Hey everyone, just wanted to share a program called Harman How To Listen that you can try if you’re interested in training your ability to discern changes in frequency bands. You essentially compare “Flat” to “EQ” for a given track and try to discern which band the EQ is affecting. This is only training a very specific subset of listener ability, but some might find it useful for discerning what effect specific peaks and valleys can have on a transducer. Something nice about this program is that you can also choose your own tracks you'd like to use.
It took me a little over an hour to reach Level 8 which is the minimum requirement to be a "Harman trained listener". I can't really offer any tricks here, but the song you choose does matter. Some songs don't have as much information under 50Hz which will make it difficult to discern whether it's "1" or "-1", and I found myself having to listen for a specific part of the song I chose (Joe Nichol's "Sunny and 75"). Auditory masking, where one frequency affects another, also makes it substantially more difficult at the higher levels. Prior to Level 11, which I tried for fun and got almost every trial incorrect, I would at least hit fairly close to the correct band. I wouldn't get discouraged if you don't get very high at first. You can train your ability to listen for these things; I think I could realistically hit Skill Level 9-10 with practice. And I'd consider myself a pretty average listener - there are people out there who can hit over 12!
Of course, stuff like this doesn't really matter if you're just listening to enjoy the music. But as a reviewer, I do feel obligated to some degree to do tests like these so that people have a baseline for what my listening ability is. It also serves as a good reality check for me sometimes
Honestly Precog, I can tell the novelty has worn off. It gets real old real fast sifting through the sand for one sparkle. Your efforts and talents would be better spent taking time with quality releases much further upmarket, and leave the low-hanging fruit to the low-end YouTube reviewers.
But hey, if you're getting compensated and it's paying your way, I can think of worse part time jobs out there for sure.
Yes, I'll be devoting less time to the budget stuff going forward, and they'll likely be aggregate posts if I do them. Probably take on less review units for that stuff. I have another priority right now (outside of all the schoolwork and the reviews).
Of course, stuff like this doesn't really matter if you're just listening to enjoy the music. But as a reviewer, I do feel obligated to some degree to do tests like these so that people have a baseline for what my listening ability is. It also serves as a good reality check for me sometimes
This is the main reason I don't write full reviews, I don't feel qualified enough to do so accurately, I only post short impressions compared to other stuff I own. I'm amazed how some reviewers can pick out if a frequency is slightly elevated etc so easily. I've been looking out for something like this to help me learn what to listen for, thank you.
Hey everyone, just wanted to share a program called Harman How To Listen that you can try if you’re interested in training your ability to discern changes in frequency bands. You essentially compare “Flat” to “EQ” for a given track and try to discern which band the EQ is affecting. This is only training a very specific subset of listener ability, but some might find it useful for discerning what effect specific peaks and valleys can have on a transducer. Something nice about this program is that you can also choose your own tracks you'd like to use.
It took me a little over an hour to reach Level 8 which is the minimum requirement to be a "Harman trained listener". I can't really offer any tricks here, but the song you choose does matter. Some songs don't have as much information under 50Hz which will make it difficult to discern whether it's "1" or "-1", and I found myself having to listen for a specific part of the song I chose (Joe Nichol's "Sunny and 75"). Auditory masking, where one frequency affects another, also makes it substantially more difficult at the higher levels. Prior to Level 11, which I tried for fun and got almost every trial incorrect, I would at least hit fairly close to the correct band. I wouldn't get discouraged if you don't get very high at first. You can train your ability to listen for these things; I think I could realistically hit Skill Level 9-10 with practice. And I'd consider myself a pretty average listener - there are people out there who can hit over 12!
Of course, stuff like this doesn't really matter if you're just listening to enjoy the music. But as a reviewer, I do feel obligated to some degree to do tests like these so that people have a baseline for what my listening ability is. It also serves as a good reality check for me sometimes
Hmm, I'm not sure. If MRS gets it, I'll at least hear it eventually!
Yes, I'll be devoting less time to the budget stuff going forward, and they'll likely be aggregate posts if I do them. Probably take on less review units for that stuff. I have another priority right now (outside of all the schoolwork and the reviews).
For anything below $300, just wait for good reviews before doing yours. I, myself, currently have 7 iems under $300 with me to review and they are on my backlog. Honestly, I've went through all of them and only Hana 21 is OK to me and the others were just hmm.
Haven't release my writing and I have been postponing them too much as they are too average for me to write.
So, I understand the boredom at this point. What sticks out on me the last few months was Yuan Li. Not really others, even Bravery is OK to me.
Hey everyone, just wanted to share a program called Harman How To Listen that you can try if you’re interested in training your ability to discern changes in frequency bands. You essentially compare “Flat” to “EQ” for a given track and try to discern which band the EQ is affecting. This is only training a very specific subset of listener ability, but some might find it useful for discerning what effect specific peaks and valleys can have on a transducer. Something nice about this program is that you can also choose your own tracks you'd like to use.
It took me a little over an hour to reach Level 8 which is the minimum requirement to be a "Harman trained listener". I can't really offer any tricks here, but the song you choose does matter. Some songs don't have as much information under 50Hz which will make it difficult to discern whether it's "1" or "-1", and I found myself having to listen for a specific part of the song I chose (Joe Nichol's "Sunny and 75"). Auditory masking, where one frequency affects another, also makes it substantially more difficult at the higher levels. Prior to Level 11, which I tried for fun and got almost every trial incorrect, I would at least hit fairly close to the correct band. I wouldn't get discouraged if you don't get very high at first. You can train your ability to listen for these things; I think I could realistically hit Skill Level 9-10 with practice. And I'd consider myself a pretty average listener - there are people out there who can hit over 12!
Of course, stuff like this doesn't really matter if you're just listening to enjoy the music. But as a reviewer, I do feel obligated to some degree to do tests like these so that people have a baseline for what my listening ability is. It also serves as a good reality check for me sometimes
Hmm, I'm not sure. If MRS gets it, I'll at least hear it eventually!
Yes, I'll be devoting less time to the budget stuff going forward, and they'll likely be aggregate posts if I do them. Probably take on less review units for that stuff. I have another priority right now (outside of all the schoolwork and the reviews).
If something isn't special enough to warrant its own review, then aggregate reviews are a great way to summarize a collective of somewhat bland or disappointing sets.
Bought these off of a friend after the suggestions to try some earbuds. Diving straight into the sound, they're about what I was expecting. The X6 has decent amounts of mid-bass, but it decidedly falls short in terms of extension and in sub-bass quantity (probably tapering off a little under 100Hz). It be what it be, but I think I'm someone who fundamentally needs sub-bass to enjoy my music. The X6's treble follows a similar pattern to the bass, sporting a fairly smooth curve through the lower treble and then rolling off after ~7kHz, maybe a tad earlier. In any case, the X6 definitely has less treble extension than the Tanchjim Tanya and sounds pretty veiled. I can hear below and above the numbers I've given on a sine sweep; however, not so much when listening to actual music. The midrange, then, is where I hear the X6 as making its mark. There's some bloat coming in from the mid-bass - too much warmth for my preferences - but there's nothing really wrong with the overall midrange tonality which is a rarity given an IEM at this price point.
Earbud imaging is alright? It's three-blob and hella fuzzy, but the X6 does sound pretty open for the price point. This is especially true for soundstage width. Technicalities outside of this, though, are all about what I'd expect: the X6 has poor resolution and no dynamics. In essence, the X6 actually reminds me a lot of my Sennheiser HD6XX with its solid midrange tonality, rolled-off bass and treble, and generally three-blob imaging. So yeah...I hate it for almost all the same reasons, but I can also see why the X6 appeals to listeners after a smooth, slightly bassy listen. A pretty safe buy for $10 or whatever it costs, even if it won't be converting me to earbuds any time soon. I also won't score the X6 for now because earbuds are fundamentally different from IEMs and I don't really have a baseline.
Bought these off of a friend after the suggestions to try some earbuds. Diving straight into the sound, they're about what I was expecting. The X6 has decent amounts of mid-bass, but it decidedly falls short in terms of extension and in sub-bass quantity (probably tapering off a little under 100Hz). It be what it be, but I think I'm someone who fundamentally needs sub-bass to enjoy my music. The X6's treble follows a similar pattern to the bass, sporting a fairly smooth curve through the lower treble and then rolling off after ~7kHz, maybe a tad earlier. In any case, the X6 definitely has less treble extension than the Tanchjim Tanya and sounds pretty veiled. I can hear below and above the numbers I've given on a sine sweep; however, not so much when listening to actual music. The midrange, then, is where I hear the X6 as making its mark. There's some bloat coming in from the mid-bass - too much warmth for my preferences - but there's nothing really wrong with the overall midrange tonality which is a rarity given an IEM at this price point.
Earbud imaging is alright? It's three-blob and hella fuzzy, but the X6 does sound pretty open for the price point. This is especially true for soundstage width. Technicalities outside of this, though, are all about what I'd expect: the X6 has poor resolution and no dynamics. In essence, the X6 actually reminds me a lot of my Sennheiser HD6XX with its solid midrange tonality, rolled-off bass and treble, and generally three-blob imaging. So yeah...I hate it for almost all the same reasons, but I can also see why the X6 appeals to listeners after a smooth, slightly bassy listen. A pretty safe buy for $10 or whatever it costs, even if it won't be converting me to earbuds any time soon. I also won't score the X6 for now because earbuds are fundamentally different from IEMs and I don't really have a baseline.
The X6's magic trick is how enjoyable the sound can be for $10 (I got them for $6). Whenever I listen to the X6 and really enjoy how it sounds I look over at my $3000 IEMs and think 'wait...what...!?'. Obviously you can't really compare, but in a way, and at a certain level, you can. The value is downright stupid good, and the reason why non-hobbyists look at us like we're deluded.
Bought these off of a friend after the suggestions to try some earbuds. Diving straight into the sound, they're about what I was expecting. The X6 has decent amounts of mid-bass, but it decidedly falls short in terms of extension and in sub-bass quantity (probably tapering off a little under 100Hz). It be what it be, but I think I'm someone who fundamentally needs sub-bass to enjoy my music. The X6's treble follows a similar pattern to the bass, sporting a fairly smooth curve through the lower treble and then rolling off after ~7kHz, maybe a tad earlier. In any case, the X6 definitely has less treble extension than the Tanchjim Tanya and sounds pretty veiled. I can hear below and above the numbers I've given on a sine sweep; however, not so much when listening to actual music. The midrange, then, is where I hear the X6 as making its mark. There's some bloat coming in from the mid-bass - too much warmth for my preferences - but there's nothing really wrong with the overall midrange tonality which is a rarity given an IEM at this price point.
Earbud imaging is alright? It's three-blob and hella fuzzy, but the X6 does sound pretty open for the price point. This is especially true for soundstage width. Technicalities outside of this, though, are all about what I'd expect: the X6 has poor resolution and no dynamics. In essence, the X6 actually reminds me a lot of my Sennheiser HD6XX with its solid midrange tonality, rolled-off bass and treble, and generally three-blob imaging. So yeah...I hate it for almost all the same reasons, but I can also see why the X6 appeals to listeners after a smooth, slightly bassy listen. A pretty safe buy for $10 or whatever it costs, even if it won't be converting me to earbuds any time soon. I also won't score the X6 for now because earbuds are fundamentally different from IEMs and I don't really have a baseline.
Pretty close to my hearing, I'd say. As for the tuning, I immediately noticed that the highs take a noticeable roll-off right after the lower treble region, with the upper midrange/lower treble area is perceived to be slightly elevated relative to the neutral level. The X6's bass extension is very good imo, but only in the context of other earbuds.
And on the technical side, leaving aside the poor in an absolute sense detail retrieval, separation, note definition and all that (honestly, I would assume even for like $10 they aren't among the best in that regard), it is worth noting the tactility, overall punchiness of the bass and its texture, which gives pretty enjoyable macrodynamics, so can't entirely agree on this one. Although some amount of muddiness is definitely there, especially after clean yet emphasized bass responses of the FDX1 and Hana 2021.
In the end, the aforementioned surprising engagement of the X6 is created as it seems to me with a combination of somewhat unusual for the IEM world imaging, certain good things about the bass and, of course, mostly appealing, easy-to-love tonality.
Bought these off of a friend after the suggestions to try some earbuds. Diving straight into the sound, they're about what I was expecting. The X6 has decent amounts of mid-bass, but it decidedly falls short in terms of extension and in sub-bass quantity (probably tapering off a little under 100Hz). It be what it be, but I think I'm someone who fundamentally needs sub-bass to enjoy my music. The X6's treble follows a similar pattern to the bass, sporting a fairly smooth curve through the lower treble and then rolling off after ~7kHz, maybe a tad earlier. In any case, the X6 definitely has less treble extension than the Tanchjim Tanya and sounds pretty veiled. I can hear below and above the numbers I've given on a sine sweep; however, not so much when listening to actual music. The midrange, then, is where I hear the X6 as making its mark. There's some bloat coming in from the mid-bass - too much warmth for my preferences - but there's nothing really wrong with the overall midrange tonality which is a rarity given an IEM at this price point.
Earbud imaging is alright? It's three-blob and hella fuzzy, but the X6 does sound pretty open for the price point. This is especially true for soundstage width. Technicalities outside of this, though, are all about what I'd expect: the X6 has poor resolution and no dynamics. In essence, the X6 actually reminds me a lot of my Sennheiser HD6XX with its solid midrange tonality, rolled-off bass and treble, and generally three-blob imaging. So yeah...I hate it for almost all the same reasons, but I can also see why the X6 appeals to listeners after a smooth, slightly bassy listen. A pretty safe buy for $10 or whatever it costs, even if it won't be converting me to earbuds any time soon. I also won't score the X6 for now because earbuds are fundamentally different from IEMs and I don't really have a baseline.
I can't really appreciate the X6. Bought one for fun. I have some experience with buds too. Midrange price etc but the X6 is a bit too low res for me. Lost a lot of treble and mids details but the bass is noticeably the biggest quantity out of all range. And the DD does have a great impact for the low note even though bass resolution is lackluster.
When i wear them, i tried to crank up because i want the mids and highs to be clearer but it still doesn't...
If something isn't special enough to warrant its own review, then aggregate reviews are a great way to summarize a collective of somewhat bland or disappointing sets.
Yeah the X6 is impressive for the money but I've not actually listened beyond the first day it arrived, I was curious what a $10 bud sounds like.
Absolutely best bang for buck ($80) bud I've tried which competes with my IEMs for ear-time is a DIY custom bud - Newbsound 32pro. An all-rounder with excellent bass (for buds), lovely lush mids and crisp but never harsh highs - coupled with fantastic wide soundstage. Seriously impressive.
In technicalities, the See Audio Bravery have below average soundstage for a close to $300 USD set (in height, depth and width), especially when the stock Xelastec eartips are used. Imaging and layering are good, though clarity, instrument separation and micro-details are not class-leading at this price range. Dynamics sound a bit subdued on this set, even with amping, so that’s one area that can be improved on.
Tonality and timbral accuracy are good for a pure BA setup.
The Bravery are decent in my book, but at this price range, the competition against other big boys is huge and there are other competitors with better technical performance, if that is the department you are looking at specifically.
Some "Scientific" Musings on BA Bass, Timbre, and Coherency
Most readers will know that while I appreciate measurements and the science behind the sound, I do have a subjectivist lean in my reviews. I believe that there are aspects of sound - intangibles - distinct from frequency response and the stuff that can currently be measured. But today, I wanted to take a look at the other side of the fence and some of the rational behind what the science says. To be clear, this is not to say stuff like "slam", "dynamics", and "resolution" do not exist. I think we can all agree that these are qualities of sound that many can hear; it's just a matter of how these qualities are being perceived that I want to understand better. As another disclaimer, most of the stuff I'll be writing about here comes from discourse with community members such as Oratory1990 and the other science-y peeps, so 1) it'll mostly be me regurgitating the conversations with some of my own thoughts, and 2) this should not be considered an authoritative post (like, at all) on the subject matter.
Okay, so to establish what the science says, it's more or less the idea that frequency response is responsible for most - if not all - of what we hear. I think there are some stipulations for stuff like imaging, but it's pretty rigid in what's responsible for what we hear. You can check out another certain forum if you'd like to know what I'm talking about.
I. BA vs. DD Bass
One of the things that's tripped me up to no end, then, is the idea that BAs and DDs should sound identical when matched to the same frequency response. As we all know, BAs...well, they kind of suck at bass in particular. No slam, no decay, no sense of authenticity - you've heard me rant endlessly in my reviews. Usually. The scientific argument for what actually constitutes what we call "BA bass" is based upon leakage tolerances. Some measure of leakage is always present when we seal IEMs in our ear, but BAs inherently leak more air than DD drivers, meaning that they cannot hit the lowest sub-bass frequencies with as much SPL. Initially, I though this was a stretch, but the more I think about it, I do think this is a large component of what qualifies sonic differences between BAs versus DDs:
Most BAs IEMs have crap sub-bass extension and go for a more mid-bass oriented response (eg. the CFA Andromeda). The ones that do have a sub-bass oriented bass curve and good extension are universally the ones I'd deem "above average" for BA bass.
If you EQ in ridiculous amounts of bass on a BA IEM - or listen to something like KZ's awful AST - you will start to hear more decay wherein a bass note takes longer to fade out.
One way to control this leakage is via a front-vent; 64A is a good example à la their apex modules. By moving the leakage to a known point, you can have it "overwhelm" the inherent leakage from the IEM in the ear canal and tune the IEM more accurately. Furthermore, most all of 64A's BA IEMs follow sub-bass curves minus the U18t which easily has the worst bass in their lineup.
The distinction in all these cases, however, is what we perceive as "slam" or physicality. Even if you managed to match a BA and a DD IEM perfectly in terms of FR, I suspect you'd still hear a difference. Similarly, even if you slap SPL for days onto BA bass and get that decay, at least in my experience, it still won't match a DD for the sense of slam to when a bass note hits. To this end, people usually like to make the argument that DDs push air to the front, whereas BAs vibrate from the side. But technically, they should be achieving the same result - vibrating air - if FR is the end-all-be-all! Ruling that out, then, I would imagine that "slam" partly has to do with non-linearities - AKA distortion. This is an idea that Resolve has proposed for some time as, for example, BAs usually have noticeably higher third-harmonic distortion than dynamic drivers (more on that later). Now, this is assuming that we can actually hear this distortion because, technically, you're not supposed to be able to hear it at the levels they're present in IEMs. But we know what distortion sounds like; it makes things sound fuzzy and sort of grainy. You might not see where I'm going until you ask, "But how is slam created in the first place?". Discourse with an IEM maker explained it better than I could: "The slam is all about the clear boundaries between each bass notes. The clearer the line, the better the slam". I'd say this goes deeper than, say, drum hit to drum hit (or stuff like the 64A Nio wouldn't slam!) and it's more about the foundation, the quality of the transients themselves.
II. "BA Timbre"
Sennheiser IE900:
Okay, to circle back to that 3rd harmonic distortion stuff - it's really interesting! If you go to "distortion" in REW, you can see a distortion breakdown by harmonics. Almost universally, full-BA IEMs exhibit higher 3rd harmonic distortion that their DD counterparts, with the Sennheiser IE900 exhibiting probably the lowest I've seen for a DD. As for whether I think it's 3rd harmonic distortion specifically that creates BA timbre, that's a lot harder to say. Honestly? I doubt it. There's a strong possibility that it's either an extraneous or a confounding variable. There are some patterns, such as BA IEMs like the U12t, U6t, and SA6 (IEMs I've praised for their lack of BA timbre) sporting noticeably lower 3rd harmonic distortion than their peers. But the consistency seems to fall apart with hybrids like the KZ ZS10 Pro, which absolutely has god-awful timbre, clocking in evenlower than the aforementioned IEMs. Another issue is that distortion behaves differently in actual music as opposed to a set sine sweep.
III. Coherency
You guys have also probably seen me rant about incoherency before. This is the idea that different driver types attack and decay at different speeds, lending to differences of timbre which can be jarring to hear in hybrid IEMs. Scientifically, though, I was informed it shouldn't be perceivable outside of an egregious crossover (which would present itself very obviously in FR). As a counter, I brought up the idea of waterfall plots because they clearly reflects differences in the decay spectrum between DDs and BAs. But technically, 5-10ms occurs faster than should be perceptible to our ear! This really makes me wonder: Are we just placebo-ing ourselves when we talk about things like coherency, at least within this context?
An idea proposed by Cineira is that, instead, these reverb trails are perceived as extra SPL. This is based on the idea that when we hear pulses in quick enough succession, the SPL is perceptively amplified. Could this be why many BA IEMs sound like what I'd deem to be upwards-compressed? Some other people I know suggest that it's more the overall tonality - so indeed frequency response - of an IEM that lends to a sense of disjointedness. For example, we see a 200Hz dip on IEMs like the Moondrop Variations and Symphonium Helios that makes the bass sound too distinct. There's also a lot of IEMs like the Moondrop Blessing 2, for example, that have holes in their treble response particularly after 10kHz. Could this possibly be lending to the "grit" and BA timbre that we hear? It's really difficult to say because it goes against a lot of the things I've told myself so far.
Furthermore, verifying these dips is next to impossible because measurement couplers aren't accurate after 10kHz. Incidentally, that's a question I've had for some time: Why aren't the couplers accurate after 10kHz? I think MadEconomist and Oratory1990 explained it pretty well. 1) There's obviously the inherent resonance peak, but 2) couplers are designed to mimic the canal of the human ear, and humans have high variability from person-to-person after 10kHz. Because of these limitations, I don't think we can ever know how something actually truly measures in the treble. And even if I run sine sweeps by ear, it's a reflection of what I hear; others won't necessarily hear a peak or trough where I hear it that high-up. For reasons like these (and the fact that funding studies requires massive $$$), I don't think we'll ever see the end of the debate on stuff like this. Luckily, I don't really have a vested interest in either side, and I'm happy to play the fence sitter. It's certainly fun learning about this stuff at least!
Some "Scientific" Musings on BA Bass, Timbre, and Coherency
Most readers will know that while I appreciate measurements and the science behind the sound, I do have a subjectivist lean in my reviews. I believe that there are aspects of sound - intangibles - distinct from frequency response and the stuff that can currently be measured. But today, I wanted to take a look at the other side of the fence and some of the rational behind what the science says. To be clear, this is not to say stuff like "slam", "dynamics", and "resolution" do not exist. I think we can all agree that these are qualities of sound that many can hear; it's just a matter of how these qualities are being perceived that I want to understand better. As another disclaimer, most of the stuff I'll be writing about here comes from discourse with community members such as Oratory1990 and the other science-y peeps, so 1) it'll mostly be me regurgitating the conversations with some of my own thoughts, and 2) this should not be considered an authoritative post (like, at all) on the subject matter.
Okay, so to establish what the science says, it's more or less the idea that frequency response is responsible for most - if not all - of what we hear. I think there are some stipulations for stuff like imaging, but it's pretty rigid in what's responsible for what we hear. You can check out another certain forum if you'd like to know what I'm talking about.
I. BA vs. DD Bass
One of the things that's tripped me up to no end, then, is the idea that BAs and DDs should sound identical when matched to the same frequency response. As we all know, BAs...well, they kind of suck at bass in particular. No slam, no decay, no sense of authenticity - you've heard me rant endlessly in my reviews. Usually. The scientific argument for what actually constitutes what we call "BA bass" is based upon leakage tolerances. Some measure of leakage is always present when we seal IEMs in our ear, but BAs inherently leak more air than DD drivers, meaning that they cannot hit the lowest sub-bass frequencies with as much SPL. Initially, I though this was a stretch, but the more I think about it, I do think this is a large component of what qualifies sonic differences between BAs versus DDs:
Most BAs IEMs have crap sub-bass extension and go for a more mid-bass oriented response (eg. the CFA Andromeda). The ones that do have a sub-bass oriented bass curve and good extension are universally the ones I'd deem "above average" for BA bass.
If you EQ in ridiculous amounts of bass on a BA IEM - or listen to something like KZ's awful AST - you will start to hear more decay wherein a bass note takes longer to fade out.
One way to control this leakage is via a front-vent; 64A is a good example à la their apex modules. By moving the leakage to a known point, you can have it "overwhelm" the inherent leakage from the IEM in the ear canal and tune the IEM more accurately. Furthermore, most all of 64A's BA IEMs follow sub-bass curves minus the U18t which easily has the worst bass in their lineup.
The distinction in all these cases, however, is what we perceive as "slam" or physicality. Even if you managed to match a BA and a DD IEM perfectly in terms of FR, I suspect you'd still hear a difference. Similarly, even if you slap SPL for days onto BA bass and get that decay, at least in my experience, it still won't match a DD for the sense of slam to when a bass note hits. To this end, people usually like to make the argument that DDs push air to the front, whereas BAs vibrate from the side. But technically, they should be achieving the same result - vibrating air - if FR is the end-all-be-all! Ruling that out, then, I would imagine that "slam" partly has to do with non-linearities - AKA distortion. This is an idea that Resolve has proposed for some time as, for example, BAs usually have noticeably higher third-harmonic distortion than dynamic drivers (more on that later). Now, this is assuming that we can actually hear this distortion because, technically, you're not supposed to be able to hear it at the levels they're present in IEMs. But we know what distortion sounds like; it makes things sound fuzzy and sort of grainy. You might not see where I'm going until you ask, "But how is slam created in the first place?". Discourse with an IEM maker explained it better than I could: "The slam is all about the clear boundaries between each bass notes. The clearer the line, the better the slam". I'd say this goes deeper than, say, drum hit to drum hit (or stuff like the 64A Nio wouldn't slam!) and it's more about the foundation, the quality of the transients themselves.
II. "BA Timbre"
Sennheiser IE900:
Okay, to circle back to that 3rd harmonic distortion stuff - it's really interesting! If you go to "distortion" in REW, you can see a distortion breakdown by harmonics. Almost universally, full-BA IEMs exhibit higher 3rd harmonic distortion that their DD counterparts, with the Sennheiser IE900 exhibiting probably the lowest I've seen for a DD. As for whether I think it's 3rd harmonic distortion specifically that creates BA timbre, that's a lot harder to say. Honestly? I doubt it. There's a strong possibility that it's either an extraneous or a confounding variable. There are some patterns, such as BA IEMs like the U12t, U6t, and SA6 (IEMs I've praised for their lack of BA timbre) sporting noticeably lower 3rd harmonic distortion than their peers. But the consistency seems to fall apart with hybrids like the KZ ZS10 Pro, which absolutely has god-awful timbre, clocking in evenlower than the aforementioned IEMs. Another issue is that distortion behaves differently in actual music as opposed to a set sine sweep.
III. Coherency
You guys have also probably seen me rant about incoherency before. This is the idea that different driver types attack and decay at different speeds, lending to differences of timbre which can be jarring to hear in hybrid IEMs. Scientifically, though, I was informed it shouldn't be perceivable outside of an egregious crossover (which would present itself very obviously in FR). As a counter, I brought up the idea of waterfall plots because they clearly reflects differences in the decay spectrum between DDs and BAs. But technically, 5-10ms occurs faster than should be perceptible to our ear! This really makes me wonder: Are we just placebo-ing ourselves when we talk about things like coherency, at least within this context?
An idea proposed by Cineira is that, instead, these reverb trails are perceived as extra SPL. This is based on the idea that when we hear pulses in quick enough succession, the SPL is perceptively amplified. Could this be why many BA IEMs sound like what I'd deem to be upwards-compressed? Some other people I know suggest that it's more the overall tonality - so indeed frequency response - of an IEM that lends to a sense of disjointedness. For example, we see a 200Hz dip on IEMs like the Moondrop Variations and Symphonium Helios that makes the bass sound too distinct. There's also a lot of IEMs like the Moondrop Blessing 2, for example, that have holes in their treble response particularly after 10kHz. Could this possibly be lending to the "grit" and BA timbre that we hear? It's really difficult to say because it goes against a lot of the things I've told myself so far.
Furthermore, verifying these dips is next to impossible because measurement couplers aren't accurate after 10kHz. Incidentally, that's a question I've had for some time: Why aren't the couplers accurate after 10kHz? I think MadEconomist and Oratory1990 explained it pretty well. 1) There's obviously the inherent resonance peak, but 2) couplers are designed to mimic the canal of the human ear, and humans have high variability from person-to-person after 10kHz. Because of these limitations, I don't think we can ever know how something actually truly measures in the treble. And even if I run sine sweeps by ear, it's a reflection of what I hear; others won't necessarily hear a peak or trough where I hear it that high-up. For reasons like these (and the fact that funding studies requires massive $$$), I don't think we'll ever see the end of the debate on stuff like this. Luckily, I don't really have a vested interest in either side, and I'm happy to play the fence sitter. It's certainly fun learning about this stuff at least!
I imagine the ideal sets for coupler measurement accuracy could be the deeper insertion Etymotic ER series IEMs and the like. Coupler resonance and ear canal variations may be non-factors for them, since they place the nozzle so close to the ear drums themselves.
This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
By continuing to use this site, you are consenting to our use of cookies.
