What is "Sound Stage?"

[fas42]

The excerpts from the microphone book? I scanned it; didn't find anything about capturing vertical information for playback. Did you?

Tim

Well, obviously pages are missing, but at one point the text states "... approach is compatible with ... if the height information is appropriately mixed into the playback array". In some magical way this vertical data must have been captured, which rather obviously occurs because there are pictures of mic setups with cartridges pointing up and down. Over the several pages there the words "up", "down" occurriing several times with respect to capturing sound in that axis.

So perhaps not the best first read, but the introduction in this paper, http://www.google.com.au/url?sa=t&rct=j&q=eigenmike&source=web&cd=50&ved=0CNYCEMUBMDE&url=http%3A%2F%2Fdocs.google.com%2Fviewer%3Fa%3Dv%26q%3Dcache%3AhlCv-PztZn8J%3Ausers.cecs.anu.edu.au%2F~thush%2Fugstudents%2FMCTChanThesis.pdf%2Beigenmike%26hl%3Den%26gl%3Dau%26pid%3Dbl%26srcid%3DADGEESjbfAKxV4jZXCbg4MG7jW7-zbJ2W89pzfYaZ_bfMGleK32toYQIZim0gQHcKsEfALmr2Wx1TZEDd_RiK4ZhJ-OgVLDXHqRFVwGkTCzbxPLWPfxjlpbGXBT5XynAudwSMur276Dw%26sig%3DAHIEtbRQa5bD8eIlicctIwmFxF4D1-xjKw&ei=Ou1AT6GkMs-jiAe9--zmBA&usg=AFQjCNFhvCcux7BZXnyVkDCZUlnXgTShOw, Theory and Design of Higher Order Field Recording, gives a nice overview.

In terms of understanding why this in fact works, the Wikipedia article: http: //en.wikipedia.org/wiki/Human_echolocation, gives some pretty good clues ...

Frank

 

[Phelonious Ponk]

Well, obviously pages are missing, but at one point the text states "... approach is compatible with ... if the height information is appropriately mixed into the playback array". In some magical way this vertical data must have been captured, which rather obviously occurs because there are pictures of mic setups with cartridges pointing up and down. Over the several pages there the words "up", "down" occurriing several times with respect to capturing sound in that axis.

So perhaps not the best first read, but the introduction in this paper, http://www.google.com.au/url?sa=t&rct=j&q=eigenmike&source=web&cd=50&ved=0CNYCEMUBMDE&url=http%3A%2F%2Fdocs.google.com%2Fviewer%3Fa%3Dv%26q%3Dcache%3AhlCv-PztZn8J%3Ausers.cecs.anu.edu.au%2F~thush%2Fugstudents%2FMCTChanThesis.pdf%2Beigenmike%26hl%3Den%26gl%3Dau%26pid%3Dbl%26srcid%3DADGEESjbfAKxV4jZXCbg4MG7jW7-zbJ2W89pzfYaZ_bfMGleK32toYQIZim0gQHcKsEfALmr2Wx1TZEDd_RiK4ZhJ-OgVLDXHqRFVwGkTCzbxPLWPfxjlpbGXBT5XynAudwSMur276Dw%26sig%3DAHIEtbRQa5bD8eIlicctIwmFxF4D1-xjKw&ei=Ou1AT6GkMs-jiAe9--zmBA&usg=AFQjCNFhvCcux7BZXnyVkDCZUlnXgTShOw, Theory and Design of Higher Order Field Recording, gives a nice overview.

In terms of understanding why this in fact works, the Wikipedia article: http: //en.wikipedia.org/wiki/Human_echolocation, gives some pretty good clues ...

Frank

Frank, I'm trying hard not to lose patience, but you keep pointing me to lengthy documents that you don't appear to have read yourself. Have you found anything that specifically refers to using microphones to capture vertical information for playback in stereo, or not? Link me to the specific page, or better yet, give me a quote. Or just concede that you haven't found anything that addresses that specific issue, the specific issue in this conversation.

Tim

 

[Soundproof]

This is the worst kind of "research" imaginable, Frank. It is comparable to someone with a regular microscope claiming it can discern electrons because the Hubble space telescope can see extremely distant galaxies. The argument becomes ludicrous.

You can not extrapolate that there is height information in regular two-channel recordings just because people are trying to find out what is required to achieve depiction of verticality in non-stereo systems. They do this because two-channel lateral stereo can't.

 

[NorthStar]

Plus the one at 299

http://www.soundfield.com/soundfield/soundfield.php

Rob

Interesting read that gives us a good insight on possible future recording mic technics (3D).

* Reminds me of the four-capsule measurement Trinnov setup microphone..
If I'm not mistaking it measures in both the horizontal and vertical planes.
{Trinnov Optimizer measures not only in the frequency response and time domain (distances, dimensions, and reflections) but it also measures and compensates for the room's 3D spatial characteristics.}

______________________

- If music (& movies) recording techniques can be improved,
then perhaps the Room EQ systems have an easier workload afterwards.

I believe that fifty years from now we will look back and find how primitive we were ...
On the recording techniks and sound propagation in our home's soundstaging.
Better mic developments (quality, + for various more specific utilities) will of course ensued.
...All in the name of realistic surround envelopment with a natural soundstage where all sounds emanating from the musicians' instruments and the venue's space will be reproduce in their full glory with that 3D spatial integrity. And the height of that venue will also be represented accurately (fairly), and so will the 'profondeur' (depth).

There are very few music halls or concert venues in the world that sound acoustically good;
in the vast majority of cases our home sound systems sound much better.

Audio (Music/Movies) is becoming more and more a personal involvement ...
Give us the tools and we'll learn how to use them ourselves.
But we do need the audio technical engineers for developing these right tools indeed.
And also the music lovers with an ear for realistic and natural pleasure; and the first people have to be of course the musicians who are aware and willing to perfection their sound art, with the recording/mixing sound engineer's contribution, through their inventive experiments from improved microphone recording technics.

Professional Sound Mixing/Recording - Soundstage - Guitar Player - ..;
you guys read those Audio/Sound magazines?

________________

Just an added throw in for good measure (on Room EQ):
* http://en.wikipedia.org/wiki/Digital_room_correction
* http://www.audiosignal.co.uk/Resources/Digital_room_equalisation_A4.pdf

-> Always helps on improving furthermore our sound stage. :b

 

[NorthStar]

Well, obviously pages are missing, but at one point the text states "... approach is compatible with ... if the height information is appropriately mixed into the playback array". In some magical way this vertical data must have been captured, which rather obviously occurs because there are pictures of mic setups with cartridges pointing up and down. Over the several pages there the words "up", "down" occurriing several times with respect to capturing sound in that axis.

So perhaps not the best first read, but the introduction in this paper, http://www.google.com.au/url?sa=t&rct=j&q=eigenmike&source=web&cd=50&ved=0CNYCEMUBMDE&url=http%3A%2F%2Fdocs.google.com%2Fviewer%3Fa%3Dv%26q%3Dcache%3AhlCv-PztZn8J%3Ausers.cecs.anu.edu.au%2F~thush%2Fugstudents%2FMCTChanThesis.pdf%2Beigenmike%26hl%3Den%26gl%3Dau%26pid%3Dbl%26srcid%3DADGEESjbfAKxV4jZXCbg4MG7jW7-zbJ2W89pzfYaZ_bfMGleK32toYQIZim0gQHcKsEfALmr2Wx1TZEDd_RiK4ZhJ-OgVLDXHqRFVwGkTCzbxPLWPfxjlpbGXBT5XynAudwSMur276Dw%26sig%3DAHIEtbRQa5bD8eIlicctIwmFxF4D1-xjKw&ei=Ou1AT6GkMs-jiAe9--zmBA&usg=AFQjCNFhvCcux7BZXnyVkDCZUlnXgTShOw, Theory and Design of Higher Order Field Recording, gives a nice overview.

In terms of understanding why this in fact works, the Wikipedia article: http://en.wikipedia.org/wiki/Human_echolocation , gives some pretty good clues ...

Frank

WoW Frank, what a great link! :b

 

[bblue]

WoW Frank, what a great link! :b

It doesn't address anything pertinent, though. It's about spherical microphone arrays for use in teleconferencing (primarily) to create a wrap around audio ambience to go hand in hand with video. As I read it (much was way over my head) a sphere of mics can mean 45 or more, with a mathematically calculated output (encoder) spanning at least three audio channels and a usable response (at this level of description) of 200Hz - 3400Hz. Presumably in each playback environment there would be a three channel decoder and some large number of speakers to reproduce the recorded environment.

Nor does echolocation, as that is a principle of ears/brain interpretation, with each ear being a multi-directional device capable of individually revealing directional cues based on direction of sound entry to the ear canal. It's interesting, to be sure, but not relevant to Stereo two channel sound.

Ambisonics and similar technology uses multiple microphones in a (usually) tetrahedral array into an encoder and on to multichannel to record. It then requires a decoder on the playing end.

All of above are a different technology than stereo recording and playback, but are significant because they try to address the LACK of height and spherical detail preservation of stereo recording.

--Bill

 

[NorthStar]

Bill, me I find the read interesting for my own personal use.
For me it is educative, and I also find it relevant enough.
And I fully respect what your personal impressions are from it, and that you just posted above.

No two people love the same woman. ...Some' like that anyway. :b

______________________________

Tim, I posted this link before: http://www.linkwitzlab.com/accurate stereo performance.htm

=> Very simple: Go to B8 regarding Height.

* Also: some stereo recordings, where the two loudspeakers, when properly positioned, perform a disappearing act, completely; and we are left with an holographic image from a sonic stage with Width and Depth (and sometimes a sense of Height).

Height is not the last bastion from the overall soundstage.
We shouldn't put too much emphasis on it, I think.

Nobody hears the same as the next person to us.
And some people's brains are more apt at deciphering, constructing, perceiving Height in music listening.
And no two people have the exact same room.
And we try to match a pair of loudspeakers to our own set of ears. :b
Then we organise them to the room (or the room to them).
And finally, we all have our own music genre preference, and our medium(s).
And intensity level ...

 

[Soundproof]

It doesn't address anything pertinent, though. It's about spherical microphone arrays for use in teleconferencing (primarily) to create a wrap around audio ambience to go hand in hand with video. As I read it (much was way over my head) a sphere of mics can mean 45 or more, with a mathematically calculated output (encoder) spanning at least three audio channels and a usable response (at this level of description) of 200Hz - 3400Hz. Presumably in each playback environment there would be a three channel decoder and some large number of speakers to reproduce the recorded environment.

Nor does echolocation, as that is a principle of ears/brain interpretation, with each ear being a multi-directional device capable of individually revealing directional cues based on direction of sound entry to the ear canal. It's interesting, to be sure, but not relevant to Stereo two channel sound.

Ambisonics and similar technology uses multiple microphones in a (usually) tetrahedral array into an encoder and on to multichannel to record. It then requires a decoder on the playing end.

All of above are a different technology than stereo recording and playback, but are significant because they try to address the LACK of height and spherical detail preservation of stereo recording.

--Bill

Unfortunately, this kind of "any factoid that fits" research is endemic in audiophilia, when one tries to fit supposition to the possible - and is why I used the Hubble Space Telescope analogy above. Ambisonics is a system that places loudspeakers corresponding to a lateral and vertical (often spheroid) microphone array, in order to recreate a soundscape that lateral 2-channel stereo is incapable of. Referring to Ambisonics as proof of some "height perception" in a regular stereo system just makes the referrer seem less than serious.

So - no - it's not a great link, it's silly to link to it, particularly when one doesn't put in an iota of effort into indicating whatever spurious relevance one feels there is.

"This car flies."
"Without wings?"
"Yes, it doesn't need wings!"
"How do you know the car can fly without wings?"
"Haven't you ever seen an airplane fly?"

 

[NorthStar]

Perhaps we should wait for Frank to give us the scoop on his own evaluation,
and the personal why of his links posting. I got a pretty good idea so far ...

Because right now we are simply giving our own take from our own perspective.
One is happy, three not particularly fond of them (links).

_______________

Anyway, let's keep it forward, and not be so addicted to only one aspect of soundstage (height),
in particular as there are many many equations in that overall space. :b

 

[NorthStar]

Listening Sessions ...

I am extremely relaxed right now and I'm listening again to:
Dead Can Dance - Into The Labyrinth from the CD (4AD record label: CAD 3013 CD).

* Now for all who have the CD or LP versions, please do this:

- Listen to track #1 and from 2:45, give special attention. ...Listen if you can hear the height.
- Check track #9. ...It is even more obvious.
- And listen to tracks #10 & #11 as well.
{And of course check the width & depth of that huge soundstage.}

-> And when you're done, please return here and share your findings.
[From your own room => most important.]

_________________________

Next I'm going to listen to: David Johansen and the Harry Smiths,
from the Chesky CD (96/24), JD196.
- It is a natural recording and I'm going to see if I can hear 'height' on this.
And also to Christy Baron - Steppin', CD (96/24), JD201.
And Clark Terry - One On One, CD (96/24), JD198.
Then more Chesky selections on SACDs and also on DVD/SAD (Super Audio Disc with true 96/24).

Then some Stereo DVD-Audios at 192kHz/24-Bit high resolution audio.

Then I got few more music selections to check out;
more from Chesky, as well Fathom, and Astralwerks, DMP, Sheffield Lab, Opus3, ECM, Reference Recordings, Channel Classics, Water Lily Acoustics, APO (Analog Productions Originals), AQ (AudioQuest Music), all on Stereo CDs & SACDs.

And another day (or perhaps later on today if I feel like it, or relaxed enough);
- EELS - Daisies of the Galaxy, from Dreamworks Records.
- The Art Ensemble of Chicago - Fanfare for the Warriors, on Koch Jazz Records [HDCD].
- Mark Isham - Miles Remembered: The Silent Way, on Columbia Records [HDCD].
- Aimee Mann - Lost in Space, on Superego Records/United Musicians.
- AIR - 10,000 Hz Legend, from Astralwerks Records.

>>> The Goal: SoundStage with Height, Depth, & Width. ...Natural if possible, and from the recordings themselves in collaboration with my room, my speakers, my gear, and my ears (brains included).
And everything from strictly Stereo listening (only two loudspeakers; front Left & Right).

 

[NorthStar]

Oh one last thing: analog connections from my digital sources. :b

Unless specified, like DSD direct from HDMI (for SACD Stereo for example).

 

[microstrip]

The excerpts from the microphone book? I scanned it; didn't find anything about capturing vertical information for playback. Did you?

Tim

Tim,
IMHO, I think we are looking in the wrong books. Unhappily I also do not have the time to research it appropriately, but I could confirm that the way we localize in height is quite different from the usual 2D localization. I think that this article and references wherein can be a goo track.

From the article The Cocktail Party Problem by Simon Haykin and Zhe Chen
Adaptive Systems Lab, McMaster University, Hamilton, Ontario, Canada L8S 4K1

2.1 “Where” and “What.” The mechanisms in auditory perception essentially
involve two processes: sound localization (“where”) and sound
recognition (“what”). It is well known that (e.g., Blauert, 1983;Yost&Gourevitch,
1987; Yost, 2000) for localizing sound sources in the azimuthal plane,
interaural time difference is the main acoustic cue for sound location at low
frequencies, and for complex stimuli with low-frequency repetition, interaural
level is the main cue for sound localization at high frequencies. Spectral
differences provided by the head-related transfer function (HRTF) are the
main cues used for vertical localization. Loudness (intensity) and early reflections
are the probable cues for localization as a function of distance.


BTW, I found that most of the research on these aspects predates the internet and is not available on-line or in electronic form - you have to get the paper versions in libraries.

BTW2, The cocktail party problem is really an interesting problem. It was formulated in 1953 and people are still debating and publishing about it. We should not feel bad because after less than 2 years post WBF creation we have not found how height information can (or can not) be included in sound reproduction!

 

[fas42]

This is the worst kind of "research" imaginable, Frank. It is comparable to someone with a regular microscope claiming it can discern electrons because the Hubble space telescope can see extremely distant galaxies. The argument becomes ludicrous.

You can not extrapolate that there is height information in regular two-channel recordings just because people are trying to find out what is required to achieve depiction of verticality in non-stereo systems. They do this because two-channel lateral stereo can't.

Well, the last time I looked this forum wasn't meant to be an extension of "Nature" magazine, I apologise if I misunderstood its intent. If we doing to indulge in a solid round of academic argie bargie, let me to request something! Where is the research that very specifically has "proven" that phase information encoded in a stereo recording can't be be used by listeners to precisely locate the vertical position of a sound source. And no "all researchers know such a thing is impossible" BS responses please!! Where is the research that verified that, please ...

Frank

 

[fas42]

BTW2, The cocktail party problem is really an interesting problem. It was formulated in 1953 and people are still debating and publishing about it. We should not feel bad because after less than 2 years post WBF creation we have not found how height information can (or can not) be included in sound reproduction!

Thanks for that input, micro, and Bob, thanks, too. I have come across quite a bit of academic "we don't really understand this" and "more research needs to be carried out on how the ear/brain system works" in my looking around. The trouble is, some people just want to be spoonfed with easy yes/no answers to it all, it seems ...

Frank

 

[NorthStar]

If any member here at WBF have some difficulty to read me (remember, I'm French);
please, just ask me in the open. In any threads (part of the forums) at all.
It would be my immense pleasure to correct myself with your help, if sometimes required.
I'm here to have the most pleasant time with EVERYONE, without any single exception.
Thank you for your time & cooperation. :b

Bob

___________

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

P.S. My pleasure Frank, & anytime. :b

 

[Phelonious Ponk]

Well, the last time I looked this forum wasn't meant to be an extension of "Nature" magazine, I apologise if I misunderstood its intent. If we doing to indulge in a solid round of academic argie bargie, let me to request something! Where is the research that very specifically has "proven" that phase information encoded in a stereo recording can't be be used by listeners to precisely locate the vertical position of a sound source. And no "all researchers know such a thing is impossible" BS responses please!! Where is the research that verified that, please ...

Frank

I can't speak for Soundproof, Frank, but I am certainly not asking you to write me an academic paper. I'm merely asking you to point to something specific and relevant instead of posting links here to massive homework assignments that run down blind alleys toward an altogether different subject. And I'm not going to try to prove a negative for you because you failed to even point to a single bit of relevant evidence to back up your statements, either.

Tim

 

[NorthStar]

Tim, at least Frank tried. :b

 

[fas42]

I can't speak for Soundproof, Frank, but I am certainly not asking you to write me an academic paper. I'm merely asking you to point to something specific and relevant instead of posting links here to massive homework assignments that run down blind alleys toward an altogether different subject. And I'm not going to try to prove a negative for you because you failed to even point to a single bit of relevant evidence to back up your statements, either.

Tim

My original post was :

As an aside, googling microphones and "height information", it appears that sound engineers have absolutely no problem with using two mics sitting next to each other pointing at different angles "magically" capturing all the height info you want, but as soon as you use just one mic to do this, all bets are off. It's the first time I understood that mics have blinkers on, anything not coming directly in front of them is completely invisible -- I'm amazed that conference mics work at all ...

Do you have a problem with that bit?

If you don't, then the next question is what to do with the 2 or more channels, tracks, of that sound information: keep them all separate and send them back individually to speakers -- one mic, one speaker, or mix them in some fashion, create an audio soup of information on a lesser number of tracks. And however you decide to do that, how do you spit the information back to the listener, where do you put the speakers, etc. In other words, there are a million ways to skin the cat, which is what all the research is fiddling around with.

And I still haven't heard of anything of experiments testing whether people can detect vertical information in stereo sound. Or, perhaps the wise old men have something in their pants that tells them everything they need to know ...


Frank

 

[NorthStar]

Earlier today I was listening to David Johanson and the Harry Smiths, from Chesky record label.

First the CD from the analog connections, and then the SACD (Stereo) from the digital HDMI connection (DSD direct bitstream, not downconverted).

I listened attentively and I found real good depth in the soundstaging, like if I was there at the time of the recording.
David's main vocals were firmly anchored in the center, between my two 'flankers',
and his voice was higher than the instruments, but not by much; perhaps a few feet or so (2 or 3).

That recording started very nicely, but near the middle there were few tunes I wasn't fond of (too piercing from David's vocals), and then it settled down at the end.

If you want to do some critical listening you have to relax, and no listening sessions longer than few hours per day. This is how you'll be in top performing condition (ears & brains) for some critical music listening.

Tomorrow I'll do some more, and the day after, and so on ...
And I will share in all honesty my main impressions from my listening sessions with you guys from anything I remark (hear) in that soundstage (width, depth, and height included); for that holographic 3D spatial imaging sense).

I already talked a lot; now I'm in the listening mode 'next gear'.
Talk talk, walk walk, listen & results!

Meanwhile you can also do your own experimenting; and it's good for the ears and also for the soul, and from your own favorite music selections.
I'm on the ACTIVE audiophile mode when I'm doing this, and I feel good;
I feel better than James Brown. :b
How do you feel?

 

[Soundproof]

Frank, now you are quoting and highlighting your own erroneous assumption and asking whether Tim has "any problem with that."
I hope he does, as that is quite something.

And as to the "Cocktail Party Problem" I do trust that microstrip is aware it's the ears they are discussing, and not microphones? And I gather you are all aware of the role of the pinnae in localization? As well as the head-related transfer function - both of which one has tried to simulate with binaural heads, btw. But it turns out our pinnae are individual, and that each of us adapts to their shape when learning to localize, through early trial and error when we're young. (It's also best when they are symmetrical, which is not always the case).

So - on this page alone a quote about how humans hear, and a reference to an erroneous and unsupported conclusion. And absolutely nothing about how microphones used in hifi-recording hear height in addition to width.