There's no such thing as headphone depth, or height.

Discussion in 'General Audio Discussion' started by BenjaminBore, Mar 27, 2017.

  1. BenjaminBore

    BenjaminBore Friend

    Pyrate
    Joined:
    May 23, 2016
    Likes Received:
    2,842
    Trophy Points:
    93
    Location:
    London, UK
    So I have been reading the words of you fine fellows for a little while, and something continued to perplex me which I think I've now sussed out. What you guys refer to as depth, in a positive light, I refer to as the mids sounding small and/or distant, and feel it is a negative trait. The first time I encountered this I tried EQing thinking it was just mids recession, it didn’t work.

    I put it you that there is no such thing, that any perceived depth is more akin to the background layers of a pop up children's birthday card, and that it is static, there is no movement from back to front. My feeling is that these depth-layers simultaneously make it harder to discern mid-range centre-stage audio as it becomes more distant sounding and/or smaller in size.

    I don’t have access to my equipment at the moment but I would like to use one of those virtualisation tests where you move the ball around a dummy head to see if anything considered as having depth demonstrates any benefit, or perhaps actually hinders. I've seen it said that an inaccurate frequency response can create this depth layering effect, I've not had the opportunity to EQ a headphone that naturally exhibits this trait but I have with a hybrid amp and it didn't change anything in that regard.

    As for stage height I've been thinking this is perhaps more appropriately described as the size of sounds within a stage. There's no positional info, just the perception of how big or small each sound is. I've not tested out this theory yet but I cannot recall a time where I noticed sound moving vertically.

    For example:
    Mjolnir 2 stock tubes vs solid state: Mids sounds distant with tubes and are harder to make out, stage sounds flat depth-wise with solid state tubes but mids conform to the rest of the stage and are easier to hear.
    Focal Elear & Utopia: Similar staging. Utopia narrower width-wise, mids smaller and more distant. I've seen the Utopia described as having more depth, this is what made everything click.
    Fostex TH900 vs HD800: Fostex's stage is very wide, wider than the HD800, but the sounds are all small or the stage is very short, it's a bit like you're listening in a horizontal tube.

    An outline to portray the concept, rather than accurately show the degree of difference between these headphones:
    [​IMG]

    EDIT:
    In summary I would put forward that soundstage height and soundstage depth are misleading terms. That sound size is more accurate for the former. That the latter is more difficult to label as centre-stage layering/depth comes at the expense of centre-stage distance/size and audio discernment.
     
    Last edited: Mar 27, 2017
  2. frenchbat

    frenchbat Almost "Made"

    Joined:
    Sep 26, 2015
    Likes Received:
    1,369
    Trophy Points:
    93
    Have you entertained the idea that maybe, just maybe, headphones are not for you?

    A few guys here are not really considering headstone as a proper stage, with regards to speakers. So maybe that's your case, and you'd be better off scratching headphones and go speakers.
     
  3. BenjaminBore

    BenjaminBore Friend

    Pyrate
    Joined:
    May 23, 2016
    Likes Received:
    2,842
    Trophy Points:
    93
    Location:
    London, UK
    Not at all. For me this is just a matter of understanding these traits better, and what others mean when they refer to them. There's a very difficult language barrier in understanding what other people mean when describing sound, even when using the glossary of terms. This is just one area I've been trying to decipher.
     
    Last edited: Mar 27, 2017
  4. landroni

    landroni Friend

    Pyrate
    Joined:
    Aug 12, 2016
    Likes Received:
    1,164
    Trophy Points:
    93
  5. BenjaminBore

    BenjaminBore Friend

    Pyrate
    Joined:
    May 23, 2016
    Likes Received:
    2,842
    Trophy Points:
    93
    Location:
    London, UK
    I'd not come across that particular thread, but I've looked through all the glossary of terms I could find, and asked related question in other threads. Even so I'm not simply asking a question with this posting, I'm saying that at the very least there's an issue with the terminology used to describe these aspects, as they're misleading.
     
    Last edited: Mar 27, 2017
  6. Serious

    Serious Inquisitive Frequency Response Plot

    Pyrate BWC MZR
    Joined:
    Sep 28, 2015
    Likes Received:
    2,594
    Trophy Points:
    113
    Location:
    near Munich, Germany
    I think a lot comes down to the recordings. Pretty much all the music is mixed on speakers. With headphones you rely mainly on the recording to provide the soundstage information, even more so with IEMs.

    I feel EQ and tonality can affect the soundstage presentation, but it's far from the only thing that matters. My modded HD800 actually has a similar midrange tone to the HD650/HD600, but it still has a much better staging. Similar story with the Utopia vs HD800, although the Utopia has a different midrange tonality.
    I don't think the image size has much to do with the perceived depth, but I agree that the stock HD800 has a more diffuse imaging.

    When I made a few recordings with binaural mics in my ears (just ambient sounds, etc.) I could clearly pinpoint where the sounds came from in all three dimensions with both the HD800 and the UERM. You don't have that ability with speakers. With regular binaural recordings with a dummy head I don't get that same sense of space, so it's limited by how accurately the dummy head matches your own head.
     
  7. BenjaminBore

    BenjaminBore Friend

    Pyrate
    Joined:
    May 23, 2016
    Likes Received:
    2,842
    Trophy Points:
    93
    Location:
    London, UK
    That's an excellent point. I listened to a binaural recording of walking through a New York City street and it was incredible.

    I agree that the context for what I am saying is how something mixed for speakers sound on a headphone, and it's an important distinction. But ultimately that's the same context everyone uses when talking about headphones.
     
    Last edited: Mar 27, 2017
  8. Hrodulf

    Hrodulf Prohibited from acting as an MOT until year 2050

    Pyrate
    Joined:
    Sep 28, 2015
    Likes Received:
    4,276
    Trophy Points:
    113
    Creating an illusion of elevation on stereo speakers is still a poorly understood concept in acoustics. In nature you always have visual cues and location is aided by small movements of head.

    With stage height people most often mean overall acoustic phantom image size. You can do DSP trickery in a array type speaker to "steer" the projection, but again - it's far from trivial.
     
  9. Thad E Ginathom

    Thad E Ginathom Friend

    Pyrate
    Joined:
    Sep 27, 2015
    Likes Received:
    14,134
    Trophy Points:
    113
    Location:
    India
    <crossposted with Hrodulf>

    My take on all this, and I think I apply it to both headphone and speakers, albeit with differences, is that stereo works in two dimensions.

    But it is also that stereo is a world of illusion. A wonderful world of illusion! It encompasses far more than the simple relationship of sounds through a right-hand transducer and a left-hand transducer, because it prompts us to let stuff happen in our own heads. I don't deny the possibility that some equipment aids illusions other than the simple right-to-left picture. And of course, the recording engineers do their bit to help as appreciate the music as more than just a flat thing, with relative levels, etc --- but there is no up/down or back/front panning control on mixer for 2-channel music.

    Enjoy the picture the brain paints! It is the most amazing piece of equipment any audiophile can own, and probably the most important.

    Contrary to what I just said!... Dimensions other than L <--> R can be manipulated in a 2-channel recording. Here are some demo/tests (IIRC, for speakers, not 'phones) that just blew my mind by patently doing what I always thought impossible: LEDR Listening Tests. (warning: best not to visit that site without at least an hour to spare. Too many toys to play with!). I since came to know that this sort of stuff is done for pilots, so that they get spacial feedback from the various voice sources on their phones. But that requires individual calibration to individual ear shape.

    I'm certain there are members who know about that stuff, and it would be interesting subject to make a thread about. Thing is, though, I don't think that that technology is used in music recording.

    Correct me if I'm wrong! \/

    .
     
  10. BenjaminBore

    BenjaminBore Friend

    Pyrate
    Joined:
    May 23, 2016
    Likes Received:
    2,842
    Trophy Points:
    93
    Location:
    London, UK
    @Hrodulf @Thad E Ginathom That's very informative, thank you.

    To be clear this wasn't meant as a criticism of headphones themselves. My intent was more to understand headphones better as they are, compare notes, and perhaps encourage a change in vocabulary. EDIT: And challenge the idea that soundstage depth as is being described is even desirable. As from my perspective, particularly as someone who especially values soundstage because I also use headphones for film, this characteristic isn't really depth and the trade off for it is a negative that outways the positive.

    In relation to some of what you guys were referring to, for those that don't already know. Our ear drums are 2 dimensional also. It's our ear canal, ears, head, shoulders, and room acoustics adjusting the signal going in, and our brain processing those adjustments along with timing differences between the two sides that allows us to perceive audio in 3D. The limitation as was stated is that everything is mixed for a pair of speakers in a room, though I'm sure there are plenty tricks to manipulate this stuff in the mix, or by headphone designers, and there's long been fairly limited technology to do so after the fact via DSP. I would hope that one day there'll be binaural recordings for everything, and our headphones can calibrate to our particular ears and adjust accordingly. Headphones with some calibration features seems to be starting to trickle in.
     
    Last edited: Mar 27, 2017
  11. Changeling

    Changeling Tube Slut

    Pyrate
    Joined:
    Nov 13, 2016
    Likes Received:
    1,019
    Trophy Points:
    93
    Location:
    Sweden
    Have you tried Sonarworks + Waves NX?
    Check out Waves NX and play with "speaker" placement. It's saved me from suffering hard panned stereo death.
     
  12. purr1n

    purr1n Desire for betterer is endless.

    Staff Member Pyrate BWC
    Joined:
    Sep 24, 2015
    Likes Received:
    89,777
    Trophy Points:
    113
    Location:
    Padre Island CC TX
    The HRTF stuff makes sense, but only the context of additional information. Slight movements of the head will result in slightly different frequency response from the HRTFs. Even bobbling the head up and down to music will do this. This is the very essence of how HRTF works - how it must work - and why only systems with a position / orientation sensor on the headphone work in being able to convey positional data. Or a system that uses visual cues, i.e., 3D rendering of the environment (with viewpoint controllable by user) with more simplified HRTF models (remember Aureal / A3D?)

    A static HRTF doesn't work because it's static. It only translates sound to a static frequency response, which could have been caused by anything, EQ, mastering, etc. This is why that Superman soundtrack DSP app from DTS failed (http://www.changstar.com/www.changstar.com/index.php/topic,1012.10.html). Our brains work on discerning location based on changes to frequency response from the HRTF of changing head positions / orientations. Even, then, sometimes we get fooled.

    Biaural recordings help in being able to convey positional information; but I've found that their success in being able to do is varies. For the most part, headphone imaging is like sitting between two speakers than in front of two speakers. Mainly because we are effectively sitting between two speakers! Not a popular statement to make, but headstage = varying levels of BS, just as with height of stage (with speakers or headphones). Specific aspects are not BS though: like hole in the center, or three blob effect.
     
    Last edited: Mar 27, 2017
  13. FallingObjects

    FallingObjects Pay It Forward

    Pyrate
    Joined:
    Oct 15, 2016
    Likes Received:
    2,235
    Trophy Points:
    93
    Here's an interesting thing that I got to experience in my undergrad, in a sensation & perception class.

    At higher frequencies, and given a specific distance, your ears are incapable of telling whether a sound is coming from directly infront of you, or directly behind you.

    Given the absence of resonance through your body/bones, or reflection of the sound from other materials (like in a wide open field, or an acoustically-dampened room), there's no queues that your mind can use to distinguish between the two. I remember sticking my head in an apparatus, and after some initial adjustments, I couldn't accurately determine whether sounds were being played through a speaker infront of me, or behind me.

    This is partially the reason why speakers and headphones will likely never be quite the same; your body 'feels' sound a lot more than may be initially obvious. I suspect that speakers may eventually reach the point where we can have fairly accurate depth/height queues on the consumer level, but I don't know if it's even possible achieving that with headphones.
     
  14. Thad E Ginathom

    Thad E Ginathom Friend

    Pyrate
    Joined:
    Sep 27, 2015
    Likes Received:
    14,134
    Trophy Points:
    113
    Location:
    India
    Try the LEDR thing in my link. You can keep your head still, and watch the sound move in an arc over your speakers. It really is astonishing.

    But not, as far as I know, part of any real music recording out there. It also may vary in impressiveness for different ears in different rooms in front of different speakers ---but it is reliable enough to be out there as a test at least.

    But it takes a brain to visualise an orchestra in front of us when headphones are playing it into the middle of our heads.
     
  15. BenjaminBore

    BenjaminBore Friend

    Pyrate
    Joined:
    May 23, 2016
    Likes Received:
    2,842
    Trophy Points:
    93
    Location:
    London, UK
    Not those particular ones. But I use EQ for the frequency response correction. I've also tried a bunch of DSP solutions for surround, as well as crossfeed. Some add a bunch of heavy handed reverb to make it sound like an auditorium, others do a decent job of creating positional sound without effecting image width or size, but degrade fidelity, softening bass and treble.

    Yes of course, this had slipped my mind. Everyone will recall a time where they've heard a mystery sound and found that while they hunted around for it they were continually moving their head to help locate it's position, like a pooch or kitty on the hunt for something unseen.
     
  16. Serious

    Serious Inquisitive Frequency Response Plot

    Pyrate BWC MZR
    Joined:
    Sep 28, 2015
    Likes Received:
    2,594
    Trophy Points:
    113
    Location:
    near Munich, Germany
    @BenjaminBore I don't think that headphone depth has much to do with mids sounding distant. I do feel that you can hear sounds moving from front to back with headphones and with speakers. I also think that headstage is comparable to soundstage with speakers. My findings when comparing equipment regarding soundstage are usually consistent on headphones and speakers. Headphones generally have a tiny soundstage that is all around your head, while speakers to me are more like a properly sized rectangle infront of you. I cannot hear sounds behind me when listening to speakers, but I can on my HD800.

    Stage height is more complicated. Generally there is no height information for me, both on headphones and speakers. Only under special conditions a height perception will work in 2 channel audio, like the binaural type stuff in @Thad E Ginathom's link. In that case I find headphones to give you a better sense of something actually being above or below you, but only with proper HRTF (read: putting microphones in your own ears - not in someone elses) I find that I can sort of hear the distance of a sound. That depends on how many reflections the space you're recording in has. If it's in an anechoic chamber you likely won't hear any distance effect.

    There is a height effect with most speakers though, but not one that's in the recording. With multi-driver speakers you can hear that the sounds aren't coming from the same point in space - you can hear that there are multiple drivers. This is especially annoying where wavelengths get shorter. Most of the time the tweeter will be above the midrange. I feel that this and the phase shifts introduced by the crossover destroy the natural imaging and coherency. I actually generally find tweeters above the midrange more annoying than a tweeter below the midrange, or on one side of it.

    @Thad E Ginathom Those LEDR tests are really interesting. They seem to be made for speakers, but I can also hear the same effect on headphones. The knock test (Stereo Perception and Sound Localization) is also pretty interesting. I actually feel that the binaural one sounds most convincing both on speakers and on headphones. I wouldn't discount binaural recording as something only for headphones.
    You can also get something in between regular recordings and binaural recordings if instead of a head you use a sphere, or something shaped similar to head, but without ears and put the microphones on its sides. This will still give you some HF rise, but it's not as colored as binaural recordings.

    @Marvey I find that moving my head with speakers actually sort of destroys the illusion - you can hear more clearly that it's just two speakers playing. HRTF stuff can't easily be applied in software since every different angle needs a different compensation. Reflections will need a different compensation based on where they came from; it's a mess to get good results in software. Some time ago I posted a link with many different HRTFs applied to a test track - all of them gave a completely different soundstage for me and most of them didn't work at all. A simple recording with microphones in my own ears is in a whole different league in terms of soundstage. Your brain almost immediately accepts the colorations from your head and gives you good imaging. With OOYH I can always hear the upper midrange rise as a FR effect, not as a soundstage effect. OOYH doesn't give me a good illusion.

    One thing that bothers me with speakers is that the tonality changes depending on if the sound is coming from the center or one of the sides in a way that doesn't accurately mimic what your head would do.* Vocals either sound correct in the center or on the sides, but not both. This is more of an issue when the speakers are placed further apart, but I can't stand overly narrow speaker imaging. The speakers have to be at least as far apart as their distance to the listening position.
    Stereo imaging (with speakers and with headphones) is an illusion that doesn't sound like real life.

    There's also the Bacch-SP for speakers that tries to reduce crosstalk with speakers, which to me sounds like it would make speakers sound more like headphones. Supposedly it's very convincing with binaural recordings.

    __
    *This is more or less what I mean:
    http://www.innerfidelity.com/content/first-crunched-data-harman-head-measurement-session
    The stereo vs left/right difference is not the same as with left/right vs mono. You can either voice your speakers to sound tonally correct for the center stage or the sides, but not both. I think most speakers are in between.
     
    Last edited: Mar 27, 2017
  17. purr1n

    purr1n Desire for betterer is endless.

    Staff Member Pyrate BWC
    Joined:
    Sep 24, 2015
    Likes Received:
    89,777
    Trophy Points:
    113
    Location:
    Padre Island CC TX
    I'm not referring to moving the head intently. There is always motion with human beings, micro movements, unless you listen to your speakers with your head clamped in a vice. Also, a recording with microphones in your head may not work with other people. This is probably why off-the-shelf biaural recordings still don't work for most people. And why that (forgot its name) headphone system requires personalized calibration to sound right, otherwise it's "echo city".

    HTRF can easily be implemented in software. You don't have to take that measurements from every different angle. You just need to take measurements at increments and then interpolate. You can also simplify things and take incremental measurements rotating around the Z (up down) axis.

    As far as midrange affecting soundstage depth, this can happen. But there is also the matter of different gear affecting the soundstage much more. Four phono preamps I've used all have the same frequency response (OK, maybe 0.1db off at most), but they all portray depth very differently.
     
  18. ultrabike

    ultrabike Measurbator - Admin

    Staff Member Pyrate MZR
    Joined:
    Sep 25, 2015
    Likes Received:
    8,960
    Trophy Points:
    113
    Location:
    Irvine CA
    @Serious, if you ever have a chance, listen to the Smith Realizer. Do so while moving your head slightly. An eye opening experience IMO.
     
  19. purr1n

    purr1n Desire for betterer is endless.

    Staff Member Pyrate BWC
    Joined:
    Sep 24, 2015
    Likes Received:
    89,777
    Trophy Points:
    113
    Location:
    Padre Island CC TX
    We need @Out Of Your Head / Darin to provide a sensor to attach at the top of the headphone, and then feed position / orientation information to his software via a USB dongle. Even if done around one axis, it bet the results would be awesome.
     
  20. Dr J

    Dr J Friend

    Pyrate
    Joined:
    Sep 28, 2015
    Likes Received:
    97
    Trophy Points:
    18
    Fun stuff, need to go through more test tracks on that site, thanks for pointing it out.

    The most extreme example of playing around with the sound-stage in a 2-channel recording I have come across is "Amused to death" by Roger Waters. On and off through the album when listening with speakers, instruments and voices sound like coming consistently from various places anywhere and on any height in the front half of the room and on occasion from behind listener's head. The opening track with a veteran recounting war experiences is pretty spooky when you hear it the first time.

    Another one with somewhat more restrained effect is his Pros and Cons of Hitchhiking, song called "4:50AM (Go Fishing)", where the forest glade with birds tweeting expands to fill the front of the room and someone starts chopping wood behind the listener.

    The above have been heard by some of my friends and also with their equipment and rooms, so I assume they are not an anomaly caused by my setup or brains :)

    Every now and then there are instruments in surprising albums that localize in odd places. Maybe incorrect phase or intentional, not sure.
     

Share This Page