frontal sounds go up instead of further in front of me with headphones. why?

[castleofargh]

  From the same reference, head diffraction and reflection are mostly within 500-1600Hz.

to my ear, the 1000-2000he area tends to impact the feeling of distance, but it doesn't really change the vertical move when going center in any significant way.
 
I should mention that not all centered sounds feel like they're way up. for a good deal of songs as the total depth is something like from the center of my brain to my face, it doesn't really matter what the axis is, it still feels very much in front of me. it's not something that is a problem on a daily basis when listening to music, I rarely even think about it(the reason why I only made the post now). but as soon as it's something specifically made to simulate reality(oh irony) like the barber shop virtual vid above, it goes straight up for me for stuff that should be in front.
 
I figure most of you guys do get the barber guy walking in front of you normally at the end of the vid right?
 

  What happens when you change the headphones around so left is now on the right etc?
 
I do not hear him walk in front of me either and with headphones this is going to sound more in your head than in front of you because of how the headphone delivers the sound to your ears.

left right inverted, but the guy still walks up the stairs when passing in front of me ^_^.
 

Sorry, no!

 
Many/most people hear that effect. Firstly, you are not really talking about mono, what you are actually talking about is a "phantom centre" within stereo. Stereo is an illusion, a clever illusion which takes advantage of how the brain processes the audio signals which arrive at both ears. This stereo illusion can be quite easily messed up though, that's why for example cinemas never have stereo sound systems, they always have a specific centre channel rather than trying to create a phantom centre. Headphones also defeat the stereo illusion to an extent. Hearing the phantom centre as being high up in your head is simply a consequence of your brain's perception, of how it calculates the position of sound sources, which tends to highlight that stereo is an imperfect illusion. Applying HRTFs can help to an extent but responses are variable from person to person and are only a step towards what might hopefully one day be a complete solution. While science is very good at understanding what hits our ears and how our ears respond to it, science is currently not so good at understanding how our brains manufacture a "perception" or therefore how to fool it.
 
G

 
there are plenty of reasons why it could fail to make for a realistic imaging, and I would be fine if like for the lack of speaker soundstage on headphones, it was something that everybody experiences. but it's not, many people(most?) have nothing weird happening when the guy walks in front of them at the end of the video. so yes it must have to do with HRTF at some level.
I need to fool around with EQ some more(I just went through to test a dozen of IEMs and despite the various differences, I still get the guy climbing stairs like in the pic of my first post. I tend to EQ only to my taste, not to try and move things around so maybe I'll end up finding a solution(I'm thinking with mid/side EQ maybe) but ATM at least I failed to work something out.
but there must be something triggering my reaction, I don't live in a bubble in space so my real life reference of "in front of me" should be pretty close to the one other people have right? the only thing a little out of norm is that I'm tall. I still have my nose and ears where they're supposed to be

 

 

[spruce music]

Just a thought on an experiment I haven't done.  Try brickwall filtering out everything above 5 khz.  See if the image no longer rises when you do that.

 

[spruce music]

In case you are interested in the Sony 7510, here is a review with some measurements.
 
http://en.goldenears.net/22546
 
I didn't see this until I owned these phones.  My reasons for trying them were some reports by others and that they were intended to be an eventual replacement for the long running 7506.  It didn't work out for Sony, but with the idea it was designed as a sonic improvement on those I took the chance.  I have been happy with them for what that is worth. 

 

[chazb11]

Very interesting thread. I've owned several different "mid fi" headphones and have never really heard anything "out in front of me". The sounds I assume should be in front of me are "in my head" even while the left, right soundstage seems very natural. Oddly enough, my newest cans (ATH900X) project some sounds directly behind me. (no, I don't have them on backwards)
Maybe my ears are wired backwards?

 

[spruce music]

  Very interesting thread. I've owned several different "mid fi" headphones and have never really heard anything "out in front of me". The sounds I assume should be in front of me are "in my head" even while the left, right soundstage seems very natural. Oddly enough, my newest cans (ATH900X) project some sounds directly behind me. (no, I don't have them on backwards)
Maybe my ears are wired backwards?

You just need to invert your earlobes. 

 

[Jon Sonne]

  so I was wondering if there was a "cure" obviously, but also how many other people get that feeling and maybe try to fnd out with them if we have a common factor?
 

 
I've been having the same problem for quite a long time, and I found it rather annoying and at times fatiguing. Voices and especially horns would stand tall in the image and feel like they where in the top center of my head. What solved my problem was the TB Isone plugin. I guess another good crossfeed application might also do the trick, but so far I've only tried TB Isone. I'm currently using my Dunu Titan 1 earphones with the TB Isone with the following settings:
 

 
With these settings, the vocals are now situated in the front center of my head. Notice I have turned off the room simulation. I feel the room simulation masks a lot of details and makes rock / metal music less enjoyable.   
 
 
33 degrees is the sweet spot for me, but I think you need to tweak speaker angle, HRTF, ear size and head size to your liking to get the best result. I have a medium to large head and medium size ears for a reference. Nevertheless, I think the best way to tweak the settings is to forget about your head size and ear size and just tweak it until the sound seems "natural". At least thats the way I found my settings.   

 

[castleofargh]

thanks, it's been a while since I've tried TB isone and at the time I didn't tweak it with that specific problem in mind. I'll try that when I come back home in a week or 2.

 

[sailboarder]

I have the exact same problem OP has.
 
I have no problem hearing a great stereo image from my speakers, but the same stereo material on headphones comes in an arc, left, top and right.  There's never anything forward with headphones for me.
 
I have been searching for a while for a solution, or an understanding and I think I'm now much closer, due to this interesting thread!
 
I learned that height perception is based on the comb filtering caused by the pinnae, but also that for some subject, getting the same signal from the 2 hears will give a sound coming from top.
 
Well, my face and my head are not symmetric.  I'm not a monster at all, people don't notice until I show them, but clearly not symmetric.  For a real frontal source, I am sure the comb filtering will be different from both ears and my brain process that as centered.  Now, if I'm fed two identical signals L and R, I'd guess my brain processing is confused and defaults to the top position.  I also think the signal from the back works well because this comb filtering doesn't happen as much
 
Does this make sense?  I'm no expert in the field, but I also find surprinsing that transfert functions are always symmetric...

 

[castleofargh]

wow thanks for reminding me of this. I've tried a bunch of other things about signatures since, but outside of being able to "push" the image more forward by pushing mids or stuff like that, I got nothing of significance to make the elements come down. in the end what has the most impact is forcing myself to think that the singer is in front of me. I close my eyes, focus my mind on a point, and the singer kind of moves toward it ^_^.

 
but of course it's not the most relaxing way to listen to music.
 
 
my next step will be to try and make an equal loudness contour of both ears, measure my headphone, and try to apply the same left right variation to see if it does something. feels like only a partial answer, but who knows?(not me)
still I have an audio todo list with stuff I promised I'd do last august

. so don't expect a result next week. castleofargh, the only guy who manages to get overwhelmed by a hobby.

 

 

[icebear]

Hi castleofargh,
 
have you ever used the HD800? Do you have the same lack of localization with this one?
I have a strange "inside the head" localization effect (in the middle between my ears) with most other headphones and it drives me nuts, I just can't stand this effect.
No problem at all with the HD800. Proper front sound stage and exact localization around the head with e.g. Chesky binaural recordings ... and the barber scissors for that matter.

 
Apart from the shape of your ears, which shouldn't matter too much as you grew up with them and the brain learned how to process the signals captured and pieces the location info together correctly when you listen to your environment and not to recorded music, there is one crucial component to be able to have proper localization : Resolution of the spatial info. That includes the recording in the first place. Only use live recordings with acoustic instruments when trying to judge, if you perceive the location of the sound source correctly. The recording must have captured echo and time differences properly and the playback chain with the headphones at the end must be able to transmit all this spatial info. If there is some loss along the way, then it will be more and more difficult for your brain to process the incoming signal and make a decision where the sound is coming from. Just my $0.02.

 

[castleofargh]

yeah hd800 is the same, most impressive distance, but the direction of mono sounds still goes up for me.
I've tried a lot of IEMs and headphones while thinking about this, and while I have different perceptions that seem to correlate with what people experience, it looks like my panoramic circle is tilted upward in front.
 
I've tried binnaural recordings, or just playing music in mono, even the razer software that lets you set up your own 3D perception to try and improve game imaging, where you get a sound and you adjust where you feel you're hearing it on the screen. everything works like a charm, but front stuff slowly go up as they approach center. it seems to be the state of things for me on headphones/IEMs.
and like sailboarder said, I don't have any of this when using speakers. that's why I'm starting to think that what RRod mentioned about mono sounds sometimes perceived as having a vertical component is my problem. my real life hearing could be imbalanced in some ways because of my body and when I get real mono from headphones, my brain only knows of stuff from above being mono? not too far fetched when I think about it.
if I could find some couplers made to go in the ear canal that don't cost an arm, I could verify that(and do so much more fun stuff), but all I've found that was made for that purpose cost several hundreds of dollars. and I suck at DIY stuff.
 
I write this and realize that my todo experiment on my previous post makes no sense. if I don't experience the effect on speakers where obviously any hearing imbalance would still be present, then compensating hearing imbalance isn't going to solve anything. it's fun I'm brain storming with myself. 

 

 

[sailboarder]

   
 
Apart from the shape of your ears, which shouldn't matter too much as you grew up with them and the brain learned how to process the signals captured and pieces the location info together correctly when you listen to your environment and not to recorded music, there is one crucial component to be able to have proper localization : Resolution of the spatial info. 

 
I think this is all wrong...  We agree that the brain learned to process the signal captured.  However, the height info comes from the exact shape of the ears, from comb filtering.  Now when we use headphones, the influence of the shape of the ear is lost.  If the HRTF used to replicate this is not similar to the user's, the brain will not be able to interpret it correctly.  You are then pasting new ears on the listener and the brain is lost.
 
I read more on the subject and it seems most people have symmertric HRTF, while others don't.  I wonder if I could fix the issue if I was able to measure mine.  I have access to an anechoic chamber, but I am to much of a newbie to do it all.

 

[icebear]

   
....Now when we use headphones, the influence of the shape of the ear is lost.  ...

I think this is all wrong...

The shape of the ear is irrelevant if you use in ear monitors but not with headphones.

 

[castleofargh]

the ear shape for vertical cues may be relevant for both IEM and headphones IMO, with headphones we do have a directional cue for that part of the ear, but a wrong one as sounds are coming from the side, so I don't know if it's any help at all.
frequency response was my main idea, because ITD for frontal sound shouldn't really matter right(maybe I'm missing something here)? but then with all the headphones/IEMs I've tried(+ all the EQ), I expected a tendency to pop out and it didn't(for me at least).
I have something like this saved from when I fooled around with frequencies some time back:
 

this needs small changes depending on the IEM signature obviously, and I can't really tell if the boost should be more around 5khz or as I have it here more around 2khz. but there is something going on for me in that area. even on IEMs with an already heavily boosted 3khz(as it should be).
both points are set to affect only the mid(as in mid/side EQ system) to avoid ruining absolutely everything in the track, but I could say it brings as many problems as it tries to solve. some voices end up pushed further away from me which always seems to help, but man what a poor choice of signature.
 
one of my hypothesis was that sounds were just coming from inside(as in not far enough from my head to feel outside of it), and my brain would only find one reasonable explanation to this, "above". doesn't seem completely stupid.
but even with stuff that seem to sound more forward, + crossfeed, or even full mono output, I still end up with some sounds higher than others and most will still make me look up to try and find them.
maybe speakers works for the simple reason that I see them being physically "over there" and my vision dominates my senses so strongly that I need to see the source of the sound in front of me to believe it exists in front of me?
 
 
 
I think I'll have to go and try the smyth realiser someday if only to try and answer a few of those questions.

 

[sailboarder]

  I think this is all wrong...

The shape of the ear is irrelevant if you use in ear monitors but not with headphones.

With headphones, the shape of the ear is not put to good use either.  The sound comes from the side and is not far from getting straight in the ear canal. But even if the pinnae is imposing it's signature, it will be for sounds fired on the side, horizontally.  And I hear them like they're coming from the top...