[Zachik]

The gist is (1) FR at ear drums is everything there is

I disagree. Yet again, this is presented as a fact.... where it should be presented as YOUR opinion.

Edit to clarify: Sound waves hitting your ear drums is everything. THAT - I agree with. But FR captures only part of the information of the sound waves that hit your ear drums. So, it is partial information. It represents frequencies that hit your ear drums, and the loudness of said frequencies. But it lacks attack / decay information (i.e. time domain) and probably other information as well (which represents timbre for example).

 

[o0genesis0o]

I disagree. Yet again, this is presented as a fact.... where it should be presented as YOUR opinion.

Edit to clarify: Sound waves hitting your ear drums is everything. THAT - I agree with. But FR captures only part of the information of the sound waves that hit your ear drums. So, it is partial information. It represents frequencies that hit your ear drums, and the loudness of said frequencies. But it lacks attack / decay information (i.e. time domain) and probably other information as well (which represents timbre for example).

Devil advocate: wouldn't note attack lies at 5kHz and 7-8kHz?

Anyhow, I would be interested in exploring linearity, group delay and that sorts of thing, to try to understand how IEMs behave against signals of changing amplitudes rather than a static snapshot (aka FR). Say, xMEMs drivers are well known for "speed". How does that look like from an objective measurement perspective? If we agree that it's not FR, then what else can show precisely the difference between xMEMs and real EST and Sonion EST and Knowles BA super tweeters and a plain old DD? Would that be a single measurement, or an interaction of different things, say peaks and valleys across FR?

Tbh, I buy the idea that PERCEIVED detail is a function of a smooth and extended FR with minimal masking. But I don't believe that FR alone can explain the difference in terms of the incisiveness of the instrument separation. It's not about brightness or less lower midrange, since we do have plenty of brighter and lean IEMs that just collapse when facing a large orchestral piece, whilst some warm and thick IEMs can plow through with ease.

 

[aaf evo]

I told aaf that he can use tools that are eminently available to better understand the things he said he wishes to learn, and he declined—which is fair, but it does remain true that anyone reluctant to mess around with EQ is hamstringing their potential understanding of how frequency response effects things like dynamics, detail, soundstage, timbre etc. As I've said, the people equalizing the music we listen to generally understand how this relationship works, as EQ is one of the many tools they use to shape these things.

My "lead a horse to water" comment was not about him being obstinate or me being fed up or anything like that, it was shorthand for the fact that:

a) there's no reason for me to force someone into doing something they don't want to do, nor berate them for not doing so. I told him what to do, he declined, that's that.

b) it's really hard for anyone—including me—to give someone an EQ point or profile and say "This is what bass texture sounds like". I can talk about what those terms mean to me as they commonly relate to subjective qualities, but there is no guarantee he agrees on which FR factors influence which quality.

We can use words to describe the experiences we're having, but the only ones who are sure of what we're saying are ourselves. Or in other words, I can lead a horse to my idea of bass texture, but I can't guarantee that he agrees.

I am still happy to give a more thorough explanation (or address your claims numbered #1 and #2) but only elsewhere, as I think it's rather obvious that this discussion isn't wanted here. Feel free to DM me here or find me on discord (listener.) if you'd like to chat more about it.

After all, this is a thread about one of the only flagship IEMs that actually has great midrange tone IMO, so it's not like there's much for a nerd like me to complain about anyway

Eh, I'd say the discussion was welcome but when it's presented as a fact or superior ideology (and not an opinion) it can come off a little sour I suppose.

I know very well how frequency response works and how it can effect aspects of sound reproduction like you have stated with mentioning dynamics, detail, staging, and timbre. Some of this I am agreeing with. I just am not agreeing that FR is what we "pay" for when we buy an IEM nor do I think it is entirely responsible for what we experience inside our ears. I have repeatedly brought up that the situation in that you can take two identical frequency responses but have them be output by different driver types and they will likely not sound the same, therefore the "experience" would not be the same. If you think this has to do with how we perceive sound at our eardrum and not the laws of physics between different driver technologies or implementation, then I really don't know what else I can add to this conversation. Again, if it were flat out that simple we would all just be EQ'ing everything and headphones.com wouldn't have a business to run or $500 IEMs to compare to a Sennheiser HE-1 and take pre-orders for.

I would love to see someone EQ an AFUL MagicOne (1 BA IEM) to output a similar "subjective bass quality" as the Symphonium Titan, the IEM might explode.

It's no different with televisions/monitors. There are different panels that have their own pros and cons. You cannot calibrate an IPS panel to have the same level of blacks as an OLED panel which flat out uses different technology to achieve what it does.

Anyways, it's been a mostly fun discussion. I really enjoy the Storm and I am thankful such a product exists. But I think I am going to spend some time EQ'ing my Apple lighting EarPods to see if I can replicate the Storm experience, that way I can save several thousands of dollars and put it towards a nice vacation instead.

 

[Zachik]

Anyways, it's been a mostly fun discussion. I really enjoy the Storm and I am thankful such a product exists. But I think I am going to spend some time EQ'ing my Apple lighting EarPods to see if I can replicate the Storm experience, that way I can save several thousands of dollars and put it towards a nice vacation instead.

Who are you kidding buddy? Vacation? pffffft.
You are coming to CanJam SoCal, and you may call it vacation if you so choose to do...

 

[Somafunk]

I would love to see someone EQ an AFUL MagicOne (1 BA IEM) to output a similar "subjective bass quality" as the Symphonium Titan, the IEM might explode.

Depends on how low the distortion on the aful magic one is at your listening level, possibly could be eq’d

 

[Resolve]

How can you tell me things that like driver coherency and bass quality are something that can be shown up in a measuring rig?

So, driver 'coherency' could refer to many different things, so this runs afoul of the private language problem - it's good reading on this topic actually. But when it comes to 'bass quality', Listener and I have both given explanations for how frequency response at the ear drum is ultimately responsible for goodness/badness here. We're NOT saying "measurements on a graph shown relative to a target exhaustively explain subjective goodness/badness for bass quality", even though you definitely can glean some of that - or predict some of that from the graph. But let's go through those points again here:

1. FR plots shown on 711-based systems don't provide an accurate picture of how the IEM performs on a human due to the less than accurate acoustic impedance, yielding inaccuracies at low and high frequencies. Simply put, you cannot expect to get a good read of the FR in the bass and upper treble for IEMs with most measurements graphs found in the wild. But that doesn't mean it's NOT FR that's responsible for your experience.

2. Tones that are dominantly experienced as in the bass still token a wide range of frequencies outside of the bass range. So when you hear a kick drum, for example, that tokens way more than just the bass frequencies, but you may associate its goodness more commonly with "bass quality".

3. Understanding this involves a more careful analysis of the FR for the relationship between various frequency ranges that impart goodness and badness to the tones you're experiencing. And quite frankly, this is not something I'd ever expect someone to do - but it is possible. A good example of this is the "soundstage effect". This is uniquely to do with a relationship between mids and lower treble. Pull back on the mids and increase the lower treble, and you get this widening effect. Do the opposite, and things sound closed in.


By all means, believe what you want - but I've also been there. What helped me understand this stuff was doing tone generated sweeps and EQ. Even if you don't agree with me on this, take this consideration in good faith - try some of this out and see how things change for you. Yes... we have the luxury of state of the art measurement rigs, but manual tone-generated sweeps will reveal a LOT about how headphones perform on your head, and the differences with how they graph. Try getting into EQ to see how changing some of these key regions affect your perception of some of these intangibles.

 

[Somafunk]

By all means, believe what you want - but I've also been there. What helped me understand this stuff was doing tone generated sweeps and EQ.

Yep, what he said ^ there

 

[aaf evo]

So, driver 'coherency' could refer to many different things, so this runs afoul of the private language problem - it's good reading on this topic actually. But when it comes to 'bass quality', Listener and I have both given explanations for how frequency response at the ear drum is ultimately responsible for goodness/badness here. We're NOT saying "measurements on a graph shown relative to a target exhaustively explain subjective goodness/badness for bass quality", even though you definitely can glean some of that - or predict some of that from the graph. But let's go through those points again here:

1. FR plots shown on 711-based systems don't provide an accurate picture of how the IEM performs on a human due to the less than accurate acoustic impedance, yielding inaccuracies at low and high frequencies. Simply put, you cannot expect to get a good read of the FR in the bass and upper treble for IEMs with most measurements graphs found in the wild. But that doesn't mean it's NOT FR that's responsible for your experience.

2. Tones that are dominantly experienced as in the bass still token a wide range of frequencies outside of the bass range. So when you hear a kick drum, for example, that tokens way more than just the bass frequencies, but you may associate its goodness more commonly with "bass quality".

3. Understanding this involves a more careful analysis of the FR for the relationship between various frequency ranges that impart goodness and badness to the tones you're experiencing. And quite frankly, this is not something I'd ever expect someone to do - but it is possible. A good example of this is the "soundstage effect". This is uniquely to do with a relationship between mids and lower treble. Pull back on the mids and increase the lower treble, and you get this widening effect. Do the opposite, and things sound closed in.


By all means, believe what you want - but I've also been there. What helped me understand this stuff was doing tone generated sweeps and EQ. Even if you don't agree with me on this, take this consideration in good faith - try some of this out and see how things change for you. Yes... we have the luxury of state of the art measurement rigs, but manual tone-generated sweeps will reveal a LOT about how headphones perform on your head, and the differences with how they graph. Try getting into EQ to see how changing some of these key regions affect your perception of some of these intangibles.

You still seem to be disregarding a point I have brought up several times now. So I will just copy and paste what I posted above:

I know very well how frequency response works and how it can effect aspects of sound reproduction like you have stated with mentioning dynamics, detail, staging, and timbre. Some of this I am agreeing with. I just am not agreeing that FR is what we "pay" for when we buy an IEM nor do I think it is entirely responsible for what we experience inside our ears. I have repeatedly brought up that the situation in that you can take two identical frequency responses but have them be output by different driver types and they will likely not sound the same, therefore the "experience" would not be the same. If you think this has to do with how we perceive sound at our eardrum and not the laws of physics between different driver technologies or implementation, then I really don't know what else I can add to this conversation. Again, if it were flat out that simple we would all just be EQ'ing everything and headphones.com wouldn't have a business to run or $500 IEMs to compare to a Sennheiser HE-1 and take pre-orders for.

It's no different with televisions/monitors. There are different panels that have their own pros and cons. You cannot calibrate an IPS panel to have the same level of blacks as an OLED panel which flat out uses different technology to achieve what it does.

 

[emdeevee]

Wow, I've read through all of this and am beyond confused. I'm no engineer or scientist. Quite the opposite, I'm a lawyer designed to argue, but I can't argue with the non-understandable.

Basically, the most clear thing for the layman to glean from this conversation is that eyes studying squiggly line patterns do not equal the experience of listening. Nor can such a squiggle really describe what you're going to hear. The second, more esoteric thing to be gleaned is how difficult it must be for scientists to deal with audiophiles

But I'm still learning from all of it!

 

[Resolve]

I have repeatedly brought up that the situation in that you can take two identical frequency responses but have them be output by different driver types and they will likely not sound the same, therefore the "experience" would not be the same.

If you're talking about two measured FRs being the same, this is not an unsolved question. They have different FRs at the ear drum, hence why they don't sound the same - this is acoustic Z or length mode and air volume effects and other such anatomy-related variables. If you're talking about two identical FRs at the ear drum, then the answer is... how do you know they're identical at the ear drum? In all likelihood, they're not.

If you think this has to do with how we perceive sound at our eardrum and not the laws of physics between different driver technologies or implementation, then I really don't know what else I can add to this conversation.

Sound at the ear drum is literally the consequence of physics. Driver technologies and implementation are largely responsible for that. Again I'm not saying the "driver story" isn't responsible for the FR at the ear drum, I'm saying you can get there in many different ways, and ultimately that's more important than the driver story.

Again, if it were flat out that simple we would all just be EQ'ing everything and headphones.com wouldn't have a business to run or $500 IEMs to compare to a Sennheiser HE-1 and take pre-orders for.

I'm not going to make that comparison, personally... that's a DMS thing. My comparison is more to do with the Dusk, which isn't sold by Headphones.com. And I'm also not necessarily saying it's better. It could be better for some, but also worse for some, depending on how people want their music to sound. But to the other point, yes... I recommend EQing everything! Not everything needs EQ but everything can be better personalized with EQ.

It's no different with televisions/monitors. There are different panels that have their own pros and cons. You cannot calibrate an IPS panel to have the same level of blacks as an OLED panel which flat out uses different technology to achieve what it does

And these things are in fact measurable.


We're not saying "FR is all there is", we're saying FR describes far more than just 'tonality'.

 

[AxLvR]

It's no different with televisions/monitors. There are different panels that have their own pros and cons. You cannot calibrate an IPS panel to have the same level of blacks as an OLED panel which flat out uses different technology to achieve what it does.

Exactly this. Very good comparison!

 

[Resolve]

Andrew (@Resolve), did any of your Storm seatings look like the dotted line in this one (below)? (I was using whatever single-dome silicone tip came installed on it, as sent to me by Subtonic.) I also did a single pass each on both the launch and standard editions (using the same eartips), and they were the same.

Not quite - I didn't get the dip that deep. Here are the results from multiple different tips, all being stock except the small white ones:

We also have measurements of a second unit that Listener measured:

 

[aaf evo]

If you're talking about two measured FRs being the same, this is not an unsolved question. They have different FRs at the ear drum, hence why they don't sound the same - this is acoustic Z or length mode and air volume effects and other such anatomy-related variables. If you're talking about two identical FRs at the ear drum, then the answer is... how do you know they're identical at the ear drum? In all likelihood, they're not.




Sound at the ear drum is literally the consequence of physics. Driver technologies and implementation are largely responsible for that. Again I'm not saying the "driver story" isn't responsible for the FR at the ear drum, I'm saying you can get there in many different ways, and ultimately that's more important than the driver story.




I'm not going to make that comparison, personally... that's a DMS thing. My comparison is more to do with the Dusk, which isn't sold by Headphones.com. And I'm also not necessarily saying it's better. It could be better for some, but also worse for some, depending on how people want their music to sound. But to the other point, yes... I recommend EQing everything! Not everything needs EQ but everything can be better personalized with EQ.



And these things are in fact measurable.


We're not saying "FR is all there is", we're saying FR describes far more than just 'tonality'.

So I've been in this hobby for what, like 15 years? Idk. I have probably heard several hundreds of sets priced anywhere from like $300 to $9000, with most of them likely costing over $1500. Throughout my time in the hobby since graphs have been easily accessible online I've had the ability to read these graphs and listen to enough sets to where I can get an idea and relation of how my HRTF will perceive certain tunings "at the ear drum" and I am able to tell if perhaps maybe something may be too shouty for me, too bright, etc etc. I think this is fairly is a fairly common ability between a lot of IEM users in the hobby at this point, no?

What I will never understand is how you can put so much weight into frequency response graphs that you can make a claim that you are paying for the tuning which gives you the "experience" you are getting. I just can't resonate with this. And this brings me back to my comparison above about the IER-Z1R and Traillii. With what data we have available to us, with the graphs most of us listeners understand how to read, and based off of those who have heard both IEMs in their own ears, I think it would be very fair to say these IEMs just sound nothing alike in the lows. However based off of looking at the graph comparisons you'd think that these IEMs may have a similar bass presentation. But you would never know this until you push play as the the 12mm DD in Z1R handles bass much differently than the dual BA in the Traillii.

From the beginning I have agreed that FR can show more than tonality, but I don't think it provides as much as you are claiming it does. Or certainly not to the extent to where a statement such as saying that Storm is only giving a certain experience because of how it is tuned.

Here is another example, these two IEMs I believe use the exact same driver but have different internal structures. One of them (and I already forgot which one because I removed the tags) has MUCH stronger sense of L/R separation and imaging whereas the other literally sounds like a blob of compressed sounds being presented in mono audio. I am not sure how anyone can look and come to a conclusion on which is which. If you think the sound at the eardrum between these sets is what gives such a difference between the perceived technical attributes between both, then it is oddly consistent amongst most people who have heard both.

Anyways, I really really must get to bed, I can't even think straight at this point and I'm not sure anything I wrote above makes sense anymore. It was a good chat. But we have different approaches and experiences regarding this topic and that is okay.

 

[Resolve]

So I've been in this hobby for what, like 15 years? Idk. I have probably heard several hundreds of sets priced anywhere from like $300 to $9000, with most of them likely costing over $1500. Throughout my time in the hobby since graphs have been easily accessible online I've had the ability to read these graphs and listen to enough sets to where I can get an idea and relation of how my HRTF will perceive certain tunings "at the ear drum" and I am able to tell if perhaps maybe something may be too shouty for me, too bright, etc etc. I think this is fairly is a fairly common ability between a lot of IEM users in the hobby at this point, no?

What I will never understand is how you can put so much weight into frequency response graphs that you can make a claim that you are paying for the tuning which gives you the "experience" you are getting. I just can't resonate with this. And this brings me back to my comparison above about the IER-Z1R and Traillii. With what data we have available to us, with the graphs most of us listeners understand how to read, and based off of those who have heard both IEMs in their own ears, I think it would be very fair to say these IEMs just sound nothing alike in the lows. However based off of looking at the graph comparisons you'd think that these IEMs may have a similar bass presentation. But you would never know this until you push play as the the 12mm DD in Z1R handles bass much differently than the dual BA in the Traillii.

From the beginning I have agreed that FR can show more than tonality, but I don't think it provides as much as you are claiming it does. Or certainly not to the extent to where a statement such as saying that Storm is only giving a certain experience because of how it is tuned.

Here is another example, these two IEMs I believe use the exact same driver but have different internal structures. One of them (and I already forgot which one because I removed the tags) has MUCH stronger sense of L/R separation and imaging whereas the other literally sounds like a blob of compressed sounds being presented in mono audio. I am not sure how anyone can look and come to a conclusion on which is which. If you think the sound at the eardrum between these sets is what gives such a difference between the perceived technical attributes between both, then it is oddly consistent amongst most people who have heard both.



Anyways, I really really must get to bed, I can't even think straight at this point and I'm not sure anything I wrote above makes sense anymore. It was a good chat. But we have different approaches and experiences regarding this topic and that is okay.

All good, I'll respond in the morning. It's an important topic. Have a good night.

 

[dleblanc343]

Good lord, too much science up in here!

I jest… but really I think there’s a couple factors people don’t talk about which breaks down critical analysis to a very efficient degree.

1. Volume-dependent HRTF

I hate that this is a thing, but it’s probably the biggest contributing factor to different sound nuance from user to user. It also takes into account user preference as well. And it’s especially more variable with more colored sets (I.E IER-Z1R, Elysian Annihilator). Sometimes just a volume click up or down on your source can lead to a transformative change - welcome to electronics!

2. Driver engines employed.

Different transducer technologies will sometimes have drastically different output even if aligned near identically on target. A different component spec even within the same driver tech (let’s say Knowles vs Sonion BA’s) also is a huge contributor. Some drivers are just better than others, and usually it’s reflected in the component cost. Now taking just the raw individual performance of a driver on bench and putting it into application is another thing altogether. Where you are crossing over your frequencies and calibrating your driver series will open up tons of variable results - this then comes down to tuner’s expertise and often more than not their subjective preference from trial and error or simply aligning to targets. Don’t forget, there was a time where manufacturers were tuning to the diffuse field target, even with iem’s, and in 2024 we know that this would sound absurdly poor.

On the topic of Frequency Response, I think it’s overrated. I won’t drive a car that performs best on a dyno test because I know that won’t translate to the track. I do blend a lot of my subjective beliefs and objective beliefs and consider myself a “measurebator” to a degree, but go figure some of my favorite gear are capitalizing on listening realism moreso than frequency realism and just do it better.

Recently had the chance to try a very hyped set from Audiosense - DT200, and while the tuning and tonality on it was superb, it’s probably the worst thing I’ve heard this year in terms of reproducing my music. Tons of transient smearing, poor dynamics, and wooly bass - just no tech all tone. This is an example of many hundreds of products, all of which having their audience - music listening is a subjective hobby really.

Now I’m motivated to measure Storm, I’ve been thoroughly enjoying it and it’s one of those elusive products that is arguably best on paper technically but also very well tuned. I think it deserves its status.