IEM insertion depth has measurable effects on...frequency response.Maybe regardless of FR it really matters a lot how far the things are stuck down your lugholes?
IEM insertion depth has measurable effects on...frequency response.Maybe regardless of FR it really matters a lot how far the things are stuck down your lugholes?
Is that measured on a rigid and unchanging electro-mechanical device, or by some fantastic new method of measuring how the typical brain responds to stimuli? Which may vary from one individual to another?IEM insertion depth has measurable effects on...frequency response.
And another thing (sorry, am on the ale): If insertion depth is genuinely significant then all the measurements which ignore it become less valid. It means we cannot compare an Etymotic with a Moondrop with a Shure. All comparisons of IEMs with different insertion depths are misleading.IEM insertion depth has measurable effects on...frequency response.
Not if you start to gradually ignore the response errors above 6 kHz as I do.A
And another thing (sorry, am on the ale): If insertion depth is genuinely significant then all the measurements which ignore it become less valid. It means we cannot compare an Etymotic with a Moondrop with a Shure. All comparisons of IEMs with different insertion depths are misleading.
Can you expand on that and clarify?Not if you start to gradually ignore the response errors above 6 kHz as I do.
This might help.Can you expand on that and clarify?
Not really, as I don't see how test results published here acknowledge or reflect that.This might help.
![]()
What (physically) causes the 7.5-8.5kHz peak in human hearing with IEMs and why it's important...
I'm sick of people saying "it's a resonance peak of the measurement rig" because it's not just that. I hear it clearly in every IEM with direct correlation to measurements and the peak's intensity. What varies of course is the exact position of the peak (7.5-9.5kHz range in 99% of cases...www.audiosciencereview.com
Maybe read the thread.Also, why should the points I raised be corralled into a narrow discussion about the response between 7.5 & 8.5 kHz?
There are too many vagaries from mounting the driver to differences in measurement rig. I don't believe in averaging to dial these out as that then doesn't represent any unit. You can see this in my measurements of left and right units even though the measurement rig is the same:Can you expand on that and clarify?
So would 3k be the limit of what different measurement rigs can reliably distinguish? After 3k what are we really seeing in measurements? Tendencies?There are too many vagaries from mounting the driver to differences in measurement rig. I don't believe in averaging to dial these out as that then doesn't represent any unit. You can see this in my measurements of left and right units even though the measurement rig is the same:
![]()
There is excellent tracking to 3 kHz but above that deviations start.
Also, the target is highly smoothed as otherwise it too would have those ups and downs. So comparisons against it are not totally valid either.
As such, I EQ to ear instead of relying on measurements alone. If the EQ in that region sounds better, then it is a good bet.
There is no hard limit. In your mind, gradually distrust the measurements from then on and focus on larger picture. Certainly above 10 kHz it is a crapshoot.So would 3k be the limit of what different measurement rigs can reliably distinguish? After 3k what are we really seeing in measurements? Tendencies?
My experience of FR measurements of stuff I actually own and use is that they very accurately describe bass response, and if it is overcooked and ruins the mids. And then there are the frequencies of the human voice. And that's what we notice, what we actually pay attention to and *hear*. The rest, unless truly awful, matters a lot less. I have EQ apps on my desktops and laptop and phone and smartphone which can import and apply AutoEQ profiles. But in the end I go back to just correcting the grossest errors (which is usually bass bleeding into mids) and not worry about minor stuff. The unusual thing about the Truthears is that the bass doesn't need correction, but the frequency range of the human voice does. They are decent enough and worth the price, and I like a lot how they engineered the crossover to be like a very well implemented low shelf filter, but I'm not convinced by them because that human voice range is not so well executed as the bass, and it matters at least as much.There is no hard limit. In your mind, gradually distrust the measurements from then on and focus on larger picture. Certainly above 10 kHz it is a crapshoot.
Ah, you EQ'd them to the same curve did you? My comment was more related to the fact that the measured frequency responses of the IEM's are different, which would explain particularly some of the points you mentioned in that post of yours I quoted earlier. (But for this post now, I don't really understand the relevance of your examples about twigs / shouting / sleeping / hair cells in your ear.)Even if they are eq'd to supposedly sound the same? I'm not a measuring apparatus. Maybe regardless of FR it really matters a lot how far the things are stuck down your lugholes? Hearing is an amazing thing. Someone right beside you can be practically yelling and you might not even notice. Or a twig might snap 20 metres away and you are instantly alert. I can't sleep well in a noisy place. I have a friend who can't sleep well outside of a busy city as he needs the noise. The hair cells in our cochlea don't always respond the exact same way to the same stimulus. We are not mechanical but bio-mechanical and our brains, where hearing actually happens, which control those hair cells, are not an open book. Everyone who is certain that they understand hearing is mistaken. This focus on FR measured, necessarily, on electro-mechanical devices is really useful for identifying differences or certain characteristics but using it as "one graph to rule them all" is a very narrow focus.
Very true and important lesson here.My experience of FR measurements of stuff I actually own and use is that they very accurately describe bass response, and if it is overcooked and ruins the mids. And then there are the frequencies of the human voice. And that's what we notice, what we actually pay attention to and *hear*. The rest, unless truly awful, matters a lot less.