[mvw2]

Yes.  All we have is a left and a right source.  It's up to our minds to create the perception from what it hears.  I do totally agree with shane55 and pizzafilms.  When we discuss the idea of sound stage, it's a bit of a goofy subject in the way that we approach it completely from the wrong direction.  We do like to say this earphone has X sound stage and that earphone has Y sound stage.  In reality this is just B.S. because the earphone doesn't create the sound stage.  Rather it is a side-effect of other things.  While our mind is the device that creates the sound stage we so heavily discuss, it is the earphone that recreates audio well enough to let our minds do it. 
 
I have in my reviews and comments linked some aspects of sound to their effect on sound stage.  Dynamic range is a good example.  The ability to portray loud or quiet well isn't always easy for earphones.  It may not be a very linear range or may be very compressed.  Any limits here affect how the earphone reproduces mixed volumes within the music.  Some sounds may be louder than they should be.  Some sounds may be quieter than they should be.  Our mind relates whatever it hears back to what it's used to and creates a mental perception.  Non-linearity and compression end up skewing the sound stage our mind makes up.  Our mind just uses what's there.  The earphone creates sound within its capabilities and limits to some level of accuracy.  At the end of the day though, the earphone has a skew of some sort that correlates back to the perceived sound stage we hear, and this sound stage is repeatable across all music. 
 
We end up creating this idea that the earphone has a particular sound stage.  Earphone X has more depth than earphone Y.  When we say this, it isn't wrong.  It isn't B.S.  It's correct for what it is.  It's a fake thing, yes, but it does correlate back to a true part of the earphone.  For example, the lack of depth relates back to dynamic range mainly.    It's not that earphone Y has less depth.  Earphone Y has a more compressed dynamic range.  This might correlate with a small linear range of the driver, high damping, being under powered, or some other factor associated with the dynamic compression of the audio signal when converted from electrical to sound waves.  We can start with a mechanical problem.  We can follow this to low dynamics in the audio wave.  Further our ear picks up this more limited range.  Our mind does what it does best, make up stuff, and we get a perceived sound stage.  In this case, the stage is a bit compressed, squished, and we lack depth.  Since the earphone is always dynamically compressed our minds will always create a sound stage with little depth.  In the end, we can say earphone Y has little depth simply because it will always be perceived that way.
 
So why can't someone just say an earphone has a limited dynamic range and leave it at that?  Well, unfortunately it's not quite so simple.  There are a lot of variations.  Compressed?  How compressed?  How much does it end up actually influencing the end sound?  Is it compressed through the entire frequency spectrum or only lower frequencies?  When it gets to the time our minds create a mental sound stage, how much does the compressed dynamics affect the perception?  In which ways is it affected?  I think it's easier for a person to go into more detail from the sound stage side than it is for the person to go into detail from the engineering side.  Part of it may be that the earphone for most is simply a black box.  Even if we know exactly how it's made, we have no real data on the earphone.  All we have is what we can hear.  Since we are stuck from the hearing side of things, we have to start from that side.  This unfortunately means we start with comments on sound stage.  Is this bad?  I don't really think so.
 
Would it be better to simply have scientific data?  Sure.  I'd love to have every earphone measured in great detail.  I'd love to break it down into just the physical operation.  It's more of an absolute.  It's like Kippel testing of home audio drivers.  They are simply built a certain way.  The motor and suspension operate in a particular way, and certain sound aspects correlate from this.  Material choice plays a role and has its own effect on sound. 

 

[Danz03]

Yes, I've not heard of the term 'soundstage' until I joined Head-Fi and I've never ever been aware of 'soundstage' in headphones before. I worked as a sound engineer too and me and my colleagues used to describe most recored music as panned mono recordings with added reverbs, because that's basically how most non-classical music is recorded and mixed, so the concept of 'soundstage' seems more like a myth than fact. However, since reading about it here, I do notice that there is a difference in the size of 'soundstage' between headphones, whether it's something that's real or not.
I remember many years ago an engineer brought in this thing called the 'dummy head', which was basically a specially designed pair of microphones imbedded inside this model of a human head. We did some recordings with it and listened to them through an ordinary pair of headphones and it was amazing, I could actually pin point where ever piece of instrument was accurately, like I was there. There were technologies developed in the past to try to recreate 3D or surround soundscape with stereo speakers, notably the Roland RSS system, Madonna's 'Immaculate Collection' was original mixed using it but it was nothing like as amazing as the 'dummy head' recordings, and the listening position is quite critical, turn your head more than 45 degrees and the effect will be lost.
I think audiophiles are not really looking for something accurate, they want something pleasurable to listen to instead, otherwise, there wouldn't have been so many tube amps around. Personally, I don't actively try to feel the 'soundstage' of my headphones, I'm more focused on listening to the performance, arrangement and production of the music rather, but I could also easily understand how one can be interested in all the psychological aspects or just the pure pleasure in listening to music.

Thanks, Shane.
 
Wow! FINALLY some reality in the definition and use of "soundstage" in this forum.
 
I've been a sound engineer for more than 25 years doing everything from mixing front of house, designing and installing sound systems and recording and mixing albums, and I've always laughed by a$$ off at the way the term "soundstage" has been thrown around here. It's so completely and utterly overused and misused when it comes to iems.
 
I mean seriously, "is the soundstage wider or taller?" or "the soundstage is not as deep as I would like" or "it places the instruments nicely"...what complete and utter horse$crap.
 
The only controls that iem manufacturers have over audio is: frequency response, dynamic range and distortion. That's it. EVERYTHING else is based on the original recording and how that recording is reproduced by the iem's frequency response, dynamic range and distortion. Period. 
 
Even then, "placement" is truly only 2D...left and right. If you think you hear depth, it's really only a function of how much high end has been trimmed or boosted and/or reverb added.
 
Sorry about the rant...but I've been wanting to get that off my chest for quite a while now. Please everyone, go back to listening to some great music.
 

 

[DaEMoNteNTAcLe]

finally..
someone brought some sense into soundstage. otherwise it's always about wide.. then later about, width and height. then later size. but it is actually soundstage AND positioning that gives the real sense of imaging. also, there is distance, forward, backward, depth. there is also size, big objects falling on ground should 'sound and feel' like big object, even though it is far and presented in low volume. small object, no matter how near, hence louder, should still 'sound and feel' like small object. this is dynamic range, that has been mentioned.
there is also echo. here, ALL closed iems will have this removed or controlled, only open iems (ur ok1 is open) will provide best echo, since sounds can travel far and dissapears. closed phones will have the sound reflected back, usually tuned so that it overlaps, thus gaining amplification, hence the illusion of more details, or, dampened with some foam/felt.. etc.. some people here mods with blutack as well.. to improve clarity, and reduce reflection, because we want to hear the sound produced by drivers, not reflected sound.
it's like pointing ur speakers to the wall behind it while playing music vs pointing the speaker directly at you.
 
i was hoping you (mvw2) would also mention about phase difference.
 
phase is like controlling how sound travels. from speakers, its straight forward, from speaker > air > ur ears, hence very very obvious.
from iem/headphone, there is a definite lack of distance. to overcome that, there will be phase difference.
a sound coming from right, should be heard by the right ears first, before it reaches the left ear. the difference in time and gain on left vs right, will also create a sense of distance, echo and position. the technology to control this effect is known as crossfade.
 
anyways sound quality aside, here the discussion is already on presentation of the earphone itself. good phones will almost always produce good sounds.. but all of them will still have different ways of presenting the sound. good phones will present the sound with more clear and precise imaging. once done properly, it will improves the experience a lot.
 
given the limitation of the space between the ears and drivers.. this is still the area where headphones will trumps earphones, and speakers will trumps headphones. the distance is needed to re-create spatial presentation, something that has been dampened and eliminated as much as possible in studio recordings (at least i have never heard or read about a recording studio that have lots of noise), since on poor phones, those echos, noise, reflections, etc. will appear together and clutter all the sounds.
 
ck100 is open acoustic, as well as ck10, ok1. that's one of the reason why the presentation is better. most iem are not, since open means less isolation, and people go iem for isolation.
 
as for me, i wont go ck100, since i found it thin and harsh sounding (to my ears), hence prefer ck10, but i believe it does just as well as ck100 in imaging.

 

[Antony6555]

Soundstage size in earphones is kind of irrelevant to me, since you're pretty much only going to get small and extra small, in my experience. Buy earphones based on tone reproduction, not soundstage. I have the ck100s and while they do better imaging than the er-4, imaging sounds like a tiny mess to me still. If you want decent imaging, you're going to have to go with fullsize phones.

 

[incongnit0]

Not too picky on soundstage as long as I do choke on the sound produced. UM3X sounds great but the soundstage leaves something to be desired

 

[james444]

Quote:

We end up creating this idea that the earphone has a particular sound stage.  Earphone X has more depth than earphone Y.  When we say this, it isn't wrong.  It isn't B.S.  It's correct for what it is.  It's a fake thing, yes, but it does correlate back to a true part of the earphone ...

 
I agree that soundstage is a fake thing, but one just has to sift through this glossary to realize that many other terms we use to describe sound are equally artificial. So what? Most of us are consumers, not engineers. As long as we have a common understanding of these terms there's IMO nothing wrong with using them, regardless of their technical background.
 
Not meaning to sound ignorant and in fact (as a layman) I find this thread quite interesting and informative. Just meaning that a term may well be dubious from a scientific POV, yet still hugely useful to help us illustrate what we hear.

 

[shane55]

[size=10pt]^^ Yeah, I get that.[/size]
 
[size=10pt]As I intimated before, ‘soundstage’ is like 'warm' or 'cold' in terms or how it relates to the actual recording, but I believe a bit more insidious. It can be liked or disliked, sought after or avoided, and like 99.9% of everything here, that's a matter of taste (and opinion).[/size]
 
[size=10pt]While adding warmth (or coolness) to a sound creates change in tonality, ‘soundstage’ anomalies pull you even farther from the original by introducing over and undertones as well as reverberations that can mask the clarity. Changing tonality or the sonic signature typically only raises or lowers certain frequencies and doesn’t add aberrant waves.[/size]
 
[size=10pt]I’m strictly talking about IEM’s here, where the sound is shot straight from the driver, through the eartip, directly into the ear canal.[/size]

 

[MGbert]

Quote:

i was hoping you (mvw2) would also mention about phase difference.
 
phase is like controlling how sound travels. from speakers, its straight forward, from speaker > air > ur ears, hence very very obvious.
from iem/headphone, there is a definite lack of distance. to overcome that, there will be phase difference.
a sound coming from right, should be heard by the right ears first, before it reaches the left ear. the difference in time and gain on left vs right, will also create a sense of distance, echo and position. the technology to control this effect is known as crossfade.
 

 
 So true.  Also, differences in frequency response between the left and right earpiece can easily muck up the perceived soundstage.  There were several records issued in the 1960's which were recorded in a final mono mix, but were "artificially enhanced" for stereo by passing the mono signal through 2 out-of-sync comb filters.  So, for instance, a narrow band around 1000 Hz would be boosted in the left but cut in the right.  That had the effect of widening the perceived sound beyond the phantom middle between the two speakers compared to the original mono, but did nothing towards allowing localization of any instrument within that "stage".  In fact, it smears the directional cues, making it harder to determine just where in the left-right continuum.  Differences in freuency response left-right would have that same effect, methinks.
 
MGbert

 

[mvw2]

The reason why I do look at sound stage as important is because if it is correct it means the earphone is presenting audio correctly.  It offers transparency, low distortion, a relatively correct (or simply good enough) frequency response, good dynamics that are relatively uncompressed and linear, and a reasonable amount of detail, not just note speed but also articulation of note.  Basically a good sound stage means the earphone is doing stuff correctly.
 
Now I do agree trying to quantify sound stage with musical recordings is hard.  I tend to focus more on less mixed recordings to help gauge an earphone, thinks like live concerts with the screaming audience in the mix or a live performance of someone using only a room mic that picks up not only the artist but also the room echos and any other noises.  I didn't even create this thread till after I started looking at sound stage relative to video which is engineered to have specific positioning.  You watch movie information it is designed to be three dimensional with specific placement.  The earphones need to present information correctly enough to have your mind put those sounds in the right spots.  This doesn't always happen, and it's due to the earphone doing something wrong in some way, although wrong might have actually been engineered into the earphone in the first place on purpose. 
 
Now this isn't to say that what is not good for video isn't good for audio.  The importance of sound stage in audio is less.  We don't need things placed correctly.  Most of the time all we want is an earphone that is clean, detailed, and quick enough simply not to be muddy and keep the sounds separated.  It ends up being nice when the dynamics are there to help space out the sounds.  It's nice when the texture/articulation is there to bring through the reverb and echo and fill out the dimensionality of the perceived space.  A lot of what this does is simply help spread out the sounds so singers, instruments, sounds are distinguished separately and we can "see" more in the music.  There are some earphones that are quite exceptional at doing this, correctly or not.  The UM3X is a great example.  Every sound you hear exists in its own place.  All the sounds are separated out and individualized.  It lets you hear a LOT of the information in the music.  It's very nice if you want it.  Translating the UM3X to video, it's a little different.  It ends up portraying information too far left and right for what's shown on the video.  This effectively shows some form of non-linearity in presentation.  A sound that should be a 30%:70% mix of left:right putting the sound "image" at 60 degrees to the right (not sure what it would be exactly) is instead deciphered by the mind as being 75 or 80 degrees to the right instead, farther than it should be.  You end up not syncing the "visual" of the sound with the visual in the video.  In a way you'd kind of want both to work really well, and I guess if the video is right, the audio is more "right."  However we are creatures of preference, and preference doesn't have to be correct.  For audio we are trying to match image placement.  We aren't so constrained, so we gain extra freedom of choice.  We are free to have more or less of various aspects to suit our tastes in sound reproduction.  Maybe we don't like high separation.  Maybe we just want the sounds blended together.  Maybe we want the sound to be more forward, aggressive, in our face.  Maybe we don't want something so aggressive.  Maybe the preference is for something more laid back, something less enveloping.  We can quite quickly correlate sound stage aspects which makes discussing sound stage nice for most people.  Really this translates back to more raw aspects of the earphone and the driver(s) used.  In reality everything we hear can be traced back to what the driver's doing, its linear range, motor strength, damping, distortion, any skew of linearity, efficiency, thermal handling, etc.  The problem is that I'm pretty sure I will never see a BL curve for an IEM in my lifetime, let alone a slew of other real test data that effectively defines what the driver does.  We really don't get measurements for any of this.  We barely get frequency response, and even if we do we really aren't told how.  In the end all of us are kind of stuck with subjective descriptors to describe what an earphone does.  Sound stage is simply one of the areas of description we go into.

 

[goodvibes]

While I believe that sound stage can exist in IEMs, I don't find it as important to determine goodness. We usually don't know how something was recorded or if perspective is natural or created. What we do know is that regardless of how it was recorded or processed, there's somebody on the other end of that microphone trying to convey and emotional message. I find this is actually the most difficult job of stereo reproduction in general. Call it goose bump factor if you will but if that's working there's very much right with the sound. With any type of segmented evaluation, it's very easy to not see the forest for the trees. Not implying that's going on here or that these aspects are mutually exclusive in any way.

 

[mvw2]

Well phase isn't really an issue outside of multi-driver earphones running crossover networks.  Even then it's not an issue unless they're implementing something wrong.  Phase isn't discussed a whole lot in car or home audio even.  Most of the discussion about phase is did you hook up your - and + wires right for all your speakers.  There is also useful discussion when talking about subwoofers, mainly ported setups where we are using the sound waves from the rear side of the driver that are 180 degrees out of phase with the port.  Very low frequencies can end up having a rather sizable delay that is hearable.  This in itself isn't a big deal unless we are trying to align everything right in the time domain and the group delay from the phase issue is high enough for us to care about.  Not a whole lot of people go into crossovers and discuss phase.  Some networks require reversing the polarity to be in phase, and it is sometimes put out of phase on purpose to help create separation between drivers by effectively creating a heavy cut at the crossover point.  I am not an electrical engineer.  I have not really looked into the crossover networks used in earphones.  They are not the same as larger systems that just use capacitors, inductors and resistors.  IEMs run very small transistors that do the work, and I'm not sure how they relate back to the large scale versions.  I just don't know what kind of capabilities/limitations exist with this form of filtering.  I simply don't know what it's doing when compared to what an inductor, capacitor, and resistor does which I do understand.  I just don't know if a transistor is the same or completely different in its operation of the signal.
 
I will say from a big picture view phase is really only an issue around the crossover frequency.  Once you get a ways above or below the x-over range, phase really doesn't do anything.  Since the frequencies vary in time, we don't even perceive things in or out of sync.  An example would be taking a a midwoofer and a tweeter.  Both are controlled through a passive crossover network, let's say 2nd order at 3kHz.  It really doesn't matter.  We purposely set the tweeter out of phase.  What we hear different is that the highs and the mids sound noticably separated from each other.  We end up losing the blending through the 3kHz range.  There is some loss of coherency between the highs and mids from this from the loss of blending.  At the same time we separate the two ranges with a distinct cut and sort of separate and individualize the frequency ranges.  Highs stand out more.  Mids stand out more.  We don't hear that the highs are out of phase from the mids though.  People do use phase to band-aid time domain issues, mainly done in car audio when time alignment isn't available.  People will set one woofer out of phase, one or both tweeters out of phase, or any mix that helps create a more dimensional sense of the music, although it's completely wrong for fixing a distance issue that is specifically a time alignment issue.
 
In earphones we really aren't working with different drivers and when we are we aren't really worrying about distancing.  Phase issues may or may not be there depending on the crossover network.  We are typically more sensitive to frequency response differences and more sensitive to how well or poorly the manufacturer blended the drivers together, not only pairing two like sounding drivers but also attenuating each properly to match output, picking an appropriate x-over point to keep both within their functional ranges, and using any additional filtering to shape the end response and cut out any potential issues like a ragged or spiked response.  Phase is simply a smaller issue in my eyes.

 

[shane55]

Quote:

While I believe that sound stage can exist in IEMs, I don't find it as important to determine goodness. We usually don't know how something was recorded or if perspective is natural or created. What we do know is that regardless of how it was recorded or processed, there's somebody on the other end of that microphone trying to convey and emotional message. I find this is actually the most difficult job of stereo reproduction in general. Call it goose bump factor if you will but if that's working there's very much right with the sound. With any type of segmented evaluation, it's very easy to not see the forest for the trees. Not implying that's going on here or that these aspects are mutually exclusive in any way.

 
[size=10pt]Over many years, and with many setups, speakers, headphones, (mine and others') etc., I've come to know a good deal of my music fairly well and have a pretty good sense of 'the best it can sound'... to my ears, tastes, etc. Some of these recordings are of a single instrument; some are symphonic, electric, etc. You are correct that we can't know how 99.99% of them are recorded (unless you actually do the recording !), I have been in many of the venues and have seen some of the performances that were recorded live. While auditory memory is usually about as good as a sieve holding water, when a very familiar recording is heard… it’s right or it’s not. It’s improved or it’s lessened. [/size]
 
[size=10pt]I will not keep an IEM with ‘enhanced’ soundstage or other spatial effect. When I test sound reproduction devices, I test them with known recordings and listen for just that… the absence of sonic aberrations (among many other things). If there is suddenly an echo or instrument placement that is different from what I’ve heard for years; something is wrong. That said, if I suddenly hear a chair creek, or that slight whistle of a squeaking bow on a string, where before I heard nothing; this is revelation! [/size]
 
[size=10pt]I want all my equipment to be as true as possible to the recording (as I've come to understand it over the years)… in terms of ‘information’. That information can be coloured slightly (my D7000’s do a great job of warming the sound beautifully), but I don’t want it otherwise altered, obscured, or added to.[/size]
 
[size=10pt]But as for mvw2's earlier point... As I don't listen to anything other than music through IEM's, so I can't really comment there, but I think I understand your point.[/size]
 
[size=10pt]shane[/size]

 

 

[Bennyboy71]

Listen to the new Dylan mono boxset. It's not about soundstage, its just about sound. Same with all music. Feel it and hear it and enjoy it. Size is arbitrary and what you think of as being large or small is entirely a construct in your head. Just ask women.

 

[Napilopez]

Quote:

I've never been big on the idea of sound stage within an earphone. It does have its purposes but I'm not a big critic of it since I see sound stage in an earphone as something that is limited by the fact that we have to stick an earphone housing in our ears and that can only do so much. Even the largest sound stage is still quite small and a closed headphone has a much larger stage than them. I find most earphones to have a general left and right but nothing really clear cut in terms of placement. There are exceptions as some earphones I've heard have much better placement but I find earphones as a whole to not have that great sound stage regardless.

I haven't read all the posts in this thread, but on this note I have to disagree somewhat. Naturally full-sizes will tend to have better soundstages, but this isn't always the case. My HJE900s, for example have a vastly superior spatial presentation compared to my RX700s or FC700s, and just slightly better than my old PX200-II. It might just have to do with transparency, but I get more convincing out of head feelings with the Panasonic's than any of my "out-of-ear" phones. I suppose that might just have to do with the relative transparency though.

 

[mvw2]

Odd, I think the HJE900 kind of sucks. <shrugs>  I know that doesn't really say much, but I'll just say there are significantly better options out there.