• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

The Physics of 3-D Soundstage

levimax

Major Contributor
Joined
Dec 28, 2018
Messages
2,348
Likes
3,462
Location
San Diego

Cars-N-Cans

Addicted to Fun and Learning
Joined
May 19, 2022
Messages
819
Likes
1,009
Location
Dirty Jerzey
A member @restorer-john made the recording linked below in his backyard with a portable recorder. I hear left right, front back, and higher lower. I think the recording which retains actual spatial information is key to the image. Fancy electronics that try to mimic real spatial information never seem as convincing to me as a simple real ambient recording.

I think "birds" rather than birds would be a better title lol... They have those weird-ass Kookaburras down there. Had one of those things escape from the zoo or something and it was in the woods behind my house for a week. Made it sound like I was in Jurassic Park.
 

Ricardus

Addicted to Fun and Learning
Joined
Mar 15, 2022
Messages
843
Likes
1,153
Location
Northern GA
, panning conversely is quite restrictive and only covers one aspect of stereo imaging, namely that imposed by the mixing engineer.
Everything in the imaging of a song is the fault of the mix engineer and the choices they made.

That's the point here. There is no magic. The rest is all "audiophile" nonsense.
 

Cars-N-Cans

Addicted to Fun and Learning
Joined
May 19, 2022
Messages
819
Likes
1,009
Location
Dirty Jerzey
Everything in the imaging of a song is the fault of the mix engineer and the choices they made.

That's the point here. There is no magic. The rest is all "audiophile" nonsense.
But my contrary point is not everything someone can listen to on a 2-channel system will be mixed, or there may be other techniques used besides conventional panning. What nomenclature do we use then? Not to be contrarian, but just because a word is used by audiophiles does not mean we have to eschew using it, only that we collectively agree on what it’s meaning encompasses in the objective realm. I would view imaging as a function of the sound reproduction system, and panning as that done by the mix engineer in the process of producing music. The two can then be a subset of the soundstage we experience as end-consumers once the mixing done is heard through the system with its intrinsic imaging characteristics, which can range from "three-blobs stereo" to binaural. I suppose you could just use imaging but then one is robbed of having a concise word to use to encompass what is heard when differing mixing or recording techniques interact with how the particular sound reproduction system images.
 

kongwee

Major Contributor
Joined
Jan 22, 2022
Messages
1,024
Likes
276
For mix engineer, it is a "stage" just in front of them to place their instrument where they wanna be. Of course, for two mic setup, the stage is really defined by the record engineering skill. There is no closer term to use soundstage. Audio engineer will be in the art rather than scientist. They are making an art, not to explore the science. Even with Atmos music, we still call it soundstage. It is an area when we pull al the sound we want in a "stage". Of course, you can use imaging to represent what you wanna deal with your client or audience.
 
Last edited:

Axo1989

Major Contributor
Joined
Jan 9, 2022
Messages
2,804
Likes
2,806
Location
Sydney
I think "birds" rather than birds would be a better title lol... They have those weird-ass Kookaburras down there. Had one of those things escape from the zoo or something and it was in the woods behind my house for a week. Made it sound like I was in Jurassic Park.

It's a great selection of Australian birdlife @restorer-john has at his doorstep.

The August-September winds came late this year but the Channel-billed cuckoos finally arrived. If you think Kookaburras sound prehistoric, imagine my enjoyment the first time a brace of these birds arrived and started their 4 am chorus in the Blackbutt trees just outside (I guess if you'd just flown down to Sydney from New Guinea you'd probably make a song and dance about it).


That one's a bit solitary, although you can hear some far-off friends after the second call. They're more fun when they get together. There's some very nice soundstaging of those bird calls and a blowfly in the "rapid piping call" example on the page that video came from. It's either great stereo or quite likely binaural.

And speaking of cuckoo, there's one in every thread I guess: if anyone thinks that all there is to assembling a layered stereo image in the studio is a bit of panning, they are also stuck in some prehistoric time.
 
Last edited:
OP
T

tallbeardedone

Active Member
Joined
Sep 3, 2022
Messages
102
Likes
215
A member @restorer-john made the recording linked below in his backyard with a portable recorder. I hear left right, front back, and higher lower. I think the recording which retains actual spatial information is key to the image. Fancy electronics that try to mimic real spatial information never seem as convincing to me as a simple real ambient recording.

Hey this is a really cool recording! I hear birds way up in my ceiling. More stuff like this please! Does anyone know where I can find other ambient recordings like this with such great spatial detail? Very very cool.
 

suttondesign

Addicted to Fun and Learning
Forum Donor
Joined
Feb 4, 2019
Messages
732
Likes
1,310
Location
Bellingham, WA
I'd like to respond to the original post not to disagree with it, but add my experience with two completely different approaches to excellent speaker imaging. I've used (and am a dealer for) Dutch 8C and all the Linkwitz designs, including the 521.4. So my comparison is dipole vs. cardioid.

The Dutch 8C, when toed in, sound like what you expect when mixing a recording: pinpoint accuracy in a soundfield defined by the speaker side-to-side limit, with depth cues those from the original recording. Live pop and well-recorded classical retain a deep, convincing soundstage from whatever vantage the recording engineer intended, whether up front or some distance back. Ordinary pop recordings with emphasized vocals put the singer very far forward but stable. In all recordings, but particularly well-engineered ones, the buttery-smoothness of the frequency response above, say, 150hz or 200hz is just pleasurable beyond belief. That is even true with the speakers near the front boundary because the design controls for this. I believe this carefully-controlled dispersion, which aims to minimize early reflections, is a very successful approach, and it is better than any marketed speaker I have ever heard. I have not heard the Genelecs -- I'm trying to become a dealer, but product is so scarce that Genelec is delaying taking on new dealers. Finally, the tweeter of the Dutch is incredible -- triangles sound amazing.

Drawbacks?

1. The Dutch deep bass sounds like speakers and does not have the low distortion of a dedicated subwoofer. Ideally, you roll them off 2d-order at 45-50hz and let a devoted pick up from there.

2. The Dutch activate room modes like ordinary speakers. Correction can only go so far. My current setup has a fearful cancellation at 63hz, for example. I can tame the reinforcing modes, but they never go away.

But now take a diametrically opposed approach, the Linkwitz 521.4 dipole. The principle, which I find interesting and insightful, is that by allowing room interaction, but with a significant delay, the secondary and tertiary reflected sound causes the brain to, in essence, correct for delayed sound and construct in the listener's mind the correct soundstage as if the speaker had the carefully-controlled dispersion of the cardioid. (That is what I think is not generally appreciated about the design principle of these speakers.) But with this important difference from a cardioid: the reverberant soundfield is all there. The practical effect, for whatever reason, is to make the soundfield recede quite a bit, and vocals, even in ordinary pop recordings, are farther back. The soundstage is broad and deep, and the speakers are difficult to perceive as distinct sound sources. The sound is utterly beguiling, with a wide-open feeling. Part of the latter is that there are zero box resonances. Part of it is also that open-baffle woofers have a more natural sound. This is not what the recording engineer intended; it's what the design accomplishes to improve 2-channel recording by manipulating sound in a certain way. Siegfried was adamant about that point.

The drawbacks of the Linkwitz dipoles are significant:

1. You must have lots of boundary distance for the speakers, and the ideal version of the speaker's placement also requires a very distant rear wall, which I suspect very few people can have in an ordinary home.

2. You have to learn how to set up a complicated 8-channel-amplified setup with an external crossover. No matter how many times I set these up, it's always tricky -- and I've been involved in audio since 1978!

3. By no stretch does the Linkwitz have the buttery-smooth response of the new breed of active speakers like the Dutch, Genelec, Neumann, and KEF. I don't know why. (And yes, you critics, it could be my build, so please don't feel the need to say as much.) You can fuss with the crossover and EQ parameters all day, but unless you're an engineer, I doubt you'll get the response linearity a company with an R&D department can produce.

4. In reasonably-large rooms, open-baffle woofers cannot pressurize the space to achieve true subwoofer levels. The drivers flop like fish but can't produce high SPL below about 35 or 40 hz. I have always used a powerful sub in a corner to pick up the bottom end. This makes the system that much more complicated. But I like recordings with super-deep stuff, plus I use the speakers in a video setup.

5. In a video setup, the recessed vocals make dialog harder to discern.

Which do I pick? Linkwitz, at the end of the day, but only because I can deal with the drawbacks to make it work. For most people, most of the time, in most ordinary rooms, the Dutch 8C (or their equivalent in Genelec, Neumann, KEF) are wonderful.
 

Blumlein 88

Grand Contributor
Forum Donor
Joined
Feb 23, 2016
Messages
20,521
Likes
37,050
I would like to see some research similar to what the NRC did and was continued at Harman for dipoles only. See if there aspects to optimize for the most preferred parameters of a dipole system. I find panel dipoles to be enjoyable despite what are usually obvious flaws. Flawed monopoles just sound flawed without much enjoyment. So would dipoles work for better scores if we didn't use dipole bass. Or maybe dipole bass only works. Or some other combination. Maybe something not a dipole. In terms like microphones maybe some super cardioid or hypercardioid pattern is a sweeter spot in between monopole and dipole.

Of course with the market the way it is there is less impetus for such research because dipoles will always be at a disadvantage in most domestic listening spaces.
 

suttondesign

Addicted to Fun and Learning
Forum Donor
Joined
Feb 4, 2019
Messages
732
Likes
1,310
Location
Bellingham, WA
Right, plus which, dipoles are expensive. Lots of high quality drivers and electronics. The LXmini are cheap, but also very limited in SPL.
 

Cars-N-Cans

Addicted to Fun and Learning
Joined
May 19, 2022
Messages
819
Likes
1,009
Location
Dirty Jerzey
For yucks I read Paul McGowan's Audiophile's Guide: The Stereo. It was, to put it mildly, light on technical data, and heavy on that other stuff. But I could not for the life of me figure out what the hell he was talking about when he said he could hear the drummer way in the back, vocalist in the front, guitar behind him off to the side, etc. The whole description is completely contradictory to what imaging is possible with conventional two-channel audio, and thought he was just coo-coo as usual until I remembered he was talking about his IRS-Vs, a dipole speaker. Then his description of what he was hearing finally made sense. And we see this theme carried on in their new speakers with the rear tweeter.

I would too agree that a diffuse field of reflections does indeed impart a great sense of openness to the system. I set mine up to have a diffuse field of reflections as well a short time after the direct sound. Otherwise it just sounds anechoic and closed-in without it. I think there are some details to how the auditory center handles the first direct sound heard vs. the early reflections that are either not fully understood, or it may be too impractical to actually implement something that better mimics what we would actually hear. My anecdotal experience is that provided they are not too separated from the direct sound and come within the first 30-35 ms or so, the reflections are what add the "context" to the environment the sound is being recorded in, if that makes sense. Of note, though, is that with dipoles the sense of "location" of the various sources in a recording will more likely be coming from the dipoles and their interaction with the room rather than the actual recording since they can potentially take such a long and circuitous path to the listening position in some setups, esp. if the room is quite specular. At that point the source of the reflection itself and not the respective ILDs from the recording will be perceived as the localization cue I would think. But if the direct sound can be made prominent enough so the imaging is not perturbed, it definitely works in my own experience. Now whether that is best from a dipole speaker or some other way, I don't know. With conventional speakers there are so many things working against you that having something like a dipole, even if its fully artificial, could be beneficial if it makes it seem more life-like.
 

audiofooled

Addicted to Fun and Learning
Joined
Apr 1, 2021
Messages
530
Likes
588
For yucks I read Paul McGowan's Audiophile's Guide: The Stereo. It was, to put it mildly, light on technical data, and heavy on that other stuff. But I could not for the life of me figure out what the hell he was talking about when he said he could hear the drummer way in the back, vocalist in the front, guitar behind him off to the side, etc. The whole description is completely contradictory to what imaging is possible with conventional two-channel audio, and thought he was just coo-coo as usual until I remembered he was talking about his IRS-Vs, a dipole speaker. Then his description of what he was hearing finally made sense. And we see this theme carried on in their new speakers with the rear tweeter.

I would too agree that a diffuse field of reflections does indeed impart a great sense of openness to the system. I set mine up to have a diffuse field of reflections as well a short time after the direct sound. Otherwise it just sounds anechoic and closed-in without it. I think there are some details to how the auditory center handles the first direct sound heard vs. the early reflections that are either not fully understood, or it may be too impractical to actually implement something that better mimics what we would actually hear. My anecdotal experience is that provided they are not too separated from the direct sound and come within the first 30-35 ms or so, the reflections are what add the "context" to the environment the sound is being recorded in, if that makes sense. Of note, though, is that with dipoles the sense of "location" of the various sources in a recording will more likely be coming from the dipoles and their interaction with the room rather than the actual recording since they can potentially take such a long and circuitous path to the listening position in some setups, esp. if the room is quite specular. At that point the source of the reflection itself and not the respective ILDs from the recording will be perceived as the localization cue I would think. But if the direct sound can be made prominent enough so the imaging is not perturbed, it definitely works in my own experience. Now whether that is best from a dipole speaker or some other way, I don't know. With conventional speakers there are so many things working against you that having something like a dipole, even if its fully artificial, could be beneficial if it makes it seem more life-like.

In my experience, dipoles aren't exactly required in order to perceive imaging as MCG. describes it. I do it all the time. Here's one video I found about ILD and ITD:


I'm no expert, but my best guess is that a loudspeaker must be void of any resonances, FR and directivity errors which would tell it's physical location. Secondly it's about setup so that early reflections are most welcome, if the loudspeakers are up for it (the reflections have very similar tonality as the direct sound). Then if there aren't any perceivable channel imbalances, the effects which are in the recorded material should make you believe they are taking place in your 3D environment. Yet another mind trick for me would be that if you physically sense the lower frequency vibrations it certainly helps the credibility of it all... Still I think there's nothing magical about stereo, but when done right it can be a very pleasurable experience.
 

Cars-N-Cans

Addicted to Fun and Learning
Joined
May 19, 2022
Messages
819
Likes
1,009
Location
Dirty Jerzey
In my experience, dipoles aren't exactly required in order to perceive imaging as MCG. describes it. I do it all the time. Here's one video I found about ILD and ITD:
Re: Locating speakers this is probably one reason that wide, or at least controlled, directivity is needed as well. Ideally you want the SPL level within the speakers listening window to be relatively uniform so both ears hear roughly the same level from the tweeter. This means there will be no ILDs related to the speaker itself being introduced. I believe this is the “halo of sound” effect since the ITDs alone related to small path differences are not very precise, giving uncertainty to the speaker’s position.

As far as stereo goes, there is the perpetual broken record statement of cross-talk being an issue, which it is. With cross-talk, some or even most of the ITDs can be corrupted. My typical experience with conventional stereo and good speakers is “moments of excellence” where there can be some spatial effects giving a deeper image. But usually it’s more like what you get with a large flat-screen TV. Nice, but not fully accurate to real life. Getting rid of the cross-talk gets you there to true immersive imaging since the ITDs can be preserved like headphones, but that is a whole new ball of wax, and something that is not always practical to implement. But the fact that it can work so well is why it’s still relevant in my opinion, anyway. Maybe someone will come up with some clever beam-steering technology or something to give a more general purpose XTC system that doesn’t need head tracking and/or extensive measurements to work. It would be nice to make it something included in a DSP speaker you could just pull out of a box from the store one day. Edit: There are the various flavors of BACCH systems as mentioned earlier, but Im not sure how well suited the more inexpensive ones would be for audiophile use. The “big boy” stand-alone systems are crazy expensive as far as I know.
 

puppet

Senior Member
Joined
Dec 23, 2020
Messages
446
Likes
284
Nice post @suttondesign ... interesting that you find a preference for the dipole vs the Dutch.
I'm with you on the dipole ability to draw you in and the negatives that you point out do become a deciding issue when choosing between loudspeaker types such as you've done. For me, as well, the bass was the problem in on one hand sounding "right" but on the other hand not enough. Often wondered if more attention was placed on the rear wave attenuation vs the pure dipole mindset a better, more room friendly, system could result. My experience with dipole mids and tweeters result in that attenuation requirement for my room. It's probably more of a personal taste thing but I find that the additional rearward output, when controlled, is really helpful in my enjoyment.

I think designers of a dipole system might do well in trying to exert more control over the overall radiation ... not in the bass area but the lower mid on up. Something that makes room placement easier without giving up too much of that open sound.
 
Top Bottom