Skip to main content

Notice

Please note that most of the software linked on this forum is likely to be safe to use. If you are unsure, feel free to ask in the relevant topics, or send a private message to an administrator or moderator. To help curb the problems of false positives, or in the event that you do find actual malware, you can contribute through the article linked here.
Topic: Soundstage and headphones (Read 33715 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

Soundstage and headphones

I've been helping my teenage nephew choose some headphones.  He's been doing some research and has picked up the concept of "soundstage". (The headphone enthusiasts sometimes call it "headstage").  It seems that soundstage is one of the things the headphone enthusiasts seek. 

As far as I can see "soundstage" is a term the headphones enthusiasts have taken from loudspeaker enthusiasts who use it to mean the placement of instruments when listening to stereo sound.

Have I misunderstood something because I would guess this "soundstage" is a function of the recording and any subsequent digital encoding rather than a function of the speakers.  In other words, the headphones only make a very minor contribution to the soundstage.

In fact, I can understand two loudspeakers needing to be matched in order to match phase but surely this doesn't even apply to headphones?

I am confused!  Can anyone kindly explain where the mix up lies.




Soundstage and headphones

Reply #1
*Grabs popcorn*

Soundstage and headphones

Reply #2
As far as I can see "soundstage" is a term the headphones enthusiasts have taken from loudspeaker enthusiasts who use it to mean the placement of instruments when listening to stereo sound.


Well, I  like this definition of soundstage I  found at about.com (I'll explain later why):

Quote
Definition: Soundstage is the ability to visualize the placement of musical instruments and vocalists in a music recording. A good soundstage also allows the listener to perceive the size and space of the performance venue in which the recording was made.

Source:
http://stereos.about.com/od/glossary/g/soundstage.htm

Actually it's the second part of  the definition that interests me. Depending of the headphone , you have the illusion , that the sound is taking more or less space around you.
I've experienced big differences between headphone models, which make me think that not everything is from the recording.

It seems that different factors have an impact on the perceived soundstage. These are some factors  I've noticed (not exhaustive) :
-  if the headphone is open or closed
-  how far are the drivers from the ears
-  the amount of treble (compared to others frequencies).

Soundstage and headphones

Reply #3
Soundstage is an illusion and I don't know of any way of measuring, quantifying, or specifying it.    With stereo speakers, you CAN reliably make the sound seem to come from near the center even though there is no center speaker.  You can also go far-left or partially-left, etc.    But other than that, positioning gets a little vague.  And in general, we aren't as good at "locating" a "phantom image" between the speakers as we think.*

With headphones, you can get something similar and some listeners seem to to experience a very good soundstage.  The big variable seems to be the listener (and the recording).  Some listeners get the impression the sound is coming from overhead or even from inside their head, etc. 

What I really suspect is that listener A might get a better soundstage from headphone A (or speaker A) and listener B might get a better soundstage from headphone B (or speaker B).    And in general, some listeners will get a better soundstage illusion than other listeners. 

You can get headphone amplifiers with a control to blend the left & right channels together.  Many headphone listeners feel this gives a more natural sound.  And of course, it does affect the "soundstage" (for better or worse depending on the recording, the listener, and the amount of blend).

Bottom line - I'd ignore what anybody says about soundstage and choose headphones that sound good to you and that are comfortable to you.

Every headphone (and speaker) sounds different.  Headphones are particularly difficult to measure, and manufacturers often fudge the specs (for both speakers & headphones).    And, different people have different tastes/preferences.    And once you get above a certain price point (maybe around $200 USD) you might find a $200 headphone that sounds better (to you) than anything more expensive.

Quote
In fact, I can understand two loudspeakers needing to be matched in order to match phase but surely this doesn't even apply to headphones?
Of course, the left & right speaker should be the same model and then the phase (and output-level and frequency response) will match.    And of course, the left & right headphones will also be identical.   

I've reversed the phase of one speaker before and I know what that does to the sound.  I've never tried that with headphones, but the result would be less dramatic because the left & right sound waves never combine.    I would not expect to hear any difference at all.

P.S.
Quote
I've been helping my teenage nephew choose some headphones. He's been doing some research and has picked up the concept of "soundstage".
90 percent of what you read from the 'audiophile community is total nonsense!  They have a whole vocabulary of words like "detail" that seem to mean something, but have absoultely no meaning in science or engineering.    There are basically three measures that affect sound qualty - Noise, distortion and frequency response.  (There are also timing & acoustical factors like reverb or echo.)  If you tell me you are loosing "detail" because the high frequencies are rolled-off, that's fine.

Ethan Winer has some good information on his website that can help your nephew with some of this audiophile nonsense.  There is also a link to a video of his AES presentation.   



* Moulton Labs has a good article about panning (placing a phantom image in the stereo field during production/mixing).    The author did some careful experiments and he was surprised how imprecise it is.    His main point is that audio engineers are wasting their time f they try to perfectly position every instrument across the soundstage.

Soundstage and headphones

Reply #4
What I really suspect is that listener A might get a better soundstage from headphone A (or speaker A) and listener B might get a better soundstage from headphone B (or speaker B).    And in general, some listeners will get a better soundstage illusion than other listeners.


There are consistent impressions from listeners to others, when comparing headphone models.
It would be hard to find someone that think , a senn hd25 have subjectively a wider soundstage than a senn hd800 (i.e usually it's the reverse).  I  could cite the akg 701 instead of the hd800 ; it's significantly cheaper, but I've  only heard the hd800.
I've also seen some people complaining that the soundstage of the hd800 sounds artificial (what sounds huge to some, just sounds fake to others).  I've been myself wondering if the manufacturer didn't use some "dirty" trick to inflate the soundstage of the hd800 ,  but it just seems that the drivers are positioned far from the ears, and are also "angled".


Soundstage and headphones

Reply #5
I've reversed the phase of one speaker before and I know what that does to the sound.  I've never tried that with headphones, but the result would be less dramatic because the left & right sound waves never combine.    I would not expect to hear any difference at all.

P.S.
Quote
I've been helping my teenage nephew choose some headphones. He's been doing some research and has picked up the concept of "soundstage".
90 percent of what you read from the 'audiophile community is total nonsense! 

I agree with most of your post except for the two things I've quoted.

A. It's not 90%, it's 95 to 99%. 

Furthermore, IMHO, the other forums where people talk about their non level matched, uncontrolled, sighted comparisons of audio products are worthless due to expectation bias, etc. Just because we all can agree that there really are audible differences between headphones doesn't mean we suddenly no longer have to guard against all the same common pitfalls we do in say a comparison of amplifiers. The same rules apply even though they are much more difficult to pull off.

B. I think you'll find relative phase reversal [not to be confused with absolute phase/polarity] quite easy to hear through headphones, even ones with nearly perfect isolation and no crosstalk (like ear canal blocking IEMs such as Etymotics), with the right material. Try the three test files here:

http://www.audiocheck.net/audiotests_polaritycheck.php

[Note: I don't fully agree with everything that author says, I'm just linking you to the test which I assume you can easily try out with headphones.]

Soundstage and headphones

Reply #6
I have tried lots of headphones and I haven't noticed a difference in 'soundstage' yet.

Soundstage and headphones

Reply #7
It's possible to significantly reduce the perceived left-right separation on HD518 by simply stuffing some cotton into the space behind the drivers. Doing this affects the frequency response significantly (less bass, more treble).
This leads me to believe that this "soundstage" effect is simply derived from amplitude/phase response of a headphone (in my case muffling reflections within the ear-cup).

 

Soundstage and headphones

Reply #8
I don't have anything to say specifically about headphones, but in the context of loudspeakers my personal experience suggests that soundstage is mostly a construct inside the listener's brain rather than some attribute of the actual sound field. In other words, how the listener happens to be feeling at the time influences how the sound field is perceived. (I'm talking here about the subtle aspects of height and depth - not the rudimentary soundstaging of a left/right image with well-defined centre, which is trivial to achieve with any competent system).

My empirical evidence for this belief? Just once in my life, I heard a solid, "holographic" soundstage. It was so vivid that it seemed I could get up and wander around between the musicians. It was a truly magical experience. I have since listened to the same recording again on the same system in the same room, but the experience was never replicated. Pretty much all of the external factors were the same (although I admit things like exact temperature and humidity wouldn't be the same), so it seems overwhelmingly likely that the difference was down to something in my head. (What's more, the source was a vinyl LP played through a pair of Linn Isobarik speakers - well known for their *lack* of pin-point imaging).

Soundstage and headphones

Reply #9
Just because we all can agree that there really are audible differences between headphones doesn't mean we suddenly no longer have to guard against all the same common pitfalls we do in say a comparison of amplifiers. The same rules apply even though they are much more difficult to pull off.


I understand your concerns. I'd still think that headphones can be differentiated regarding their soundstage. 
With these limits:
- Soundstage is often the least of concerns from people seeking "good" headphones  (a particular frequency response is usually the top priority). Lot of people just don't care.
- Since soundstage cannot be measured, it's difficult to assess its importance. However I invite people to listen models such like the hd25 II 1 and the hd800. If after that, they persist to say "perceived sounddtage depends only of the recording" , I'd be at least surprised.
- The definition of soundstage is relatively vague, and often listeners have their own idea of what soundstage means. In my particular case, I've explained that it's an illusion of "wideness" (i.e how much space the sound seems to take around you).

This leads me to believe that this "soundstage" effect is simply derived from amplitude/phase response of a headphone (in my case muffling reflections within the ear-cup).


Thanks for this pertinent comment. The "level of muffling" coming from  reflections inside cups is an idea worth to be explored. In the two examples I've cited, the  reflections are quite minimal with the hd800, while in comparison they are significant with the hd25 II 1. But imho not all  kind of "reflections" are disliked by people (leading to different "colorations" of the sound) , so it might be a bit more complex than that.

I just got also an idea, regarding some "soundstage experiments". I suggest you try to the vst sheppi, and get familiar with the "deep" and "wide" switches.
http://dallashodgson.info/articles/OpenAmbienceProject/
The impact of these switches is not always easy to notice , I'm not sure why (it might depend of the headphone used, and it could be psychological too...). Anyways, the deep switch, sounds a  bit like a subtle reverb . Hence something  subtly reverby,  can contribute to a subjective "deepness" in the sound. And there's the wide switch, that invert the phase between the left & right reflections, and that contributes to a subjective "wideness" (hence phase matters too).  Now , maybe we could find some similarities between how a headphone affect perceived soundstage, and that sheppi effect.

In other words, how the listener happens to be feeling at the time influences how the sound field is perceived.


I agree to some extent. I think, this could be compared to some optical illusion that can be sometimes be "reversed" depending of how you are feeling (this one for instance: http://en.wikipedia.org/wiki/File:Two_silh...white_vase.jpg ) .
There's something happening at the psychological level.

Soundstage and headphones

Reply #10
I don't perceive a soundstage with headphones.   It's not something I think about, and it doesn't detract from my enjoyment of music.  But given a choice I'll listen to speakers, and in fact I rarely listen to headphones.

So this morning, I decided to put on a pair of headphones* to see if I could analyze & describe the way I experience it.

My first impression is:  It sounds like I'm listening to headphones.  There are hard-panned sounds that clearly sound like they are coming directly from the "speaker" right next to my ear.  I guess this is why some listeners like a blend control on their headphone amp, but again this doesn't bother me... I like the fact that I can hear some sounds clearly from the left or right.

Most of the sound seems to be coming from somewhere else, but I can't pinpoint where it's coming from.  It does sound very nearby, within inches of my head.

Then I tried a mono file...  In theory mono should sound like it's coming from the center of the soundstage.    But, that's not what I "heard".  Again it sounded very close, but I wasn't really hearing the "little speakers" next to my ears.  It was sort-of like the sound was coming from all directions.  Or, more like the sound is coming from no particular direction.  Almost like when you have one ear plugged-up and you can't tell where sounds are coming from, but not unpleasant like that. 

What I'm hearing (in mono or stereo) is unlike anything I hear in the "real world".  I certainly wouldn't "forget" that I'm listening to headphones. 

There are Binaural Recordings made in a real acoustic space (i.e. a music hall) with microphones stuck in an artificial head.  These are supposed to sound very realistic (although unless you have head-tracking headphones, you are still missing the ability to turn your head to triangulate the location of a sound).  I've never heard a binaural recording, and that's the big problem...  None of my favorite music is available in that format.

With stereo speakers, I get "better" soundstage, but it's far from perfect.    Again, I can clearly hear hard-panned sounds coming directly from the left & right speakers (not very "realistic" either).  But if there is a singer in the center, I can't pinpoint the exact location.  Sure, it sounds "centered", I'm not locating the sound coming from two separate speakers (like it really is), but it's more like there's an array of speakers across most of the "stage" and the sound is coming from all of them.   




* I was listening to my Grado SR225 (open back) headphones.  These are my "best", "most natural" sounding headphones.  ...Which seems like a funny thing to say after I just said they don't sound like anything in the real world.

P.S.
The fact is, you can't really "pinpoint" sound location on a real live stage either!  Except, you can usually get a general direction and the see where he sound is coming from.    Once you see where the sound is coming from, your ears seem to better able to locate the sound... 

There are 2 situations - Purely acoustic music is normally performed in reverberant halls where the sound is bouncing all around you.    Your brain still gives you a direction where the sound comes from first,  but it's harder to pinpoint in a reverberant environment.  We are not always aware of the reverb unless you are thinking about it.  But, the "room sound" becomes vary apparent, even distracting, if you make a recording from your listening position.  When you play-back the recording and all of the sound, including the reverb, is coming from the same direction, it sounds unnatural.

And, most live music is mono from a PA system (with maybe some direct sound from the stage in smaller venues, especially drums).  Even with symphonies, there is "sound reinforcement".

Soundstage and headphones

Reply #11
I guess this is why some listeners like a blend control on their headphone amp, but again this doesn't bother me... I like the fact that I can hear some sounds clearly from the left or right.


For some old recordings, the stereo separation might seem excessive, and even annoying. In particular I've listened to the track "Within You Without You" from the Beatles, and the separation between the instrument played on the left and the instrument played on the right can be a bit overwhelming with headphones (or at least it's not so musical ).

When I listen to  music with headphone, most of the time I'm not interested to locate each sounds.
However I care,  if the sounds seems distant, or "on my face" ( I prefer when it's distant).

Quote
I've never heard a binaural recording, and that's the big problem..


There's a popular example of binaural recording example called "Virtual Barber Shop".
You can find it on youtube:
http://www.youtube.com/watch?v=8IXm6SuUigI

If you were talking of a regular music album, then I've listened to  the one titled  "Dr Chesky's Sensational, Fantastic, and Simply Amazing Binaural Sound Show!".  It's an interesting listening experience, however I do not wish that every albums sounds that way (it grabs too much attention).

Quote
We are not always aware of the reverb unless you are thinking about it.

Yes, but unconsciously  it  gives a clue regarding the size of the room.
With dsp, reverb with "long tails", emulates sound bouncing inside big halls/rooms.

Soundstage and headphones

Reply #12
To discount soundstage as "subjectivist rhetoric" is nonsense. Soundstage is the perception of depth and width (and sometimes height) of a recording. Some speakers do a great job of extending the sound field behind and outside of their physical placement. Others have a soundstage firmly anchored to the loudspeaker. Same with headphones, really great ones (usually open backed) externalize the recording so it sounds like parts of the recording are emanating from outside the headphones. Not the best example, but bose 901 speakers, when properly placed (NOT IN CORNERS) and using the included eq project a sound field that extends far beyond their actual location. High end speakers, properly placed can do the same. Unfortunately its not something that can be measured quantitatively so objectivists claim it doesn't exist.
For your nephew I recommend determining first if open backed would be appropriate for his purposes. If so he cant go wrong with sennheiser HD 650 or the like. If not, start with the regularly recommended closed audio technica athm50 or sennheiser HD380 pro.

Soundstage and headphones

Reply #13
Quote
I've never heard a binaural recording, and that's the big problem..


There's a popular example of binaural recording example called "Virtual Barber Shop".
You can find it on youtube:
http://www.youtube.com/watch?v=8IXm6SuUigI

Friends of mine claimed they could locate some sounds coming from the front.

I wonder why I can't. For me the coronal plane is the limit. The soundfield never extends in front of my head.


Soundstage and headphones

Reply #14
Try some binaural recording with headphone, for example:
http://www.youtube.com/watch?v=OSgxBlkHjAg
This kind of recording will tell you exactly what soundstage is.
No matter high priced gear you choose, you will never achieve true soundstage from ordinary stereo recording that is meant for loud speakers.

Soundstage and headphones

Reply #15
Oh, kind of duplicates.
I should have read previous posts.

Soundstage and headphones

Reply #16
Friends of mine claimed they could locate some sounds coming from the front.

I wonder why I can't. For me the coronal plane is the limit. The soundfield never extends in front of my head.


If you are talking specifically of the "virtual barber shop" recording, I locate every sounds  behind me. Otherwise, on "regular" music recordings ,  it's unreliable.

Soundstage and headphones

Reply #17
I wonder why I can't. For me the coronal plane is the limit. The soundfield never extends in front of my head.

Same here. The problem with binaural recordings is that they sound different for each person. To hear the right placement, your outer ear needs to exactly the same shape as the outer ear of the (dummy) head used in the recording. That's also why binaural never really got successful.

Considering soundstage, the problem with headphones and earphones is that the sound always seems to come from inside my head. I've never heard any headphone (besides while listening binaural recordings) with a soundstage in front. There's a soundstage 'inside my head', I can locate left and right, but the middle is right in the center of my head. Usually that's not really a problem, but when I mix recordings and listen closely, it is pretty annoying. Therefore I always check my mixes for stereo balance and issues on speakers.

Music: sounds arranged such that they construct feelings.

Soundstage and headphones

Reply #18
Soundstage is an illusion and I don't know of any way of measuring, quantifying, or specifying it.
True to a point. You can however measure that the speakers are well-matched, which is important for the stereo illusion to work well.

You can perform objective analysis using a dummy head to capture the output of the speakers. I've only done this in simulation, and looked at the effect of various issues but not speaker quality.


Quote
With stereo speakers, you CAN reliably make the sound seem to come from near the center even though there is no center speaker.  You can also go far-left or partially-left, etc.    But other than that, positioning gets a little vague.  And in general, we aren't as good at "locating" a "phantom image" between the speakers as we think.*
In the sweet spot, with decent speakers, you can place the sound anywhere between the two speakers reliably. If this doesn't work for you, it's either you or your speakers that are at fault. Using other tricks you can move it slightly beyond the speakers unreliably, and forwards/backwards unreliably.

To get a similar effect with headphones requires a binaural recording / simulation and an expectant listener. Some "normal" recordings have certain features which help to create a similar illusion.

It's very easy to diminish or destroy the effect of real binaural recordings by listening via bad headphones. Interestingly it's very hard to do the same with bad lossy coding - contrary to audiophile belief, true binaural cues are very resilient to audible lossy coding artefacts (except intensity stereo etc).

Chasing soundstage with normal stereo recordings and heaphones is far harder, because the recording was never designed to give that effect over headphones in the first place. Hence, at least in theory, headphones that change the sound may give a better soundstage than those which do not.

In practice, those which give the impression of the flatest and most natural frequency response seem to be best able to convince at least some listeners that there is a real soundstage somewhere (sometimes beyond their own heads), and a perceived flat/natural frequency response is typically something to aim for anyway.

Cheers,
David.

P.S.
Quote
* Moulton Labs has a good article about panning (placing a phantom image in the stereo field during production/mixing).    The author did some careful experiments and he was surprised how imprecise it is.    His main point is that audio engineers are wasting their time f they try to perfectly position every instrument across the soundstage.
He did a test with clicks. The basic theory of level-only panning doesn't work with clicks, only tones. He then tried time-only panning and found it worked well. There are arguments for/against both - the "best" (audiophile beloved) recordings seem to have both level and time cues, and these sound especially good over headphones. Mainstream recordings mostly avoid this, and it's not just laziness: inter-channel time differences are a killer for mono compatibility. FM radios the world over blend to mono as the signal fades, so mono is still relevant.

Soundstage and headphones

Reply #19
@ktf
Thanks for the info.

If you are talking specifically of the "virtual barber shop" recording

Yes, I'm talking about the barber shop recording. My friends claim to hear specific sounds in front of the coronal plane.


Again, I can clearly hear hard-panned sounds coming directly from the left & right speakers (not very "realistic" either).  But if there is a singer in the center, I can't pinpoint the exact location.

I listen to B2031A in a near field situation(equilateral triangle) in my living room. When I listened to them the first time I was stunned by the phantom center and the phantom sources. Depending on the recording, they are narrow and defined.

On the PC I listen to Nubert 311, a more classical constructed passive loudspeaker. When I place them on top of the B2031A and switch between them, they produce a soundstage that is more behind the speakers and extends in depth. The phantom center and the phantom sources are not as definied compared to the B2031A, the soundstage sounds more diffuse.

You might wanna give these speakers or the smaller version a try.

Soundstage and headphones

Reply #20
While obviously no kind of conventional headphones could hope to reproduce space in stereo recordings the way a pair of speakers does, their presentation does vary a fair bit. Some pull the middle of the stereo image apart to an extreme degree, while others give more sensible results. There are two things that you should look at when trying to understand this:
1. Directional hearing, and Blauert's directional bands in particular.
2. The influence of the pinna and outer ear in general.

I believe that ultimately it's down to (perceived) frequency response in one way or another. The less the outer ear is involved in reproduction, the higher the spread between listeners would be expected to become. Whatever the listener hears may then trigger directional cues. Remember that headphone listening is rather unnatural (at least if the material was not intended for it) and often has the brain grasping for any kind of straws. Frequency content is such a straw.

It probably is no accident that headphones described as "spacious" usually have a peak at ~9 kHz (which lines up with an "up" band) and above (which is usually interpreted as "behind"). I have a pair of old HD420SLs which exhibit a fairly extreme case of "three blob syndrome" (even worse than closed cans) - must be their mid-treble suckout, as that just about lines up with a "front" band. I guess if the brain doesn't know where to place a sound, it ultimately falls back to level difference and just maps it to one ear.

So assuming all of this applies... there will never be the perfect headphone for everyone, at least when not using any processing of some kind. Give it an accurate frequency response, and it won't be the most spacious with normal recordings. Make it more spacious, and it'll deviate from neutral. (Whether you want cans to be dead neutral is another matter. I tend to listen at low levels if possible and prefer a bit of built-in loudness - though I do mean a bit, not +10 dB at 100 Hz or such.)

BTW, the only form of crossfeed that I've ever found satisfactory is the Meyer crossfeed in Rockbox. It merely removes the annoying "in your ear" sensations while leaving the rest alone. I have no idea what the special magic of this implementation is, and the code is all Greek to me since I don't really understand how RB handles its int32 samples.

Soundstage and headphones

Reply #21
The way I see is is that headphones eliminate phase problems and the entire issue of room acoustics - reverberations and such. So your brain gets a playback of the recording that technically has more fidelity, since it eliminates those alterations to the sound.

Of course, for some things like heavy bass music (like dub) I prefer speakers since, if you want to feel the bass and not just hear it, you need a big speaker moving a room full of air.

Soundstage and headphones

Reply #22
If the music has been engineered to sound good with headphones then room acoustics could be considered a detriment, but since most music has been engineered using speakers, then the engineer is basically depending on room acoustics to complete the sound that he himself is hearing.

Soundstage and headphones

Reply #23
If the music has been engineered to sound good with headphones then room acoustics could be considered a detriment, but since most music has been engineered using speakers, then the engineer is basically depending on room acoustics to complete the sound that he himself is hearing.


Although I never tested this myself, I've seen in some discussion people reporting , that playing some recordings inside an anechoic chamber, doesn't sound so great.
I guess, a minimal amount of reverberation from the room is necessary to enjoy some recordings.

I'm also thinking that the current drivers limitations from speakers/headphones, have also an impact on how thing are mastered.
The hd800 have "ring drivers", which is unusual. I believe, because of this,  music with "artificially produced" sounds (psytrance, techno etc..) ,
are often not so enjoyable through the hd800 (although I think it's fine with classical, or nature recordings).

Soundstage and headphones

Reply #24
Based on my observations, headphones with elevated treble (bright) tend to sound great ("crisp", "detailed" or whatever) for acoustic instruments. But all sorts of artificial/distorted sounds (electric guitar through a guitar effect with too little low-passing, synthesized "instruments", "C64 music" and whatnot) are fatiguing or outright painful on those.

Could this be also the case for HD800?