Harman preference curve for headphones - am I the only one that doesn't like this curve?

[watchnerd]

but headphones present special difficulties that make defining a "flat" response highly contentious in the first place.

@Ilkless summarised some of the difficulties and attempted resolutions in this article.

This is one of the reasons why I find dogmatic adherence to any particular preference curve for headphones to be, frankly, gross over-simplification bordering on ignorant fundamentalism.

 

[Robin L]

I have my weekend's stout in a reusable growler filled at one of the local breweries.

 

[ZolaIII]

I don't like the bass bost, I prefer it as flat as possible and less boost in highs. Find it less distorting that way (especially in low mids and male vocals while still being able to hear all it can be in lows) and of course less sibling. But I hear rather good. Both scales should go up if you don't and with more headroom towards lows boost. Contrary to this speakers simulated on the headphone would have roll-off both in highs and lows (distance - energy fading not the natural roll-off).

 

[Robin L]

Yeah, I don’t know if it’a due to measurement validity, or just my mood (I usually wear headphones when winding down, not getting amped up; that’s my speaker system’s job), but I tend to like a bit less bass than what the Harman curve would suggest; for sure not flat though.

The curve is a mix of seasoned listeners and general folk (after being screened and whatnot), and it’s not a surprising conclusion that those people that put 18” subs in their SUVs and blast out the neighborhood are likely to like more bass than the people that visit audiophile forums.

It's the audio equivalent of The Olive Garden restaurant.

I'd say more along the lines of a soul food/bbq restaurant, and I can get down with that.

I like big bass and I cannot lie . . .

 

[watchnerd]

I'd say more along the lines of a soul food/bbq restaurant, and I can get down with that.

I like big bass and I cannot lie . . .

The analogy wasn't with the cuisine type, but traditional (family owned Italian) / niche vs mass market compromises thereof (Olive Garden)

Think local soul food / bbq restaurant vs the McRib.

 

[Zensō]

I use EQ on a variety of headphones for both casual listening (Harman Curve via Roon) and music production (flat via Sonarworks 4). I find the Harman Curve enjoyable for casual listening, but the flat Sonarworks EQ is better for creating masters that translate well on various systems.

 

[Robbo99999]

Well, I'm a fan of the Headphone Harman Curve (I use the 2018 version), I think it sounds very natural, pretty much like my flat reference speakers. It's possible my speakers have a tad more energy in some places north of 1kHz, and it's possible I may experiment with tweaking parts of the Headphone Harman Curve to see if it can match my speakers closer. I have indeed experimented with tweaking the Headphone Harman Curve for my HE4XX headphones, as it's the headphone that sounds the least like my speakers in comparison to my other headphones (once it's been EQ'd to the Headphone Harman Curve), less energy around 1-3kHz I think, but it would require some more experimentation. My AKG K702 / NAD HP50 & Senn HD600 sound a lot closer to my speakers in tonality when EQ'd to the Headphone Harman Curve.

EDIT: during the course of my experimentation I created some EqualiserAPO configuration file EQ's that can flip between the 2018 Harman Curve to the 2015 Curve as well as the 2013 Curve, feel free to activate the following files in EqualiserAPO if you've already got your headphones EQ'd to the 2018 Harman Curve.....so you can see which one you prefer, and I even created the last one as an extra bonus which is a mix of the 2013 Bass with the 2018 Treble (find all attached).
Here's a graphical representation of the different Harman Curves:

 

[watchnerd]

Well, I'm a fan of the Headphone Harman Curve (I use the 2018 version), I think it sounds very natural, pretty much like my flat reference speakers. It's possible my speakers have a tad more energy in some places north of 1kHz, and it's possible I may experiment with tweaking parts of the Headphone Harman Curve to see if it can match my speakers closer. I have indeed experimented with tweaking the Headphone Harman Curve for my HE4XX headphones, as it's the headphone that sounds the least like my speakers once EQ'd to my speakers in comparison to my other headphones, less energy around 1-3kHz I think, but it would require some more experimentation. My AKG K702 / NAD HP50 & Senn HD600 sound a lot closer to my speakers in tonality when EQ'd to the Headphone Harman Curve.

I'm curious:

What's your biological age and geographic ancestry?

 

[blse59]

That's perfectly valid with respect to electronics and (to a limited extent) loudspeakers, but headphones present special difficulties that make defining a "flat" response highly contentious in the first place.

@Ilkless summarised some of the difficulties and attempted resolutions in this article.

What is the difficulty in defining a flat response? I would say the ety er4s has a technically and objectively flat response.

 

[Robbo99999]

I'm curious:

What's your biological age and geographic ancestry?

I'm 44 and white Caucasian, indigenous to the UK (England)!

 

[watchnerd]

I'm 44 and white Caucasian, indigenous to the UK (England)!

So you fit the expected preference profile based on age and likely ear shape.

 

[Robin L]

The analogy wasn't with the cuisine type, but traditional (family owned Italian) / niche vs mass market compromises thereof (Olive Garden)

Think local soul food / bbq restaurant vs the McRib.

I was thinking of 18" dual subs in a SUV or a van as a good thing.

I get what you're getting at. The pop music with really deep bass that has deep market penetration at the moment has absurdly exaggerated bass. On purpose. The AKG K371 serves that type of music very well. I'm glad I've got the 6XX Topping combo, it works better with anything acoustic. I'm using EQ to cut back the boom and turn up the bottom and top. However, even that's not 100%: some Baroque Chamber music has more of an edge with the AKG 'phones. For large ensembles the 6XX's are better, but there is some acoustic music that's more believable to these ears via the K371s, they reproduce more rosin 'n' pick.

I suspect if I had the bucks I'd get pricer headphones, but that would presume things going back to something approaching normal. Meanwhile, I'm happy with a cheapish set of headphones that adhere, more or less, to the the Harman curve, circa 2021.

 

[Robbo99999]

So you fit the expected preference profile based on age and likely ear shape.

I don't know what ethnicities it's been modelled on.

 

[Feelas]

Issue is it creates a thunderous sub bass (like an out of control sub in a speaker system) that muddies up the bass in many songs.

Only if the IEMs in question don't fare well in regards to boosted bass. If they do, they won't spray whole midrange with THD, alas it'll all be clean & tight. Common misconception.

The top selling beers in the world are:

#1 Snow
#2 Budweiser
#3 Tsing Tao
#4 Bud Light
#5 Skol

Ergo, a beer made the "right way" should taste like those pale lagers.

Now, someone will point out that beers are not reproducers, and rightly so. But it's illustrative of the dangers of equating "what sells best" with "what is right."

Not quite. You'd have to make a blind testing w/ 20 different kraft styles & popular beers and assess how much each of them are liked, and then choose the ones that have least std-dev from the profile each way. Harman was computed that way. Besides that, headphones used w/ preference testng were LCD-2s and different "end-games", the results showed that it doesn't matter. One of famous beer vloggers in Poland made a tasting video while suffering from COVID-19 (note: he mentioned that drinking while being sick is dangerous), when having problems with smell. What occured is that he had problems differing between stouts, lagers and IPAs, all of which are incredibly different, but when lacking smell it doesn't actually matter. The difference is mostly in expectance.

How pretentious to call anything popular bollocks. Prefering flat response on headphones or on speakers is not some sign of higher intelligence, it's just a matter of different perception, but I'd stray from saying that it's somehow better - there's simply no real reason to do so.

The pop music with really deep bass that has deep market penetration at the moment has absurdly exaggerated bass. On purpose.

But... pop music is produced in a way that'll have most presence in the mid/upper bass region, simply because most people don't own anything that goes under 40-50Hz or that is all unhearable (e.g. in cars)...

Guys, I'm not trying to be smart, but we're delving very, very close to straw man realms... I think it's better that we stop and try to concentrate on facts. The fact is that flat bass response is not something that happens in rooms. Just like Toole showed that people in general prefer flattest speakers in blind testing, we can estimate that the same is true for headphones. In fact, that was the reason why Harman worked on blind-testing rig for headphones.

 

[watchnerd]

I don't know what ethnicities it's been modelled on.

I don't know the populations used in the listening tests, but we know that various dummy heads have ears / face shapes / HRTFs that correlate more with particular ethnicities.

https://www.****************/wp-content/uploads/2020/04/Example-of-a-dummy-head-and-torso.jpg

vs

https://www.****************/wp-content/uploads/2020/04/A-Chinese-dummy-head.jpg


Discussed here.

 

[watchnerd]

How pretentious to call anything popular bollocks. Prefering flat response on headphones or on speakers is not some sign of higher intelligence, it's just a matter of different perception, but I'd stray from saying that it's somehow better - there's simply no real reason to do so.

No bollocks -- I like Budweiser just fine.

Although in that macro-lager genre I prefer Miller.

Beers of their style are my preference to go with pizza and taqueria food.

 

[andreasmaaan]

What is the difficulty in defining a flat response? I would say the ety er4s has a technically and objectively flat response.

Did you read the article I linked? I can explain, but doubt I can do a better job than the article.

This a raw response of the ER4S (one of many possible raw responses, each one dependent on measurement conditions):

In what respect is this "flat"? And why should it be?

 

[watchnerd]

Guys, I'm not trying to be smart, but we're delving very, very close to straw man realms... I think it's better that we stop and try to concentrate on facts. The fact is that flat bass response is not something that happens in rooms. Just like Toole showed that people in general prefer flattest speakers in blind testing, we can estimate that the same is true for headphones. In fact, that was the reason why Harman worked on blind-testing rig for headphones.

How do you define "flat" in a headphone context?

That's not trivial.....

 

[Robbo99999]

I don't know the populations used in the listening tests, but we know that various dummy heads have ears / face shapes / HRTFs that correlate more with particular ethnicities.

https://www.****************/wp-content/uploads/2020/04/Example-of-a-dummy-head-and-torso.jpg

vs

https://www.****************/wp-content/uploads/2020/04/A-Chinese-dummy-head.jpg


Discussed here.

Yeah, that top one, that's how I normally walk around!

 

[watchnerd]

Yeah, that top one, that's how I normally walk around!

He's a bit of a mouth-breather, isn't he?

I'm probably the same.