[sander99]

@PhonoPhi:
These are all very complicated subjects. If you want to come anywhere close to understanding what is being said here in terms of science/physics/mathematics/theory you could start reading this:
https://en.wikipedia.org/wiki/Sound_localization

And once you have a grasp of how human sound localization works you can take a closer look at different types of audio recordings. What kind of localization cues are in there, and what kind are not. (And sometimes it is a mix of different sounds that each can have a different set of localization cues and/or a different perspective).

With the above information maybe you will be able to understand that different types of recordings (in particular even different types of stereo recordings) need different delivery methods for optimal results (with respect to sound localization).

PS: I'll admit that I know almost nothing about designing IEMs. But I do know a lot about the above subjects. And so does @castleofargh and a few other people here.

 

[bigshot]

He doesn't understand.

 

[PhonoPhi]

He doesn't understand.

No "claiming victories, I am not "invested" in the subject in any sense (not trying to sell here anything, not claiming superiority, just that it can be done and should be much more cost-effective), and would love to hear good scientific arguments.

Instead, as above, all the arguments are that I do not understand something and should "learn something", and again my assumption/hypothesis was formulated (and reformulated/repeated) physically/ mathematically.

So it feels like trying to discuss automobiles with the tribe of the proud horse owners. I do have all the respect for your great horses, so "no victory" claims/feeling again, just little point of arguing further...

Yet, the complete analogy with "cable discussions" is fully deserved (Identical subjective arguments that your opponents have not really experienced it and should learn more; do try as a gedanken experiment applying your arguments as "cable believers", marvelous fit indeed ).

 

[bfreedma]

No "claiming victories, I am not "invested" in the subject in any sense (not trying to sell here anything, not claiming superiority, just that it can be done and should be much more cost-effective), and would love to hear good scientific arguments.

Instead, as above, all the arguments are that I do not understand something and should "learn something", and again my assumption/hypothesis was formulated (and reformulated/repeated) physically/ mathematically.

So it feels like trying to discuss automobiles with the tribe of the proud horse owners. I do have all the respect for your great horses, so "no victory" claims/feeling again, just little point of arguing further...

Yet, the complete analogy with "cable discussions" is fully deserved (Identical subjective arguments that your opponents have not really experienced it and should learn more; do try as a gedanken experiment applying your arguments as "cable believers", marvelous fit indeed ).

Nice insults.

I see this in Sound Science from time to time. Someone visits with a concept they’ve decided is “the way things are”, and continues to argue a narrow slice of the technology while refusing to acknowledge clear issues/deficiencies as they are pointed out (Expected room interaction, phase coherency, wave propagation, etc.).

Enjoy your multi driver IEMs, but standing on your soapbox and shouting isn’t making a compelling argument and/or changing minds. It isn’t changing mine at least

 

[PhonoPhi]

Nice insults.

I see this in Sound Science from time to time. Someone visits with a concept they’ve decided is “the way things are”, and continues to argue a narrow slice of the technology while refusing to acknowledge clear issues/deficiencies as they are pointed out (Expected room interaction, phase coherency, wave propagation, etc.).

Enjoy your multi driver IEMs, but standing on your soapbox and shouting isn’t making a compelling argument and/or changing minds. It isn’t changing mine at least

"Scientific forum"??
"Deficiencies?

Anyone formulated what can be missing in terms of the reconstruction of the wave front??

We record the sound near each ear as it "should be, let it be in the most perfect/calibrated room and reproduce.

OK for a given "head anatomy", but then more substantial/influential than the "head anatomy is the anatomy of individual ear channel, so no perfect "Beattles" reproduction "as intended" for anyone other than the sound engineer, right?

But then why indeed I should argue here in vain...

So safely back to your elite horses, gentleman
I need to stop writing here.

P. S. Your first line is a perfect example of an unsubstantiated perception bias, so dominant in this thread...

 

[bfreedma]

"Scientific forum"??
"Deficiencies?

Anyone formulated what can be missing in terms of the reconstruction of the wave front??

We record the sound near each ear as it "should be, let it be in the most perfect/calibrated room and reproduce.

OK for a given "head anatomy", but then more substantial/influential than the "head anatomy is the anatomy of individual ear channel, so no perfect "Beattles" reproduction "as intended" for anyone other than the sound engineer, right?

But then why indeed I should argue here in vain...

So safely back to your elite horses, gentleman
I need to stop writing here.

P. S. Your first line is a perfect example of an unsubstantiated perception bias, so dominant in this thread...

Thanks for validating my post. Goodbye.

 

[PhonoPhi]

Thanks for validating my post. Goodbye.

It is all about individual perception and its limitations as per Einstein' famous quote.
Goodbye indeed.

 

[redrol]

I like this thread because folks talk about actual sound science rather than emotional opinions. This is how we get actually better sound playback.

 

[sander99]

Anyone formulated what can be missing in terms of the reconstruction of the wave front??

Depends on what signal you feed the IEMs.

In the case of a typical stereo recording based on panning sounds between the two channels:
There will be no controlled/intended spectral localization cues whatsoever (no monaural spectrum cues, no binaural spectral cues), let alone spectral cues matching with your personal Head Related Transfer Function. (In real life spectral cues are the result of direction-dependend filtering of sound by body, head, ear flaps, pinna that the brain learns to interpret during life)(Sometimes someone's brain might pick unintended coïncidental spectral cues that could place (in the listeners perception) sound a little bit out of the head, or a little bit forward or backward inside the head.)
There can be ITD (Interaural Time Difference) and ILD (Interaural Level Differences) cues but without supporting (and non-contradicting!) spectral cues you will never perceive sound far outside your head at a pinpoint location. (And some recordings do not contain any ITD cues at all, or only for some sounds in the recording.)
Also there will be no correct reaction to moving the head. The brain can initiate involuntary head movements to get another perspective to solve ambiguities in sound localization.

In the case of a dummy head binaural recording:
There will be spectral cues, but generally not matching your personal HRTF. And there will be supporting ITD and ILD cues. This may place sounds a bit further out of the head, but unless you are very lucky in matching HRTF these will not be placed in the correct directions at the correct distances.
(In the case of a binaural recording the correct directions and distances would be the actual directions and distances relative to the dummy head during the recording.)
Also in this case there will be no correct reaction to moving the head. The brain can initiate involuntary head movements to get another perspective to solve ambiguities in sound localization.
(The barber shop example is convincing for many/most people because a number of aspects are all favorable to success:
-The sound of the scissors is close to the head, and close-outside-the-head sound is easier to achieve than far-outside-the-head sound
-The sound of the scissors is behind the head, and you can not see what's behind you so visual cues won't contradict the auditory cues
-The sound of scissors close behind your head is familiar to most people, )

So that's what's "missing in terms of the reconstruction of the wave front".

When listening to a typical panning-based stereo recording over loudspeakers (real or well implemented personal HRTF based and headtracked virtual speakers) all above types of localization cues are added to the sound waves before they reach your ear drums. Admittedly, the perceived distance is determined mainly by the distance from the listener to the loudspeakers, and not controlled by the recording (although secondary cues like reverberation can add some depth).
Again: there is nothing wrong with choosing to listen to such a recording with IEMs, but many people will prefer to perceive sound as coming from a distance outside of the head, like >90% of all sounds that we hear in "nature", including the sounds of all life acoustical instruments and voices.

 

[sander99]

@PhonoPhi:
I forgot a few things and continue here.
Most localization cues are binaural, and depend on what happens in both ears. With multichannel over loudspeakers each ear hears all the loudspeakers. Because of that your brain can localize all the loudspeakers around you. Next sounds are panned between 2 or more loudspeakers and the panning effect will make you perceive the sound somewhere between those loudspeakers. That can be in any direction, only thanks to the fact that your brain can localize all the loudspeakers, only thanks to the fact that both ears hear all loudspeakers and get all the real physical localization cues from all loudspeakers.
For example blowing the left front channel and the left rear channel directly into your left ear with different channels inside IEMs and without additional processing is meaningless. Your brain won't be able to perceive the sound from the "front drivers" as coming from a significantly different direction as the sound from the "rear drivers" (Even monoraul spectral localization cues will not be present because the sound of both the "front-drivers" and the "rear-drivers" has skipped the direction-dependend filtering by your body, head and outer-ear.). So panning between those two drivers (or those two sets of drivers) will not work. Might as well just add the signals together and put them in one set of drivers (conventional downmix to stereo). And that is fine too if you prefer that. But most people will prefer multichannel sound to come from all directions. Otherwise it wouldn't be any different from stereo!

 

[bigshot]

He’s just being argumentative for argumentative’s sake. The only thing he has to say is that the IEMs he bought are better than anything else. Validation bias. Nothing else there.

 

[bigshot]

I see this in Sound Science from time to time. Someone visits with a concept they’ve decided is “the way things are”, and continues to argue a narrow slice of the technology while refusing to acknowledge clear issues/deficiencies as they are pointed out

That’s a bingo. Remember our friend who had definite opinions on compression without having a clue about how compression is used? It was the only subject he could talk about. The same misconceptions over and over again. Not a good use of our time.

 

[bigshot]

I need to stop writing here.

Finally something we can agree on!

 

[PhonoPhi]

He’s just being argumentative for argumentative’s sake. The only thing he has to say is that the IEMs he bought are better than anything else. Validation bias. Nothing else there.

Sorry, factually wrong statement!
(Not worth my lawyers to contact yours of course..)
I never said that my IEMs are "better", I enjoy them, but they are not. I do have all the respect for calibrated room setups, different from the science though...

P. S. I ignore the rest.

 

[sonitus mirus]

With only 2 speakers on or in a person's ears, what is it that distinguishes between a sound coming from in front or behind the listener? Same goes for sounds perceived as originating above or below the listener's position. Is it only that we expect, based on experience, that certain sounds should originate at specific locations? What if one person's hearing does not perceive these sounds the same as another? Take the exhaust of a rally car in a video game. Common sense would suggest the sound should be coming predominantly from the rear of the car, but logically, with only two dimensions, this could just as easily appear to be coming from the front of the listener when positioned in the driver's seat.

I found an interesting article that talks about some to the things recently discussed here. It is mentioned how multiple drivers may be somewhat advantageous to spatial sound sensations, with some caveats.

https://www.tomshardware.com/reviews/how-we-test-gaming-headsets,5212-2.html

" 5.1- or 7.1-channel sound delivered through headphones is always imaginary, and can only be achieved by the brain relying on previous experience. "