Driver story, the acoustic system and the limits of EQ

We often talk, btw, like EQ is somehow actually altering the physical FR of our headphones or speakers, when in fact it is really not.

All EQ does is vary the volume or amplitude of the signal that you’re listening to at different frequencies. This is why the other characteristics of your tansducers, and how those react to or are effected by the changes in response to the signal, are also important.

1 Like

When it comes to “driver story”, my mind always goes back to this THpS video:

I think I will not be ready to be as derisive and dismissive of estat “driver story” as the podcast team seems to be every time the topic comes up, until someone can get experienced listeners to make those same faces when listening to EQ’d dynamics or planars costing the same or less than the Stax or HE-1 the EQ was based on. This kind of material and similar Stax reviews that exist tell me there’s something there that happens because of estat design, and it’s not marginal in intensity, and it’s prrrobably not captured in constant-FR features you can get with plain EQ, or we’d all already be running Stax-like EQ on our respective potatophones and enjoying “resolution nirvana” or w/e.

1 Like

:ok_hand:

20 additional characters

Have you peeped the acoustic design and frequency response of these headphones? They have FR colorations that absolutely lend to them being hyper-detailed (flat and reduced bass, forward 1 kHz, relaxed ear gain and low treble, forward upper treble) as well as an incredibly open acoustic design that lends to them sounding incredibly spacious and black-of-background.

None of these things are necessarily specific to the performance of the driver itself, they are the effect that the overall sound and design of those headphones have for those listeners. This isn’t “driver story,” in terms of a driver type simply being better technology or whatever, but it is driver story in that many Stax electrostatic designs tend to have similar design, and thus similar sonic traits.

Andrew himself says in that video he’d probably still EQ it, Cameron says he’s missing sub-bass, and Chrono still prefers other headphones for tone, which indicates the estats are making some sort of tonal tradeoff in order to get the technical performance they’re getting.

it’s prrrobably not captured in constant-FR features you can get with plain EQ, or we’d all already be running Stax-like EQ on our respective potatophones and enjoying “resolution nirvana” or w/e.

We can absolutely do this, the problem is that “max resolution” typically has pretty awful timbre. Additionally, the understanding of “technicalities through frequency response” is barely understood (if understood at all) in the audiophile space, so most people armed with EQ tend to focus on optimizing timbre instead of courting the set of compromises that optimizing for a “technicality” often entails (the compromise usually being timbre).

4 Likes

That “timbre” and “detail” are tradeoffs may itself be a strong indication that they both are qualities emerging from frequency response.

I liken EQ to what the old Ohio State coach Woody Hayes said about throwing the football:
“Three things can happen, and two of them are bad”.

I messed around with EQ for too long and eventually decided to dump EQ altogether. There are enough headphones available nowadays that are “close enough” to neutral that attempting to apply EQ can do more harm than good.

I find more benefit from using PGGB upscaling than I do from EQ.

As always, YMMV.

1 Like

All I’m seeing is idle hypothesizing / throwing stuff at the wall. Has anyone got their mind blown by some imitation of these FR elements and acoustic design aspects in a headphone with a different driver technology, just like in the “brain-melting” video? Has anyone heard such an imitation attempt who could compare it to actual Stax, and have they said it came very close or was indistiguishable? Without experimental results, it looks to me like you don’t have the prerequisites to be so confident in your hypotheses and certainly not to be so derisive of “driver story” (for estats) as you are on the podcast.

None of these are relevant to what I highlighted from that video: estats can be mind-blowing without having correct timbre across the board. But can non-e-stats reproduce that mind-blowing quality? I’d even settle for “90% of the way”, but I don’t know any example where anyone has done it, and therefore proven that they understand correctly what it is that produces those sound qualities to quite such a degree as reported by all the historically wowed reviewers.

If we’re just going to throw around guesses, my guess is that the “extreme resolution” is created by exceptionally low distortion, which is created by exceptionally planar movement by the estat membrane which has the same force acting on it at every point on its surface, unlike e.g. planar magnetics where only the electrical trace is pulling back and forth and the rest of the membrane is passive, plus there’s a current that takes non-zero time to pass through that trace and start generating a magnetic force at different times between the start and the end of the trace, so it’s not all quite as uniform, quite as instantaneous, quite as planar. Maybe this brings to light a routinely obfuscated tail end of the distortion experiment: maybe below a certain amount of THD we no longer perceive it as irritating ‘extra’ sounds imposed ‘over’ the music but as a lack of resolution that only becomes apparent when it’s even further removed AND when listened to through supremely resolution-capable estats. :smiling_face_with_sunglasses: Who has compared the distortion audibility results from people who did the test with potatophones vs. people who did it with the Shangri-La Sr. or HE-1? :wink:

1 Like

Do you mean “experiential”? If so, I think the NectarSound Hive hits the mark for me. Also an eStat, and with a philosophy? tambre? design ethic? not far from the STAX SR-007 (Mk II). Yes, I have a pair, and some old STAX and have heard some but not all of the new STAX.

And I do agree with your guess about what’s different. Even the old SR-5n that I have are ethereal in the mid and upper (they’re a tad…ahem…bass shy) in a way that neither my Rossons nor my mid level HiFiman planars manage.

1 Like

No, I’ve heard estats and those are the colorations and design factors that elicit the response of speed, spaciousness, and detail for me. That’s not hypothesizing, that’s pointing to tangible aspects of the headphone and drawing a line of causation based on observations I’m making of the sound and experience. I’m doing what everyone else does when they describe a headphone, but instead of attributing it to magic, I’m basing it on what I know to be true about the headphone.

To be clear, estats never melted my brain (because they’re not that good), but there are non-estat headphones that I quite enjoy—and actually prefer—for similar reasons, like the HEDDphone V2 or the Moondrop Cosmo. I wouldn’t say they’re quite the same in terms of openness relative to many estats, but I actually preferred their presentations of texture (Cosmo) and space (HEDD V2) vs. the estats I’ve heard despite the tuning having very similar hallmarks (flat and reduced bass, forward 1 kHz, relaxed ear gain and low treble, forward upper treble).

I don’t need prerequisites to state my opinion on what I’m hearing? I like the estat presentation, I assume for the same reasons others do, but its not the end-all-be-all for me. Novel, sure, but not the peak.

Also, driver story deserves to be derided considering there has been no exhibited correlation between driver and sound quality while people who barely understand what they’re hearing swear up and down that they’re hearing something that cannot be captured by current data collection methods. It’s really tough to believe that a cohort of enthusiasts who have so often demonstrated their lack of discernment would be able to sense things that our standard measurements would not—in almost all cases, the measurement instruments are more precise and reliable than the listeners, and in my years in this space I have literally only once seen the inverse demonstrated (though this one case would be generous to ascribe to skill on listener’s parts, when it was actually a failure of the measurement instruments).

Sure, this is broadly true of headphones in general though. People can and do enjoy colored headphones.

Why do you think estats are the only thing that can produce this level of sound quality/reaction? Why are you putting so much stock into 3 listeners who like quite bright headphones saying “wow” at a headphone as if that’s going to be the zenith of sound quality even for these three listeners, even when all of them pointed out issues? Furthermore, why do you take this as evidence of some broader trend that other headphones/technologies cannot emulate? Throw a rock on YouTube or HeadFi and you’ll find a video of someone having an equally effusive reaction to a dynamic driver, planar magnetic, ribbon, AMT, or balanced armature headphones/IEMs.

If you’re looking for a continuously identifiable “good sound quality” trait across reviewers, you simply will not find it. The closest analog, the only thing I can think of that consistently leads to more positive reviews, is unironically probably price.

And I’m the one doing idle hypothesizing? :sweat_smile: Do you have any evidence to suggest that a lack of distortion typical in “lo-fi dynamic driver headphones” (which we already have a decent idea isn’t detectable by listeners and doesn’t affect preference all that much) is more likely to affect people’s impressions than consistently present FR colorations and acoustic characteristics—both of which are measureable and easily identifiable—found in estats?

Like, my impressions of what causes the estat presentation is at least rooted in the things we know to be true (the frequency response is frequency response-ing, the openness is open). If anything your guess stands counter to the established science, and thus would need to meet a much higher burden of proof than someone saying “this headphone is very open, I think this gives it a cool spaciousness effect” or “this headphone has basically no bass, so its a very clear and unencumbered sound presentation.”

Idk dude, even large amounts of THD (or, more accurately, IMD) may not perceived as extra sounds. You’re taking pretty massive logical leaps here, assuming estats are inherently the best playback system when we don’t really have evidence to support that.

I will admit to having more faith in @Torq’s thoughts on headphones and chains that he’s heard than in instrument measurements. And close to that confidence in @generic’s, @SenyorC’s and several others.

1 Like

I used to feel that way about others, but no longer.

I certainly feel you on this, SSN757. Maybe I can cut the EQ cord when I get some new HPs. :slight_smile:

Re estats and driver story. I have not personally listened to any estats, but I believe they generally have exceptionally low distortion. And probably also exceptional modal performance, which can improve the clartiy and detail in higher frequencies. And I suspect that is a big part of what people respond to when listening.

Most content is authored on speakers or headphones that are probably not as resolving though. So there might be some questions of fidelity that come into play there.

Axel Grell has also said that the Sennheiser HD 800 (a dynamic driver headphone) was designed to sound like an estat. And the HD 800S continues to be quite a popular and well-regarded headphone.

I don’t experience distortion as extra sounds btw. I experience it mainly as a lack of clarity, detail, spaciousness, and precision in imaging. Headphones that are higher in distortion sound more narrow or closed in to me, and the imaging is less distinct and more diffused. Bass also sounds less tight and precise.

I really don’t see any conflict between saying “it’s all FR” and “this particular driver type tends to have these sound characteristics”. There is only so much a manufacturer can do to tune a headphone’s FR after all. And while theoretically we could use PEQ to tune a planar to sound like an electrostatic, we all know that practically this is extremely difficult (impossible?).

4 Likes

If only he continued to make his thoughts public :cry:

2 Likes

At the same time, like @donjoe0, I suspect there may be more to distortion than science has shown so far. Admittedly I don’t read all of the latest papers coming out of AES, but I didn’t think THD or IMD is considered a good representation of what happens when playing full spectrum music - especially when it comes to human perception.

2 Likes

Cam (GoldenSound) recently did a pretty good overview of the different driver types.

I think there are some general trends that can be identified in the sound sigs of different types of headphones btw. Some of them seem to be a little more brand specific though, like the treble on Audeze vs. HFM planars for example. I don’t know if estats have a particular sound signature, other than maybe low distortion, and rolled off bass on many of the Stax.

The Sennheiser HE-1 is fairly neutral based on Resolve’s graphs. But a bit laid back in the mids (around 1.5 to 2 kHz) like an HFM planar. And a little bright in the treble around 10k or so. It also appears more rolled off in the very high treble, not unlike some studio monitors.

1 Like

I have been listening w/o EQ for a week now as an experiment (Arya Stealth and DCA Aeon X Closed) So far, I rather like it. Not totally ready to give up EQ yet…

2 Likes

Great point, yeah. And you can see it in the “brain melting” video too: the same people go from bad tonality estats (L700) to better tonality estats with more bass (X9000, Shangri-La) and they’re more blown away, not less. This is fatal to the “well clearly it has to be the bright tonality” explanation.

As for the openest-open housings: I saw some discussions once of the HE-400i getting better with a more transparent rear grille - or even using it without one - so I tried it for myself, just taking the grilles off and listening. No big improvement to my ears. The most that stood out was that I lost the precision of instrument positioning along the depth/radial direction, especially for hard-panned sounds, where yes you could say they seemed to be farther out, but also less distinct as to where farther out exactly.

I can’t say I liked it, and more importantly I don’t associate such a change with the notion of “higher resolution/detail” - quite the contrary, higher resolution should provide better imaging / crisper positioning everywhere in the stage. So I’m also unconvinced of this openest-open-back explanation for why estats sound so resolving. (Of course there’s always the possibility that my planar experiment proves nothing because I didn’t remove the magnetic stator, nor replaced it with an acoustically transparent one - could be that’s the specific kind of openness that’s needed in this context.)

1 Like

The longer I’ve used PEQ, the less I use. For one thing, you never do exorcize the FR demons completely. Each headphone still has it’s own unique sound even after EQing to a target, so why not roll with it and just correct the big problems instead? Besides, I don’t want all my headphones to sound alike anyway.

There are a couple of headphones I currently own that don’t need EQ. One has uneven treble, but I don’t bother to address it as I like using it on any device without caring about EQ. The other is near perfect for me, but does benefit from a sub-bass boost.

3 Likes