[Joe Bloggs]

Alright, picture the following:

Your audio signal is like the one at the bottom (simplified). This is the exact pattern the physical diaphragm traces out in the air to play the music. From here, it should be clear that a lighter, more nimble diaphragm would be able to trace out the bottom waveform more precisely than a heavier, thicker diaphragm, simply due to inertia. Changing the frequency response only changes the amplitudes of the input sinusoidal tones (which will change the shape of the output wave), but at the end of the day, a lighter and more nimble diaphragm can trace out the arbitrary waveform shape better than a heavier and slower diaphragm.

... no. Any diaphragm that is capable of the whole audible frequency range will be capable of any combination of frequencies in any visually complex combination. You will e.g. notice that the bottom waveform does not change direction any more steeply than the 300Hz and is thus not any more demanding of high speed changes than the 300Hz waveform by itself.

 

[JohnYang1997]

Alright, picture the following:

http://data:image/png;base64,iVBORw...Z9/WvgP+qQtZT4P8Bky1uIluam4UAAAAASUVORK5CYII= http://data:image/png;base64,iVBORw...Z9/WvgP+qQtZT4P8Bky1uIluam4UAAAAASUVORK5CYII=

Your audio signal is like the one at the bottom (simplified). This is the exact pattern the physical diaphragm traces out in the air to play the music. From here, it should be clear that a lighter, more nimble diaphragm would be able to trace out the bottom waveform more precisely than a heavier, thicker diaphragm, simply due to inertia. Changing the frequency response only changes the amplitudes of the input sinusoidal tones (which will change the shape of the output wave), but at the end of the day, a lighter and more nimble diaphragm can trace out the arbitrary waveform shape better than a heavier and slower diaphragm.

Do you know that lighter diaphragm can cause other issues if other aspects are not concerned? Why not go ahead and read some about transducers. We can manufacturer sub 1um diaphragm no issue and at very low cost but we hardly see that. Do you know why that is? and why we can use such technology in planars and electrostatic headphones?

 

[WhatToChoose]

That's not the case either. First of all, the ability to get frequency response correct just almost takes everything we need. On top of that if we want sensitivity and low distortion is just very difficult.
What you described is NOT frequency response but rather tonal balance. So you should really understand what frequency response is and how is everything related to it. It is 99% or even more.

Also, to evaluate a driver, there are frequency response, impedance, sensitivity, distortion, maximum output. Expensive headphones/earphones don't necessarily have better in those regard either. All the technology in driver/transducer in these days are cheap. How much do you think a dynamic driver cost? A few cents? A dollar? Most expensive ba drivers are worse in distortion and cost more but even that you don't see more than 10 dollars. So what's the point? Did you prove anything?

Also eq is essentially cheating in physics. It's hard to understand the true laws that it breaks. Surely it's not unlimited, it decreases sensitivity and may decrease snr of the system. But the fact digital processing exists is much more than you think. How to increase high frequency response without changing resonance frequency or lose bass response in drivers? How to match 3k hump without introduce distortion? How to calm the driver without using too much acoustic resistance that lowers the transducer's air driving ability? Eq essentially breaks all those laws with little sacrifice.

You're absolutely correct, most transducers (except for custom developed ones) cost next to nothing. with high end IEMs though, you pay for development costs, and probably the brand name. Perhaps I am misunderstanding your interpretation of frequency response, but to your other points, I suggest you research something called "Fourier Transform." It will provide you with the knowledge to understand the fundamentals of audio reproduction, and bypass the manufacturing limitations that you mention.

Here is a video for you:



Lastly, if you think EQ is cheating physics, you are not looking at the whole picture. Your points are correct, but you are limiting your thinking to manufacturing and tangible limitations. With all else held constant, the inertial property of the diaphragm remains.

It is interesting you mention distortion of BA drivers too, I was thinking that BAs have a higher power/weight ratio, at the expense of maximum power input, which is why BA bass is usually anemic. I'll leave that for a more knowledgeable person though haha.

 

[Joe Bloggs]

You're absolutely correct, most transducers (except for custom developed ones) cost next to nothing. with high end IEMs though, you pay for development costs, and probably the brand name. Perhaps I am misunderstanding your interpretation of frequency response, but to your other points, I suggest you research something called "Fourier Transform." It will provide you with the knowledge to understand the fundamentals of audio reproduction, and bypass the manufacturing limitations that you mention.

Here is a video for you:



Lastly, if you think EQ is cheating physics, you are not looking at the whole picture. Your points are correct, but you are limiting your thinking to manufacturing and tangible limitations. With all else held constant, the inertial property of the diaphragm remains.

It is interesting you mention distortion of BA drivers too, I was thinking that BAs have a higher power/weight ratio, at the expense of maximum power input, which is why BA bass is usually anemic. I'll leave that for a more knowledgeable person though haha.

I have been working with the Fourier transform to the extent of using it to analyze stereo music on the fly and play it through N number of speakers, each speaker playing frequency components appropriate to the direction of the original stereo music per the analysis except widening the stereo image from 60 degrees to 360 (pm me if anyone is interested). And I fail to see the relevance of the Fourier transform to the topic at hand. Or at least, to how it negates any points we have made.

 

[JohnYang1997]

You're absolutely correct, most transducers (except for custom developed ones) cost next to nothing. with high end IEMs though, you pay for development costs, and probably the brand name. Perhaps I am misunderstanding your interpretation of frequency response, but to your other points, I suggest you research something called "Fourier Transform." It will provide you with the knowledge to understand the fundamentals of audio reproduction, and bypass the manufacturing limitations that you mention.

Here is a video for you:



Lastly, if you think EQ is cheating physics, you are not looking at the whole picture. Your points are correct, but you are limiting your thinking to manufacturing and tangible limitations. With all else held constant, the inertial property of the diaphragm remains.

It is interesting you mention distortion of BA drivers too, I was thinking that BAs have a higher power/weight ratio, at the expense of maximum power input, which is why BA bass is usually anemic. I'll leave that for a more knowledgeable person though haha.

You obviously didn't grasp the idea. The issue happens when the driver breaks up at 16khz but you try to eq the high frequency over the break up point.
We have already passed the era that transducer can't reproduce high enough frequency response. It's the shape at lower frequency that truly matters.

Also the idea of ba drivers are anemic has zero to do with inertial property. It's the Xm parameter and distortion at higher level. Also in frequency response you don't see much energy boost under 200hz compared to typical dynamic driver.
And it has something to do with natural resonance frequency, you lose frequency response under that frequency when you don't get 100% air tight seal. But it's all the same and coherent. You can check frequency response in your ears as well.
Also the implication that I don't know ft is hilarious.

 

[JohnYang1997]

One last thing before I go to sleep. 99% people who talks about inertia don't know physics or science. It's called mass. Quote me on that in the future.

 

[Joe Bloggs]

Also the implication that I don't know ft is hilarious.

Easy does it... at least he starts to know about it now so we're getting to be on the same level

 

[WhatToChoose]

... no. Any diaphragm that is capable of the whole audible frequency range will be capable of any combination of frequencies in any visually complex combination. You will e.g. notice that the bottom waveform does not change direction any more steeply than the 300Hz and is thus not any more demanding of high speed changes than the 300Hz waveform by itself.

You are right, both Dynamics and BAs for example will reproduce any tone designed for....at steady state. I am not very good at providing explanations, but let me try to give a scenario instead to paint a picture:

Let's say you have an audio file that is plays a 5kHz tone for 1000 microseconds, and then plays a 10 kHz tone from 1000 microseconds to 2000 microseconds. During that frequency transitional period at 1000 ms, a diaphragm with a higher power to weight ratio will be able to change frequencies faster than a diaphragm with a lower power/weight ratio (transient period). Given this, the best diaphragm would be one that has the shortest transient period physically possible, which would be one with the highest power/weight ratio. This example is simple, but in a music file with different tones starting and stopping constantly, transients must come into play.

 

[Joe Bloggs]

You are right, both Dynamics and BAs for example will reproduce any tone designed for....at steady state. I am not very good at providing explanations, but let me try to give a scenario instead to paint a picture:

Let's say you have an audio file that is plays a 5kHz tone for 1000 microseconds, and then plays a 10 kHz tone from 1000 microseconds to 2000 microseconds. During that frequency transitional period at 1000 ms, a diaphragm with a higher power to weight ratio will be able to change frequencies faster than a diaphragm with a lower power/weight ratio (transient period). Given this, the best diaphragm would be one that has the shortest transient period physically possible, which would be one with the highest power/weight ratio. This example is simple, but in a music file with different tones starting and stopping constantly, transients must come into play.

If the change in frequency happens instantaneously, it would involve all frequencies up to infinity and be beyond the capability of any driver on earth to reproduce. Luckily the ultrasonic components would also be beyond the capability of any human ear to discern. You would know this if you really knew the Fourier transform.

 

[WhatToChoose]

One last thing before I go to sleep. 99% people who talks about inertia don't know physics or science. It's called mass. Quote me on that in the future.

I suggest you and @Joe Bloggs learn some respect. Neither of you have provided evidence to the contrary, and have not been able to formulate any logical refutation. Your juvenile behavior reveals your true nature.

 

[WhatToChoose]

If the change in frequency happens instantaneously, it would involve all frequencies up to infinity and be beyond the capability of any driver on earth to reproduce. Luckily the ultrasonic components would also be beyond the capability of any human ear to discern. You would know this if you really knew the Fourier transform.

We are talking about physical limitations. Read the context, and try again.

 

[Joe Bloggs]

If the change in frequency happens instantaneously, it would involve all frequencies up to infinity and be beyond the capability of any driver on earth to reproduce. Luckily the ultrasonic components would also be beyond the capability of any human ear to discern. You would know this if you really knew the Fourier transform.

For example.
100Hz sine wave on the left, 500Hz sine wave on the right

Spectrogram:

The frequency spike at the change point goes up past 500kHz, the max frequency that can be analyzed (because the sampling rate tops out at 1000kHz)

I suggest you and @Joe Bloggs learn some respect. Neither of you have provided evidence to the contrary, and have not been able to formulate any logical refutation. Your juvenile behavior reveals your true nature.

I don't speak for JohnYang but I don't see where I have been disrespectful.

We are talking about physical limitations. Read the context, and try again.

Read the above, and understand that no driver on earth can handle the transient you envision, nor does any driver need to, for fidelity to humans.

 

[Joe Bloggs]

Zoomed in comparison pic of the transition point from 100Hz to 500Hz, before and after removal of frequencies above 24kHz (by downsampling to 48kHz and back up, if you want to know). Note that for all intents and purposes there *is no transient* anymore. The "transient" part of the transition can't actually be heard by any living human being. If a driver is capable of up to, say, 40kHz instead of 24kHz, it would be able to make the transition marginally sharper, but no human would be any the wiser for it.

 

[491838]

@Dealux forgive me for asking: is what you mean as dynamic is like in musical term of the difference between the softest/quietest passage and the loudest? and also do you mean that the more dynamic earphones are the ones with faster attack and decay in that?

Yes, for the first question. I don't know what makes a headphone more dynamic.

 

[chinmie]

I don't speak for JohnYang but I don't see where I have been disrespectful.

well, I'm only reading, so i put myself as outsider of this whole conversation, but from where I'm standing... the way you and john deliver the message... kinda does feel putting down others as not knowledgeable enough and beneath you. maybe you both are right on that, but still kinda sting a bit