This is good advice, just reminded me, the audiologist that made my custom attenuators (ear plugs) said the exact same thing years ago, and I totally forgot, wee flashback, ha. If I remember correctly, he also suggested you'd be less likely to push wax up into your ear doing it this way...
Yep, also there is a resonance at 5 kHz that needs to be killed and the bass re-shaped:
I think you'll struggle but have a look here https://www.audiosciencereview.com/...acos-ios-ipados-and-android.18450/post-602403
That's even more of a difference than GRAS's example, especially around 2-4 kHz, with the response below that of the RA0045. If this difference is similar for the Truthear Zero, its elevation above the Harman target in that frequency range could be even higher, with a peak ~3 dB over target instead of the 2 dB measured on here with the RA0402, which could further explain why so many are finding it 'shouty'.
It kinda did, but details weren't as prominent as stock and if I try to attenuate the mids/highs any further, I just get less details with tinny sound. I played around with the EQ for hours but I was going nowhere since everything I did just won't make it sound good.
Only if I use Comply tips but still, those shoutiness won't go away.
If you try to EQ based on music only, this will be very difficult in my experience. If you still got the motivation, I'd recommend using some online equal loudness tester. Tune in 1 kHz to your normal listening level, then start to tune up and down from there by comparing each frequency to the 1 kHz reference and adjusting the EQ step by step.
You shouldn't EQ frequency response by equal loudness:
I used a tone sweep and confirmed with dense, full range music.
Disagree. Tones first, music second.
High frequency details won't be as prominent because they aren't supposed to jump out of the frequency band like the stock tuning. Does it sound like neutral speakers in a room? What's your reference?
I do feel that my EQ could use another -0.5 dB from 2.5 to 5 kHz. I might try that later. Otherwise I'm not sure what else to do.
Well, we can go off Amir's measurement though instead of Crinacle's, if we think Crinacle's is not reflecting the most accurate picture. Not been following all the in's & out's of your conversation, but Amir's are done on a bonafide GRAS.
Still not using the same coupler (the GRAS RA0402) as Harman used though (the RA0045), which as seen in the post by @staticV3 you quoted above, can produce significant differences in the measurements. The point is neither of them use the coupler Harman used when developing the in-ear target (Crinacle using a fake RA0045 clone), so measurements from both will result in a less accurate approximation of the Truthear's (or any IEM's) deviation from the Harman target than measurements using a (genuine) RA0045, as Oratory does.
He's not only saying that:
So he's saying they're only maybe useful in terms of judging transducer frequency response within a frequency range that is flat on your equal loudness contour (otherwise you would not be able to differentiate between a change in loudness due to a change in your ear's sensitivity with frequency, and a change in loudness due to the frequency response of the transducer):
There's a perceptual difference between listening to a sine sweep of one frequency after another, which we do not intuitively know how it should sound because no natural sounds are anything like this, and we cannot differentiate between dips in our equal loudness contour and transducer resonances this way; and broadband sound like music composed of many different tones simultaneously, which we do know how it should sound, and for which our brain intuitively 'filters out' our equal loudness contour when listening for how 'correct' it sounds.Also, his comment about how using a tone sweep would lead to corrections that look like an equal loudness curve, so use music instead - what do you think music is? The result would be the same. It all has to pass through our ears.
