upper mids and some of the treble a bit more than most speakers. But I doubt that is affecting your comparison much.

I just thought of something... What measurements of your cans did you use to calibrate to Harman? Could the curve you used be compensated?