[castleofargh]

It's where Harman's work and little statistical analysis gave 3 main preference groups. Is there a market for a non-dominant group? Probably, but depending on tuning, it will rapidly aim at a niche within a niche, which would still be interesting as a market population for sure, if said population had any clue about what they need/look for. Which they almost never do.
Just looking at the push against the Harman target(mostly from a point of view of ignorance), when their work suggests it should be favored by nearly 2/3 of the population, gives food for thought about the risks of aiming at a target FR that pleases even fewer people in a hobby where everything has to be dumbed down to being either right or wrong.


Also, welcome all to Sound Science, or sorry that an interesting discussion, unusually full of pertinent information that all IEM/headphone users could benefit from, got moved here to die. I found myself identifying(a little too much) with the "fun at parties HRTF guy" in @Resolve's video. And I do think, people who stay long enough in the hobby will end in that group, because once you learn and understand how much impact the HRTF and FR in a simplified way, have on our subjective experience, it makes no sense to ignore it. With the obvious caveat that someone must actually want to learn about hearing, acoustic, and really, himself, which is strangely rare in a hobby that prides itself on having a passion for sound and listening.

A FR graph from a given coupler, taken by a given guy, is telling us about that system and that IEM(not the model, that pair in particular). Another coupler, IEM pair, tips, or another placement's habit will show some differences in the graph. We need to know that, where it's more likely to change something, and be cautious about directly associating a given graph with our subjective experience, because we're also "different couplers" with different insertion habits, different tips and different IEMs.
The issues found with graphs are found with listeners, it's an acoustic fact.
But listeners also have differences in perception from the missing HRTF cues, plus a sensitivity to psychological biases (audio and non audio), plus hearing loss, plus, in this hobby, quite the vague and flowery subjective ways of describing the frequency response and how it changed that listener's experience.

So when wondering if we'll purchase an IEM(so we can listen to it!), the feedback of a buddy or a review is in practice even less reliable about the FR than a graph. And given how FR is a main variable when it comes to preference and subjective experience, it feels really silly to argue against FR graphs. Sure listening is better, but again, that almost always happens later for the average Joe. In the meantime the false argument is to say that not relying on graph data is better than knowing more, which is a little dumb.
Of course FR isn't everything and as I just said a FR graph isn't perfect by any standard. But almost all audiophiles underestimate how much of their experience really is just a matter of FR. Of that much I am sure.

 

[nephilim32]

That’s exactly the point I was making There are many more outliers beyond the objective measurement at play, that could yield a better experience with a different calibrated vehicle/ IEM.

Like you mention, the suspension, the brakes, are all very crucial secondary components (think eartip, distance from bore to eardrum, some might downplay). Then comes the user error/ control, tons of variables at play, all must be accounted for.

Yeah I like the performance car analogy. It is a good way to look at it.

 

[bigshot]

Looks like someone started to talk about objective standards of sound reproduction where that wasn’t welcome and the whole discussion got scooped up and deposited in sound science.

 

[gregorio]

My guess is that now the thread has been moved to the Sound Science subforum, where we’re allowed to challenge/refute assertions and claims, many of the participants will decline to participate any further. This is despite having made various assertions specifically regarding science/research, as some of them are (to be overly polite/charitable) somewhat misleading. For example, the claim that there isn’t much research or the quote below, which appears to be more of a re-worded marketing myth than a factual assertion?

Most of the research that's referenced in the objective side of the audio space are all white papers that are made by institutions and organisations that are very much more academically inclined than hobby focused. What I mean to say is the the bias towards subjectivity in the hobby spaces likely has next to no impact on the current pool of research in this field, despite it's commonality in discourse in our spaces.

G

 

[KinGensai]

Too much weight is given to how much information is conveyed by an FR graph IMO, FR graphs don't factor in the time domain at all and don't measure how accurately the transducer reproduces complex signals like music (as compared to the single tone sweep FR measurements use in general). It's essential to look at metrics like THD, IMD, and CSD to get a better idea of what's going on with the fidelity and tonality of a transducer.

My experience with IEMs across the budget range points to CSD and IMD being the most important and improved metrics that set apart junk from quality because of the reduced noise caused by intermodulation errors they denote at the analog stage.

 

[sghound]

I have stayed out of this till now. What the heck, as you all gave me a headache ( yet I keep reading ) I should be allowed to give someone else a headache.

I am no expert, but I have done some research in this area and what has always bothered me about these simple graphs is that they are exactly that; simplified graphs. Here is what I mean. Pick a note or notes on that graph and play them with a piano. Now a guitar. The graphs of the notes played by the piano and the guitar will be different as the overtones will be different which creates the unique timbre of each. For example:
Below are three charts of a flute, piano and trumpet all playing middle C.



All 3 are different accounting for the different timbre of each instrument at middle C.

In an IEM graph all we see is a single line going across all the frequencies. There is no timbral relationship depicted. Therefore how can one make any guess about the timbral sound of said IEM? This is just one example of a lack of data needed to understand how an IEM sounds.

whole discussion shoulda ended with this but noooooooooo the nubs still dun get it.

 

[castleofargh]

whole discussion shoulda ended with this but noooooooooo the nubs still dun get it.

I personally don't understand what his argument is. I tried, and I don't get it.
Musical instruments can even have harmonics that are louder than the fundamental frequency. If we were to look at the harmonic content of an instrument like we do for an IEM, that would mean values near and sometimes above 100% THD. Hans get the flammenwerfer!!!
The usually accepted amount for headphones and IEMs is around 1%(harmonic content 40dB below the main tone) or lower, which is subjectively quiet and does not give a subjective feeling similar to timbre because the fundamental is just so much louder.


What will affect the timbre of a recorded instrument through the IEM is how that IEM will change the amplitude of some harmonics. But how can we represent that change in amplitude at different frequencies caused by the IEM? With a FR graph! Back to square one.

 

[VNandor]

I have stayed out of this till now. What the heck, as you all gave me a headache ( yet I keep reading ) I should be allowed to give someone else a headache.

I am no expert, but I have done some research in this area and what has always bothered me about these simple graphs is that they are exactly that; simplified graphs. Here is what I mean. Pick a note or notes on that graph and play them with a piano. Now a guitar. The graphs of the notes played by the piano and the guitar will be different as the overtones will be different which creates the unique timbre of each. For example:
Below are three charts of a flute, piano and trumpet all playing middle C.



All 3 are different accounting for the different timbre of each instrument at middle C.

In an IEM graph all we see is a single line going across all the frequencies. There is no timbral relationship depicted. Therefore how can one make any guess about the timbral sound of said IEM? This is just one example of a lack of data needed to understand how an IEM sounds.

All of these sounds can be broken down into a sum of singular frequencies called harmonics as the graph you provided notes it. The sum of these frequencies is called the "spectrum" of the sound. The frequency response shows how the IEM manipulates the spectrum of sound. Pressure changes over time can be calculated from the spectrum and the other way around also holds true, it could be even done on pen and paper for simple cases. Sound can be described by either as a function of pressure over time or as a function of pressure over frequencies. They are equivalent in a mathematical sense but certain things can be seen and be understood easier in one form than in the other. Just by looking at the squiggly line, you can't really tell how a trumpet would sound, can you? Even though the pressure changing over time perfectly describes sounds since that is the definition of sound. The data is there by definition, it just isn't easy to read it.

 

[theveterans]

So I do hear these things, but what people need to realize is that treble features at the ear drum have a massive influence on how things come across. Tones that dominantly token midrange frequencies still have their upper harmonic content meaningfully influenced by treble features. The same is true for bass. It really is worth watching Precog's video on this, as he does a good job of explaining that part in particular.


Moreover, those of us who are doing manual tone gen stuff and using it to EQ to various features of our HRTF, we are all saying essentially the same thing here. If you haven't done that, I recommend giving it a try. Again, not to a target, but to the features you've identified as being part of your own HRTF.

I think timing/accuracy/jitter of harmonic reproductions as well as inherent amplification is needed to reproduce those treble features at the eardrum. A simple sine sweep for FR and 1KHz FFTs won't show any of those timing of harmonic reproduction differences (only amplitude of a specific fundamental frequency is meaningful as harmonics are typically masked from very low distortion figures) that differentiates how soundwaves of an instrument/timbre/detail/resolution/spatial information is perceived at the eardrum

 

[theveterans]

I have never heard of anyone describe bass texture using the words "forward" or "intense" but rather adjectives like "rich, dry, deep, plasticky, one noted, bloated" etc.

These bass texture terminology don't exist in the audio science hence why you're such on different wavelengths. For an audiophile these are reproduced through "system synergy" while for audio science it's a combination of FR + impedance matching or damping factor + SPL listening levels

 

[Somafunk]

Also, welcome all to Sound Science, or sorry that an interesting discussion, unusually full of pertinent information that all IEM/headphone users could benefit from, got moved here to die.

I figure there is, and always will be, a certain percentage of the audio hobby that do not care to be informed that there is no magic pixie dust or special sauce in their chosen iem’s, cables, dacs, they do not care for graphs or talk of applying eq to change the sound but would rather talk of a cable purchase that gives them a sheen of air to the vocals, or increasing the low end heft etc.

There is a recent review of an $8000 iem cable on the forum, I’d happily pay to see a genuine blind test using db matching with a selection of iem cables, perhaps a sound science stall/booth at every canjam with various setups and a big neon sign proclaiming “get your golden ears badge here”.

As someone with a mechanical/electronic engineering background I have come to conclusion that I should’ve went into the cable game 25 years ago.

 

[Resolve]

I think timing/accuracy/jitter of harmonic reproductions as well as inherent amplification is needed to reproduce those treble features at the eardrum. A simple sine sweep for FR and 1KHz FFTs won't show any of those timing of harmonic reproduction differences (only amplitude of a specific fundamental frequency is meaningful as harmonics are typically masked from very low distortion figures) that differentiates how soundwaves of an instrument/timbre/detail/resolution/spatial information is perceived at the eardrum

When it comes to this type of stuff, the question usually becomes, at what point is it audible? And the problem there is that the answer is a bit of a moving target because it'll likely depend on the individual. It's kind of similar to harmonic products in general, because in practice... you won't find two products that measure identically in this respect. So if we say "there's an effect there at the ear drum", that might be true, but is it really responsible for subjective goodness/badness of the experience? It might be for one person, but not for someone else. I would love to see more research done on audibility thresholds for stuff like this. At the very least, when isolating FR, we can be sure this is massively influential on the subjective goodness/badness, and so that's where I think it makes sense to look - at least initially.

 

[theveterans]

When it comes to this type of stuff, the question usually becomes, at what point is it audible? And the problem there is that the answer is a bit of a moving target because it'll likely depend on the individual. It's kind of similar to harmonic products in general, because in practice... you won't find two products that measure identically in this respect. So if we say "there's an effect there at the ear drum", that might be true, but is it really responsible for subjective goodness/badness of the experience? It might be for one person, but not for someone else. I would love to see more research done on audibility thresholds for stuff like this. At the very least, when isolating FR, we can be sure this is massively influential on the subjective goodness/badness, and so that's where I think it makes sense to look - at least initially.

Assuming two different devices/DAPs with completely different architectures (Cayin RU7 1-bit delta sigma against A&K SP3000 multibit, delta sigma) where both have low distortion measurements at least 70dB down from fundamental (no way to hear the harmonics physically next to fundamental especially with audio masking effects) and ample power to drive an IEM comfortably (70dB SPL calibrated volume matched to 0.01 dB) and a measured ruler flat FR, would both device sound identical under A/B testing with the same IEM (e.g. let's take the Subtonic Storm in this case)? If not, do you think despite the measurements of the devices, these two different devices influence the Storm to deliver a different FR at the eardrum where a lot of manufacturers claim that different op-amps have their own signatures even though those op-amps are well below the threshold of audible masking?

 

[Resolve]

Assuming two different devices/DAPs with completely different architectures (Cayin RU7 1-bit delta sigma against A&K SP3000 multibit, delta sigma) where both have low distortion measurements at least 70dB down from fundamental (no way to hear the harmonics physically next to fundamental especially with audio masking effects) and ample power to drive an IEM comfortably (70dB SPL calibrated volume matched to 0.01 dB) and a measured ruler flat FR, would both device sound identical under A/B testing with the same IEM (e.g. let's take the Subtonic Storm in this case)? If not, do you think despite the measurements of the devices, these two different devices influence the Storm to deliver a different FR at the eardrum where a lot of manufacturers claim that different op-amps have their own signatures even though those op-amps are well below the threshold of audible masking?

It would probably depend on the rest of the measured response, like if there's an effect on FR, OI and so on. But, supposing all else is equal, I suspect the sound would be identical in blind A/B tests for most people. Like 70dB is a lot lower, and provided the harmonics are within the masking window I can't imagine it would have an effect. At the same time, I do know that Cameron can hear differences in ABX tests that most people cannot. So I don't want to say this is guaranteed to be below the audible threshold. It also probably depends on what the result is for distortion vs frequency as it rises, which often isn't tested. But that's all stuff within Cameron's wheelhouse, not mine.

I basically came to the conclusion on comparably well-performing source gear that while there are times I feel I hear a difference (with solid state sources), whenever I've run an ABX... I've never found that it made the experience better/worse to the point where I felt there was a reason to care. Tube-based sources is a different matter entirely.

 

[gregorio]

When it comes to this type of stuff, the question usually becomes, at what point is it audible? … I would love to see more research done on audibility thresholds for stuff like this. At the very least, when isolating FR, we can be sure this is massively influential on the subjective goodness/badness, and so that's where I think it makes sense to look - at least initially.

I don’t know why we keep seeing stuff like this in the audiophile world. I can only assume that audiophile marketing has a vested interest in keeping the audiophile community in the dark, through “lies of omission” and misrepresenting the facts, etc.?

To answer your point, “yes” that is indeed where “it makes sense to look - at least initially”, which is exactly why the early psycho-acousticians looked there initially in the 1890’s, then did so again in far greater detail in the 1930’s when advances in audio electronics allowed it and at numerous times since! So, it makes no sense to do even more research when there’s already been so much going back over 90 years. Nevertheless, we keep seeing points like this in the audiophile community.

G