Is it time for a new Harmon Curve survey?

[Killcomic]

Not necessarily, you don’t seem to understand what an average is. For example, the average number of children per family in the USA was 2.4 but of course not a single family actually had 2.4 children.

You don't seem to understand what averages are either. If someone likes 1, but another likes 10, the average is 5.5, meaning neither of the samples are satisfied with the final result.

It is not supposed to be a “true neutral measuring curve” it is supposed to be a preference curve!

Gosh, you don't say!

That’s not the problem with the Harman Target, that’s the whole point! That’s like saying the problem with a voting exit poll is that it’s just a survey of which candidate the voters preferred.

It is a problem if you start putting too much faith on something that a small group of people is being sampled as a representative of a large pool.

There’s no evidence of which I’m aware that suggests people on the other side of the planet average a significantly different preference curve. The reason it may not be completely relevant to you is because you might be part of the 50% or so of people who have a somewhat different preference.

Oh really? Then why is it that Asian IEMs have elevated treble whereas American ones seem to lean toward thick, mid centric sound?

Trust Head-Fi to have people aggressively defending a target curve

 

[sendler2112]

Everyone seems to want someone else to dictate their target response curve. They don’t want to invest time into determining their own personal curve for themselves.

The reason I started this discussion is because someone IS dictating the response curve to us. And I have read many reports by expert reviewers that are using programs that allow them to EQ that don't like the Harmon curve as it is. There is way too much bass below 150Hz. And the reviewers usually talk about bringing the highs back up. I personally also find the Harmon curve to be too high in the 800 Hz to 2,000Hz range which muddies the fine detail. farther up in the 4-7kHz range.

I don't have a multi-thousand dollar measurement system and my chosen player (HQ Player) doesn't offer a graphic EQ that I can easily adjust by ear on my own. Most players and DACs only have parametric EQ. So I have to rely on the third party measurements, such as AutoEQ.

https://autoeq.app/

And take their parametric correction as it is and applies to bring about the Harmon curve. Because this is the current standard.

But based on the reports and feedback by many different professional users of EQ, I think we can do better.

 

[sendler2112]

You don't seem to understand what averages are either. If someone likes 1, but another likes 10, the average is 5.5, meaning neither of the samples are satisfied with the final result.

The sample size would be in the high hundreds or 1,000 and include listeners from all over the world with just one year of rounds of the CanJam festival.

Talking about a mean average with a sample size of 2 extremes isn't a good example.

Firstly, with this many samples, whatever few extreme outliers there were would get washed out. Anyone that is serious enough to travel to a CanJam and buy a ticket,to the event is already pretty serious.

I honestly believe the vast majority of listeners would end up choosing similar settings. With this many samples we would see modes of preference develop. A 15 band graphic EQ would be sufficient. And I totally expect most participants would turn down the 50Hz and 100Hz sliders from the Harmon curve and possibly turn up 10kHz and 20kHz if nothing else.

We then throw out the obvious outliers which are provided by the few people that think it is amusing to spoil a test, and average the values for each band. Apply this to the original Harmon correction and the new Head-Fi curve is born.

 

[bigshot]

My ears are average. Harman sounds great to me, except with a little adjustment in the upper mids. To you, the problem is the bass. One size doesn’t fit all. There is no perfect curve for everybody because our ears are all shaped differently. My preferred curve may be different than yours. If you’re too lazy to grab an equalizer and experiment, you get what you get.

 

[Whazzzup]

Is it time? i want a double blind scientific analysis with at least 98.5% statistical significance.

and who eq anyway. straight line not good enough?

 

[sendler2112]

My ears are average. Harman sounds great to me, except with a little adjustment in the upper mids. To you, the problem is the bass. One size doesn’t fit all. There is no perfect curve for everybody because our ears are all shaped differently. My preferred curve may be different than yours. If you’re too lazy to grab an equalizer and experiment, you get what you get.

It's not a matter of lazy. HQ Player does not offer a graphic EQ. And it's not just me. I see other professional reviewers questioning the desirability of the Harmon curve and making additional corrections. And: Is your headphone set up currently adjusted to Harmon? How so?

 

[gregorio]

You don't seem to understand what averages are either. If someone likes 1, but another likes 10, the average is 5.5, meaning neither of the samples are satisfied with the final result.

Huh? You state I don’t understand averages but support that with an example which is effectively the same as mine? No one claims that if you take two individuals the target curve will necessarily be ideal for either of them!

Gosh, you don't say!

Gosh I do say and gosh why did you bring up a “true neutral measuring curve” when that has nothing to do with the Harman target curve or what it demonstrates?

It is a problem if you start putting too much faith on something that a small group of people is being sampled as a representative of a large pool.

If YOU “start putting too much faith on something” that it is not, then that’s YOUR problem, not the problem of the Harman curve! Your apparently misguided interpretation of what the Harman curve is and does, is the problem.

Oh really? Then why is it that Asian IEMs have elevated treble whereas American ones seem to lean toward thick, mid centric sound?

Yes really, I was really not aware of any such evidence! If you can provide reliable evidence of what you claim here about Asian IEMs and causation rather than just a correlation, then I will really be aware of such evidence.

Trust Head-Fi to have people aggressively defending a target curve

You mean: Trust the Sound Science subforum to defend the demonstrated science against some audiophiles with unscientific and/or impractical suggestions and some false assertions based on what they believe their preferences to currently be and misinterpreting the facts? If so, then “yes” and thanks!

The reason I started this discussion is because someone IS dictating the response curve to us.

Someone has always got to dictate a response curve to you, how can a manufacturer make a set of headphones without dictating a response curve? However, Harman is NOT dictating a response curve, it is merely presenting a majority preference of a sample group under controlled conditions. If it is becoming a standard, it’s because it is a valid study and many others have found that it works. If it didn’t, then they wouldn’t use it and it wouldn’t become a standard.

It's not a matter of lazy. HQ Player does not offer a graphic EQ.

Then use a player that does offer an EQ.

And it's not just me. …

Sure, that’s what a majority preference means, there won’t be just one person for whom it’s not ideal.

G

 

[sendler2112]

Then use a player that does offer an EQ.
G

I can try the graphic EQ plugin in Foobar to hear how my preference compares to the Harmon curve.

 

[XTF1]

HQ Player does not offer a graphic EQ.

I don’t understand your point here. I don’t know how many PEQ filters HQ Player supports, but you could perfectly set AutoEQ to optimize the target with say 5 filters, and use a set of other filters to modify the response to your taste, a mix of low-Q and high-Q filters for example. Yes, it may appear more complex than a GEQ, but HQ Player is no simple piece of software either: this seems inconsistent…

 

[bigshot]

It's not a matter of lazy. HQ Player does not offer a graphic EQ. And it's not just me. I see other professional reviewers questioning the desirability of the Harmon curve and making additional corrections. And: Is your headphone set up currently adjusted to Harmon? How so?

You can get an outboard EQ.

I make additional corrections too, Harman just gets me very close. Most people will find Harman sounds good.

I used measurements of my cans to determine corrections to bring the native curve to Harman.

I don’t think you understand what the Harman curve is, and how it’s intended to be used.

 

[sendler2112]

I don’t understand your point here. I don’t know how many PEQ filters HQ Player supports, but you could perfectly set AutoEQ to optimize the target with say 5 filters, and use a set of other filters to modify the response to your taste, a mix of low-Q and high-Q filters for example. Yes, it may appear more complex than a GEQ, but HQ Player is no simple piece of software either: this seems inconsistent…

I think I already mentioned that I have the AutoEQ filters for the HiFiMan HE1000SE for the Harmon curve installed in HQ Player and did try it. Which started me on this whole discussion. The Harmon curve has way too much bass. even after cutting the suggested boost in half. The reduced highs became dim and the prescribed elevation in the 1-2kHz range sounds muddy.

 

[XTF1]

I think I already mentioned that I have the AutoEQ filters for the HiFiMan HE1000SE for the Harmon curve installed in HQ Player and did try it. Which started me on this whole discussion. The Harmon curve has way too much bass. even after cutting the suggested boost in half. The reduced highs became dim and the prescribed elevation in the 1-2kHz range sounds muddy.

It looks like the Harman target does work well for you… Fine! Have you tried another target in AutoEQ, one that would be a better starting point for you? But again, you may have to fine-tune the results, either modifying the filters calculated by AutoEQ or adding other filters. There is nothing wrong with adding you own filters on top of the AutoEQ ones.

We don’t necessarily need a new Harman target, what we always want is larger population samples, a better understanding of individual preferences and how they “cluster” around one or multiple targets (e.g. Harman and other targets).

 

[bigshot]

Because Harman doesn’t work for you, it doesn’t mean it doesn’t work well for most people.

 

[Killcomic]

You don't seem to understand what averages are either. If someone likes 1, but another likes 10, the average is 5.5, meaning neither of the samples are satisfied with the final result.

Gosh, you don't say!

It is a problem if you start putting too much faith on something that a small group of people is being sampled as a representative of a large pool.

Oh really? Then why is it that Asian IEMs have elevated treble whereas American ones seem to lean toward thick, mid centric sound?

Huh? You state I don’t understand averages but support that with an example which is effectively the same as mine? No one claims that if you take two individuals the target curve will necessarily be ideal for either of them!

Gosh I do say and gosh why did you bring up a “true neutral measuring curve” when that has nothing to do with the Harman target curve or what it demonstrates?

If YOU “start putting too much faith on something” that it is not, then that’s YOUR problem, not the problem of the Harman curve! Your apparently misguided interpretation of what the Harman curve is and does, is the problem.

Yes really, I was really not aware of any such evidence! If you can provide reliable evidence of what you claim here about Asian IEMs and causation rather than just a correlation, then I will really be aware of such evidence.

You mean: Trust the Sound Science subforum to defend the demonstrated science against some audiophiles with unscientific and/or impractical suggestions and some false assertions based on what they believe their preferences to currently be and misinterpreting the facts? If so, then “yes” and thanks!

Someone has always got to dictate a response curve to you, how can a manufacturer make a set of headphones without dictating a response curve? However, Harman is NOT dictating a response curve, it is merely presenting a majority preference of a sample group under controlled conditions. If it is becoming a standard, it’s because it is a valid study and many others have found that it works. If it didn’t, then they wouldn’t use it and it wouldn’t become a standard.

Then use a player that does offer an EQ.

Sure, that’s what a majority preference means, there won’t be just one person for whom it’s not ideal.

G

Ha ha, take a chill pill XD

 

[bigshot]

I saw a graphic that plotted the test subjects for the Harman curve once. If you’re confused about Harman’s usefulness, that graphic will help you understand. I don’t have that graphic but you should be able to google it.