• 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.

Why is audio so angry?

JohnYang1997

Addicted to Fun and Learning
Joined
Dec 28, 2018
Messages
872
Likes
350
#81
I don't think the SR is bass light so much as they are not bass boosted. I think that most tastes have been moving towards a bassier sound because of trends in mastering and also tuning of headphones in particular. Preference is a dynamic property which evolves, it is always interesting (generally disappointing) to revisit the tastes of earlier periods in life. I find that bass is a bit like salt in that if you're used to a high salt diet food with no salt or just a pinch of salt tastes bland and tasteless but once you adjust to a new taste then eating something high in salt just feels like the salt is overpowering everything else. The Harman research is a statistical analysis of preferences, and preference is a subjective parameter which is subject to change.
I agree.
 

JohnYang1997

Addicted to Fun and Learning
Joined
Dec 28, 2018
Messages
872
Likes
350
#82
Well, according to the studies I cited, the standard model genuinely doesn't have enough bass. This makes the bassier version objectively better, and I'm glad they released it.



They made assumptions about what the target response should be. When I bought my ER4PT 7 years ago it came with a nice card showing a custom frequency response measurement, hand-signed and everything, and it was quite flat according to their target (which was probably diffuse field or something). That shows that they care much more about objective performance, as opposed to listening by ear. The problem is, they used the wrong target, but they couldn't have known that at the time - the research wasn't there yet.
The problem is your ears are not trained properly to accept that type of sound. Also, insertion is important. Ear canal resonance at wrong frequency and poor seal will definitely kill the sound completely, because they use BA drivers. Use big tips and insert as deep as you can. And listen to it for a week.
Hearing preference is not born is trained.
 

edechamps

Active Member
Joined
Nov 21, 2018
Messages
174
Likes
408
Location
London, United Kingdom
#84
The thing is harman target sounds considerably worse than etymotic er4. There is no mean to choose 200hz and below as bass. Causing a lot of confusion. And the high frequency roll off is also problematic which is not the case of real world.
To prove, just use a pair of eqed studio monitors and use ear to eq the earphone. The end result will be much closer to etymotic than harman.
If you say so. With all due respect, I'm not going to dismiss double-blind, statistically rigorous studies because of your personal opinion :)

The process we used is use a collection of songs to define bass and mid range (probably 7k and below) and to use sine wave generator to eq 7khz + .
Can you provide a link to the details of this study?

I use to like harman and sean olive. But they proceed to make n5005 and n700 which sound horrible and don't even represent harman target at all.
Yes, that's a shame, but that's mostly because most of Harman doesn't seem to care about Olive's results, sadly. My hypothesis is that they're driven more by marketing than by science. Harman is a large company - left hand, right hand, and all that. Just because Harman makes some (arguably) bad products doesn't necessarily mean the studies that their researchers put out are worthless - these are different departments and they might not even talk to each other.

The problem is your ears are not trained properly to accept that type of sound.
What do my ears have to do with it? I don't care what my ears prefer for the purpose of this discussion. I don't discuss my personal preferences because this is "Audio Science Review", not "Audio Feelings Review". What I care about is results that are statistically relevant across the population as a whole.
 

JohnYang1997

Addicted to Fun and Learning
Joined
Dec 28, 2018
Messages
872
Likes
350
#85
If you say so. With all due respect, I'm not going to dismiss double-blind, statistically rigorous studies because of your personal opinion :)



Can you provide a link to the details of this study?



Yes, that's a shame, but that's mostly because most of Harman doesn't seem to care about Olive's results, sadly. My hypothesis is that they're driven more by marketing than by science. Harman is a large company - left hand, right hand, and all that. Just because Harman makes some (arguably) bad products doesn't necessarily mean the studies that their researchers put out are worthless - these are different departments and they might not even talk to each other.



What do my ears have to do with it? I don't care what my ears prefer for the purpose of this discussion. I don't discuss my personal preferences because this is "Audio Science Review", not "Audio Feelings Review". What I care about is results that are statistically relevant across the population as a whole.
1, We don't publish paper, but we compare directly to studio moniters in well treated room and eqed to some of the hifi curves/ small room curves. The main difference of these curves is sub 200, and 1k-2k. Some curves, like BK hifi recommendation has rolling off starts from low frequency while some other curves like etymotic used small room curve that started from 2khz. None represents harman curve, original objective or subjective tweaked. The increased slop high frequency roll off just doesn't exist in any objective measurements.
2, N700 and n5005 are praised by olive himself(check olive's facebook). And are used as good example of matching harman target curve, which is a joke. I(as a university student) helped "moondrop"(chinese iem company) to make their target. And what's funny? "Blessing"(model name) that followed our target actually matched harman target much better than n5005 and n700. But not exactly matched because that will sound bad.
3, Please re-evaluate er4( er4sr and er4xr also er2se) in comparison with flat studio monitors in well treated room. If it's close then that means it's good. Objectively good.
4, Harman target is 50% subjective. And both objective part and subjective part have issues resulting the un-natural artificial ungly sound. In any way, Harman target is just a reference that need to be taken with grain of salt or just as inspiration instead of objective target.
5, Peope tend to like the sound they are used to. So personal preference can be changed and would change over time. Without a reference, subjective preference doesn't mean anything.
Some links:
https://m.blog.naver.com/gre_nada/221327191729
https://m.blog.naver.com/gre_nada/221375778465
https://www.rtings.com/headphones/reviews/akg/n700nc-wireless (read raw measurement instead of their target replied)
 

Soniclife

Major Contributor
Joined
Apr 13, 2017
Messages
1,332
Likes
829
Location
UK
#86
The thing is harman target sounds considerably worse than etymotic er4. There is no mean to choose 200hz and below as bass. Causing a lot of confusion. And the high frequency roll off is also problematic which is not the case of real world.
To prove, just use a pair of eqed studio monitors and use ear to eq the earphone. The end result will be much closer to etymotic than harman.
The process we used is use a collection of songs to define bass and mid range (probably 7k and below)
and to use sine wave generator to eq 7khz + . I use to like harman and sean olive. But they proceed to make n5005 and n700 which sound horrible and don't even represent harman target at all. And there is no reason to believe that the original speaker eqed curve has rolled off highs. That's just wrong. More so, the newer iem target is just getting worse and worse. Completely out of controlly wrong. Yes I'm angry. Because they doesn't fucking know shit.
This post illustrates the original thread question very well, someone has their belief system calmly challenged, and suddenly it's toys everywhere but in the pram.
 

PierreV

Active Member
Joined
Nov 6, 2018
Messages
294
Likes
614
#88
Well, according to the studies I cited, the standard model genuinely doesn't have enough bass. This makes the bassier version objectively better, and I'm glad they released it.
Well, these are attempts to develop predictive models. The method is commendable, but the global result should not be treated as gospel as it often his. The authors seem well aware of that and I imagine this is going to be published next.

https://www.researchgate.net/public...ir_Preferred_Headphone_Sound_Quality_Profiles

Significance should be relatively low (compared to other fields with larger Ns) but that is probably the best we are going to get in this field. But this does not automatically means objective/subjective should be systematically opposed.

One can objectively say things like

- most people prefer that target.
- most people in segment A prefer another target.
- the segments are clearly different (or not).
- etc

A subjectivist can perfectly say, objectively

- it seems I do not belong to the overall majority.
- it seems I belong to another segment.
- etc

Where things fall apart is when the objectivist tells the subjectivist that he shouldn't like some response or the subjectivist claims his preferred response is some kind of absolute best response.

So there are, imho, two very different aspects in the objectivist/subjectivist debate

The first one is just when taste is the dominant issue (archetype: response curve). It is perfectly OK not to like what the majority has been proven to like. If that is not OK, I would suggest a blind registration test on this board and to forbid people who don't like the Harman curves from joining ;)

The second one is when claims that can't be measured or blind tested are made (archetype: fancy cables). There, obviously, objectivism calls for zero tolerance, especially if those bogus claims are used in daylight robbery schemes.

PS: segmentation should be of great interest to manufacturers who, no doubt, would love to see their products well received by their target market. The recent archimago blind test hints that demographics play a significant role in preferences imho.
 

JohnYang1997

Addicted to Fun and Learning
Joined
Dec 28, 2018
Messages
872
Likes
350
#89
Well, these are attempts to develop predictive models. The method is commendable, but the global result should not be treated as gospel as it often his. The authors seem well aware of that and I imagine this is going to be published next.

https://www.researchgate.net/public...ir_Preferred_Headphone_Sound_Quality_Profiles

Significance should be relatively low (compared to other fields with larger Ns) but that is probably the best we are going to get in this field. But this does not automatically means objective/subjective should be systematically opposed.

One can objectively say things like

- most people prefer that target.
- most people in segment A prefer another target.
- the segments are clearly different (or not).
- etc

A subjectivist can perfectly say, objectively

- it seems I do not belong to the overall majority.
- it seems I belong to another segment.
- etc

Where things fall apart is when the objectivist tells the subjectivist that he shouldn't like some response or the subjectivist claims his preferred response is some kind of absolute best response.

So there are, imho, two very different aspects in the objectivist/subjectivist debate

The first one is just when taste is the dominant issue (archetype: response curve). It is perfectly OK not to like what the majority has been proven to like. If that is not OK, I would suggest a blind registration test on this board and to forbid people who don't like the Harman curves from joining ;)

The second one is when claims that can't be measured or blind tested are made (archetype: fancy cables). There, obviously, objectivism calls for zero tolerance, especially if those bogus claims are used in daylight robbery schemes.

PS: segmentation should be of great interest to manufacturers who, no doubt, would love to see their products well received by their target market. The recent archimago blind test hints that demographics play a significant role in preferences imho.
The thing is objectivists aren't all actual objectivist. Many, more like most of them, are subjectively opinionated biased with objective means assisted. However, this doesn't mean subjectivists are better. In fact, we need to be as objective as possible but with a doubt on everything. And truly non biased, is the way to go. And be able to face errors and mistakes.
 

edechamps

Active Member
Joined
Nov 21, 2018
Messages
174
Likes
408
Location
London, United Kingdom
#90
Well, these are attempts to develop predictive models. The method is commendable, but the global result should not be treated as gospel as it often his. The authors seem well aware of that and I imagine this is going to be published next.

https://www.researchgate.net/public...ir_Preferred_Headphone_Sound_Quality_Profiles
Thanks for the link. It's already published, actually. I just read the paper and the results are that listeners can be clustered into three classes: (paraphrasing the paper)
  • Class 1 (64% of listeners) like the Harman curve as-is
  • Class 2 (15% of listeners, predominantly male) prefer 3-6 dB more bass below 300 Hz
  • Class 3 (21% of listeners, all untrained females and old people) prefer 2-4 dB less bass below 100 Hz

I do agree that there is some personal variation here, but it all points to the Harman curve as the correct "default" curve in the absence of any information about the target listener.
 

Xulonn

Active Member
Joined
Jun 27, 2018
Messages
257
Likes
418
Location
Boquete, Chiriqui, Panama
#91
Interesting how this conversation is focusing a lot of attention on IEMs, which to appear to me to be the most difficult to measure and correlate to what people actually hear - and like on- or over-ear headphones, do not lend themselves to ABX blind testing. I can conceptually see that there would be a big difference between filling a room with sound and filling a pair of ear canals with sound - and totally different sets of physical parameters dealing with wavelengths, reflections, resonances, etc.

Do fans of IEMs here see this area area as particularly contentious?

I don't use headphones or IEM's because I have this [apparently unusual] issue of not being able to focus on music when I am cut off from the sounds around me - I cannot relax into the music. The few times I've tried headphones, I frequently pull one headphone away from my ear because I think I hear something that might need attention.
 

Xulonn

Active Member
Joined
Jun 27, 2018
Messages
257
Likes
418
Location
Boquete, Chiriqui, Panama
#92
Class 2 (15% of listeners, predominantly male) prefer 3-6 dB more bass below 300 Hz
...and this is the group that is most likely to argue about subjective audio experiences and preferences vs measurements on internet forums. Apparently it all started with our primate ancestors...
Chimp Headphones.jpg


Regulars here at ASR art mostly objectivists with a moderate to expert level of education and/or experience in related technologies, and a few of us have an interest and/or background in psychology/sociology and other "people sciences".
 
Joined
Apr 22, 2019
Messages
16
Likes
6
Location
Valley of the Sun
#93
Oh come on! Like there's any question about this!

And what kind of idiot would live in the Valley of the Sun, anyway?
Both companies products seem to work see Amirm's comment in the Okto DAC8 review.

Well, I got an offer it wouldn't have seen smart to refuse so I moved to the same place you say you're from.
 

SIY

Major Contributor
Technical Expert
Joined
Apr 6, 2018
Messages
2,431
Likes
4,000
Location
Phoenix, AZ
#94
Well, I got an offer it wouldn't have seen smart to refuse so I moved to the same place you say you're from.
Heh, I'm not FROM here, just live here. And it was so nice in February when I arrived...
 
Joined
Apr 22, 2019
Messages
16
Likes
6
Location
Valley of the Sun
#95
Heh, I'm not FROM here, just live here. And it was so nice in February when I arrived...
So you could offer a cable burn in service; say $250 to leave them out in the summer sun, $500 to leave them out in the summer sun at a vortex near Lake Havasu City or $1,000 to leave them out in the sun at a Sedona vortex. When it actually gets hot of course.
 

SIY

Major Contributor
Technical Expert
Joined
Apr 6, 2018
Messages
2,431
Likes
4,000
Location
Phoenix, AZ
#96
Not as funny as it seems- that's exactly how I'm testing coatings that I've been working on.

dry-heat-2-1.jpg
 

tmtomh

Active Member
Joined
Aug 14, 2018
Messages
102
Likes
202
#99
Women just don't have the intellectual capacity or perseverance to spend hours arguing with strangers about the audibility of -120db distortion artifacts.
Audiophilia, particularly the part focused on technical minutae like -120dB distortion artifacts, certainly is mostly the province of men. But it's got nothing to do with intellectual capacity or perseverance.
 

617

Active Member
Joined
Mar 18, 2019
Messages
114
Likes
98
Location
Somerville, MA
Audiophilia, particularly the part focused on technical minutae like -120dB distortion artifacts, certainly is mostly the province of men. But it's got nothing to do with intellectual capacity or perseverance.
I was making a joke. Anyway I feel superior to most people here because I design speakers so I argue about distortion artifacts at -40db.
 

Similar threads

Top Bottom