Harman is working on a new IEM target curve

[solderdude]

Preference part is another aspect and also included in the target curve which consists of correction of the earcanal and 'simulation' of the outer ear / head changes that are not present with IEM's.

I don't know if people evolved over time to get more bass or whether this demand for more bass came from better recordings and transducers going deeper and louder these days. I expect it to be the latter.
I do not expect that aspect to change even more as real life sounds are still the brain's reference point for realistic reproduction.

 

[Yorkshire Mouth]

I think some of the criticism is unjustified.

They've done some research, and they're continually tweaking it.

Let's say we end up with a 'perfect' IEM curve. That won't have been reached with one set of experiments.

It will be end of a journey. Some people are criticising them for going on the journey, believing they should have managed to go from nothing to 'perfect' in one step.

Life isn't like that.

Just as an aside, I'd like to see more research into finding ways to insert IEMs 'perfectly', or as close as possible. At the moment, if the position of the IEM in my ear changes even slightly, the sound changes quite a bit. More developments in this area would bring greater benefits, in my opinion.

 

[Multicore]

As I have said before, we need one target and that is that. It doesn't need to be perfect (even if it can be). We just need one that the listener then modifies to taste -- just like we do with speakers and target room curve. Folks keep agitating for new targets so not surprised that Sean is searching for one but in my opinion, no more work was needed.

We and Harman have, I imagine, different purposes. I agree with you that what we (at ASR) need can be arbitrary in its particulars so long as it is not grossly off. At ASR we are in the business of using measurements and reference curves to help us choose products for individual use. Ideally we can develop from experience our personal preference's deviation from the reference curve. So what we already have is probably good enough for us and changing it will introduce some potential for confusion.

Harman, otoh, is in the business of selling mass produced products. Their developers, I imagine, are busy segmenting the market and devising cost optimized production model designs that please buyers in the segments. If it's big business, and I guess it is, then doing some more research makes sense.

 

[lashto]

...
I welcome the desire to improve on the science as that needs to evolve. It just does not make it clearer for the general public.

a good observation. But since ASR is supposed to be both "science" and "for the general public", what exactly is the right answer/course? Stopping science because it's confusing for "the public" is not much of a solution, is it?!

More generaly, I would say there is a bit too much negativity around here lately. Complaining, bickering, 'vendettas', a lot of the not-invented-here syndrome, pretty dogmatic attitudes like "this&that is enough/done/perfect, why would anyone ask for more/change" ...
Interesting threads with lot of potential end up drowned in noise, e.g. the related one about BK 5128. Its last comment resumed it quite well:

We had a chance for a fruitful and tough discussion that could have progressed the science for everyone's benefit and what is the result instead? Acrimony.

unfortunatelly, this thread has quite a tendency to go the same way...

P.S.
and my complaining-about-complaining post is of course very 'helpful'

 

[solderdude]

But since ASR is supposed to be both "science" and "for the general public", what exactly is the right answer/course? Stopping science because it's confusing for "the public" is not much of a solution, is it?!

I never suggested such a thing... the proper way to go about this is to educate the public and believe this is what ASR is about.

 

[IAtaman]

I never suggested such a thing... the proper way to go about this is to educate the public and believe this is what ASR is about.

I don't think anyone suggested anything about stopping the science really. What was actually said was quite the opposite if I understood it correctly. General objection to the BK5128 measurements and targets developed for that equipment was that they did not have enough science to support them mostly.

 

[IAtaman]

More generaly, I would say there is a bit too much negativity around here lately. Complaining, bickering, 'vendettas', a lot of the not-invented-here syndrome, pretty dogmatic attitudes like "this&that is enough/done/perfect, why would anyone ask for more/change" ...

I read earlier this week I believe a member saying that they like it here because people here can actually change their mind. So yes, there are people here whose sole purpose seems to be to create drama and negativity, and insult others, but there are also a lot of people who are open minded, sensible and willing to listen and understand. If you see too much hatred, then I suggest you look elsewhere.

 

[solderdude]

How can science progress unless you start working with newer test gear and explore its limits and capabilities ?

The BK5128 will eventually just become yet another standard with targets developed by different people based on different research.
More confusion for the 'general public' but who knows what could come from this. Education is key.
Harman has one and it would be surprising if they just played around with it and then decided not to use it and only use older standards. We'll have to wait and see.

After a while some other test fixture manufacturer will come with yet another fixture with its own promises of higher accuracy which then too will become a new standard and targets will be developed again.

That is how measurement science progresses.

Haters will always be around.
They never show any of their own measurements or disclose what their competences are or what they published themselves. Keyboard warriors ...

 

[IAtaman]

The BK5128 will eventually just become yet another standard with targets developed by different people based on different research.
More confusion for the 'general public' but who knows what could come from this.

I don't think it will create more confusion for the general public, and that is mostly because the general public does not really care about headphone FR targets in my opinion. At the end of the day all an average consumer looks at is that whether a product is recommended / got a high score or not.

I guess the main concern about multiple standards is that they make it very hard to standardize tonal balance across products, product categories and recording/reproduction. But then again, I don't think it was possible anyway due to commercial considerations, and frankly I personally don't think it is that important anyway if you have a bit of a dip here and bit of a rise there. Nothing about music is perfect or absolute anyway.

After a while some other test fixture manufacturer will come with yet another fixture with its own promises of higher accuracy which then too will become a new standard and targets will be developed again.

Smart and knowledgeable people tried to explain this to me before yet I still don't understand why we can not have a "diffuse field", non-fixture specific target. I understand the relationship is not linear and FR dependent, but I can not understand how it is not possible to reverse the fixture's transfer function.

 

[IAtaman]

To continue my barrage of recent posts on this topic, let me say that I believe all this discussion around targets and measurement gear is more damaging to the public than having multiple targets, simply because it is exaggerating the importance of the tuning of a headphone while there are many other parameters about a headphone that matters to people which can not be measured and correlated at the moment. What happens then is that you end up people trying to sell headphones by replicating Susvara's FR and others buying it thinking they are getting a cheap Susvara. I wish more thought and discussion and research went into quantifying and defining or even debunking detail, sound stage, imaging and all that other parameters instead of having multiple iterations of the same one thing we already know how to correlate to..

 

[alumnicesar]

targets developed for that equipment was that they did not have enough science to support them

agreed 100%

 

[markanini]

To continue my barrage of recent posts on this topic, let me say that I believe all this discussion around targets and measurement gear is more damaging to the public than having multiple targets, simply because it is exaggerating the importance of the tuning of a headphone while there are many other parameters about a headphone that matters to people which can not be measured and correlated at the moment. What happens then is that you end up people trying to sell headphones by replicating Susvara's FR and others buying it thinking they are getting a cheap Susvara. I wish more thought and discussion and research went into quantifying and defining or even debunking detail, sound stage, imaging and all that other parameters instead of having multiple iterations of the same one thing we already know how to correlate to..

It's a common misconception that ear coupled transducers work like RCA connectors, when interpreting graphs. In reality the tech and/or evolution is not there. On it's own an established target doesn't cause too much trouble even if it elevates the idea of absolute adherence, for those who are inclined to think that way. But if a large enough amount of peopled don't like an established target like Harman IE currently it does end up playing a part in a lot of pointless and noisy discourse.

 

[solderdude]

Smart and knowledgeable people tried to explain this to me before yet I still don't understand why we can not have a "diffuse field", non-fixture specific target.

Then I probably shouldn't even bother to try as I am not smart nor very knowledgeable in that specific field being an electronics guy.

The fixtures all have different 'ear canals' and pinnae and different 'heads' and even different microphones and that is what determines what the mics pick up.
When the exact same stimulus is applied (DF) all these different fixtures thus all will have a different frequency magnitude response so all need a different 'correction'.
Add to that 'room curves' and you end up with many different 'targets' for different fixtures even though the sound reaching the fixtures is exactly the same.

DF (multiple speakers all around meters away from the mic in a large room) also is not the same as 2 small speakers, very close to the ear sitting in a small sealed amount of air.

I hope this simple explanation makes some sense.

At the end of the day all an average consumer looks at is that whether a product is recommended / got a high score or not.

Agreed... the problem still is that the score is based on a measurement and how that headphone fits with the target.
Mix up test fixture raw data or use the wrong target and those numbers will even be less meaningful.

But yes, most people will buy based on some positive comments of their favorite YT reviewer's subjective findings.

Of course that should not stop Harman with refining/improving/revisiting/checking their IEM curve and does not invalidate any research on headphones in the least.
One thing I am sure of... it won't be based on DF as a starting point

 

[Travis]

Even though there is a preference part in the target which may change a bit there should not be a discrepancy between perceived tonal balance between speakers and IEMs (within the tolerance band of it).

Why is that? That seems logical at face value, they should be the same.

However, knowing a little about the development of the research on the preference for loudspeakers, which incorporated DI, I’m thinking, why would they be the same at all?

IEMs bypass the function of the ears in sound perception. God and/or evolution designed the shape of our ears for a reason. I have seen lots of scholarly articles on how ear shape is part of the perception chain.

I have had custom made IEMs, for many years and wearing them at first is always an adjustment, the sound is vastly different, but you adjust.

I have heard the same about in-ear hearing aids vs. external. Some love in -ear, others could never adjust and describe sound as being under water.

What’s the research on this in terms of in ear monitors and perception of music?

 

[solderdude]

The research on this probably falls short. Sean Olive admitted as such in an interview with Crin. So hence the 'looking at the IEM target again'.

There are a few reasons why IEMs are problematic.
1: insertion depth
2: ear canal shapes differ from a tube
3: the pinna is an important part of perception and differs from person to person. This is bypassed by earbud/IEM so an individual factor that cannot be included in a target.
4: perfect seal must be obtained which is not easy

So for those reasons that what may follow a target created basically on a tube of described size and diameter (the standard) could be perfect.
That does not mean the exact same IEM will deliver the exact same FR on your ear drum and even if it did just that it may not be what (due to personal pinna shape) your brain expects to hear based on 'reality'. Personal EQ will certainly help there.

Another thing is that the brain is highly adaptive. You can get used to a certain presentation and the more often you use it the sooner the brain 'knows' what to do.
When you listen at first it may sound 'off' a while. This 'adjustment' time seems to get shorter when using it more often and when used really often it almost becomes instant.
That is... when the tonal balance is not too far off.

 

[IAtaman]

Then I probably shouldn't even bother to try as I am not smart nor very knowledgeable in that specific field being an electronics guy.

Same here, mostly an electronics guy. Maybe one little difference between you and I is that you are being humble and I am not Appreciate the effort.

The fixtures all have different 'ear canals' and pinnae and different 'heads' and even different microphones and that is what determines what the mics pick up.
When the exact same stimulus is applied (DF) all these different fixtures thus all will have a different frequency magnitude response so all need a different 'correction'.
Add to that 'room curves' and you end up with many different 'targets' for different fixtures even though the sound reaching the fixtures is exactly the same.

DF (multiple speakers all around meters away from the mic in a large room) also is not the same as 2 small speakers, very close to the ear sitting in a small sealed amount of air.

I hope this simple explanation makes some sense.

Yes it does, and I understand all that.

Let me try it this way - we all understand rooms affect the measured tonal balance of a speaker. And our solution to that is the Near Field Scanners that takes room out of the equation so we can see the "raw" performance of a speaker, and correlate it to tonal balance and preference etc. With all these fixtures and different pinnae and ear canals etc, I feel like we are trying to build a room that more accurately represents the room of an average human, while what we should be trying to do is to find a way to take the room out of the equation - an NFS for the headphones. Does this make sense?

 

[192kbps]

As I have said before, we need one target and that is that. It doesn't need to be perfect (even if it can be). We just need one that the listener then modifies to taste -- just like we do with speakers and target room curve. Folks keep agitating for new targets so not surprised that Sean is searching for one but in my opinion, no more work was needed.

"We need to move on and use distortion measurements that are perceptually meaningful."

An updated post

www.audiosciencereview.com

How to understand Dr. Sean's views on THD, have we misunderstood him?

 

[solderdude]

Same here, mostly an electronics guy. Maybe one little difference between you and I is that you are being humble and I am not Appreciate the effort.


Yes it does, and I understand all that.

Let me try it this way - we all understand rooms affect the measured tonal balance of a speaker. And our solution to that is the Near Field Scanners that takes room out of the equation so we can see the "raw" performance of a speaker, and correlate it to tonal balance and preference etc. With all these fixtures and different pinnae and ear canals etc, I feel like we are trying to build a room that more accurately represents the room of an average human, while what we should be trying to do is to find a way to take the room out of the equation - an NFS for the headphones. Does this make sense?

The problem is the room is part of the sonic experience and is included.
The idea most people have is that headphones should behave/sound similar to speakers in a room. More specifically near perfect speaker in an optimal listening room. Not your speakers in your room but could potentially be your speakers in your room at the listening position EQ'ed to the speaker target.

Personally I don't agree that headphones should sound like speakers in a perfect room.
They should sound tonally accurate at a specific listening level to a 'perfect' recording.

To me Harman is not that. It is a bit more 'impressive' than the actual recording. This is what most people prefer it seems. I am not most people.

 

[digitalfrost]

Well this be interesting.

https://x.com/seanolive/status/1713261982680535044?s=20

 

[isostasy]

The problem is the room is part of the sonic experience and is included.
The idea most people have is that headphones should behave/sound similar to speakers in a room. More specifically near perfect speaker in an optimal listening room. Not your speakers in your room but could potentially be your speakers in your room at the listening position EQ'ed to the speaker target.

Personally I don't agree that headphones should sound like speakers in a perfect room.
They should sound tonally accurate at a specific listening level to a 'perfect' recording.

To me Harman is not that. It is a bit more 'impressive' than the actual recording. This is what most people prefer it seems. I am not most people.

What would sounding tonally accurate mean without the context of the speakers in a room though? I don't understand how it could be any other way.

To add another perspective to the conversation, I was intrigued that Jermo, the product manager from Sennheiser, said recently that for them IEM tuning isn't as simple as trying to match the frequency response of a headphone. I think in one of those videos they said they felt the IE900 was the closest to a HD800S in IEM form which is interesting given how differently they measure. Makes me want to give one of the cheaper Sennheiser IEMs a go despite how far off the measured response is from my current preference (Etymotic ER2XR).