[KinGensai]

But then I'm left doing EQ entirely by ear. I have no problem trusting my own judgment, but I like to EQ with at least some "objective" reference in mind.

Right, that's why I mentioned squig.link and graphic visualization. The objective reference is your preference in comparison with some target response, usually the Harman IE target.

IEMs weird me out because they are so physiologically dependent. Each person literally hears each IEM different from how any other person hears that same IEM. Ultimately, I'd much rather leave that technical craft to IEM designers than to my own ear. That seems like a key region for a designer. I trust them more than I trust myself to that work.

The problem with this reasoning is the very variability of HRTF and it's stochastic nature, which you note in the first half of your statement. The designers will do one of these things: go for the Harman IE target, go for a house sound, or go off of in house acoustics research. This works better for HPs because those take your pinna into account, so the tuning has more leeway. IEMs have no such luxury, so the final tuning is up to the end user to do. As long as the IEM itself is built well, EQ will fit it to your preferences and make it shine.

 

[firesign]

I'm new to IEMs. I've been a dedicated EQ junkie for all of my over-ear headphones for years, but I'm finding that EQ for IEMs just doesn't... work? I mean, yes, my EQ adjustments are all audible, but everything I try sounds off. I can pretty confidently improve the sound quality on any pair of over-ear headphones with EQ, but I'm finding the same process leading to dead end after dead end on IEMs. My sample size of IEMs is admittedly very low, so it's possible the issue is just with the two particular IEMs I own (Truthear Hola and Kiwi Ears Orchestra Lite). But after a few weeks devoted to the task of EQing these IEMs, I might be ready to give up and just leave them without any EQ at all.

I'm curious what others think: is EQing IEMs a fool's errand, while EQing over-ear headphones is perfectly manageable? Why would there be such a big difference between the two?

Maybe you don't know how to EQ well...

All my EQed IEMs sound better than un-EQed.
All, without exception. The ones I currently have and the ones I have had throughout my life.

 

[Hypops]

Maybe you don't know how to EQ well...

This isn't helpful. Care to elaborate? If you read my post, you'd know I have extensive experience with EQ for both headphones and loudspeakers. My point was that IEMs require a very different approach to EQ than those others. What's your approach?

 

[jirams]

Is EQuing an exercise that would result in all head/ear phones being fed a tweaked signal to make them all sound the same?
Or am I being cynical?

 

[KinGensai]

Is EQuing an exercise that would result in all head/ear phones being fed a tweaked signal to make them all sound the same?
Or am I being cynical?

EQ is a tool used in audio engineering that also has utility for consumers. In mixing/mastering EQ is used to address noise and interference when mixing with other tracks or fine tuning stems. Consumers obviously can't do this normally once a song is compiled and shipped, but EQ can fix problem areas in transducers that cause perceptual errors or irritation in use.

It's just a tool. Use or don't as you please.

 

[Hypops]

Is EQuing an exercise that would result in all head/ear phones being fed a tweaked signal to make them all sound the same?
Or am I being cynical?

Theoretically possible, practically impossible. I'm sure you could get tunings on two different IEMs to sound very similar, but other differences will rear their heads: fit, insertion depth, tip choice, nozzle length, ear physiology, subjective "technicalities," etc. It's in these latter differences that I'm encountering limitations/obstacles to EQing IEMs (as compared to EQing full-size headphones, which is relatively straightforward).

 

[firesign]

This isn't helpful. Care to elaborate? If you read my post, you'd know I have extensive experience with EQ for both headphones and loudspeakers. My point was that IEMs require a very different approach to EQ than those others. What's your approach?

The title of the thread is not helpful either...

My approach is the same that with headphones:
- if a hear frequency that I don't like, I try to EQ it up or down playing with the bandwidth (or Q).
- if I have the IEM frequency response graph, I can use it to try to get my favorite graph.

And I get results immediately, just like with headphones.

 

[Hypops]

Was watching an old Q&A episode of The Headphone Show, and Resolve was suggesting something along the lines of one of my thoughts above. In a discussion of how borked the Harman in-ear target is, he was hypothesizing that it might be something to do with group delay and non-minimum phase issues in multi-driver IEMs. All guess-work, of course, since no one seems to know exactly why the in-ear target seems so off.

At the very least, it's becoming clearer to me that the IEM world is so weirdly obsessed with FR measurements that listeners and reviewers almost never pay attention to other equally important measurements like distortion, phase, and impulse response. For single-driver IEMs and full-size headphones, I get why these measurements are overlooked since (with the exception of distortion) they're not really relevant to single drivers. But for multi-driver IEMs, this info could be really useful. When I EQ loudspeakers, this info tells me what can and can't be corrected by minimum-phase EQ. Not sure why IEM reviewers with fancy measurement rigs don't publish these other data points. Combine that with the wild variations in HRTF, and it's no wonder EQ works so differently on IEMs.

 

[o0genesis0o]

Was watching an old Q&A episode of The Headphone Show, and Resolve was suggesting something along the lines of one of my thoughts above. In a discussion of how borked the Harman in-ear target is, he was hypothesizing that it might be something to do with group delay and non-minimum phase issues in multi-driver IEMs. All guess-work, of course, since no one seems to know exactly why the in-ear target seems so off.

At the very least, it's becoming clearer to me that the IEM world is so weirdly obsessed with FR measurements that listeners and reviewers almost never pay attention to other equally important measurements like distortion, phase, and impulse response. For single-driver IEMs and full-size headphones, I get why these measurements are overlooked since (with the exception of distortion) they're not really relevant to single drivers. But for multi-driver IEMs, this info could be really useful. Not sure why reviewers with fancy measurement rigs don't publish these other data points. Combine that with the wild variations in HRTF, and it's no wonder EQ works so differently on IEMs.

Measuring those things without a proper isolating chamber would likely lead to bad data, wouldn't it? Personally, I rather create and give no data than misleading ones.

Btw, didn't some audio science big shots insist that IEMs are minimum phase devices and thus everything = FR?

 

[Hypops]

Measuring those things without a proper isolating chamber would likely lead to bad data, wouldn't it? Personally, I rather create and give no data than misleading ones.

This seems to be another big issue with IEMs. IEM measurement data is still kind of a wild west of different targets, tools, and methods. Nevertheless, if people are so free and easy about publishing their FR measurements, why not do the same for other measurement data? If we collectively think that FR measurements are "good enough," wouldn't those other measurements be similarly "good enough"? I'm no acoustics expert, so I'm genuinely curious.

Btw, didn't some audio science big shots insist that IEMs are minimum phase devices and thus everything = FR?

I've seen answers going in all directions. For sure, the consensus seems to be that single-driver headphones/earphones are minimum-phase devices, but I can't help but wonder if things go a bit wonky with so many drivers, driver types, etc. going on in modern IEMs. Maybe the wavelengths are so large compared to the size of IEMs and ear canals that phase interactions are irrelevant, but I haven't seen anything other than educated conjectures. Again, I'm no expert, so I'm just guessing. Every attempted search I've done just leads back to random posts on Head-Fi.

 

[o0genesis0o]

I've seen answers going in all directions. For sure, the consensus seems to be that single-driver headphones/earphones are minimum-phase devices, but I can't help but wonder if things go a bit wonky with so many drivers, driver types, etc. going on in modern IEMs. Maybe the wavelengths are so large compared to the size of IEMs and ear canals that phase interactions are irrelevant, but I haven't seen anything other than educated conjectures. Again, I'm no expert, so I'm just guessing. Every attempted search I've done just leads back to random posts on Head-Fi.

Many manufacturers work hard to align the phase of multi driver IEM. However, when I watch the old materials from Jerry Harvey, by aligning the phase, he overcomes the gaps and rolls off in frequency response, so, I don't know, it seems everything traces back to FR

Getting the data you mentioned is not hard with REW, btw. It gets everything with just one sine sweep. Can share some in the next review.

 

[Hypops]

Many manufacturers work hard to align the phase of multi driver IEM. However, when I watch the old materials from Jerry Harvey, by aligning the phase, he overcomes the gaps and rolls off in frequency response, so, I don't know, it seems everything traces back to FR

The question then is what happens when you start applying minimum-phase EQ filters to that nicely aligned phase response...

Getting the data you mentioned is not hard with REW, btw. It gets everything with just one sine sweep. Can share some in the next review.

I figured as much. I'm so used to seeing all of this data for loudspeaker measurements (and in some headphone measurements), so I've always wondered why IEM folks fixate so much on just the FFT graph. At this point, there's a whole feedback loop since now IEM listeners just want FR graphs, so reviewers keep producing them, creating more demand from listeners, and so on, and so on. But there may be a good reason for it.

 

[KinGensai]

https://reference-audio-analyzer.pro/en/report/hp/qdc-anole-vx.php

https://www.audiosciencereview.com/forum/index.php?threads/7hz-x-crinacle-zero-2-iem-review.50534/

It's out there, just not much of it. IEMs are a PITA to measure.

 

[Hypops]

https://reference-audio-analyzer.pro/en/report/hp/qdc-anole-vx.php

https://www.audiosciencereview.com/forum/index.php?threads/7hz-x-crinacle-zero-2-iem-review.50534/

It's out there, just not much of it. IEMs are a PITA to measure.

Exactly and exactly.

Really interesting to see an impulse response for a multi-driver IEM. Seeing more of them would help contextualize what “good” looks like. And phase may indeed be a non-issue if these two are at all representative of all multi-driver IEMs

 

[Hypops]

Now that I've recently added the Symphonium Meteor to my collection, I'm starting to think that my issues with EQing IEMs are actually specific to all-BA sets. While it's relatively hard to come by phase measurements for many IEMs (let alone, all-BA sets), those that I have found have impedance and phase graphs that look quite different from other sets, including hybrids/tribrids. This has me suspecting that multi-BA--or possibly just all-BA--sets are susceptible to phase smear when using (minimum-phase) EQ.

Admittedly, a sample size of two IEMs is hardly definitive. Without a measurement rig of my own or widespread impedance and phase (or even distortion) plots of many IEMs, I'll just have to rely on my ears. And my ears are telling me not to EQ either of my (all-BA) sets. I'll have to wait until I get another IEM with a different driver makeup to compare.

Curious if others have noticed a difference when EQing all-BA sets vs other driver combinations?