[yaluen]

Well it seems like the 7kHz resonance peak issue is a major sticking point for you, so I'd once again suggest you look into the Sennheiser IE 800 (http://rinchoi.blogspot.ca/2013/03/sennheiser-ie800.html).
 
Not sure how it fares in terms of isolation and leakage however, but very possible it won't be up to your requirements unfortunately. Its impedance curve is pretty much flat so you shouldn't run into any problems with changing response due to different sources. Channel mismatch could be a point of concern looking at various measurements.
 
Also, those plots for the Titan further illustrate the unreliability of IEM measurements, at least the ones I've seen so far. The green trace is captioned "8mm away from the reference plane", where reference plane refers to the plane parallel to the entrance of the cylinder meant to simulate an ear canal in the measurement coupler, which is a specified distance away from the microphone (http://www.comsol.com/model/download/218141/models.aco.generic_711_coupler.pdf). It is not a vague reference to the "bony part of the ear canal" as @briskly said previously. This distance is a standardized 12.5mm as I understand it. Thus the green trace should represent the Titan as measured with a sealed ear canal length of 20.5mm. One can calculate the expected half-wavelength resonance frequency using the following equation.
 
f = v / (2 * L) where v = 343m/s and L = sealed ear canal length in m
 
Yet the green trace seems to imply that the resonance peak is at 7.5kHz when it should be around 8.4kHz. It seems to me measurement takers should be checking results for plausibility, similar to how you would do the same after solving a math problem.

 

[briskly]

The reference plane in the human ear canal is defined as corresponding to the position in the canals where plugs would normally terminate in the ear. See here.
This is around the second bend of the canals, where the cartilage makes way for the skull proper, as stated in the 711 simulation you linked. Hence the reference to the bony part of the ear canal.
 
As to @speakerphone and distance to reference plane measurements, a snippet regarding insertion and reference plane concerning a different IEM.

Comply tip is much longer.
 
These two measurements, lengths from ref plane are the same but depths are different.
So if match the both tips at same length from reference plane(measuring the length of ref plane ~ end of the eartip), insertion depth of comply tip is little bit more shallow. 

 

[Lunatique]

  Well it seems like the 7kHz resonance peak issue is a major sticking point for you, so I'd once again suggest you look into the Sennheiser IE 800 (http://rinchoi.blogspot.ca/2013/03/sennheiser-ie800.html).

It doesn't look too bad. I'd have to pull the entire bass region down about 6 dB or so, and fill in that big recess near the 3 KHz region. 6 to 10 KHz region is a bit more irregular than ideal, so EQ'ing it to be smoother will require a bit more work. I'm not sure at that price point if it offers so much more than the RE-400 using my EQ curve though.

 

[Lunatique]

Question: Shouldn't these measuring devices try to also mimic the surface texture of the ear canal, with some irregularity (a little bit of ear wax) and a layer of fine hairs? I'm sure the irregularities and the fine layer of hair will sound different from  a smooth and polished metallic tube?

 

[james444]

In my experience, the RE600's treble can be easily improved by reduction of front damping:
 

   
Don't know about the others, but I removed my filters altogether and replaced them with a small amount of cotton swab. Then I used small pieces of dried alcohol swab to protect the nozzle before I put the tips back on.
 
The result is a RE600 with considerably better treble presence than stock, but still sounding very smooth and devoid of high frequency peaks.
 

 
 
As for measurements: very helpful indeed, but keep in mind all compensation curves are results of statistical procedures that inevitably involve variance.
 

   
Let's face it, there's simply no neutral that will fit all listeners. Diffuse field equalization was averaged out of 47 individual curves that deviated noticeably from each other. Same goes for Harman's OW equalization, anyone who cares to read the AES paper will see that measured responses (taken with probe mikes at ear drum level) deviated noticeably among individual subjects. "Absolute" or "true" neutral simply doesn't exist, there's just a lot of individual curves that share a fair amount of similarities.
 
jm2c

 

 

[yaluen]

  The reference plane in the human ear canal is defined as corresponding to the position in the canals where plugs would normally terminate in the ear. See here.
This is around the second bend of the canals, where the cartilage makes way for the skull proper, as stated in the 711 simulation you linked. Hence the reference to the bony part of the ear canal.
 
As to @speakerphone and distance to reference plane measurements, a snippet regarding insertion and reference plane concerning a different IEM.

Comply tip is much longer.
 
These two measurements, lengths from ref plane are the same but depths are different.
So if match the both tips at same length from reference plane(measuring the length of ref plane ~ end of the eartip), insertion depth of comply tip is little bit more shallow. 

 
Ah, that document better explains what "reference plane" refers to when talking about couplers, thanks. If I'm reading this correctly, it refers to the plane that bounds a specific standard equivalent volume, thus acoustic impedance, between it and the microphone, derived from the ear canal sealed at about the second bend. My mistake, please ignore my previous comment about standardized dimensions and geometry as those seem to apply to 711 couplers only. Inferring from that document, it seems there's no standard that dictates the geometry of couplers should be anatomically correct, which is interesting. Given all this, insertion depth shouldn't be stated as a value in reference to the reference plane then, since the distance the reference plane is from the microphone will vary from coupler to coupler. Also, the premise for defining the reference plane does sound inappropriate for couplers designed for IEM measurements since the vast majority of IEMs are shallow insertion and don't go nearly as deep as the second bend. Etys are perhaps the only ones I can think of that goes that deep.
 

 

  Well it seems like the 7kHz resonance peak issue is a major sticking point for you, so I'd once again suggest you look into the Sennheiser IE 800 (http://rinchoi.blogspot.ca/2013/03/sennheiser-ie800.html).

It doesn't look too bad. I'd have to pull the entire bass region down about 6 dB or so, and fill in that big recess near the 3 KHz region. 6 to 10 KHz region is a bit more irregular than ideal, so EQ'ing it to be smoother will require a bit more work. I'm not sure at that price point if it offers so much more than the RE-400 using my EQ curve though.

 
I suppose I should explain what the IE800 is trying to achieve in case you didn't read that whole article. You may want to refer to Tyll's measurements as well (http://www.innerfidelity.com/images/SennheiserIE8002013.pdf). Looking at the IE800, you can observe that it is constructed to be worn with shallow insertion. As you've experienced with other IEMs, shallow insertion would result in a sharp resonance peak at ~7kHz. The IE800 employs specially tuned resonators, what Sennheiser is calling "dampened two chamber absorbers", that act as notch filters attenuating 7kHz and 8kHz in its response, thus counteracting said resonance peak due to shallow insertion. AFAIK, it is the only IEM that does this. The dips you see between 6kHz and 10kHz are desirable for IEMs, particularly for you since you seem to be very sensitive to canal resonance. The bass is perhaps a tad too much but a boost below 200Hz is recommended to compensate for perceived lack of impact compared to an out of head sound source. The recess around 3kHz is due to the compensation curve applied. Personally, I don't understand why a target curve formulated for headphones is being applied to IEMs since IEMs aren't affected by the same pinna and head related effects. Thus, the Golden Ears target curve (http://en.goldenears.net/419) is in my opinion much more suitable for IEMs. If you compare that target to the raw curve for the IE800, which can be seen more clearly in Tyll's measurements, it's pretty much ideal. I haven't heard the IE800 myself, but I'd hazard a guess if you do a sweep, you'll hear them as relatively flat. You were lamenting how manufacturers continue to produce products with questionable treble tuning. The IE800 is the only IEM that I can think of where its engineers are trying to address such issues.
 

  Question: Shouldn't these measuring devices try to also mimic the surface texture of the ear canal, with some irregularity (a little bit of ear wax) and a layer of fine hairs? I'm sure the irregularities and the fine layer of hair will sound different from  a smooth and polished metallic tube?

 
I'd assume such parameters are impossible to standardize and build into measurement devices that need to produce widely applicable data.

 

[Breezy]

Might want to try Empire Ears Supra II and Spartan, I keep hearing good things about them. I think if you ask nicely they'll let you test them.

 

[Ming Lu]

I guess I am looking for the same sound signature iem as you do.
Have you tried Earsonics sm3 (1st version)? It is the most neutral sounding iem I had, but it could have slightly more bass response (maybe 2-3db) to make it more enjoyable.
 
For comparison with other iem I own:
UE TF10: worst fitting iem, mid is reccessed, lack of clarity
Earsonics s-em6: very good clarity, no sibilant, very neutral sound overall but I found it slightly mid forward

 

[GoodToGo1]

So did you choose a headphone? 

 

[Lunatique]

  I guess I am looking for the same sound signature iem as you do.
Have you tried Earsonics sm3 (1st version)? It is the most neutral sounding iem I had, but it could have slightly more bass response (maybe 2-3db) to make it more enjoyable.
 
For comparison with other iem I own:
UE TF10: worst fitting iem, mid is reccessed, lack of clarity
Earsonics s-em6: very good clarity, no sibilant, very neutral sound overall but I found it slightly mid forward

I've read about Earsonics a lot in the past, but the impression I get is that they tend to be mid-range emphasized. I considered getting one to try, but it seems none of the high-profile expert headphone gurus championed them.
 

  So did you choose a headphone? 

I'm still waiting for the 64 Audio U5 to arrive (it's the model they told me as being the most neutral), and for now, the RE-400 is the one I'm using, since it's so similar in sound to the RE-600 but less than half its price. I only listen to the RE-400 with my EQ curve, otherwise, it's quite boring and lacking in high and low frequencies (not full enough to get close to the Harman Target Response Curve. But at least it isn't offensively bright, which is my #1 pet peeve). For you RE-400 owners, here's my EQ curve for it (makes it much more neutral and full-range sounding):
 

 
But I made it for medium sized black foam tips that's not from Hifiman though, so if you're using different tips, the range above 2KHz would likely be different for you, and you'll have to do your own tweaking to match what you're hearing. 

 

[Deftone]

is a tough trek up the headfi mountain isnt it, my perfect endgame headphone would be the 650 with more air and sub bass but i dont thnk i will ever find it.

 

[Lunatique]

  is a tough trek up the headfi mountain isnt it, my perfect endgame headphone would be the 650 with more air and sub bass but i dont thnk i will ever find it.

Tell me about it! That would be my ideal too, if Sennheiser ever releases an updated model that does exactly what you described. I just made a new "ideal response curve" to enhance the HD650 recently, to get it closer to the Harman Target Response Curve:

 
EDIT (2016/1/31):
I just updated the HD650 EQ curve, bringing 8KHz down since it's too prominent compared to the relative amplitude of 6.3KHz and 10KHz. 
 

 
 
Here's the actual setting copied from the .xml file
 
<?xml version="1.0" encoding="UTF-8"?>
<Equalizer PatchFormat="2">
  <Band Mode="Low Shelving" Frequency="44.7744228" Gain="6" Bandwidth="2.44"/>
  <Band Mode="High Shelving" Frequency="14000" Gain="5" Bandwidth="1.92"/>
  <Band Mode="Low Shelving" Frequency="119.132429" Gain="5" Bandwidth="1.82"/>
  <Band Mode="Peak/Dip" Frequency="1200" Gain="3" Bandwidth="2.5"/>
  <Band Mode="Peak/Dip" Frequency="8000" Gain="-6" Bandwidth="0.25"/>
</Equalizer>

 

[Lunatique]

I just posted this in the thread for 64 Audio's IEMs, about my experience with the U5 and U10. I'm reposting it here too:
 
As an audio professional, I want to share my experience with the U5 and U10. 
 
I had high hopes because 64 Audio market the U5 and U10 as very neutral and accurate, tuned specifically for audio professionals who need to do mixing and mastering. That is unfornately NOT the case. Not even close. Below are some of the emails I sent to 64 Audio regarding the frequency response of the U5 and the U10:
 
"I just got the U5, and have been testing it. There is a sharp peak at around 6~8 KHz, and it's excessively bright and sibilant. On all songs that have slightly brighter mastering, it's unbearably sibilant. I tried all the tips including tips from other IEMs, and I can't get rid of that sibilant peak. What do you advise?"
 
"After some in-depth testing with log sweep tones, sine wave tones, pink noise, and a large variety of musical material, this is the custom EQ curve I created in order to get the U5 close to sounding more neutral and accurate (see attached image). 
 

 
 
When you recommended the U5 to me, you claimed it is neutral/accurate, but it is far form being that, and I cannot imagine any audio professional who would listen to it and think they sound neutral. The upper-mids region is so bright and sibilant that it's like sharp daggers stabbing the eardrums. The bass is bloated and way too thick. And there's a strange downward slope starting at around 1000~2000 Hz that makes the mids sound recessed, and then starting at around 6 KHz, it suddenly gets unbearably bright and remains so past 10 KHz and on. 
My reference is a mastering grade 2.1 system (Klein + Hummel O 300Ds and Neumann KH805) in an acoustically treated studio, and I've also compared it to other headphones like the Stax SR-007MK2, Audeze LCD-2, Sennheiser HD650, Audio-Technica M50, JH Audio Angie, Noble Audio Kaiser K10U, Westone 4, Westone W60, Hifiman RE-600/RE-400, Ultimate Ears UE900, etc. The U5 is one of the worst offenders in terms of excessive brightness--to the point of being excruciating. 
 
I'm willing to give you guys the benefit of the doubt that there's nothing wrong with the manufacturing, so I'm going to have to assume your tuning of the drivers is the problem. I don't know how you guys arrived at the conclusion that the U5 is neutral/accurate, but I'm extremely disappointed. After all that marketing effort, I can't believe this is what the actual product sounds like.
 
I'm at a complete loss here. In fact I'm dumbfounded by how bad the U5 sounds. 
 
I think I need to speak to one of your engineers about this, because I cannot understand why this is happening. And if the U5 is indeed supposed to sound like this, then I must say I feel duped by your marketing. The product sounds nothing like what was promised. "
 
When 64 Audio asked about my source, this was my reply:
 
"My sources are PC with high-end pro audio interface, or high-end soundcard, or Samsung Galaxy Note 3, going through Objective O2+ODAC amp (which has .05 impedance and reference quality sound). 
 
I've tried all the tips, and I always make sure the tips are fully seated on the stems all the way to the housing.
 
I always insert as far as I can in order to get the best seal, but I've also tried using more shallow insertion too.
 
This isn't my first pair of IEMs. I've been using them for almost a decade and have owned several of them, as well as tested many more. I'm also an audio professional and experienced audiophile, so I'm well aware of all the issues related to IEMs such as getting a proper seal, insertion depth, source, impedance, etc."
 
64 Audio then asked me to send the U5 back to them to check the tuning, and I did, and I asked them to keep me informed on what the problem was in this email:
 
"Please keep me updated on the issues you guys discover and how you fix them. I want to know if it was indeed a tuning issue with the drivers or some other problem, and I need to know for sure that the frequency response it perfectly neutral before you send it back to me. And if you cannot get the tuning to measure perfectly neutral, I'd rather you just refund me."
 
They then sent it back to me without any correspondence. But they actually sent me the U10 without telling me first, and I had no idea it was the U10 since the IEMs look pretty much identical, and it's only on the barcode sticker that you can identify the model number, and I had no reason to check the barcode sticker. After testing it, I sent them this email:
 
(Keep in mind that when I wrote this next email, I had no idea they sent me the U10 instead of sending back the U5).
 
"I just received the U5 you sent back to me after retuning the drivers, and it is still NOWHERE near sounding neutral and is not in any way acceptable for professional audio work. There's a little bit of change in the upper-mids in individual frequencies, but the overall sonic signature remains the same, with way too much sub-bass and lower-mids that sounds bloated and muddy, and the entire upper half of the frequency range is sunken and dull, with individual frequencies in the upper-mids and treble being way too sharp and bright. 
 
When your office assistant (Jessica Ilkevich) emailed me on December 28th telling me that you received the U5 I sent back, I had specifically given her this instruction in my reply:
 
'Please keep me updated on the issues you guys discover and how you fix them. I want to know if it was indeed a tuning issue with the drivers or some other problem, and I need to know for sure that the frequency response it perfectly neutral before you send it back to me. And if you cannot get the tuning to measure perfectly neutral, I'd rather you just refund me.'
 
I never got a reply, and then the U5 was sent back to me without any notice, as if my entire instruction was completely ignored. 
 
Then when I tested the retuned U5, it pretty much sounds very similar to before the retuning. I have attached two screenshots showing my EQ correction for both before and after versions and how they largely are the same. 
 

 
I have to wonder how your audio engineers are measuring and tuning the headphones, because this is NOT what neutral/accurate sounds like--not by a long shot. Do your engineers even test the IEMs using logarithmic sinewave sweeps, individual frequency sinewave test tones, pink noise measurements, and professional headphone measurement equipment? I can't imagine that they do because if they did, it'll be clearly obvious how far from neutral/accurate the U5 is, yet you are advertising it as something for audio professionals. 
 
At this point, I doubt any additional effort to get what you advertised/promised is necessary, as I don't feel it's even possible, so please just give me a full refund."
 
64 Audio then informed me that the U5 I had sent back to them didn't have tuning issues and it sounds as they intended, and they had sent me the U10 to try but forgot to tell me about it. If I liked the U10 they'd give me a discount. 
 
And finally, everything concluded with me sending back the U10 and getting a refund for the U5.
 
Now, my main point here is this:
 
I'm sick and tired of audio product companies claiming that their products sound neutral/accurate and are designed specifically for audio professionals that do mixing and mastering, when the products sound nowhere near neutral/accurate in frequency response. And it's not as if it's impossible to achieve frequency response that's very close to neutral/accurate, because it's been done before. Even if those attempts by the other companies weren't perfectly neutral/accurate, they were at least within an acceptable range of neutrality/accuracy that I would not raise an eyebrow if those companies claimed their products are suitable for audio professional who need neutral/accurate audio reproduction. 
 
The problem here is that companies that don't produce products that are neutral/accurate yet advertise them as so, and it really, really pisses me off. That is false advertising, and it wastes our time and energy when these products don't perform anywhere near what the companies promised. I have no doubt that their audio engineers know their products don't actually measure neutral/accurate, and I have to believe that they do test and measure their products properly. The problem here, I think, is overzealous marketing deparments trying to sell the products as something they are not, in order to appeal to wider range of customers who would otherwise not be interested in their products. When you make a claim that your products are neutral/accurate and meant for professional audio engineers doing mission critical mixes and masters, you better be able to back it up with the performance of your products, otherwise you're just another consumer audio company using hyperbolic marketing language/false advertising to sell your products as something they are not.
 
64 Audio is far from being the only offender in this, as we all have seen similar behavior from other companies. It is my hope that we can do something about it--to force these companies to change their behaviors and be more honest about their products. But the consumer audio industry is notorious for being opaque, and to ask them to show actual measurements of their products would be met with nothing but silence. Pro audio industry is better in this regard, but it's usually the high-end products that provide measurement data for their products. 
 
Anyway, I'm sharing this for those of you who need neutral/accurate frequency response and are hoping to find it in the U5 or U10. The bottomline is, you won't find it in those products. 
 
Also, I want to say that I'm not specifically angry with 64 Audio or anything--I'm just disappointed in the entire consumer audio industry (and also some pro audio companies) for how common this type of behavior is. 64 Audio is not worse than other companies, since many other companies are the same way, and we the consumers simply need to fight against it and force them to change their ways. 
 
And one more thing--none of this has anything to do with the wonderful ADEL technology or the work that's done by the folks behind that technology. I'm only talking about the frequency response.

 

[ForceMajeure]

Thank you for updating this thread. I am sure it has been a difficult and frustrating journey for you to find your IEM.
I am curious though, what over ear headphone would you consider to be neutral to your liking? Maybe having another point of reference as an overear headphone might help people around here guide you to your goal.
 
As I am sure you are aware by now neutral is different for everyone. Some signatures are close one to another having "quirks" here and there about sibilance or slightly mid bass hump, other are quite far from neutral... Also along the years the so called "neutrality" have evolved including more bass into account than in the 70s and 80s, add the fact that some of the HiFi trend was to add more lower trebble to add to the illusion of more detail and you have neutrality all over the place.
I must say that I am not that sensitive to sibilance (and I still have good ears perceiving up to 17.5khz somtimes 18khz at higher volumes) and my neutral definition have evolved along the years with the current trend.
Music that includes more bass and subbass is more present, having also different textures and layers of bass thus making this part of the spectrum ever more relevant than before to me. Changing my point of view on neutrality over the years.
 
Now I am sure you also heard about the Harman Kardon target response trying to make an educated guess for the masses about neutrality or should I say naturality. Unfortunately you have greater sensitivity to certain part of the audio spectrum that must be taken into account and making your choice more difficult.
 
To be sure I am not justifying any side here I am just pointing out facts that affected your journey from an outside perspective. I am sure you your view have some solid point regarding the frequency response on the products you have tested so far and I am thankful to you for adding your impressions. It have a great value to me with your honest review showing flaws that are not easily apparent from other reviewers or not exposed enough. 
 
I also think that you should consider your journey a point of envy by many headfiers here, you have been fortunate to test some of the best IEMS out there, though none of them is justifying it's status in your eyes.
 
Hope you find your holly grail. I also advise you to take this journey lightly.

 

[Lunatique]

  Thank you for updating this thread. I am sure it has been a difficult and frustrating journey for you to find your IEM.
I am curious though, what over ear headphone would you consider to be neutral to your liking? Maybe having another point of reference as an overear headphone might help people around here guide you to your goal.
 
As I am sure you are aware by now neutral is different for everyone. Some signatures are close one to another having "quirks" here and there about sibilance or slightly mid bass hump, other are quite far from neutral... Also along the years the so called "neutrality" have evolved including more bass into account than in the 70s and 80s, add the fact that some of the HiFi trend was to add more lower trebble to add to the illusion of more detail and you have neutrality all over the place.
I must say that I am not that sensitive to sibilance (and I still have good ears perceiving up to 17.5khz somtimes 18khz at higher volumes) and my neutral definition have evolved along the years with the current trend.
Music that includes more bass and subbass is more present, having also different textures and layers of bass thus making this part of the spectrum ever more relevant than before to me. Changing my point of view on neutrality over the years.
 
Now I am sure you also heard about the Harman Kardon target response trying to make an educated guess for the masses about neutrality or should I say naturality. Unfortunately you have greater sensitivity to certain part of the audio spectrum that must be taken into account and making your choice more difficult.
 
To be sure I am not justifying any side here I am just pointing out facts that affected your journey from an outside perspective. I am sure you your view have some solid point regarding the frequency response on the products you have tested so far and I am thankful to you for adding your impressions. It have a great value to me with your honest review showing flaws that are not easily apparent from other reviewers or not exposed enough. 
 
I also think that you should consider your journey a point of envy by many headfiers here, you have been fortunate to test some of the best IEMS out there, though none of them is justifying it's status in your eyes.
 
Hope you find your holly grail. I also advise you to take this journey lightly.

The closest to neutral/accurate headphones I've heard to date is maybe the HD650, but it needs more deep sub-bass and a bit more highest treble to be truly neutral/accurate. I posted my EQ curve to make it more neutral just a couple of posts back. There are newer headphones that measure even more neutral than the HD650, but I haven't heard or tested them, so I can't vouch for them. The newer ones that are very close to the Harman Target Response Curve are likely even more neutral, and I generally trust Tyll's (of InnerFidelity) opinion, so his Wall of Fame selection is where I often check for the best of the current headphones. 
 
I don't agree with those who say neutral/accurate is subjective. If that's the case, then there would be no professional audio standards, and audio engineers would have nothing to base their assessment on. When a pair of speakers measure perfect flat in an anechoic chamber, that is not subjective--that is hard scientific fact. When you measure the frequency response and time domain data at the listening position in an professional audio studio or a living room, the measurements and the corrections you can apply to it using DSP (such as the IK Multimedia ARC System 2 and similar technologies), is also not subjective--that is all based on science. 
 
And assuming you have no hearing loss problems and can hear all audible frequencies with accuracy, and you are familiar with how to use logarithmic sweep tone, pink noise, sinewave test tones, etc., and know how they are supposed to sound on audio reproduction devices that are neutral, then when you assess a pair of headphones, you can get quite close to objectivity, because you know you're not supposed to hear that obvious spike during the log sweep or pink noise, and when you play sinewave test tones at regularly intervals from 20 Hz to 20 KHz, you're not supposed to hear drastic dips and spikes from one frequency to the next. That is how you can objectively assess audio. 
 
Finally, when you actually have done all the corrections and listen to the neutral/accurate results, it's like a revelation. No frequency range is out of place. You don't hear the equipment anymore--you just hear the music.