• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Has the meta on IEMs moved away from Harman's "pinna gain"?

Has the meta on IEMs moved away from Harman IE and "pinna gain"?

  • Yes

    Votes: 17 51.5%
  • No

    Votes: 16 48.5%

  • Total voters
    33
That means you don't like JM-1 with 10dB tilt. Nothing strange, JM-1 is just a starting point to add your preference adjustments on top of.

You are right I didn't like it in the past but I think there are some inconsistencies that I just encounter, I tried with the Zero:2 and like the result with both tools but with the Zero: RED was different, really good result with the Hangout tool and with only 3 filters and too bassy with PW squig and measurements.
Also the imported Trio measurements worked really well with the Hangout tool and gave me somewhat similar results to the EQed Zero: RED.

I think that all targets and measurements should be interchangeable between sites, is what the 5128 promised, more reliable and accurate measurements it should not be dependent on individual couplers (it has been done for a long time with the 711 with good results).

Here are some inconsistencies in measuring or displaying the measurements.

The Zero: RED from 1kHz and on is close but below has too much difference that when EQed even with the correct tool is going to sound different.
graph(10).png
graph(15).png



The Zero:2 is much more close and requires more or less the same EQ from both tools.
graph(17).png
graph(13).png


Also using imported measurements with the Hangout tool gives better results (so far tried 2 IEMs) I think is the way the target is implemented but I don't really know.

EDIT.
After some more experimenting with well documented IEMs like the Zero:2, Zero: RED and Chu II, I'm getting different sounding EQ form different sites, maybe I prefer a more Harman-like sound because I like the results form the Hangout site by a large margin.

Something is wrong, either the Hangout JM-1 tilt is miss-calculated or the JM-1 target with -1 tilt is. Also AutoEQ with Harman filters is the same (with a difference of around 2dB bass shelf) as the Hangout with Harman filters applied.

graph(18).png
graph(19).png

I personally trust AutoEq and Crinacle more with how the target is implemented also the Hangout site offers a lot more flexibility to tailor the target to preference.

All that said I might be completely wrong in all my assumptions but as a user I use the sites the same way the majority do and is something is counterintuitive it should be pointed out.
 
Last edited:
That means you don't like JM-1 with 10dB tilt. Nothing strange, JM-1 is just a starting point to add your preference adjustments on top of.
Could also be a consequence of the inaccurate acoustic impedance of 711 couplers in the lows.
 
If possible, it's always better to use an actual 5128 measurement for EQ, all of the "Δ" (Delta) versions of JM-1 on 711 are helpful when a 5128 measurement is unavailable, but a static compensation from the 5128 to 711 can only help so much.

FWIW, I think a 10dB tilt is too much tilt, I've been enjoying a tilt in the range of -0.6dB to -0.8dB per octave plus 4-5dB of bass, but adjust to taste. I shared this on reddit, but this filter might be helpful to anyone looking to (very crudely) attempt to account for HRTF variation.
 

Attachments

  • Adjustment Filters.png
    Adjustment Filters.png
    164.7 KB · Views: 76
If possible, it's always better to use an actual 5128 measurement for EQ, all of the "Δ" (Delta) versions of JM-1 on 711 are helpful when a 5128 measurement is unavailable, but a static compensation from the 5128 to 711 can only help so much.

FWIW, I think a 10dB tilt is too much tilt, I've been enjoying a tilt in the range of -0.6dB to -0.8dB per octave plus 4-5dB of bass, but adjust to taste. I shared this on reddit, but this filter might be helpful to anyone looking to (very crudely) attempt to account for HRTF variation.

Welcome! Just out of interest, do you have a graph of Harman 2019v2 compensated to 5128 (so the reverse of the above) overlayed on JM1? Thanks
 
If possible, it's always better to use an actual 5128 measurement for EQ, all of the "Δ" (Delta) versions of JM-1 on 711 are helpful when a 5128 measurement is unavailable, but a static compensation from the 5128 to 711 can only help so much.

FWIW, I think a 10dB tilt is too much tilt, I've been enjoying a tilt in the range of -0.6dB to -0.8dB per octave plus 4-5dB of bass, but adjust to taste. I shared this on reddit, but this filter might be helpful to anyone looking to (very crudely) attempt to account for HRTF variation.
Welcome Joel - loved your video - very informative. I have a Kiwiears KE4 arriving next week so looking forward to properly comparing with a few of my favourite IEM's.
 
I have a Kiwiears KE4 arriving next week so looking forward to properly comparing with a few of my favourite IEM's.
Me too . The reviews on comfort and fit sold me .
 
Welcome! Just out of interest, do you have a graph of Harman 2019v2 compensated to 5128 (so the reverse of the above) overlayed on JM1? Thanks

Here you go. I also included Harman IE 2019v2 with JM-1 DF as the compensation, so that we're just seeing the "preference adjustment" of Harman IE against a human-like HRTF. I think it's a helpful in illustrating just how weird and unnatural the Harman IE target really is.

And, just so we're clear, the preference adjustment I've applied here to JM-1 is not my "official target" by any means. It's an adjustment I personally enjoy, and is consistent with the general trends we see in the literature, but I really encourage people to experiment and adjust to their own taste. We'd need to do proper method of adjustment testing to dial in the tilt and bass that is preferred on average.
 

Attachments

  • JM-1 DF with Preference Adjustment vs. Harman IE.png
    JM-1 DF with Preference Adjustment vs. Harman IE.png
    120.2 KB · Views: 102
  • Harman IE, JM-1 DF Compensated.png
    Harman IE, JM-1 DF Compensated.png
    85.2 KB · Views: 93
Here you go. I also included Harman IE 2019v2 with JM-1 DF as the compensation, so that we're just seeing the "preference adjustment" of Harman IE against a human-like HRTF. I think it's a helpful in illustrating just how weird and unnatural the Harman IE target really is.

And, just so we're clear, the preference adjustment I've applied here to JM-1 is not my "official target" by any means. It's an adjustment I personally enjoy, and is consistent with the general trends we see in the literature, but I really encourage people to experiment and adjust to their own taste. We'd need to do proper method of adjustment testing to dial in the tilt and bass that is preferred on average.
Many thanks !
 
Here you go. I also included Harman IE 2019v2 with JM-1 DF as the compensation, so that we're just seeing the "preference adjustment" of Harman IE against a human-like HRTF. I think it's a helpful in illustrating just how weird and unnatural the Harman IE target really is.

And, just so we're clear, the preference adjustment I've applied here to JM-1 is not my "official target" by any means. It's an adjustment I personally enjoy, and is consistent with the general trends we see in the literature, but I really encourage people to experiment and adjust to their own taste. We'd need to do proper method of adjustment testing to dial in the tilt and bass that is preferred on average.

Hello. I'm a fan of your work and your influence in the field, as you may have noticed. Do you have anything to comment on OE targets? Thank you.
 
Mine arrived today - it's excellent very comfortable and so far I am super impressed with the sound. I do hope @amirm gets one soon, though I suspect its differences from Harman 2019 IE target will be controversial here.
Mine arrived . Comfort and fit are great. The sound doesn't excite me though . Nothing bad , just a 'safe' tuning that I'm very slightly tweaking with eq.
 
Mine arrived . Comfort and fit are great. The sound doesn't excite me though . Nothing bad , just a 'safe' tuning that I'm very slightly tweaking with eq.
I mostly like it - a little too much bass for my liking - but it is very safe as a new population average curve should be IMO
 
Topping Hane seems like a great iem to someone test their preferences. Lot's of switches.
 
Another JM-1 based IEM is being worked on, codename Project Reference. This time with a moderate 4dB bass boost like Joel recommends(very close also to Soundguys target, which is part or S. Olives preference research). For now, I believe Truthear Hexa is the next closest measured IEM unless I'm overlooking something.

1727998459470.png
 
Last edited:
Topping Hane seems like a great iem to someone test their preferences. Lot's of switches.
And I would guess that the available options are closer to the preference of the masses than the IEM Harman Target (more lower mids, less lower treble, more upper treble).

Bildschirmfoto_2024-10-04_01-55-49.png

 
I must say I only now (after an intense 5 IEMs bought and 2 returned couple of months :D ) understood what the Harman critique for IEMs means.

I am an audio professional and have been dealing with sound in rooms for years, always had a few headphones with me for when I'm away from them, and a year or so ago stumbled on AutoEq and the idea of "tuning" headphones for producing. The Sundara has accompanied me since 2019, and just completely opened up in a great way, and after a few more deliveries and the idiosyncrasies thereof I realized that I needed more definition in the higher mids and highs to "control" my control room (after working longer than I should on projects) and I landed on an LCD-2c that felt "right" after 3 mere EQ points and roughly 20mins.

Not happy with that, since I was messing with my monitoring anyway I just started diving into Chi-fi, and have had a few impressions as well. The idea that I could get a lot of sound on the go for much less than over-ears and studio monitors was too good to pass on. And it seems very true, it just blows my mind how far 80€ can get you. This is the brief journey (always cross-checking all my references with the full range tuned room setup and the Audeze):

1) 7Hz Zero 2 is just a great "fun" balance, that to me is almost a benchmark in enhanced bass "neutrality", I realized I can make decisions with it after accepting that amount of bass, but that's the ceiling of low-end that is still "usable" for full-range inteligibility. The weird thing to me is that even with a lot of bass, the mids are there. The only downside is a lack of definition per se, it cannot keep up with the Audeze (40 times the price) or my room setup (about 160 times the price for the speakers alone :D ). But it is super valid as a tool for mix/master check to see how everything sounds in IEMs. Not parting with that.

2) Truthear Zero Reds: this was to me a trainwreck. Overwhelming bass, everything buried underneath it. I could get no info or insight on anything the moment any low-end came into play. Kept it for 2 weeks. Happy to get rid of.

3) Wan'er Studio: a little thin, but super in your face without splashy or zingy highs, and lean bass, but with subs! I got it for 15€, and realized I can use that for tracking or playing or for when musicians come to record, the mids feel "correct", and the overall sound reminds me somewhat of the scooped NS10 sound and its paper cones in a less fatiguing way, 100% usable, not parting with that.

4) Simgot EW300: impressive for a first listen, incredible low-end, but I realized the 7Hz Zero was much more insightful. The Simgots kept up with the lows of the Audeze, had similar low-end presence with obviously worse transient response, but I would be able to understand how they relate, except that EVERYTHING else was "off". The mids were just gone, with the presence area still existent (so that was better than the Reds), but that was it. And the highs were just ridiculous, high percussion just turned into paper zings, with high-highs boosted into tinnitus inducing chaos. Off they went. If that is their "tame version of a house sound", I don't want to know what their not-tame version is because that just feels like what a wavy mess on a graph sounds like. "Cut here, boost that, cut here, boost that, cut cut boost boost boost", and you're left with a super unnatural mess.

5) Truthear Hexa: I decided to overcome my trauma from the Reds, and ordered them when returning the EW300s. As soon as I started listening to them with their foam-tips, I remembered the first time I had HD600s on my head. Same lean lows, with more subs, and very clear mids, present highs. Very impressive, for a 10th of what the Audeze cost me. I just felt mid-bass was lacking, but that roller-coaster on the spectrum from the Simgots was nowhere to be heard here, lows aren't as candy-like, but everything else provides me enough insight and information to complement my setup for sure. I raised around 80Hz just 0.5db and it already felt "right" in an impressive way. Then I opened auto-eq and saw the target, some absurdity aside, 1db up at Hz76 and then bringing 200Hz down by 0.5db and up again at 5.8Hz by 0.5db, and it fits like a glove with the speakers and Audeze. I don't think I am parting with that.

Just thought I'd share, because what the "spot on" Harman chasing feeling brings to over-ears, is quite not the mark with IEMs, and around a month ago I thought this was just the "in" crowd trying to make weighty statements. But the Hexa with those 2 points, the Zero 2 and the Wan'er beg to differ, they all feel "closer" to neutral than every other IEM I tried (I am omitting wireless stuff, that I own and owned and others).
 
#TLDR The Hexa, the 7Hz Zero 2 and the Wan'er Studio are all closer to what I perceive as neutral and feel very equivalent (with varying deviations) from Harman for over-ears (against Sundara and LCD-2c tuned spot over it). So I am positive a new "target" is needed. Maybe that's the new "meta"?

edit: what I meant, the new "meta" could bring a similar feeling to IEMs as what Harman does to over-ear? Because trying to equate both makes for horribly bassplosive/dull/shouty/zingy IEMs... If I input what AutoEq tells me on the Hexa I have to find the nearest garbage bin. But WAY LESS of that seems coherent and inline to what I changed by ear before even loading that site, and gives me a very similar feeling to over-ear Harmans...
 
Last edited:
Another JM-1 based IEM is being worked on, codename Project Reference. This time with a moderate 4dB bass boost like Joel recommends(very close also to Soundguys target, which is part or S. Olives preference research). For now, I believe Truthear Hexa is the next closest measured IEM unless I'm overlooking something.

View attachment 396447

I still see to much 3K energy to sound neutral to me. But it does graph good.
 
Back
Top Bottom