• 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!

The headphone show interview with special guest Dr. Sean Olive from Harman International.

MayaTlab

Addicted to Fun and Learning
Joined
Aug 15, 2020
Messages
924
Likes
1,512
Thanks to everyone involved.

@Mad_Economist, the article mentioned at the 2:11 mark, is it this one ?

The article below also re-uses the "free air equivalent coupling" and "PDR" concepts :
 

Mad_Economist

Addicted to Fun and Learning
Audio Company
Joined
Nov 29, 2017
Messages
543
Likes
1,618
Thanks to everyone involved.

@Mad_Economist, the article mentioned at the 2:11 mark, is it this one ?

The article below also re-uses the "free air equivalent coupling" and "PDR" concepts :
It was either that one or " design criteria for headphones" - pardon my poor memory.

Thanks for the link! I'll give it a read
 

ADU

Major Contributor
Joined
Jul 29, 2021
Messages
1,584
Likes
1,086
At one point toward the end of this discussion, the question of whether the Harman target (and headphone response curves similar to it) should go more up or down in the 1.5 to 2 kHz range of the upper mids came up.

The answer to this may seem fairly obvious for anyone who's spent a fair amount of time comparing the measurements of different headphones in that range. I think there are several factors which could potentially influence this though. And that it might tend to vary a bit depending on the listener's age and hearing, and the recording, and also the graphs that one is looking at. Because some people have more hearing loss in the upper midrange frequencies. And some gear used for mastering music content can also have more or less of a dip in that 2k range (due to differences in their cross-over design and directivity in that range). And because the response and resonances in the equipment used for measuring the headphones might also vary a bit in that range, from measurement rig to measurement rig.

For me personally though, it would definitely be down, rather than up on Oratory's graphs... at least for listening purposes anyway. And perhaps a little more along the lines of some of the HPs shown here...

UPPERMIDDIP.jpg


For sound mastering and production though, I might possibly prefer to use a slightly different response there.

The preferences in this range will probably depend to an extent on some of the other variables already mentioned though, such as a listener's age, midrange hearing loss, the type of content they listen to and so forth. Some longtime audiophiles seem to prefer a response that's a little more forward, and perhaps somewhat closer to the Harman response curve in that range. And feel that it gives their music a bit more presence or liveliness, for example. And they may have a legitimate point on that.

The content I listen to is probably not mastered as well though. So it often seems to get rather harsh-sounding to my ears when the level in that range is more forward.

Something like this might be a bit too much of a dip though in that area, imho...

NEUMANNNDH20.jpg


I think this is (or maybe was?) one of Z's favorite HPs though. So your mileage could possibly vary on this. And maybe somethin like the response on the NDH-20 could also be worth a try. (?)
 
Last edited:

Jimbob54

Grand Contributor
Forum Donor
Joined
Oct 25, 2019
Messages
11,066
Likes
14,697
Good chat. Interesting when they hit soundstage. Dr O is like Solderdude, doesn't "hear" it. Explains why no research.

Love @Mad_Economist casually rocking the HD800.
 

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
44,368
Likes
234,394
Location
Seattle Area
I don't hear any soundstage either. But I hear a localized, separation of instruments that I consider "fun" when it is at the extreme (e.g. on 800s). And dull when it is not there at all.
 

GaryH

Major Contributor
Joined
May 12, 2021
Messages
1,348
Likes
1,804
At one point toward the end of this discussion, the question of whether the Harman target (and headphone response curves similar to it) should go more up or down in the 1.5 to 2 kHz range of the upper mids came up.

Note they were specifically discussing frequency response's influence on perceived soundstage. In terms of tonality though, the research says the Harman target is the most accurate / neutral / preferred for the majority.

It seems most listeners, like you, do find that a reduction in response in the 1-2kHz range often increases either the width or depth of the perceived soundstage, partly probably because it's mostly the center content of a stereo mix that contains those frequencies - reduce those, and the center is either perceived to be pushed back (more depth) or made more diffuse (more width), which can make for a more speaker-like sound (depending on the recording), as evidenced by the headphones you've mentioned that are known for their soundstage and exhibit this dip.

This will be at the expense of tonality though, so there's a trade-off between soundstage and tonality. Unless you have well-inplemented crossfeed DSP that is, which mimicks the natural interaural time and amplitude difference (ITD and IAD) heard between the two channels of stereo speakers. The AKG N90Q's 'Studio' active DSP mode does just this, attempting to produce more of a studio speaker-like response (and does quite well at it too). Tyll of Innerfiidelity measured the N90Q's response with and without this mode engaged (as well as a 'Surround' mode which adds simulated room reflections), and it did show a familiar dip in response in the upper mids with the Studio mode engaged:

AKG_N90Q_Graph_SpatialFR.jpg


Tyll surmised:
I was not able to get confirmation of this, but I assume the notch at 700Hz is the IAD showing the acoustic shadow of the head, and the smaller notch at 1.6kHz is the comb filter effect of the ITD signal time delay.

And noted:
I've heard a lot of similar crossfeed filters over the years—including a lot of time tweaking and tuning the HeadRoom crossfeed filters—and they all seem to create too much tonal change. I have to say I think the N90Q does it very well; I hear very little tonal change.

He caveats to the graph above:
Bear in mind this test is done with a mono sweep to both ears simultaneously and will draw out the maximum amount of tonal change as the crossfeed signals combine. A normal stereo signal will usually have much more difference between left and right, only the mono (same on both channels) signal will be effected as shown in the plot. In other words, the tonal change will not be nearly as much as shown with normal music being played.

Despite this caveat, I think this is more evidence supporting the upper midrange dip => increased perceived soundstage hypothesis for passive headphones, albeit being an inferior blunt FR-tool method that is at the detriment of tonality, in comparison to proper active crossfeed DSP that takes into account both ITD and IAD.
 

ADU

Major Contributor
Joined
Jul 29, 2021
Messages
1,584
Likes
1,086
Very interesting post, GaryH. Thank you.

You are correct that this subject of the upper midrange dip was brought up in the context of soundstage. I deliberately avoided referencing that in my previous comments though, because I think it's a potentially much more complicated subject that could encompass a number of other factors than just FR. And I think Resolve, Mad_Economist and Dr. Olive may have also touched on a few of those in the above livestream, including the openness of the headphone, and also the angle and distance of the drivers from ear canal. I've also speculated that it could possibly be related to the impulse response, and maybe also distortion as well. Which might possibly effect the clarity or accuracy of spacial cues in a recording.

I think both the ambient noise (As Dr. Olive metioned) and maybe also the cross-feed in the higher frequencies in open headphones could possibly play some role too. I still have a little bit of trouble separating the concepts of imaging and soundstage though. And at a little bit of a disadvantage on the latter subject, because I've never owned any decent open HPs, other than some lower cost on-ear Koss headphones with foam earpads that were open on the sides. Those did seem to have a more open and a bit more "3D" sound in some ways than many of the closed studio type HPs I've used since then. And they were quite impressive for their price imo. (Though their extension in the bass left something to be desired, since there was no real pressure inside the cups. And the drivers were so small.)

I think this is the first time that I've heard the headshadowing theory though, which is also interesting. But still believe there's probably more to this subject than just applying a different EQ to a pair of headphones.

As a general rule, I am resistant to more exotic explanations where there are more prosaic and tangible ones available. So I mostly attribute my preferences for a more withdrawn level in that 1.5 to 2k range to some of the more mundane explanations mentioned in my previous post above, and the fact that I might still have somewhat greater sensitivity than some other listeners in my age range in that area. And also to proper decoding of recordings mastered on poorer quality gear, with more pronounced cross-over and directivity issues. When I get a chance, I'll try to post a few more examples to (hopefully) illustrate some of this a bit better.
 
Last edited:

ADU

Major Contributor
Joined
Jul 29, 2021
Messages
1,584
Likes
1,086
Since you invoked Tyll H. It should probably also be mentioned that as general rule, he seemed to like headphones with a bit more of a rise in that upper midrange area, perhaps a bit more comparable to the Harman curve... At least up to a point anyway.

Here he is with Bob Katz discussing a possible EQ profile to better achieve something like that on the Sennheiser HD800, which is one of the HPs that is fairly well-known for having a bit more of a dip there. It should perhaps also be noted that before the Harman curve came along, alot of headphones had a somewhat flatter and lower response in that upper midrange area. In fact it was probably more the norm up to that point. And that rise really only became a thing that alot of headphone manufacturers wanted to emulate after Tyll and the Harman curve began to champion that idea.

If I remember right though, I think Tyll himself speculated that his preference for a brighter sound in that range might've been somewhat related to his own progressive hearing loss in that range. You probably can't review speakers or headphones for too long as a living without it effecting your hearing in some ways.


Alot of folks were already experimenting with EQ-ing headphones, in various ways, when this video came out btw. But I believe Tyll was one of the first headphone reviewers to think about and include some suggestions on how to do this in some of his headphone reviews. (Which is now, of course, a much more commonplace type of thing.)
 
Last edited:

ADU

Major Contributor
Joined
Jul 29, 2021
Messages
1,584
Likes
1,086
Fwiw, I was pretty much against this idea of a higher slope or level in the upper mids almost from the beginning, after trying it out on a few of my own headphones at the time. Because it has never sounded right with the type of content that I listen to.

I listen mostly to heavily compessed pop, hip-hop, and other more mainstream things of that sort though, which often has its dynamic range significantly compressed to boost its loudness. Versus the more lossless fullrange type of content that alot of other audiophiles or enthusiasts seem to enjoy.

I'm not sure if its really the compression which is to blame though. Or just inferior engineering and equipment used for producing alot of the content.
 
Last edited:

thewas

Master Contributor
Forum Donor
Joined
Jan 15, 2020
Messages
6,747
Likes
16,186
It seems most listeners, like you, do find that a reduction in response in the 1-2kHz range often increases either the width or depth of the perceived soundstage, partly probably because it's mostly the center content of a stereo mix that contains those frequencies - reduce those, and the center is either perceived to be pushed back (more depth) or made more diffuse (more width), which can make for a more speaker-like sound (depending on the recording), as evidenced by the headphones you've mentioned that are known for their soundstage and exhibit this dip.
According to the directional frequency bands research of Blauert which started in 1969(!) a reduction around 1 Khz makes the sound more direct and frontal:

Directional-bands-in-the-median-plane-The-angles-0-90-and-180-indicate-front-up_W640.jpg

frontal sound = present in the sound, close, direct, foreground - This can be achieved by raising the frequencies 300 to 400 Hz and 3 to 4 kHz and by lowering frequencies by 1 kHz.
rear sound (and from top) = diffuse, distant and spatial in sound - This can be achieved by raising the frequencies by 1 kHz.

 

ADU

Major Contributor
Joined
Jul 29, 2021
Messages
1,584
Likes
1,086
Fwiw, this PDF from 2011 also shows how free-field in-ear measurements will vary in their tonal balance based on the angle or direction of the source to a measurement rig. These measurements were done with three different rigs by GRAS, B&K and Head Acoustics.


So yes, it's logical that the perceived timbre of a sound source will vary with the angle to the listener. Because the sound is interacting differently with shapes of the head, body, and ear before reaching the eardrum. And I would not be surprised if some sound engineers use these differences to enhance the sense of space and direction in their mixes.

Although I have tried using three different diffuse field curves to compensate Oratory1990's raw plots, I haven't tried the one shown in this PDF yet. It looks like this was done with a different rig, and maybe also pinna though, the 45BM.

I should have mentioned this before, but a more accurate FR on a pair headphones might also do a better job of relaying some of those timbral spatial and directional cues (if any) to the listener. Along with perhaps lower distortion, and a clean, precise attack-decay/impulse response. This is what I mean by good "decoding" of the content. And it probably encompasses all of the above, to some extent.

I think it might be possible to still have a good sense of soundstage on a pair of headphones that is not as tonally accurate though. If it has some of the other qualities or characteristics mentioned above, including an open design, angled drivers, and so forth.
 
Last edited:

Dealux

Active Member
Joined
Aug 31, 2019
Messages
175
Likes
195
Location
Arad, Romania
I don't hear any soundstage either. But I hear a localized, separation of instruments that I consider "fun" when it is at the extreme (e.g. on 800s). And dull when it is not there at all.
That's confusing as hell to me. It's like finding out that some people have no internal monologue.

There is a definite difference in terms of stage width between headphones but also differences in depth and sharpness/clarity of the images (i.e. instruments).
 

Dealux

Active Member
Joined
Aug 31, 2019
Messages
175
Likes
195
Location
Arad, Romania
What kind of difference?
The extreme case is comparing an IEM like the ER4 to the likes of the DT 990, K612, or the HD800. The ER4 sounds quite narrow (in your head) by default but can sound wide in some situations like binaural. Just not as wide as most full size headphones.

For full size open backs, the HD800 probably sits at the top as the widest though with odd imaging (to my ears, too spread out) and something like the HD600 is narrower and more in your head sounding. I'm well aware this is likely due to tuning differences. On some headphones it seems like instruments can sound as if they're coming from outside the earcup. However, if you really focus on it you can kinda tell it's a sort of illusion.
 

maverickronin

Major Contributor
Forum Donor
Joined
Jul 19, 2018
Messages
2,527
Likes
3,308
Location
Midwest, USA
That's confusing as hell to me. It's like finding out that some people have no internal monologue.

I'm going to second this as well. I have a hard time understanding how we could perceive something so differently.

Keeping the program material constant, different headphones have a different areas of space within or around one's head in which apparent sources can be located. The size and shape of this area can vary enormously between headphones. Do you not perceive it this way @amirm?

IME, part of it is adherence to a generic target curve with a decent amount of pinna gain, like Harman, Etymotic, or DF. The more important factor though is that the physical relation of the drivers to one's ears should allow more of one's individual pinna transfer function to be added to the final FR reaching the eardrum. This is usually achieved with some combination of making the drivers large, at an angle to the pinna, or more distant from the pinna.
 
Top Bottom