• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

IEMs Single DD vs Multiple Drivers

I was thinking of a more generalized principle, but didn't tell. The DD was found to be simple, successful, flexible, and cheap. So the attention gravitates towards it, which accelerates its adoption and optimization until a point where alternatives can't compete anymore. In single or multiple criteria, as with the Magnepan, alternatives are quickly left behind.
I wouldn't mind more optimization to the point where alternatives can't compete. I'm not sure what you mean with your speaker analogy, the majority of well-measureing speakers are two-way, not single driver. For IEMs a concern with multiple DDs is added bulk which can affect fit. So adding the compact BA seems reasonable if FR goals are hard to achieve with 1DD.
 
. For IEMs a concern with multiple DDs is added bulk which can affect fit. So adding the compact BA seems reasonable if FR goals are hard to achieve with 1DD.

I give you the mighty big sounding but small sized Sony XBA-N3, one DD + one BA, tuned well and still one of my fav listens. Sony were utterly daft to discontinue this model

53613521489_523298c45b_c.jpg


53640658100_2b5eb95374_c.jpg


54306000947_3ce2761fe1_c.jpg
 
I wouldn't mind more optimization to the point where alternatives can't compete. I'm not sure what you mean with your speaker analogy, ...
Was it me who introduced the Magnepan? Planar speakers don't play a major role today, because the dynamic driver took over for various reasons. I dislike planars altogether, because the companies using them most often use its exotics as an advertizing tool, and make me cringe.

Anyway, my explanation of the principle was to close to the topic. Take several rows of shopping carts. People take their's from and bring back the cart to the longest row. Because it is easier. Same with technological development. Once a convenient, see criteria above, solution is halfway established more effort it put into it. The effort makes the solution more attractive, bringing more effort and success to it, making it even more attractive and so forth. It doesn't mean alternatives are bad, but.

The DD in IEMs is that successful, no reasonable improvement is expected further. It is plain perfect. There is no need for 'planar' anymore. Now designers are asked to work on the internals of the IEM's shell and nozzle. The HRTF related stuff, the theory of operation if you will, was discussed before. I leave it at that as not to see any super-targets no more. No bad feelings :cool:
 
I disagree with almost everything you say and doubt that you are very versed in the subject.
You see, you obviously need to learn the language better. The IE600 graph is not due to a worse control of the treble, but on the contrary it is due to a careful tuning. All those depressions between 2 and 10 kHz are not there by casuality, especially the depressions at 6 kHz and 8 kHz, but controlled and obtained through the Helmholtz resonance chambers placed on the nozzle and you go to destroy them through the E.Q.?
No, I don't want to use it in the first place, as it does not look at all like a good IEM. You say it has resonances that are created by purpose? Oh my god.
Basically you take a significant slice of the cost of that IEM and throw it out the window. Buying a Maserati to make it sound like a Golf seems smart to you?
Haha, the IE600 is a Maserati? For me it is just an overpriced Polo.
If you want to have that kind of post E.Q. tuning (very simple to obtain) why spend more than 150€?
That is what I am saying, one does not need to spend much money at all for a good IEM, when using EQ. And the Truthear Nova is almost perfect WITHOUT EQ, so absolutely no need to ever spend more than 150€.
It is obvious that a much more complex system will be much more complex to transform, but it is not a defect as you show it, on the contrary...
Also regarding Variation, that peak at 3 kHz is wanted and you are paying for the implementation of more drivers to have that.
Maybe Moondrop wanted this peak, but I don't, I EQ it away.
Why spend all that money to have a distinctive feature (features that you may or may not like) and then eliminate it to have a "normal" curve that any good IEM between 25€ and 150€ can have? It's crazy! It's like spending 4000€ for a DanClark headphone whose cost is all in the excellent engineering to get to have those features and then eliminate them with the E.Q.
You get a HE400se for 110€ that has "standard" tuning, low distortion and do whatever you want with it. What are we talking about?
Headphones are a different story, there you can have soundstage and other effects,
For me it's bad audio knowledge + bad financial management in one post.
That is what you are doing 100%
 
Then it goes onto your ear canal, and blocks it which makes specific resonances that depend on the former, and then last on your very personal physiology. Your physiology determines the shape of treble curves at the ear drum when listening naturally too, but under quite different circumstances. It is never alike the infamous target curves.

The 5128 claims to be a much better replica of the human ear canal than the 711 and thus the measurements already entail, at least to a good extend, all the individual resonances. They might shift a bit, but if a hugh peak appears in the measurement, it will for most individuals most likely also be audible.
 
I see what you saying. I might have not been completely true to you, or maybe now after you point out those stuff, there are few things that aren't "perfect" so to speak with the PEQ on the Zero 2.

I had to cut down the 15000 Peak from 7 to 5 db, and the 9500 to from 5 to 4.5 db. Still of course missing the extension, but on quite of few songs the high can get a bit harsh. You cant really see the difference on the graph, but those slight cuts make it a bit more less sharper. It's like slight painful-pleasure to listen to, if you get me.

Blue 2 treble extension looks good, but overall the harman bump needs to be trimmed down for me and probably that dip in the 200s to be raised (which i havent tested, but by judging from your replays i think you are getting a grip of my likings). Maybe the slightly calmer air will be to my taste?

The Performer 5 looks great overall, its even on sale atm for 180$, but should i risk and blind buy it, until i'm not sure exactly where i am in the upper range? Reviews look great though.

The Hexa somehow matches pretty well the Zero's PEQ, just no bass, there is another Truthear set called Pure, which is basically Hexa 2. Feedback about it is very positive, but no idea what's its signature.

From what i can understand from you, if the FR isn't really far from your preferred taste (give or take 4-5 db) and mostly if above 10k its in your sweet spot, you can make any IEM (if the drivers allow it) to be EQ in 98% to each other, right? Maybe with silicone tips the graph would of look differently, or no? My Zero's with those Foam tips from Aliexpress sound very muffled, completely different and unenjoyable.



P.S - If the other fellow forum member doesn't take your invitation offer, i might! I love pizza (had a girlfriend which was joking me that i can eat pizza for a living :D ), vauge speed limits ( I'm in love with 458 Italia, still kicking my butt missing the chance to drive one) and in the process you might actually really help me find my IEMs, oh and i'm not that far from Milan, just 2 hours flight.
If you are unsure about which IEM to buy, find one that seems to suit you and then see if it is available on Amazon which has a no questions asked return policy. I never use foam tips because yes, many times they ruin the sound, at least to my ears. Regarding the FR, once you have figured out which FR is right for you and how it relates to your ear, then it will be a good tool to consider. Unfortunately the only thing you should do in my opinion is to EQ the IEM you have and it takes a bit of time to fine tune it. Keep in mind that there are other factors to take into account and an important one is the fit. Changing the insertion depth also changes the FR of your ear due to some resonances above 7000 Hz that move when inserted deeper. Some people with the same IEM, if the shell allows it, prefer to have a small tip and deep fit and others a large tip and shallow fit. That interaction changes the FR to your ear (nothing dramatic eh), even though the FR of the IEM is identical. It can happen that you simply won't be able to have a sufficient seal or your ear will hurt after 1 minute (for me Zero RED). So in my opinion you will hardly be satisfied with the first purchase even knowing your tastes. You can narrow the field, but at the beginning you will still have to shoot "randomly" within a more or less large group.
 
I appreciate your professional clarification. One point I have still. An IEM circumvents the head's shape, especially the pinna. It has to simulate it in terms of frequency response. But that shape is an individual trait of humans. IEMs need to be tuned or equalized to the user. If tuned in the factory a variety of options needs to be offered, that a person can chose from, given that an optimum is the goal.

This applies to the resonances in the upper treble likewise. The problem is harder, but luckily of lesser impact. As it seems to me, people are less sensitive to tamed peaks and dips, even if they do not match the individual HRTF perfectly, because the common noises present in the upper register are for one pretty broadband, and second mostly short lived. Not the least the HRTF changes a lot with incidence angle, and more rapidly the higher the pitch.

It is well understood, that multi driver designs could be tuned to more complicated shapes more easily. Question is, if that helps any individual, and in case a tighter match is desired, if it wasn't better done with electronic equalization--to taste, if that is really available.
True. The FR that will reach your eardrum (eliminating the psychological part from the equation) has small individual variations also due to how we are used to inserting them in the ear, in fact the FR tells the truth if you have understood how your ear works in function of the FR measured on that system. There are infinite different choices on the market precisely to be able to take a different slice of the market. As for "ad personam" customization, this is what some professional companies can do on request, exactly like for the tips. If I had purchased an IE600 in Germany and not in Italy, Sennheiser offered me a service to customize the tip with the mold of my ear, so hardware customizations exist and in all sauces, but with limitations and with significant costs. Nowadays even hearing aids are performed via DSP after the first hardware customization, which is the most intelligent solution
 
I disagree with almost everything you say and doubt that you are very versed in the subject.

No, I don't want to use it in the first place, as it does not look at all like a good IEM. You say it has resonances that are created by purpose? Oh my god.

Haha, the IE600 is a Maserati? For me it is just an overpriced Polo.

That is what I am saying, one does not need to spend much money at all for a good IEM, when using EQ. And the Truthear Nova is almost perfect WITHOUT EQ, so absolutely no need to ever spend more than 150€.

Maybe Moondrop wanted this peak, but I don't, I EQ it away.

Headphones are a different story, there you can have soundstage and other effects,

That is what you are doing 100%
Ahaha, ok :)
 
The 5128 claims to be a much better replica of the human ear canal than the 711 and thus the measurements already entail, at least to a good extend, all the individual resonances. They might shift a bit, but if a hugh peak appears in the measurement, it will for most individuals most likely also be audible.
@RogerSmith, in case to are more engaged in the topic than the average Joe (sic!), you might lookup the topic of head related transfer function, HRTF. It was measured for many individuals. All are different not only to some degree, but quite different. Not only quantitatively, but qualitatively different. The HRTF changes with the angle of incidence of the sound in individual ways, and more. This shows especially in graphs that are not smoothed to much. B&K did an investigation in order to present a paper on their new model of the head and torso simulator, HATS--a must read.

In short all people have individual HRTFs, and the 5128 has an individual HRTF also, just like a human.

Question is: how to evaluate the right pinna gain, as that is circumvented by the IEM's application to the ear
Question is: how to dicriminate peaks 'n dips originating in the HRTF of the 5128 from those introduced by the IEM's faults
Question is: how to translate the interaction of the IEM with the 5128 to the interaction with other ears

This should be enough food for thought me thinks. I came to disregard the "target curves" that many guys take for gospel. This is more than just confusion. It is fallacy.

Add.: on "... if a hugh peak appears in the measurement, it will for most individuals most likely also be audible." People today have the easy handed opportunity to equalize such mishaps. Nobody does that with such peaks. Because of the excellence of modern DDs it is less of a tech problem. It is because people cannot find their optimum. Conversely, it is not that much of a problem then, right?
 
Last edited:
@Sebby wont bother you more, just one last opinion on these Kinera Celest Phoenixcall

I've never had it in my hands and since I'm not a reviewer I doubt I'll ever have it in my hands, so I can't give you an opinion about this IEM. Reading around it seems like a pretty good IEM with some winning features. From the graphs it seems very suitable for some genres and less for others, but knowing my tastes I know it wouldn't be suitable for me. After I try to equalize 2 IEMs on this curve to at least understand the flavor, but for now I agree with Jimbob54
 
@RogerSmith, in case to are more engaged in the topic than the average Joe (sic!), you might lookup the topic of head related transfer function, HRTF. It was measured for many individuals. All are different not only to some degree, but quite different. Not only quantitatively, but qualitatively different. The HRTF changes with the angle of incidence of the sound in individual ways, and more. This shows especially in graphs that are not smoothed to much. B&K did an investigation in order to present a paper on their new model of the head and torso simulator, HATS--a must read.

In short all people have individual HRTFs, and the 5128 has an individual HRTF also, just like a human.

Question is: how to evaluate the right pinna gain, as that is circumvented by the IEM's application to the ear
Question is: how to dicriminate peaks 'n dips originating in the HRTF of the 5128 from those introduced by the IEM's faults
Question is: how to translate the interaction of the IEM with the 5128 to the interaction with other ears

This should be enough food for thought me thinks. I came to disregard the "target curves" that many guys take for gospel. This is more than just confusion. It is fallacy.
I agree, in fact the Target curves are reliable as long as you understand how your ear hears that particular FR, but they are not gospel in general, because that particular curve may please you but not me and not only for physiological reasons but also psychological. I myself have more than one IEM and all different, some with FR curves that apparently seem like a disaster.
 
@Sebby wont bother you more, just one last opinion on these Kinera Celest Phoenixcall

Well, I confirm some things said previously and that is that it is not an "all-purpose" IEM for me and we enter the field of the specific IEM and despite having tried to match the FR using up to 13 bands, there remains the variable of the fit that makes the difference and the timbre inconsistency that many multidrivers have, in addition to the variable of the measurement after 10 kHz not being 100% reliable.

Having said this, with you I can use a bit of subjectivism because we have similar tastes, more or less: taking my "killer tracks" in my opinion it is not very suitable for male voices, tracks with a lot of "Loudness" or deep bass. It is not terrible but I do not hear them very natural and in general on the voices in my opinion there is a bit too much focus that works in some tracks with female voices, but loses in some male ones making some sibilants emerge that are a bit laborious and also lacking a bit of thickness in the voice itself.
It sits on a line that is sometimes crossed by tracks that have been produced with more compression than necessary, but in general it stays close to that line without straying too far like in the track "The court" by Peter Gabriel.
In general, otherwise I would make a long comment telling you track by track what I like and what I don't like, it sacrifices a bit of timbre especially of the voices in exchange for technical aspects that I think are very good, such as the soundstage and layering. It brings out many details that are in the background of the track (the details are always there if the track contains them, but some FRs are better at bringing out some things and others less because of auditory masking), the soundstage is quite wide, nicely separated and good imaging. I'll tell you, I'd be very curious to try one...
With some tracks, like those by Parov Stelar, I had a lot of fun and sometimes it almost sounds "spectacular".
Just based on the FR, would I want one as my only IEM? NO. However, if my impressions were to be reflected also with the real IEM and I had bought it with my eyes closed, I would be happy to have one to pair with the others but, having an E.Q. I would stick with Zero2 at that point, unless the experience with the IEM really turns out to be very different and positive; at which point I would buy it.
:)I'll give you a hint: take a good look at Meze Alba
 
Last edited:
...in addition to the variable of the measurement after 10 kHz not being 100% reliable.
It is reliable, actually. Only that one doesn't know what to do with it. The IEM interacts with the ear canal. It closes it, while when hearing naturally the ear canal is open.

The nozzle and the cavity inside the IEM prolong the effective canal length. The whole arrangement of ear drum, canal, the nozzle and the IEM's cavity supports resonances at frequencies X, Y, Z. The pattern is determined by said properties in addition to the whole thing being closed.

Natural hearing involves the ear drum, the canal and the pinna, while the latter can be seen as basically open in first approximation. Giving a different pattern of X, Y, Z etc. Or, ideally wondrously a similar pattern

O/k, this is all a bit complicated. But so it is. That's the least to understand before evaluating an IEM's match to the various target curves. Smoothing, averaging are other considerations. Individual HRTFs for even the measurement rig are another one discussed above.

The measurments are perfectly reliable, the interpretation is not. Some folks fail to get it right, fixated on other than personal preference. Some times the latter doesn't exist, but an urge to stick to curves instead. The 'objectively' best stereo, approved by science ... misunderstood up to appearing grotesque.
 
Last edited:
Well, I confirm some things said previously and that is that it is not an "all-purpose" IEM for me and we enter the field of the specific IEM and despite having tried to match the FR using up to 13 bands, there remains the variable of the fit that makes the difference and the timbre inconsistency that many multidrivers have, in addition to the variable of the measurement after 10 kHz not being 100% reliable.

Having said this, with you I can use a bit of subjectivism because we have similar tastes, more or less: taking my "killer tracks" in my opinion it is not very suitable for male voices, tracks with a lot of "Loudness" or deep bass. It is not terrible but I do not hear them very natural and in general on the voices in my opinion there is a bit too much focus that works in some tracks with female voices, but loses in some male ones making some sibilants emerge that are a bit laborious and also lacking a bit of thickness in the voice itself.
It sits on a line that is sometimes crossed by tracks that have been produced with more compression than necessary, but in general it stays close to that line without straying too far like in the track "The court" by Peter Gabriel.
In general, otherwise I would make a long comment telling you track by track what I like and what I don't like, it sacrifices a bit of timbre especially of the voices in exchange for technical aspects that I think are very good, such as the soundstage and layering. It brings out many details that are in the background of the track (the details are always there if the track contains them, but some FRs are better at bringing out some things and others less because of auditory masking), the soundstage is quite wide, nicely separated and good imaging. I'll tell you, I'd be very curious to try one...
With some tracks, like those by Parov Stelar, I had a lot of fun and sometimes it almost sounds "spectacular".
Just based on the FR, would I want one as my only IEM? NO. However, if my impressions were to be reflected also with the real IEM and I had bought it with my eyes closed, I would be happy to have one to pair with the others but, having an E.Q. I would stick with Zero2 at that point, unless the experience with the IEM really turns out to be very different and positive; at which point I would buy it.
:)I'll give you a hint: take a good look at Meze Alba
Thank you for taking the time! I asked about this IEM because i might get it from a guy on a decent price and reading about it, the IEM responds to EQ quite well. So fixing the lean bass and the "botox" smile doesn't look that hard, anyway ill be running it constantly with JM12/JA11 and the 5-band PEQ should be enough to tune it to taste. What the IEM does after 8k, i'll need to test in person. Probably its on my lower limit of treble extension or it might actually be on my sweetspot (hopefully).

P.S - Just saw the nozzle size is quite big too, I am one of those with big canals so this is another plus for good fitment.
 
@RogerSmith, in case to are more engaged in the topic than the average Joe (sic!), you might lookup the topic of head related transfer function, HRTF. It was measured for many individuals. All are different not only to some degree, but quite different. Not only quantitatively, but qualitatively different. The HRTF changes with the angle of incidence of the sound in individual ways, and more. This shows especially in graphs that are not smoothed to much. B&K did an investigation in order to present a paper on their new model of the head and torso simulator, HATS--a must read.

In short all people have individual HRTFs, and the 5128 has an individual HRTF also, just like a human.

Question is: how to evaluate the right pinna gain, as that is circumvented by the IEM's application to the ear
Question is: how to dicriminate peaks 'n dips originating in the HRTF of the 5128 from those introduced by the IEM's faults
Question is: how to translate the interaction of the IEM with the 5128 to the interaction with other ears

This should be enough food for thought me thinks. I came to disregard the "target curves" that many guys take for gospel. This is more than just confusion. It is fallacy.

Add.: on "... if a hugh peak appears in the measurement, it will for most individuals most likely also be audible." People today have the easy handed opportunity to equalize such mishaps. Nobody does that with such peaks. Because of the excellence of modern DDs it is less of a tech problem. It is because people cannot find their optimum. Conversely, it is not that much of a problem then, right?
Thanks, I do know most of the relevant literature and am in no need to be taught by you, so keep it calm, please. All people have different HRTFs and PRTFs and all headphones different HpTFs and all that. But large peaks in a measured FR cannot be desireable and might be difficult to EQ away, especially as these peaks will vary from individual to individual in size and variation, so become tedious if not impossible to EQ.
 
Thanks, I do know most of the relevant literature and am in no need to be taught by you, so keep it calm, please. All people have different HRTFs and PRTFs and all headphones different HpTFs and all that. But large peaks in a measured FR cannot be desireable and might be difficult to EQ away, especially as these peaks will vary from individual to individual in size and variation, so become tedious if not impossible to EQ.
... as you say :rolleyes: Can't accept the hostile approach, good by.
 
Last edited:
... as you say :rolleyes:
You Gents realize you are passionately disagreeing about a shared interest that you share, yet 99.99% of the world could care less of a crap about?
I always find it a bit sad about audio discussions online. The passion you share will drive you further apart in a microcosm no one outside of this small community gives a goddamn about. I went through this kind of crap two days ago with someone I though I respected, but in the end we drive each other apart.
 
You Gents realize you are passionately disagreeing about a shared interest that you share, yet 99.99% of the world could care less of a crap about?
I always find it a bit sad about audio discussions online. The passion you share will drive you further apart in a microcosm no one outside of this small community gives a goddamn about. I went through this kind of crap two days ago with someone I though I respected, but in the end we drive each other apart.

Well, people seem more interested in pizza than in transducers for IEMs in this particular microcosm.
 
Back
Top Bottom