The No.1 Website for Pro Audio
 Search This Thread  Search This Forum  Search Reviews  Search Gear Database  Search Gear for sale  Search Gearslutz Go Advanced
Why don't we worry about vertical imaging? Virtual Instrument Plugins
Old 4 days ago
  #1
Gear Head
 

Thread Starter
Why don't we worry about vertical imaging?

There is a LOT of emphasis placed on the stereo image in the horizontal plane. In both the recording and the reproduction aspects of audio, people work very hard to get this right. What amazes me is that I can barely even find a reference to vertical image placement.

A few years ago I was listening very carefully to a well known historical vocal performance through a very excellent stereo system. I was puzzled because the sonic image simply did not match my mental image. I expected the singers vertically above the orchestra, but my perception kept demanding that this was not the case. The orchestra seemed to be "above" the singers. Months or years later I saw a photograph of that session, and to my astonishment, the singers were in what I would normally think of as the orchestra pit and the orchestra was above them, on the stage! In other words, the sonic image had correctly communicated to me the vertical placement of the sound sources. I don't even remember which performance this was, and I certainly have never understood the very unusual placement, but that's beside the point.

The whole experience has made me wonder why we don't seem to care about imaging in the vertical. I suppose that part of this must have to do with the nature of stereo reproduction. There are two speakers, whose design varies widely, and I'm sure that not all of them - horns, for example - are well suited to producing an accurate vertical image. But it seems that there's more to it than that... Thoughts?
Old 4 days ago
  #2
Lives for gear
 
Unclenny's Avatar
I have always thought of the vertical component as being simply gain.

Curious to hear some thoughts on this.
Old 4 days ago
  #3
Lives for gear
 

I hear the vertical as being strictly a frequency response issue, even when speakers may only have one speaker. Guess its from the years working with pro PA gear cabs stacked up. You'd have the bass bins on the bottom, mids center and horns on top. Even most Hi Fi Speakers usually have Tweeters on top, mids center and Woofers on the bottom. Reason being is tweeters tend to project in a straight line and the woofers are producing wave lengths so wide sometimes that don't even fully form so you're dealing with 1/2 and 1/4 wavelengths. Bass cabs and speakers are usually larger and much heavier so they are usually placed on the bottom.

Anther reason is bass vibrations travel along the ground or through the floor to your feet. I used to love playing gigs on a wooden stage because you could feel the bass and kick and keep a solid groove playing with others.

So when I mix, the stereo width is a matter of panning, Height is mostly frequency response and depth, the third dimension is a combination of frequency response and time based effects like chorus, reverb and echo. The time based effects have to be combined with frequency roll off because sound looses treble and bass as it travels over a distance till there's usually nothing but midrange left. The Reverb/reflective sounds of the time based effects can be EQed separately too to produce rooms made from different materials. Wood will reflect differently then stone or sheetrock.

The balance between absorbed and reflected frequencies takes a bit of brain power to grasp. You have to exercise your imagination and place yourself in an actual room when you mix and have the abilities to change not only the size and depth of that room but to change its construction materials and instrument placement within that room. That ability doesn't just appear some day, you have to train yourself to think three dimensionally in order to produce three dimensional aural images at will in any shape of form needed.

You can gain allot of insight by cross referencing an acoustic perspective with a visual perspective, especially since sound waves are invisible like light beams are.



Think of how this would sound as you move an amp farther away from the front of the stage. If that musician down the hall going to sound the same as someone right in front of you, even if the room reverberates? How does pre delay and frequencies change when close or far, left or right, top to bottom?

Old 4 days ago
  #4
Lives for gear
 

Quote:
Originally Posted by blwatlongwood View Post
The whole experience has made me wonder why we don't seem to care about imaging in the vertical. I suppose that part of this must have to do with the nature of stereo reproduction. There are two speakers, whose design varies widely, and I'm sure that not all of them - horns, for example - are well suited to producing an accurate vertical image. But it seems that there's more to it than that... Thoughts?
It's not that the speakers aren't suited for vertical imaging, it's that there's only two of them in a stereo system. All you have to do to see where the problem is is imagine how the horizontal image is created in a stereo system.

A signal that only exists in the left speaker is perceived as coming from the left speaker. A signal that is going into both speakers at the same time at the same level is perceived as coming from between them, and if you are seated with the same distance to both speakers it'll be the "center" between them - the "phantom center". And any other horizontal position between fully left and fully right will be a combination of delay/level differences...

But you don't have that with vertical imaging, because you only have the two speakers on the horizontal plane. If you had a 'standard' where both speakers were right in front of you but one was on top of the other then you'd have vertical imaging, with a phantom vertical center - but no left/right.

Yes, people are talking imaging and it typically falls into two categories; either using frequency to imply that high/low relationship, or post-production audio, meaning film and TV and the newer "3d audio" standards such as Auro3D or Dolby Atmos or binaural imaging for games and VR/AR. For VR/AR you use a transform to create a response that is similar to real life but played back in a stereo headphone configuration. For Auro3D/Atmos you end up having speakers either in the ceiling (Atmos) and/or on a second surround plane (Auro3D).
Old 4 days ago
  #5
Gear Guru
 
Brent Hahn's Avatar
 

Dolby Atmos' main attraction is that it has "up."

It's funny to me how so many people hear records in a starkly visual way, and talk about "stereo images" and "soundstages" and such, and how left/right matters to them. I could be fine in a mono world. But at the same time I'm really keen on getting the right sense of forward vs rearward. And the OP is just as keen on up vs. down. Go figure.
Old 4 days ago
  #6
Gear Maniac
 
Poinzy's Avatar
 

Quote:
Originally Posted by blwatlongwood View Post
There is a LOT of emphasis placed on the stereo image in the horizontal plane. In both the recording and the reproduction aspects of audio, people work very hard to get this right. What amazes me is that I can barely even find a reference to vertical image placement.

A few years ago I was listening very carefully to a well known historical vocal performance through a very excellent stereo system. I was puzzled because the sonic image simply did not match my mental image. I expected the singers vertically above the orchestra, but my perception kept demanding that this was not the case. The orchestra seemed to be "above" the singers. Months or years later I saw a photograph of that session, and to my astonishment, the singers were in what I would normally think of as the orchestra pit and the orchestra was above them, on the stage! In other words, the sonic image had correctly communicated to me the vertical placement of the sound sources. I don't even remember which performance this was, and I certainly have never understood the very unusual placement, but that's beside the point.

The whole experience has made me wonder why we don't seem to care about imaging in the vertical. I suppose that part of this must have to do with the nature of stereo reproduction. There are two speakers, whose design varies widely, and I'm sure that not all of them - horns, for example - are well suited to producing an accurate vertical image. But it seems that there's more to it than that... Thoughts?
I think if people had ears on the tops of their heads and underneath their jaws, "vertical imaging" might get more attention.
Old 4 days ago
  #7
Lives for gear
 

Quote:
Originally Posted by Poinzy View Post
I think if people had ears on the tops of their heads and underneath their jaws, "vertical image" might get more attention.
We do perceive vertical positioning though, it's just that it isn't as clear as horizontal positioning. Our skulls, shoulders, ears etc all filter the incoming signals differently depending on position and our brains then decode that to determine position, even vertically.

I'd guess that there hasn't been the same evolutionary advantage to have as high 'resolution' vertically as horizontally since most of our foes have been on the ground. Predators we worried about in the past were mostly on our plane, not in trees or in the sky. But that said, of course there's some benefit to resolving the vertical plane at least somewhat.

What's interesting though on that note is that the format competing with Dolby Atmos makes the point that as we move 'higher' and 'higher' our 'resolution' drops, and it's therefore more beneficial to have a second surround layout above the first. So a 7.1 system above a 7.1 system, and pretty much nothing exactly overhead (although it was eventually expanded). This way, if you're watching a movie where the knights are about to storm a castle, you get the ground level on the first level 7.1, and the people on the wall on the higher level. You don't get the same pin-pointing in Atmos because you "only" have overheads, not a second level in the front. So it ends up being a "phantom elevation" in Atmos.
Old 4 days ago
  #8
Quote:
Originally Posted by mattiasnyc View Post
What's interesting though on that note is that the format competing with Dolby Atmos makes the point that as we move 'higher' and 'higher' our 'resolution' drops, and it's therefore more beneficial to have a second surround layout above the first. So a 7.1 system above a 7.1 system, and pretty much nothing exactly overhead (although it was eventually expanded). This way, if you're watching a movie where the knights are about to storm a castle, you get the ground level on the first level 7.1, and the people on the wall on the higher level. You don't get the same pin-pointing in Atmos because you "only" have overheads, not a second level in the front. So it ends up being a "phantom elevation" in Atmos.
I would have to hear it to be convinced. It seems like a bunch of marketing hype... but I'll admit, it's intriguing. Our brains can process both horizontal and vertical planes for incoming sounds. Like if you're in the forest and you hear a twig snap, somehow you can tell by the sound if it's coming from the ground, or up in the trees. I have no idea how our brains can process the vertical placement.

As far as mixing though, it's impossible to replicate this effect. Generally, in my own mental image, I picture bass parts to be physically lower, and higher parts to be physically higher, but I think that's just my imagination.
Old 4 days ago
  #9
Lives for gear
 

Quote:
Originally Posted by AnalogOverdose View Post
I would have to hear it to be convinced. It seems like a bunch of marketing hype... but I'll admit, it's intriguing.
Convinced of what?
Old 4 days ago
  #10
Gear Head
 

Thread Starter
Quote:
Originally Posted by mattiasnyc View Post
We do perceive vertical positioning though, it's just that it isn't as clear as horizontal positioning. Our skulls, shoulders, ears etc all filter the incoming signals differently depending on position and our brains then decode that to determine position, even vertically.
Indeed. The thing that made me wonder in the first place was the revelation through demonstration that vertical images can in fact be conveyed as I related i the original post.

I guess that part of the issue is that while most stereo reproduction on the market will yield at least some moderate amount of left-right stereo imaging, my experience has been that it takes a pretty outstanding rig to accurately convey up-down images.

The point was made somewhere further up that there's also depth cues - basically the other dimension - and I think those are conveyed somewhere between the left/right and up/down dimension.
Old 4 days ago
  #11
Gear Guru
 
Brent Hahn's Avatar
 

Quote:
Originally Posted by blwatlongwood View Post
The point was made somewhere further up that there's also depth cues - basically the other dimension - and I think those are conveyed somewhere between the left/right and up/down dimension.
Check this out. It's not music, it's a radio commercial. Totally mono, mostly recorded through a phone. And "depth cues" are the be-all and end-all.

Old 4 days ago
  #12
Lives for gear
 

Quote:
Originally Posted by blwatlongwood View Post
Indeed. The thing that made me wonder in the first place was the revelation through demonstration that vertical images can in fact be conveyed as I related i the original post.

I guess that part of the issue is that while most stereo reproduction on the market will yield at least some moderate amount of left-right stereo imaging, my experience has been that it takes a pretty outstanding rig to accurately convey up-down images.
It basically can't be done as far as I can see. Just adjusting frequencies won't cut it. It's really as simple as having more playback channels to which you can send sounds. The more pin-point sources you have for reproduction the better you can localize audio. But it becomes an issue of standards and budget.

On top of that; think of every home environment you've been in where there's been even just a stereo setup - how many of them had a) the speakers set up so that there was a sweet spot, and b) treated walls/ceiling/floor? In a lot of cases speakers just go where it's convenient and doesn't look bad. So the imaging is off just from the initial positioning. Then you have the architecture of the room and the lack of acoustic treatment. So now imagine 5.1/7.1 surround systems and how poorly they're often set up. Now add ceiling speakers.....

Quote:
Originally Posted by blwatlongwood View Post
The point was made somewhere further up that there's also depth cues - basically the other dimension - and I think those are conveyed somewhere between the left/right and up/down dimension.
Well.... really just a different set of tools to convey that sense. Depth is in a way not that hard to emulate, since we know how sound changes through a medium given distance. Lightning and the thunder that goes with it is the easiest example. In movies it's almost always physically wrong, but in reality it tells us a lot about the nature of sound as far as "depth" (or distance) is concerned.
Old 3 days ago
  #13
Lives for gear
 
Owen L T's Avatar
The short-form, direct answer to the topic is: we don't worry about it, because there is absolutely nothing we can do about it! If it was an integral part of experiencing music, then technology might have evolved differently. But: 4 musicians on a stage? They're all at the same height. Sure, an orchestra has raked seating. But sitting 40 feet away from them (still pretty close) the perception of relative height on stage is approaching zero. It's just a non factor in how we experience and enjoy music.
Old 3 days ago
  #14
Lives for gear
 

Quote:
Originally Posted by Owen L T View Post
The short-form, direct answer to the topic is: we don't worry about it, because there is absolutely nothing we can do about it! If it was an integral part of experiencing music, then technology might have evolved differently. But: 4 musicians on a stage? They're all at the same height. Sure, an orchestra has raked seating. But sitting 40 feet away from them (still pretty close) the perception of relative height on stage is approaching zero. It's just a non factor in how we experience and enjoy music.
I don't agree with that actually. I think there is a difference between why 3d audio (for lack of a better term) hasn't been adopted for music and whether or not music reproduction benefits from it.

On the first point I think it's pretty simple; it's hard enough to convince people to spend money on a 5.1/7.1 surround system for film. We've been conditioned to mono for decades followed by stereo for decades. The potential benefit of listening to music in a higher than two channel count isn't obvious to the average listener, whereas it is for film. People did and do go to the movies where they almost always these days experience surround sound. So they have a frame of reference. There's an argument to be made there. But try to say that they should spend a bunch of money for a bigger system to listen to music and it becomes harder to justify.

That of course leads to the second point which is whether or not music benefits from it, and that's where I disagree with you. In my opinion it does absolutely benefit. More speakers - assuming proper production and reproduction environments - is better, at least to a degree. So if you have a concert recording of a symphonic orchestra the argument for surround is that we want the reflections from the sides/rear as well as direct sound from the front. Well, we can easily extrapolate from that that we want the "upper level" as well.

Now, at AES this year Genelec had a little mixing clinic on the floor where this engineer went through how he mixed a song for Dolby Atmos. I think the setup was 7.1.4. As many probably were I was skeptic at first. However, there's something to be said for using the elevation layer. Just as you get more 'real estate' for your instruments when spreading out into surround relative to stereo you actually also get more space when elevating, giving the instruments more room to breathe.

In addition to that, I think that aesthetically there are interesting things one can do. If you imagine the average concert we're actually not on the same plane as all the musicians, but often there's an elevated stage. Now, I wouldn't want a band hanging in the air so to speak, but by elevating some elements you do actually get a different sense of space. On top of that, from an aesthetic standpoint, we also have more conceptual aesthetics that aren't just mimicking "here's a band in a room".

So again, I think I disagree a bit. There are benefits to music in an immersive format, and the reason we aren't seeing it is because people don't want to spend the money (yet) for various reasons, and because people have been conditioned to playing back far worse quality mixes/masters in far worse conditions.
Old 3 days ago
  #15
Lives for gear
 
norfolk martin's Avatar
 

Humm....

So, with an 8 speaker system,one in each corner of a room :

if the idea of the "stereo image " from 2 speakers holds, one could theoretically locate a point on any position on the six walls? Could one also locate a point inside the room , rather than on a wall? e.g, if you started with the four back speakers and cross-faded to the four front, would the sound seem to start in the center of the back wall, pass through a listener in the center of the room , and end up in the center of the front wall? Or would the source point appear to split in two and travel down the side walls and rejoin at the front?
Old 3 days ago
  #16
Lives for gear
 

Quote:
Originally Posted by norfolk martin View Post
Humm....

So, with an 8 speaker system,one in each corner of a room :

if the idea of the "stereo image " from 2 speakers holds, one could theoretically locate a point on any position on the six walls? Could one also locate a point inside the room , rather than on a wall? e.g, if you started with the four back speakers and cross-faded to the four front, would the sound seem to start in the center of the back wall, pass through a listener in the center of the room , and end up in the center of the front wall? Or would the source point appear to split in two and travel down the side walls and rejoin at the front?
Pass "through the listener".

But again, you do make gains when having more speakers. Just as your phantom center in stereo is subject to your own position as a listener between those two speakers so does any phantom image in a higher-channel system. The whole reason there's a center speaker in theaters is because of the need to avoid a phantom image for dialog in movies. The same applies anywhere else in the room really, except the need isn't exactly the same since we get worse directional perception when we move away from our field of vision basically.
Old 3 days ago
  #17
Gear Guru
 

People interested in this topic should check out the LEDR test with your speakers.

I have the Chesky test CD with this test and I have to admit I was skeptical but it really is freaky how they can create an "up" illusion from two speakers.

It is based on pinna transforms, the way the pinna 'tells' your brain where a sound is coming from. I do an exercise with my students where I move across the front of the room and ask them to point to me with their eyes closed. I move left and right and center, obviously no problem. Then I quietly stand on a chair and they all point UP to me, even though I am still in the 'middle' of the room and my voice is still arriving 'equally' at both ears. Not one student points straight ahead. They all point up.

It may be that in the example in the OP, the recording was made using the dummy head system - which has realistic pinnae.


Quote:
Who's behind our LEDR?
The LEDR signals were generated by Northwestern University's Spatial Reverberator, a software program running on an RISC mainframe computer. The spatial reverberator was designed by researchers Gary Kendall and Bill Martens of the Computer Music Department, while the trademarked LEDR test signal is the brainchild of acoustician Doug Jones. These psychoacousticians have been researching what are called "pinna transforms," the way in which the shape of the head and outer ear permit us to hear signal direction.

By programming the filter characteristics of the pinna transforms into the spatial reverberator, the Chicagoans can literally move sound around the room, even behind the listener, using only a single pair of loudspeakers placed in front of the listener (footnote 4).

Grading the LEDR paths
The first path, Up, will amaze your friends and quiet your enemies. It's hard to believe that a sound can appear to travel from a loudspeaker up to 6' above the speaker! This path is generated first in the left speaker, then in the right. The sound should begin at about eye level and then travel as straight as possible up in the air about 6'. You should grade the system on how vertically straight the path is, how high the image goes, if it is continuous (unbroken), whether it approaches or recedes from the listener (it should not), and if the left and right paths are symmetrical.

I like to call the second path the "Rainbow," but Doug officially calls it the Over path. The sound should begin at one speaker and travel in a smooth arc to the other speaker, from left to right and then returning. The top of the rainbow should be as high as the previous Up signal (about 6' above eye level). Judge the Over path by how smooth, continuous, symmetrical, and rainbow-shaped the arc is.

The last path, Lateral, tests left-to-right stereo imaging. This consists of four elements.

First, the sound moves from left to right, between the acoustic centers of the speakers. Since a speaker's acoustic center may not be its physical center, you should use the first Lateral test to adjust your speakers until the sound traverses a 60 degrees angle from the listener's point of view. Second, the sound moves from beyond the right loudspeaker to beyond the left (about 1' out from acoustic center). The next two signals are the mirror image of the above; third, from right to left speaker, and fourth, from beyond the left to beyond the right.

Again, grade your system by how straight, continuous, and symmetrical this path is. Grade the beyond path by how far out from the speaker it appears to go (about 1' to the left or right of the requisite speaker, according to Doug Jones), and that it does not approach or recede from the listener.
Topic:
Post Reply

Welcome to the Gearslutz Pro Audio Community!

Registration benefits include:
  • The ability to reply to and create new discussions
  • Access to members-only giveaways & competitions
  • Interact with VIP industry experts in our guest Q&As
  • Access to members-only sub forum discussions
  • Access to members-only Chat Room
  • Get INSTANT ACCESS to the world's best private pro audio Classifieds for only USD $20/year
  • Promote your eBay auctions and Reverb.com listings for free
  • Remove this message!
You need an account to post a reply. Create a username and password below and an account will be created and your post entered.


 
 
Slide to join now Processing…
Thread Tools
Search this Thread
Search this Thread:

Advanced Search
Forum Jump
Forum Jump