[WhatToChoose]

Ok, I was going to send you a pm, but I can't because of your privacy settings. Ok, so I'll try to tame this down a bit and cut through a bit less hype

Here's my take on 64 Audio.

The issue is primarily how one determines product pricing. You can figure out your R&D, manufacturing, distribution and marketing costs and add a reasonable profit margin. Or... you can simply charge the maximum price you think the market will tolerate. It doesn't take Einstein to figure out which of those two camps 64 Audio sit firmly in.

And 64 Audio products are really nothing extraordinary. They use standard off-the-shelf components that you can buy for pennies from sites like mouser.com or diyearphone.com. TIA (tubeless in-ear audio) is what many other manufacturers have been doing for ages. FLC, Campfire Audio, Beyerdynamic, Sennheiser, Shure, etc, don't make a song and dance about it or charge a 1000% mark-up for it. A couple of $0.50 balanced armatures without a driver tube doesn't justify a $3500 price tag.

And then there's the even sketchier side of their business. 64 Audio came to prominence during a collaboration with Stephen Ambrose. They licenced his 'ADEL' module (the Ambrose Diaphonic Ear Lens). The claim was that this device protected your hearing by limiting your exposure to 'pneumatic pressure'. In hindsight, this is almost comical. Anyway, here is the background...

The sound pressure level at your eardrum is just a function of the normal stress imparted by the motion of the air molecules against your eardrum (tympanic membrane). Those air molecules create a force, which - if different from the pressure on the other side of your eardrum - creates a net force causing the eardrum to move, in turn moving the fluid and cilia in the inner ear and registering sound in your brain. Ambrose made a big deal about a mechanism called the stapedius reflex, which causes the tympanic membrane to tighten and protect the cilia against loud sound - and even occlusion noise when you vocalize. This is a good evolutionary trait because current science knows no way to regenerate the cilia once you lose them.

When you insert an IEM (or put on an over-ear can), your eardrum should be maintained at equilibrium - in other words, in an ideal world nothing should happen just by putting on a pair of headphones that causes you to experience compression (pressure) or expansion (vacuum). With well-sealing tips on non-ported IEMs, it is possible to push the tip in far enough to create a seal, and then further still to raise the mean pressure on the eardrum, or conversely, create a seal and then have the IEM tug slightly due to the weight of the cable, creating what some call a vacuum seal. This might be slightly uncomfortable, but it's very unlikely to be harmful unless you're jamming or yanking your IEMs in and out of your ear canals extremely quickly. Inserting an IEM creates only a tiny fraction of the mean pressure difference you'd experience if you went on a plane or dived a few feet underwater. You can always equalize that pressure again by swallowing or yawning. It's often a good idea to open your jaw when inserting an IEM and then you'll never have this issue anyway.

What is going to adversely affect your hearing long term are large sustained oscillatory motions of the cilia. I'll agree with Ambrose that we all want to avoid that because we want to protect our hearing.

So far so good. Here's the problem with the ADEL marketing. They claim there's this terrible thing called "pneumatic pressure". This is garbage. There is simply no such thing. Pressure is pressure - the same normal force created by the same molecules. An increase in pressure simply means you've squeezed more air molecules into the same space at the same ambient temperature. Ambrose's video illustration of a garden hose being held with a thumb over the end shows a complete ignorance of fluid mechanics. Water flow in a garden hose is a fixed mass-flow rate of an essentially incompressible fluid. Air is compressible and there is no mechanism in dynamic, balanced armature, planar magnetic, electrostatic or any other acoustic driver to create a constant flow rate or a net (mean) pressure shift. That is simply not how headphones work. All headphones are zero net-mass-flow oscillators.

Now imagine the following. Imagine you have a driver in a tube which splits or forks into two identical tubes. At the end of one is your eardrum; at the end of the other is the ADEL membrane. Let's first imagine that the structural response of the ADEL membrane is identical to that of your eardrum. (And the pressures are already the same on the other side of both, since your eustachian tube will be at ambient pressure.) Now we'll move the driver with a sound source. We can express the driver motion as A.cos(B.t+C), where A is the amplitude, B is the frequency and C is a phase shift. Both eardrum and ADEL membrane move in sync because we have a perfectly symmetrical setup. Let's call this IEM 1:


Now consider IEM 2. In IEM 2, I have a setup which is exactly half of IEM 1. Let's say I couldn't afford an ADEL module, so I have just half a driver tube which runs to only one secondary tube that goes directly to my eardrum. So the first scenario (IEM 1) is completely symmetric with the driver tube splitting and running to two identically-responding eardrums/membranes. Again, I'm going to move the driver in IEM 2 with the exact same A.cos(B.t+C) motion. What do I hear differently from IEM 1? Absolutely nothing. Mathematically, the two scenarios create absolutely identical responses at the eardrum. The only difference is IEM 2 is moving a driver that has half the area of IEM 1, so it's simply moving half the amount of air and needing a lower power, i.e., a lower volume. That is absolutely the only difference.

Now let's consider what happens in IEM 1 if you use a nicely compliant membrane (one which is structurally less rigid and more flexible than the eardrum) to absorb all those nasty "pneumatic pressures". What happens now? Well, yes, to some degree, one could argue it will "protect your hearing", because if the membrane can flex more easily than your eardrum, it would somewhat lessen the SPL measured at the eardrum. But you would simply perceive this as a quieter sound - and most people would just turn up the volume to compensate. And there's a problem, because this membrane is now responding to the pressure forces at a different rate to that of your eardrum; the acoustic waves in the two tubes are no longer symmetric and you're going to get phase errors and comb-filtering interference effects between the two. Also, you're now peak-limiting the loudest sounds, which adds a second form of signal distortion - and depending on the membrane properties, you're more likely to be preferentially peak-limiting the lower frequencies, which will also skew the overall frequency response. (Peak-limiting is what sketchy mastering engineers do to win the loudness wars by adding compression and limiting the useful dynamic range for folks that need uniformly-loud pop music.) If you really want to peak limit, that can be done in software too, but using a mechanical or software device to peak limit are both terrible ideas. There's a much better way of saving your ears which won't degrade your audio signal with clipping, limiting or phase-errors and interference effects - just turn your volume down.

Best case scenario, the ADEL module has done nothing but empty your wallet. Unfortunately, the most likely scenario is that it's also been slightly degrading your sound quality while doing nothing to protect your hearing that you couldn't have achieved much more easily by simply listening at a safe volume.

At some point 64 Audio and Ambrose decided they wanted to part ways. There was a fair amount of skepticism in the community about ADEL (and rightly so), which might also have played a role. But 64 Audio didn't abandon this potential revenue stream - they just created some new nonsense of their own that would be even more profitable, as they wouldn't need to pay licensing fees for it. So these days you're also paying extra for an APEX module.

APEX and ADEL are both snake oil solutions to a problem that doesn't exist, and yet another potential source of error to the acoustic signal reaching the eardrum. I guarantee the 64 Audio folks are smart enough that they already know all this. So the fact that they continue to push it shows they're disingenuous. I would avoid 64 Audio like the plague. They are amongst the worst value-for-money audio products you can buy right now.



I also own (and like) the ER4XR



Etys are the kings of isolation, but to get that isolation you really need to use something like Comply foam tips, which tend to roll off the treble. They're very different headphones from the FLC8D - they have a much heavier mid-range boost and a relatively anemic bass. The argument for the Ety mid-range bump is that they should look pretty much flat after a diffuse-field correction, but as I mentioned above, whether that works for you or not will depend on your own ear canal anatomy, and according to those equal loudness curves, flat is actually not what you want anyway. I use the Etys for their isolation in a noisy environment, but in every other situation I'd choose the FLC8D. Hope that helps!

My bad haha, I must have changed that privacy at some time for some reason, re enabled it

I couldn't have asked for a better reply, thanks for the information.

I always enjoy learning the inner workings, and your description of their overmarketing of generic off the shelf components and the ADEL module helps clarify quite nicely (and save my future self $3500!). From my brief research, I can't find any evidence to the contrary of your points, so I think the conclusions drawn are quite fair. Excellent.

The power of marketing at its finest, eh? To be fair, Tia (tm) and ADEL (tm) sound like impressive technological acronyms, until examined more closely.

 

[EagleWings]

@csglinux , thanks for this nice post and raising some nice arguments. Here is my response:

@EagleWings - good, comprehensive set of impressions there! Obviously our preferences are different (and you're using gold filters and I'm using gunmetal). I agree with you that the overall sound of the Andromeda is pretty good. I'd say disappointingly good. Disappointing in the sense that you have to suffer extreme torture from the stupidly-designed outer shell in order to listen to it. Unlike you, however, I do not prefer the overall sound of Andromeda to the FLC8D. (And I'm not shilling. If you've read my previous posts here and my review of the FLC8N you'll see I've not been shy of criticising FLC in the past.) There are loads of points to discuss here, but briefly (or as brief as I can be!):

So true. Apart from the impedance issue, the edgy design a big heart burn about this IEM. I definitely believe you. I prefer my $10 Mi earbuds over my $350 Sony 1000XM3. For me Andro's fit is not too bad. My only 2 annoyances are its non-linear impedance and its high sensitivity that it gets too loud and hisses with a lot of equipment.

1) I think the "treble extension" that people appear to perceive with the Andros is simply that they have an overall brighter sound signature by virtue of a rolled-off bass and slightly recessed mids (which you'll definitely see after applying a diffuse-field compensation). The perception of these couplers not being reliable beyond 10 kHz is I think largely due to the fact that the 711 standard only has specifications for the input and transfer impedance up to that point. But the mics are typically ruler flat all the way out to 20 kHz, and there's no indication that anything below 10 kHz really matches the average human ear anyway. It likely represents somebody's ear, but that might not be you or I. The newer (and badly-named) "hi-res" couplers do nothing but add a half-wave resonance damping to make THD measurements look better; there is zero evidence they offer any more physical realism than standard 711 couplers above 10 kHz. You can be fairly confident that the Andromedas do roll off just beyond 9 kHz. Campfire Audio's own measurements of the Andromeda show the exact same behavior, and I have plenty of other IEMs that don't roll off this early, so this isn't a limitation of the coupler or mic.

Yes, I have seen that behaviour on head-fi, where people confuse lower-treble brightness with upper-treble extension. But I definitely hear the Andro to have a better extension than the 8D and the 8N. Also, the flaw in comparing the treble extension of 2 earphones by overlaying the FR on top of each other and placing the cursor at 1kHz is, it may not representative of the nominal volume level, at which we listen to these earphones. Because Andro's response is more flat, you probably need to move Andro's response up by 5dB, to bring it close to the volume levels of how I listen to these 2 earphones. When you do that, the upper treble of the Andro would have a higher average/consistent response than the 8D.

I am not saying your coupler is inadequate or flawed. I am just saying that measurements beyond 9kHz even on the best 711 couplers and 20kHz flat mic might not actually represent what one might hear for reasons such as:
1) an individuals ear canal characteristic might be quite different from the characteristics of the 711 couplers. For example, unlike the Etys which are inserted deeply into the canal, Andro and 8D are comparatively inserted in a shallow way. Which means, the soft tissue part of the ear canal comes into play and we need to factor in the effect the soft tissue region might have on the sound waves. But the 711 doesn't have the soft tissue area
2) these sort of differences between an individual's ear canal and the 711 coupler might lead to considerable variations in resonant peaks and dips beyond 9 kHz
3) difference in sensitivity to treble peaks between individuals. For example: I am sensitive to 6kHz and 10kHz peaks. I am sensitive to 8kHz peaks too, but not to the same level of 6 and 10kHz peaks. A couple of my friends are not that sensitive to 6 and 10kHz peaks but are sensitive to 8kHz peaks

As for whether measurements below 9kHz is reliable or not, as long as the measurer is using reliable gear and is methodical and consistent, I would trust the measurement until 6kHz point and would take the 6-9kHz with a grain of salt, as the half wavelength of the sound waves start to approach the length of the ear canal, which results in resonant peaks and dips. But I like to see the 6-9kHz region to know if the peaks exist and how big the peaks are. I have seen your measurement work and you are knowledgeable, so I would trust your until 9kHz. But beyond 9kHz, I would not read too much into it, even if the measurement is from Tyll. Which brings me to my point. While the measurements show that the Andro has a roll off after 9kHz, it is possible that in my ear, it produces a more perceivable extension than the 8D.

But for the sake of the argument, let's consider your measurement is an accurate representation of what I perceive. If you look at 8D's response after 10kHz, it has a peak at one area and the surround areas are completely dipped. The overtones of various notes of various instruments, lie throughout the 10-18kHz region. So while the overtones in the per-treble region of some notes of some instruments might be heard better on the 8D, Andro would reproduce overtones in the upper-treble region more consistently.

2) As for transparency, soundstage, layering, instrument separation, rhythm, refinement, realism and other such nebulous things... If you take a frequency sweep and capture amplitude and phase, you can reconstruct everything about the IEM, including phenomena like transient square wave or impulse response. I don't discount the possibility there's some aspect of the sound we aren't yet prioritizing in our analysis, but unless and until we discover that magical 5th dimension, I don't know what else there can be that differentiates these headphones. I never underestimate placebo effect - especially on my own limited brain. When reproducing sound recorded on stage mics for playback on stereo speakers, an expansive soundstage via any headphone can be, at best, a cheap illusion. Playback through something like a Smyth realizer is needed to correctly account for soundstage.

I am going to disagree with this. For example, on the forum that must not be named here, they measured something called the burst response of a headphone. What they did was play a 50 or 100 cycles of monotone on the headphone and measured the SPL level of the tone (100Hz sine wave) using a mic. The difference between the headphones in their transient response before the driver reaching a stable state of playing the tone at the correct SPL level was very interesting. One headphone exhibited overshoot, while the other exhibited undershoot. I don't think anyone else has measured something like this. This is just an example of how we have not explored measured everything we can about a headphone system. And once we come up with measuring all these various aspects and capture the various characteristic of a headphone (construction type, driver type, orientation size, no of drivers, crossover, distance from driver and eardrum, average air volume between driver and ear drum etc) and if we feed such data of thousands of headphones into a super-computer, then we might start seeing some kind of pattern. Just because the current measurement methods do not show a pattern with respect to those "nebulous" aspects, it is not conclusive that these nebulous aspects cannot be measured or need a 5th dimension to measure them.

Soundstage is a loose term. What we typically refer to as soundstage is the ability of a headphone to place an instrument farther from you, but at the same time retaining the harmonic information of the notes to recreate accurate timbre. Every recording has some spatial information contained in them. So I expect the headphone to not only create the tonal aspect of the instrument accurately, but also the location and placement aspect of it, relative to each instrument in the ensemble.

3) No worries if anybody else prefers a rolled-off bass. There's no right or wrong in terms of subjective preference. But in terms of equal loudness, I would argue FLC8D is right and the Andros are, well, too rolled-off:

It would be hard to argue around this one, as preference plays a huge role. And my preference seems to be in line with the B&K curve derived in 1974. See Fig-5 in this paper: https://www.bksv.com/media/doc/17-197.pdf

3) No worries if anybody else prefers a rolled-off bass. There's no right or wrong in terms of subjective preference. But in terms of equal loudness, I would argue FLC8D is right and the Andros are, well, too rolled-off:

https://en.wikipedia.org/wiki/Equal-loudness_contour

... and I shouldn't have to apply EQ to a $1000+ headphone

You should not take the Fletcher Munson cure into consideration for frequency response of IEMs, as long as the frequency response of the IEM is consistent for the volume range at which we humans listen to music. A frequency response should only try to match a headphone target curve or get close to it. Here is why. As you are aware, headphone target curve is derived based on a speaker system. When you are listening to the speaker or the IEM at 80Hz average loudness, Fletcher Munson curve is the same for both the cases. So if you are going to try and compensate for the Fletches Munson curve only on your IEM by adding more dBs in the insensitive regions, you will be essentially creating a response that moves away from how you hear the music from a speaker system.

5) That's possible. A flatter mid-range at the eardrum would tend to better preserve the ratio of the relevant harmonics and so the timbre of individual instruments. But note the "at the eardrum" part. That's going to vary from individual to individual according to anatomy and the proportion of the ear canal occluded by the IEM. In my experience, the amount of mid-range (~3-4 kHz) bump preferred to give the correct timbre can have wild variations from person to person (see for example here: https://www.head-fi.org/threads/rha-cl2-tuning-preferences.894595/) so it's difficult to extrapolate this conclusion to others.

Yea, I'd have to agree with this one as well. I myself prefer a 10db rise there. I know a few people who prefer a smaller bump there.

 

[EagleWings]

Also, about the ADEL. I have owned 2 ADEL Custom IEMs and 2 non-ADEL custom IEMs. Here is my experience:

- When it comes to Custom IEMs I did feel ADEL helped alleviate some listening fatigue when listening for extended time. I noticed this, when I sold my ADEL CIEMs and moved to non-ADEL CIEMs, as I am not able to listen to my non-ADEL CIEMs as long as I used to listen to my ADEL-CIEMs.
- Also, there is this possibility of the non-ADEL CIEMs having minor treble peaks that doesn't stand out, but causes fatigue over time. So my experience is not anything conclusive. But based on my experience, I do want to give ADEL the benefit of the doubt.
- I don't think I experience similar fatigue when using universal IEMs. So I am not sure ADEL would help alleviate fatigue in universal IEMs

Q: Does it help relieve fatigue when it comes to Custom IEMs?
A: Probably

Q: Does it save your hearing from permanent hearing damage and provide long term health benefits?
A: I doubt it

 

[csglinux]

Good stuff @EagleWings - I think we're pretty much on the same page here. Just a few little nuances below...

For me Andro's fit is not too bad. My only 2 annoyances are its non-linear impedance and its high sensitivity that it gets too loud and hisses with a lot of equipment.

I hear you about the impedance swings. I'm not sure I'd really blame the Andros for the hiss though. I place the blame there squarely on the shoulders of the DAC/amp OEMs, very few of whom seem to care about noise floor levels (which, of course, never feed into their THD+N specs, because they're always masked by a good, strong 1 kHz signal).

Yes, I have seen that behaviour on head-fi, where people confuse lower-treble brightness with upper-treble extension. But I definitely hear the Andro to have a better extension than the 8D and the 8N. Also, the flaw in comparing the treble extension of 2 earphones by overlaying the FR on top of each other and placing the cursor at 1kHz is, it may not representative of the nominal volume level, at which we listen to these earphones. Because Andro's response is more flat, you probably need to move Andro's response up by 5dB, to bring it close to the volume levels of how I listen to these 2 earphones. When you do that, the upper treble of the Andro would have a higher average/consistent response than the 8D.

I agree about the arbitrary nature of normalizing the curves. We have the same problem doing SPL-matched listening tests. There's no unique test signal you can use to SPL-match two IEMs that have different FRs. I don't think there's an answer to that, because it depends on the dominant frequencies in the particular piece of music you're listening to.

I am not saying your coupler is inadequate or flawed. I am just saying that measurements beyond 9kHz even on the best 711 couplers and 20kHz flat mic might not actually represent what one might hear for reasons such as:
1) an individuals ear canal characteristic might be quite different from the characteristics of the 711 couplers. For example, unlike the Etys which are inserted deeply into the canal, Andro and 8D are comparatively inserted in a shallow way. Which means, the soft tissue part of the ear canal comes into play and we need to factor in the effect the soft tissue region might have on the sound waves. But the 711 doesn't have the soft tissue area
2) these sort of differences between an individual's ear canal and the 711 coupler might lead to considerable variations in resonant peaks and dips beyond 9 kHz
3) difference in sensitivity to treble peaks between individuals. For example: I am sensitive to 6kHz and 10kHz peaks. I am sensitive to 8kHz peaks too, but not to the same level of 6 and 10kHz peaks. A couple of my friends are not that sensitive to 6 and 10kHz peaks but are sensitive to 8kHz peaks

As for whether measurements below 9kHz is reliable or not, as long as the measurer is using reliable gear and is methodical and consistent, I would trust the measurement until 6kHz point and would take the 6-9kHz with a grain of salt, as the half wavelength of the sound waves start to approach the length of the ear canal, which results in resonant peaks and dips. But I like to see the 6-9kHz region to know if the peaks exist and how big the peaks are. I have seen your measurement work and you are knowledgeable, so I would trust your until 9kHz. But beyond 9kHz, I would not read too much into it, even if the measurement is from Tyll. Which brings me to my point. While the measurements show that the Andro has a roll off after 9kHz, it is possible that in my ear, it produces a more perceivable extension than the 8D.

But for the sake of the argument, let's consider your measurement is an accurate representation of what I perceive. If you look at 8D's response after 10kHz, it has a peak at one area and the surround areas are completely dipped. The overtones of various notes of various instruments, lie throughout the 10-18kHz region. So while the overtones in the per-treble region of some notes of some instruments might be heard better on the 8D, Andro would reproduce overtones in the upper-treble region more consistently.

There are two other issues we should consider. The eartips used make a huge difference. I usually use SpinFits for measurements and I guess most other people here are not using SpinFits. Tips definitely change the FR in the 10 kHz+ region. (I also find the gold filter has more mids AND more upper mids/lower treble, so treble extension with gold filters may appear less by comparison.) The other significant effect is insertion depth. This can significantly shift the resonance peaks and is almost certainly different from one individual to another. I'd be willing to bet my mother in law's life that the soft tissue in the ear canal does nothing to accentuate high frequency peaks. The only effect I'd expect from soft tissue is damping, and that always preferentially affects the highest frequencies. Oh, I just thought of a third source of potential discrepancy: unit variation. That's definitely not a negligible effect with FLC's earphones.

I am going to disagree with this. For example, on the forum that must not be named here, they measured something called the burst response of a headphone. What they did was play a 50 or 100 cycles of monotone on the headphone and measured the SPL level of the tone (100Hz sine wave) using a mic. The difference between the headphones in their transient response before the driver reaching a stable state of playing the tone at the correct SPL level was very interesting. One headphone exhibited overshoot, while the other exhibited undershoot. I don't think anyone else has measured something like this. This is just an example of how we have not explored measured everything we can about a headphone system. And once we come up with measuring all these various aspects and capture the various characteristic of a headphone (construction type, driver type, orientation size, no of drivers, crossover, distance from driver and eardrum, average air volume between driver and ear drum etc) and if we feed such data of thousands of headphones into a super-computer, then we might start seeing some kind of pattern. Just because the current measurement methods do not show a pattern with respect to those "nebulous" aspects, it is not conclusive that these nebulous aspects cannot be measured or need a 5th dimension to measure them.

I'm going to disagree with your disagreement. The only way the measurement of an impulse or burst is going to provide new information that we didn't already have from the transfer function of the frequency sweep with amplitude and phase, is if the driver is experiencing some significant non-linearity, and this shouldn't be an issue unless you're blasting your speakers/headphones at unsafe volumes. Which, of course, none of us are doing, right?

Soundstage is a loose term. What we typically refer to as soundstage is the ability of a headphone to place an instrument farther from you, but at the same time retaining the harmonic information of the notes to recreate accurate timbre. Every recording has some spatial information contained in them. So I expect the headphone to not only create the tonal aspect of the instrument accurately, but also the location and placement aspect of it, relative to each instrument in the ensemble.

I don't think soundstage should be defined that loosely though. Timbre is something that could/should be considered separately from soundstage. Positioning information comes from amplitude and timing differences between the two channels. Next to the source - a correct binaural source - the importance of the headphones pales into insignificance. It worries me that we collectively obsess about soundstage when we're all playing back, on headphones, recordings that were made for speakers. There's simply no way the correct soundstage is ever going to be reproduced by a pair of headphones given a fundamentally flawed source. I'm not saying the transient and decay response vs f isn't important, but it feels to me a bit like obsessing over a scratch on your car's paintwork after its transmission and engine have just fallen out.

You should not take the Fletcher Munson cure into consideration for frequency response of IEMs, as long as the frequency response of the IEM is consistent for the volume range at which we humans listen to music. A frequency response should only try to match a headphone target curve or get close to it. Here is why. As you are aware, headphone target curve is derived based on a speaker system. When you are listening to the speaker or the IEM at 80Hz average loudness, Fletcher Munson curve is the same for both the cases. So if you are going to try and compensate for the Fletches Munson curve only on your IEM by adding more dBs in the insensitive regions, you will be essentially creating a response that moves away from how you hear the music from a speaker system.

Part of the point of Fletcher Munson is that you do need to move away from what you'd hear from a speaker system - or rather, you also need to adjust the response from the speakers, because the optimal FR at your eardrums varies with dB level - and it's never flat, even if you account for the ear canal resonances, reflection/refraction effects from the head/torso, etc. After some convincing from @jude, I moved away from using any kind of headphone target curve for compensation. They're all somewhat arbitrary and definitely not one-size-fits-all and as a result they're just a source of confusion. As the saying goes - the great thing about standards is there's so many of them to choose from.

BTW, has anybody seen this post?:

https://www.head-fi.org/threads/flc...ressions-thread.777158/page-511#post-14863426

This was totally out of the blue to me. Anybody know how this is supposed to differ from FLC8n/d/s? I love FLC's headphones, but dang their marketing and communication with their potential customers sucks. @Flcforrestwei - why don't you just let one of us here on headfi know about your product releases? I'd happily write up a post or create a new thread and proof-read any product-release information for you. It seems we're all in the dark here.

The above model seems to have the same set of drivers and a full set of ULF, LF, M/H filters as in the 8s and 8n models (albeit with the blue filter now becoming silver). Is it supposed to be an improvement over 8s and 8n? If so, in what way? And why is it so much cheaper than the 8n?

 

[Flcforrestwei]

Good stuff @EagleWings - I think we're pretty much on the same page here. Just a few little nuances below...


I hear you about the impedance swings. I'm not sure I'd really blame the Andros for the hiss though. I place the blame there squarely on the shoulders of the DAC/amp OEMs, very few of whom seem to care about noise floor levels (which, of course, never feed into their THD+N specs, because they're always masked by a good, strong 1 kHz signal).


I agree about the arbitrary nature of normalizing the curves. We have the same problem doing SPL-matched listening tests. There's no unique test signal you can use to SPL-match two IEMs that have different FRs. I don't think there's an answer to that, because it depends on the dominant frequencies in the particular piece of music you're listening to.


There are two other issues we should consider. The eartips used make a huge difference. I usually use SpinFits for measurements and I guess most other people here are not using SpinFits. Tips definitely change the FR in the 10 kHz+ region. (I also find the gold filter has more mids AND more upper mids/lower treble, so treble extension with gold filters may appear less by comparison.) The other significant effect is insertion depth. This can significantly shift the resonance peaks and is almost certainly different from one individual to another. I'd be willing to bet my mother in law's life that the soft tissue in the ear canal does nothing to accentuate high frequency peaks. The only effect I'd expect from soft tissue is damping, and that always preferentially affects the highest frequencies. Oh, I just thought of a third source of potential discrepancy: unit variation. That's definitely not a negligible effect with FLC's earphones.


I'm going to disagree with your disagreement. The only way the measurement of an impulse or burst is going to provide new information that we didn't already have from the transfer function of the frequency sweep with amplitude and phase, is if the driver is experiencing some significant non-linearity, and this shouldn't be an issue unless you're blasting your speakers/headphones at unsafe volumes. Which, of course, none of us are doing, right?


I don't think soundstage should be defined that loosely though. Timbre is something that could/should be considered separately from soundstage. Positioning information comes from amplitude and timing differences between the two channels. Next to the source - a correct binaural source - the importance of the headphones pales into insignificance. It worries me that we collectively obsess about soundstage when we're all playing back, on headphones, recordings that were made for speakers. There's simply no way the correct soundstage is ever going to be reproduced by a pair of headphones given a fundamentally flawed source. I'm not saying the transient and decay response vs f isn't important, but it feels to me a bit like obsessing over a scratch on your car's paintwork after its transmission and engine have just fallen out.



Part of the point of Fletcher Munson is that you do need to move away from what you'd hear from a speaker system - or rather, you also need to adjust the response from the speakers, because the optimal FR at your eardrums varies with dB level - and it's never flat, even if you account for the ear canal resonances, reflection/refraction effects from the head/torso, etc. After some convincing from @jude, I moved away from using any kind of headphone target curve for compensation. They're all somewhat arbitrary and definitely not one-size-fits-all and as a result they're just a source of confusion. As the saying goes - the great thing about standards is there's so many of them to choose from.

BTW, has anybody seen this post?:

https://www.head-fi.org/threads/flc...ressions-thread.777158/page-511#post-14863426

This was totally out of the blue to me. Anybody know how this is supposed to differ from FLC8n/d/s? I love FLC's headphones, but dang their marketing and communication with their potential customers sucks. @Flcforrestwei - why don't you just let one of us here on headfi know about your product releases? I'd happily write up a post or create a new thread and proof-read any product-release information for you. It seems we're all in the dark here.

The above model seems to have the same set of drivers and a full set of ULF, LF, M/H filters as in the 8s and 8n models (albeit with the blue filter now becoming silver). Is it supposed to be an improvement over 8s and 8n? If so, in what way? And why is it so much cheaper than the 8n?

FLC8p supposed to be released next month, but Canjam Singapore is hold in this weekend, LMUE ask us to make some units for show, so we build 16 pcs for the show.

Here are the differences between FLC8s, FLC8n and FLC8d.

FLC8s: with very special armatures from Knowles, we asked a friend to buy it out of China and then ship to China, the price is very high, the cable is 7N single crystal copper cable, the price of the conductor is very high.

FLC8n: We cooperate with a Chinese armature company to develop the armatures, price of the armatures is cheaper, the cable is the same cable with FLC8s, the problem of FLC8n is industry design, the housing is metal, the finish is polish, that makes the yield very low, we published the product in Feb last year, but the product came out three months later, just because of the issue, polish finish needs a very pure aluminium, otherwise, there will be defective when the product was anodized, yield at the polish process is very low. That's the main reason to develop FLC8p even FLC8n was released less than one year.

FLC8p: 1,The armatures same as FLC8n(Cheaper armatures).
2. plastic housing(Cheaper housing).
3. Cheaper cable, cable changed from 7N single crystal copper to oxygen-free copper(OFC) cable, price of 7N crystal copper is about 20 times over the OFC's, we choose OFC for FLC8p but the conductor is three times bigger, bigger conductor make the sound thicker, but treble extension is little weaker then the 7N single crystal copper.
4. Change the blue nozzle to Silver nozzle, the Silver nozzle enhance the Mids and cut off somw the treble extension, to focus on Vocals.

We are not a big company, we are not good are marketing and products promotion, FLC8n's market performance is very bad, since cost of FLC8p can be brought down, so we bring down the price of FLC8p, that's the way to survive from the economic winter.

Hope your understand!

 

[csglinux]

Hope your understand!

I totally understand. Thanks so much for the info

 

[csglinux]

P.S. @Flcforrestwei - please do let me know if there's anything I can do to help get your future product information up on headfi. You make great headphones. I'd hate to see FLC suffer a prolonged economic winter

 

[originalsnuffy]

FLC8s: with very special armatures from Knowles, we asked a friend to buy it out of China and then ship to China, the price is very high, the cable is 7N single crystal copper cable, the price of the conductor is very high.

FLC8n: We cooperate with a Chinese armature company to develop the armatures, price of the armatures is cheaper, the cable is the same cable with FLC8s, the problem of FLC8n is industry design, the housing is metal, the finish is polish, that makes the yield very low, we published the product in Feb last year, but the product came out three months later, just because of the issue, polish finish needs a very pure aluminium, otherwise, there will be defective when the product was anodized, yield at the polish process is very low. That's the main reason to develop FLC8p even FLC8n was released less than one year.

FLC8p: 1,The armatures same as FLC8n(Cheaper armatures).
2. plastic housing(Cheaper housing).
3. Cheaper cable, cable changed from 7N single crystal copper to oxygen-free copper(OFC) cable, price of 7N crystal copper is about 20 times over the OFC's, we choose OFC for FLC8p but the conductor is three times bigger, bigger conductor make the sound thicker, but treble extension is little weaker then the 7N single crystal copper.
4. Change the blue nozzle to Silver nozzle, the Silver nozzle enhance the Mids and cut off somw the treble extension, to focus on Vocals.

Forrest,

First of all, thank you for your post. Most of us here really like what you have created and want you to succeed.

Where does the 8D fit in to this discussion? Is the housing hard to make also? If so, perhaps that can be updated the way the 8P was updated.

Personally, I really suggest you drop the green filter and create a filter in between the gold and the gunmetal that has more treble extension than the gold and perhaps a bit less mids. Since gold and gunmetal are the two most popular options something in between should do well for you.

We would also be interested in your new 7 range and what that is all about.

I think it would be a good idea to personally reach out to CSGLinux and EagleWings and take their advice on how market your products. Having a few people here who really know and like your products could really be helpful to your company. Its a shame that some of your early enthusiasts are no longer very active on head-fi (NMatheis and Brooko in particular) but CSG and Eagle would be very suitable people to work with. I am not nearly as good at explaining matters as those two though I love the two products I have from you.

You really should be and can be one of the top IEM companies on head-fi and some improved communications here can make a difference. You now have a fine distributor in the US (Music Teck) so some more attention on head-fi can result in more sales in the US in particular.

 

[EagleWings]

I hear you about the impedance swings. I'm not sure I'd really blame the Andros for the hiss though. I place the blame there squarely on the shoulders of the DAC/amp OEMs, very few of whom seem to care about noise floor levels (which, of course, never feed into their THD+N specs, because they're always masked by a good, strong 1 kHz signal).

Well, there are some DAPs that hiss with almost every multi-BA IEM. For example, the Sony A25, A35, A45. I am not talking about such DAPs. Most current day DAPs are silent enough for most IEMs. Andro, Zeus, SE846 are some examples of hypersensitive IEMs that hiss out of almost everything. With the Zeus, when you connect a DAP to your PC and transfer files, you can literally hear the electronics whirring during the data transfer.

I'd be willing to bet my mother in law's life that the soft tissue in the ear canal does nothing to accentuate high frequency peaks. The only effect I'd expect from soft tissue is damping, and that always preferentially affects the highest frequencies. Oh, I just thought of a third source of potential discrepancy: unit variation. That's definitely not a negligible effect with FLC's earphones.

Exactly, soft tissues would help dampen some upper treble frequencies. And 711 couplers won't help dampen these frequencies.

I'm going to disagree with your disagreement. The only way the measurement of an impulse or burst is going to provide new information that we didn't already have from the transfer function of the frequency sweep with amplitude and phase, is if the driver is experiencing some significant non-linearity, and this shouldn't be an issue unless you're blasting your speakers/headphones at unsafe volumes. Which, of course, none of us are doing, right?

Like I said you can't simply dismiss those measurements. For all you know data captured in those measurements might correspond to a non-tonal aspect of a headphone. For example, a BA IEM and a DD IEM might show a similar measurement in the bass but would be perceived in a different way. FR can reveal non-linearity in the response domain. But think about modal break-up type of issues. How could you even detect that on your FR or Phase plot or even on distortion measurements, CSD, Square Wave or Impulse Response measurements?

I don't think soundstage should be defined that loosely though. Timbre is something that could/should be considered separately from soundstage. Positioning information comes from amplitude and timing differences between the two channels. Next to the source - a correct binaural source - the importance of the headphones pales into insignificance. It worries me that we collectively obsess about soundstage when we're all playing back, on headphones, recordings that were made for speakers. There's simply no way the correct soundstage is ever going to be reproduced by a pair of headphones given a fundamentally flawed source. I'm not saying the transient and decay response vs f isn't important, but it feels to me a bit like obsessing over a scratch on your car's paintwork after its transmission and engine have just fallen out.

Well then let me to swap the term soundstage to headstage. Like I said earlier in one of my posts, some IEMs like U18 create reproduce the space between a centre stage instrument vs a instrument placed on the side of the stage. IEMs like the ER4SR can't recreate the same dense of space, although I prefer the stock FR of ER4SR. The sense of space pro

Part of the point of Fletcher Munson is that you do need to move away from what you'd hear from a speaker system - or rather, you also need to adjust the response from the speakers, because the optimal FR at your eardrums varies with dB level - and it's never flat, even if you account for the ear canal resonances, reflection/refraction effects from the head/torso, etc. After some convincing from @jude, I moved away from using any kind of headphone target curve for compensation. They're all somewhat arbitrary and definitely not one-size-fits-all and as a result they're just a source of confusion. As the saying goes - the great thing about standards is there's so many of them to choose from.

While I will agree with you that an IEM need not exactly follow a headphone target curve, I don't think Frequency Response should account for Fletcher Munson curve. By adding additional dBs to frequencies that typically you wouldn't hear to the same level through a speaker, you are starting to deviate from how you would hear the same track out of a set of speakers. There is a reason why we are using speaker as the bench mark. Engineers and artists use studio monitor speakers to finalise how the track should sound like. I am not advocating that an IEM should strictly follow any specific headphone target curve. But straying too far away from a speaker's response is going to reproduce the music in a way that was not intended by the artist. I have some basic criteria for an IEM's response:

1) A 2-3dB rise in the bass staring from 300Hz and then it should plateau below 100Hz, or no rise in the bass at all
2) Preferably a flat response from 300 upto 700Hz
3) No dips between 500hz to 1kHz
4) A 7-11dB bump in the 2.7kHz to 3.5khz region with the response starting to rise from around 800Hz
5) After the 3khz bump, the response should start a gradual downward slope into the treble
6) This slope should continue as a gradual roll-off after 10khz

Slight deviations from these criteria are fine. But huge deviations will start affecting the timbre of instruments.

 

[Flcforrestwei]

Forrest,

First of all, thank you for your post. Most of us here really like what you have created and want you to succeed.

Where does the 8D fit in to this discussion? Is the housing hard to make also? If so, perhaps that can be updated the way the 8P was updated.

Personally, I really suggest you drop the green filter and create a filter in between the gold and the gunmetal that has more treble extension than the gold and perhaps a bit less mids. Since gold and gunmetal are the two most popular options something in between should do well for you.

We would also be interested in your new 7 range and what that is all about.

I think it would be a good idea to personally reach out to CSGLinux and EagleWings and take their advice on how market your products. Having a few people here who really know and like your products could really be helpful to your company. Its a shame that some of your early enthusiasts are no longer very active on head-fi (NMatheis and Brooko in particular) but CSG and Eagle would be very suitable people to work with. I am not nearly as good at explaining matters as those two though I love the two products I have from you.

You really should be and can be one of the top IEM companies on head-fi and some improved communications here can make a difference. You now have a fine distributor in the US (Music Teck) so some more attention on head-fi can result in more sales in the US in particular.

Thanks for the suggestion !

Forrest,

First of all, thank you for your post. Most of us here really like what you have created and want you to succeed.

Where does the 8D fit in to this discussion? Is the housing hard to make also? If so, perhaps that can be updated the way the 8P was updated.

Personally, I really suggest you drop the green filter and create a filter in between the gold and the gunmetal that has more treble extension than the gold and perhaps a bit less mids. Since gold and gunmetal are the two most popular options something in between should do well for you.

We would also be interested in your new 7 range and what that is all about.

I think it would be a good idea to personally reach out to CSGLinux and EagleWings and take their advice on how market your products. Having a few people here who really know and like your products could really be helpful to your company. Its a shame that some of your early enthusiasts are no longer very active on head-fi (NMatheis and Brooko in particular) but CSG and Eagle would be very suitable people to work with. I am not nearly as good at explaining matters as those two though I love the two products I have from you.

You really should be and can be one of the top IEM companies on head-fi and some improved communications here can make a difference. You now have a fine distributor in the US (Music Teck) so some more attention on head-fi can result in more sales in the US in particular.

Thanks a lot! We ' ll do as your suggest .

 

[csglinux]

Hi @EagleWings - I find the same issues as you do when using my SE846. I still blame the DAP OEMs though. I shouldn't have to use a high-impedance, low sensitivity headphone to mask weaknesses and poor shielding in my DAPs. Yep, it is a problem with almost every current DAP.

Exactly, soft tissues would help dampen some upper treble frequencies. And 711 couplers won't help dampen these frequencies.

That's an interesting point. I wonder if GRAS might one day use that argument as a way to justify their newer couplers? The problems I see are: 1) their half-wave resonance damping would also apply to deep-insertion IEMs (e.g. Etys) where there's little justification for adding a damping from the fleshy parts of the canal, and 2) if we added such a damping to current 711 measurements of the Andromeda, we'd see even less energy in the 10 kHz+ region. I'm not suggesting the FLC8d is perfect in regards treble extension, just that, with my particular units, eartips, insertion depths, etc., Andromeda doesn't add a significant or meaningful amount of treble extension beyond that of the FLC8d. There are other headphones that do - just not the Andros. But it's always possible that our Andromedas would measure differently? ¯\_(ツ)_/¯

Like I said you can't simply dismiss those measurements.

Ooooh, but I can. Just watch me

For all you know data captured in those measurements might correspond to a non-tonal aspect of a headphone. For example, a BA IEM and a DD IEM might show a similar measurement in the bass but would be perceived in a different way. FR can reveal non-linearity in the response domain. But think about modal break-up type of issues. How could you even detect that on your FR or Phase plot or even on distortion measurements, CSD, Square Wave or Impulse Response measurements?

We do account for non-tonal aspects through phase response - and mode-shape effects are there to see in THD. If you have driver break-up, then yes, I agree with what you're saying, but that's operating beyond design conditions in a place that has no interest to me. Another lame analogy: You could fly a 737 Max 8 with the MCAS and anti-stall and autopilot all engaged simultaneously, but that's just silly. Our goal shouldn't be to win a Darwin award.

Well then let me to swap the term soundstage to headstage. Like I said earlier in one of my posts, some IEMs like U18 create reproduce the space between a centre stage instrument vs a instrument placed on the side of the stage. IEMs like the ER4SR can't recreate the same dense of space, although I prefer the stock FR of ER4SR.

I have to admit that's still a bit of a mystery to me. I don't own the ER4SR, but I have the XR. Its THD is nothing to write home about, but its driver has a very fast response (almost up there with the KSE1500 electrostats). It just seems that anything you do to push the sound farther out (moving the drivers farther away or venting) tends to widen the soundstage a tiny little bit. It's still a weak illusion though in comparison to a proper binaural recording or the use of a Smyth realizer.

While I will agree with you that an IEM need not exactly follow a headphone target curve, I don't think Frequency Response should account for Fletcher Munson curve. By adding additional dBs to frequencies that typically you wouldn't hear to the same level through a speaker, you are starting to deviate from how you would hear the same track out of a set of speakers. There is a reason why we are using speaker as the bench mark. Engineers and artists use studio monitor speakers to finalise how the track should sound like. I am not advocating that an IEM should strictly follow any specific headphone target curve. But straying too far away from a speaker's response is going to reproduce the music in a way that was not intended by the artist. I have some basic criteria for an IEM's response:

1) A 2-3dB rise in the bass staring from 300Hz and then it should plateau below 100Hz, or no rise in the bass at all
2) Preferably a flat response from 300 upto 700Hz
3) No dips between 500hz to 1kHz
4) A 7-11dB bump in the 2.7kHz to 3.5khz region with the response starting to rise from around 800Hz
5) After the 3khz bump, the response should start a gradual downward slope into the treble
6) This slope should continue as a gradual roll-off after 10khz

Slight deviations from these criteria are fine. But huge deviations will start affecting the timbre of instruments.

I agree with almost all the above. I think we're maybe just looking at things from a different angle.

No two people are going to share the same FR preference. (Case in point - I don't share your preference for item 1 above.) And even the same individual can have different experiences depending on the source. Here's an example. Imagine you're at a live music concert, a large distance back from the stage. When you go home afterwards, you listen (on your headphones) to the live recording the engineers made of that concert using on-stage mics piped through their mixing desk. You hear this "the way the artist intended" right? Let's be extremely generous to Dr. Dre for now and says that's true. It is still not what you heard at the concert. Mixing and mic placement aside, the FR would be different because higher-frequency waves attenuate faster over distance due to air viscosity. It's not that either is right or wrong - it's just different perceptions of something that can't have an objective reality. Fletcher Munson simply says the human ear needs more of a boost at the extremes when listening at lower volumes - via headphones or speakers (which are also never completely flat). I don't listen to my Symphony X albums at the same SPLs generated when they play live. Given my preference for listening at lower volumes, I find the extra low end in the FLC8d pretty good. The harmonics of the human voice and musical instruments are way above this low-frequency range, so I don't see this would have any noticeable impact on timbre. I'd agree that we want flat (whatever flat means to an individual's ears and brain) through the mid-range and above.

BTW, thanks for the interesting and thought-provoking discussion. I hope we've not wandered too far off topic...?

 

[marcusd]

One of our team spent a few months with the FLC8n and completed his review on it today

https://headfonics.com/2019/04/flc-technology-flc8n-review/

 

[csglinux]

Hello @Flcforrestwei,

Any chance that FLC might have a presence (or a distributor) at the 2019 SoCal CanJam?

 

[Moonstar]

Hi, here are my short impressions about the FLC 8N


The FLC Technology FLC 8N is like a Chameleon with its great sound tuning ability, which is not a truly gimmick! The detail level is the best I have heard from any In-Ear Monitor in this price category and the beautiful midrange presentation and solid build quality makes it one of my new favorites under the $500 USD barrier.

Highly recommended!



Pros and Cons:

• + Wonderful detail level and beautiful midrange presentation (Gold Filter)
• + Great Bass response, nearly on a bass-head level with the right plug combination
• + Lots of sound and fine-tune options
• + Nice accessories package
• + Good comfort

• - Tuning plugs are very small, which needs extra attention
• - The chancing of the small tuning plugs need sensitivity

Full Review
https://moonstarreviews.net/flc-technology-flc8n-review/

..and some Shots

 

[jambaj0e]

Oh no! @Flcforrestwei my worst fear just happened. When I was switching back and forth between the black and grey LF, the white part inside the grey filter popped out!

And that's the filter I want to use, too. How can I get a replacement? I live in Los Angeles.