• 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!

Sound stage depth?

Duke

Major Contributor
Audio Company
Forum Donor
Joined
Apr 22, 2016
Messages
1,523
Likes
3,745
Location
Princeton, Texas
I am not sure about front wall reflections add anything other than confusion cues to the reflections recorded from the auditory scene. The natural reflections that is part of the listening lounge all come from side walls, floor and ceiling. Having speakers close to a wall with damping behind gives to me a better illusion of what is going on at the scene.

If you can eliminate (or re-direct) the front wall reflections, or at least the initial ones, without introducing some other problem, imo that would be ideal.
 

Peluvius

Addicted to Fun and Learning
Forum Donor
Joined
Aug 5, 2021
Messages
513
Likes
543
I am not sure about front wall reflections add anything other than confusion cues to the reflections recorded from the auditory scene. The natural reflections that is part of the listening lounge all come from side walls, floor and ceiling.

The timing and volume level of the refelction in relation to the direct, wherever it comes from, has had the biggest impact from my experience in relation to imaging and soundstage.

Having speakers close to a wall with damping behind gives to me a better illusion of what is going on at the scene.

Doesn't that depend on the speakers and the room?
 

Thomas_A

Major Contributor
Forum Donor
Joined
Jun 20, 2019
Messages
3,423
Likes
2,408
Location
Sweden
If you can eliminate (or re-direct) the front wall reflections, or at least the initial ones, without introducing some other problem, imo that would be ideal.
Well at least minimizing the primary reflections that cause trouble. Using the wall for bass reinforcement is a plus with near wall placement.
 

dualazmak

Major Contributor
Forum Donor
Joined
Feb 29, 2020
Messages
2,823
Likes
2,951
Location
Ichihara City, Chiba Prefecture, Japan
I believe that, for excellent reproduction of 3D sound image/perspective, precision (0.1 msec level) time alignment between all the SP drivers measured at our listening position should be critically important as I summarized here in my project thread.
 

kthulhutu

Active Member
Joined
Mar 15, 2022
Messages
132
Likes
113
I believe that, for excellent reproduction of 3D sound image/perspective, precision (0.1 msec level) time alignment between all the SP drivers measured at our listening position should be critically important as I summarized here in my project thread.
I agree. Time alignment improves depth. Time alignment + BACCH makes for realistic sounding images that have a 3D presence to them.
 

Thomas_A

Major Contributor
Forum Donor
Joined
Jun 20, 2019
Messages
3,423
Likes
2,408
Location
Sweden
The timing and volume level of the refelction in relation to the direct, wherever it comes from, has had the biggest impact from my experience in relation to imaging and soundstage.



Doesn't that depend on the speakers and the room?
I would say keep the signal "clean" from reflections behind/near speakers, e.g. when measured at short distance (below about 60 cm distance, speaker close to wall). After that make sure reflections at LP is clean within 2 ms and those >2ms 10-15 dB lower in level.

etc 60 cm.png
 

Peluvius

Addicted to Fun and Learning
Forum Donor
Joined
Aug 5, 2021
Messages
513
Likes
543
I agree. Time alignment improves depth. Time alignment + BACCH makes for realistic sounding images that have a 3D presence to them.

I think this is one of the best things about Dirac. The time aligned EQ makes a big difference to the presence of these "stereo effects".
 

dualazmak

Major Contributor
Forum Donor
Joined
Feb 29, 2020
Messages
2,823
Likes
2,951
Location
Ichihara City, Chiba Prefecture, Japan
I think this is one of the best things about Dirac. The time aligned EQ makes a big difference to the presence of these "stereo effects".

I did exactly the same using DSP (XO/EQ/delay control) "EKIO", DAC8PRO, multiple amplifiers directly dedicatedly driving SP drivers. My latest system setup (as of May 30 2022) can be found here on my project thread.
 

NiagaraPete

Major Contributor
Forum Donor
Joined
Jun 23, 2021
Messages
2,185
Likes
1,953
Location
Canada
The Revel F208 is an interesting modern speaker because it has relatively poor image depth and projection despite measuring very well.There is very little projection in front of the plane of the speakers although there is some depth back behind them.There is a small dip in the 1500Hz -2500Hz range that some reviewers have commented on and which they have suggested might account for this recessed soundstage but it is only very slight.Very odd that Revel would release a speaker that sounds like that and yet they sold lots of them.They would probably be OK with orchestral music but sound quite odd with closely mic'ed vocals.They have now discontinued them.That speaker was measured here but nothing too unusual was found.
Another common speaker that has poor mage depth is the Yamaha NS1000.Many people seem not to hear how poor the imaging is on them.They have a peanut shaped soundstage and you will observe experienced listeners moving their heads from side to side trying to get a central image to lock in but it never happens.
Some people seem to only hear left/right and tone.
Sounds like subjective drivel. Wasn’t the Yam NS1000 made in the 70’s?
 

goat76

Major Contributor
Joined
Jul 21, 2021
Messages
1,270
Likes
1,390
Ime the distance to the back wall DOES have an effect on perceived soundstage depth.

The ear/brain system judges the perceived dimensions of an acoustic space primarily by three sets of information: The FIRST reflections; the reverberant tails; and the temporal center-of-gravity of the reflections.

When the speakers are pulled out from the wall, the first reflections in that dimension are pushed back in time and become a bit weaker, both of which indicate to the ear/brain system greater distance in that direction. The temporal center-of-gravity is also pushed back in time a little bit, which is indicative of a larger acoustic space.

The influence of distance from the wall on soundstage depth varies with the specifics, and presumably varies somewhat from one individual to the next.

The playback room's "small room signature" cues are most strongly conveyed by the first reflections, specifically their arrival times (relative to the direct sound) and arrival directions. Good loudspeaker/room interaction, usually enabled by professional attention to room acoustics, can weaken those "small room signature" cues to the extent that the venue cues on the recording become perceptually dominant, which (depending on the recording) can result in a great deal of soundstage depth.

Well, I'm totally on your side that by purely minimizing early reflections from the listening environment and maximizing the direct sound from the speakers, the reflections in the recorded space will of course be more easily heard, as well as everything else in the recording. We simply want to maximize the direct sound to be able to hear the recorded direct sounds, the early reflections, and the reverberant tails which will all give us the necessary clues that contain the depth and spaciousness of the recorded space.

But let's say we already have minimized the early reflections from the wall behind the speakers, and reduced them to just containing the low "omnidirectional bass frequencies", adding further distance to that wall will not in any significant way help us hear the recorded depth better or deeper.
And I think most of us already and normally have the speakers placed far enough from the back wall (or have solved the problem with damping like Thomas_A has done), so pulling them even further from the wall will probably not, in most cases, add any more perceived depth of the recorded space.
 

dualazmak

Major Contributor
Forum Donor
Joined
Feb 29, 2020
Messages
2,823
Likes
2,951
Location
Ichihara City, Chiba Prefecture, Japan
....
Another common speaker that has poor mage depth is the Yamaha NS1000.Many people seem not to hear how poor the imaging is on them.They have a peanut shaped soundstage and you will observe experienced listeners moving their heads from side to side trying to get a central image to lock in but it never happens.
Some people seem to only hear left/right and tone.

I completely disagree with your subjective comments with no proper citation nor validation.

In my system setup, I renovated my Yamaha NS-1000 into fully active (by eliminating all the LCR network and attenuators) in my multichannel multi-driver multi-amplifier stereo audio project. You can find the latest system setup here. Also, you would please read carefully my post here on the project thread.

Even before starting my multichannel project, my NS-1000 in intact configuration (with original LC network and attenuators) had really nice 3D sound stage/perspective driven by single amplifier, in my case Accuphase E-460. I have been keeping this "intact single amp setup" as my reference sound system throughout my project.
 
Last edited:

Duke

Major Contributor
Audio Company
Forum Donor
Joined
Apr 22, 2016
Messages
1,523
Likes
3,745
Location
Princeton, Texas
We simply want to maximize the direct sound to be able to hear the recorded direct sounds, the early reflections, and the reverberant tails which will all give us the necessary clues that contain the depth and spaciousness of the recorded space.

My understanding (from Floyd Toole's book) is that the WORST direction for a reflection to arrive from is, the exact same direction as the first-arrival sound. So imo we don't want to minimize ALL of the reflections, only the FIRST ones. If we go overboard with absorbing reflections, we can end up with an overly dead presentation.

The first reflections are the worst offenders as far as degrading imaging precision, soundstage depth, and tonal balance, and the later reflections (assuming they are spectrally correct) are effective as "carriers" for the reverberation tails on the recording, delivering them from many directions, which is desirable if we wish to have a sense of immersion.

But let's say we already have minimized the early reflections from the wall behind the speakers, and reduced them to just containing the low "omnidirectional bass frequencies", adding further distance to that wall will not in any significant way help us hear the recorded depth better or deeper.

Agreed.

And I think most of us already and normally have the speakers placed far enough from the back wall (or have solved the problem with damping like Thomas_A has done)...

Maybe so, I don't know how common it is for the wall behind the speakers to constrain the soundstage depth.
 

Tks

Major Contributor
Joined
Apr 1, 2019
Messages
3,221
Likes
5,494
I went to college for sound recording technology for a few years in the 90's. I admittedly didn't pursue a career behind the console, and my career in front of the console didn't pan out, but I have a fairly good understanding of what goes into making a recording (or what used to go into a recording anyway, lol.) I am a little confused at all the reviews of speakers that discuss the depth of the soundstage and how they can "hear the drums in the back" or "the guitar up front." When recording, there is only panning and placement of tracks across the soundstage left and right. There is no front and rear, or up and down. Especially in the situation of most modern recordings where instruments are close mic'ed and recorded at separate times I'm not sure what they're talking about. One scenario that I could potentially grasp the concept in would be something like an XY Stereo pattern mic'ing of a group of instruments where you could possibly get subtle timing cues that could create the illusion of depth, but I don't even know if that makes any sense. Can anyone shed light on how 'depth' is created or presented from a 2 channel recording being played back on 2 speakers? I'd love to hear Amir's take on this as well.

With 2 channel things like headphones or IEMs especially, there can never actually be "soundstage", because if there were, it would be a tangible metric that can be toggled in some sense even in mono. So your suspicions are correct. Soundstage is nothing more than channel panning, post processing effects like reverb, and natural physical recording realities like the setting that either creates or eliminates reflections. Especially egregious is when people talk about verticality in a stereo situation.. That makes utterly no sense without visual placebo, or contextual reference telling you which of the two sounds being AB'd is up or down first, and then only a sound that is being recorded in a reflective setting or studio that would give you an idea based on time delay.

All the folks that claim soundstage is a thing, I welcome with open arms for a demonstration either on-paper, or in-demo of how I could increase something like soundstage in already recorded material, or remove soundstage without post processing effects on the same listening device EQ'd to the same output, and similar fidelity metrics.
 

Matthias McCready

Active Member
Joined
Jul 2, 2021
Messages
209
Likes
273
With 2 channel things like headphones or IEMs especially, there can never actually be "soundstage", because if there were, it would be a tangible metric that can be toggled in some sense even in mono. So your suspicions are correct. Soundstage is nothing more than channel panning, post processing effects like reverb, and natural physical recording realities like the setting that either creates or eliminates reflections. Especially egregious is when people talk about verticality in a stereo situation.. That makes utterly no sense without visual placebo, or contextual reference telling you which of the two sounds being AB'd is up or down first, and then only a sound that is being recorded in a reflective setting or studio that would give you an idea based on time delay.

All the folks that claim soundstage is a thing, I welcome with open arms for a demonstration either on-paper, or in-demo of how I could increase something like soundstage in already recorded material, or remove soundstage without post processing effects on the same listening device EQ'd to the same output, and similar fidelity metrics.
Agreed, but again frequency (EQ) can equate "depth."

The best example I can think of is: A Thunder Storm

Think of a close "clap" of thunder vs distant rumbling thunder, the difference in the sound you hear is primarily due to high-frequency air loss.

----

Processing in this sense is traditionally source specific rather than mix wide. Example: If I want to make the lead vocal more "upfront," I might give it a dB or more in the mix, or I might pull back the competing upper-mid frequencies from the string section that are masking the vocal (bringing the vocals "forward" by pushing the strings back).

----

As far as increasing the "sound stage" overall, I would concur, I haven't experienced such a device.
 

Duke

Major Contributor
Audio Company
Forum Donor
Joined
Apr 22, 2016
Messages
1,523
Likes
3,745
Location
Princeton, Texas
Soundstage is nothing more than channel panning, post processing effects like reverb, and natural physical recording realities like the setting that either creates or eliminates reflections.

Of course soundstage is an illusion, whether picked up by the microphones or created by the engineer or both. Is it worth pursuing? Maybe for some:

According to a paper written by Wolfgang Klippel and cited by Floyd Toole in his book (third edition, pages 185-186), the “feeling of space” makes a 50% contribution to "naturalness” (realism and accuracy), and a 70%(!) contribution to "pleasantness" (general satisfaction or preference). Here is how Toole sums it up (page 186):

"Sensations of sound quality and spaciousness contribute equally to impressions of "naturalness", and spatial quality dominated the impression of "pleasantness". Therefore whether one is a picky purist or a relaxed recreational listener, the impression of space is a significant factor."

Especially egregious is when people talk about verticality in a stereo situation.. That makes utterly no sense without visual placebo, or contextual reference telling you which of the two sounds being AB'd is up or down first, and then only a sound that is being recorded in a reflective setting or studio that would give you an idea based on time delay.

It still doesn't make sense even after you've experienced it!

A couple of decades ago I bought a CD by a local musician. On one song his voice seemed to come from about where the speaker met the floor. We were friends so the next time I saw him, I asked him about that.

Turns out the album was recorded in one day, just before he left for a European tour, because his manager decided at the last minute that he needed to have a "new album" to go along with the tour. On that particular song they were so rushed the microphone in the sound booth wasn't lowered into position so it was ABOVE this head somewhat. Apparently somehow that resulted in the listener's perspective corresponding with where the microphone was in relation to his mouth. I don't even have a theory for how that could happen. And no it doesn't make sense.

All the folks that claim soundstage is a thing, I welcome with open arms for a demonstration either on-paper, or in-demo of how I could increase something like soundstage in already recorded material, or remove soundstage without post processing effects on the same listening device EQ'd to the same output, and similar fidelity metrics.

Well I claim that "soundstage can be a thing", but I'm not sure just what you are asking for a demonstration of. Could you clarify?
 

duckworp

Member
Joined
Mar 12, 2021
Messages
19
Likes
6
There is a Stereophile test CD track recorded in a large room with stereo mics. The person moves around the room clapping. As he move to the back of the room the brain instantly and correctly recognises the increased room reverb as greater distance behind the mic, and he sounds a few metres behind the speakers, which is exactly where he is in the room.

It is exactly the same phenomena when recording live music very simply with a pair of stereo mics in a large room with natural reverb - the instruments will sound in a three dimensional space.

A mixing engineer will often add reverb to recordings not made in this way to recreate the effect of soundstage depth.
 

kongwee

Major Contributor
Joined
Jan 22, 2022
Messages
1,024
Likes
276
Soundstaging is alway thing even you stereo capture it naturally.
Like the recording above, the choir is space away mic so they won't shout at you inline with your speaker.
Technically, it is just direct and indirect source of sound that create soundstage depth. Placebo indeed that how your brain process it naturally.
It is like thunder. You will heard it far away because the ground reverberation of the thunder sound. If you don't hear much reflection sound, you know you are in danger. The placebo effect.

Decades back you can the back of your sweet-spot with your front stereo. However, your head is to be straight and stiff tight to narrow sweetspot to get the effect. There was even an IP for this.
 

Matthias McCready

Active Member
Joined
Jul 2, 2021
Messages
209
Likes
273
One other avenue of processing that could affect, "the sound-stage" would also be phase.

Phase, roughly, is time.

Caveat here, is that to hear phase, you cannot perceive this from one source, you must have two (or more) of something, a reference and a signal or signals. For example two mics on a single source will have a timing relationship, based on their distance from the source(s), and the directionality of the source.

Example 1)
Distance: If you are recording a drum kit, a cardinal sin for the classical aficionados I know, and have two overhead mics one of which is 4ft from the snare, and one of which is 8ft away from the snare. There will be a timing and phase offset in the mics, in how they pickup the snare. The mix engineer, being on the receiving end of this "dilemma," could decide to time align them, giving the snare drum a tighter transient response, while causing more problems for the rest of the kit or they could leave the mics as they are, which could be perceived as giving greater "space" to the snare. Which is "right" would depend on the context.

Example 2) Directionality: If you are recording a snare, and you have two mics that are equidistant to the center of the drum, but one of them is underneath, they will be getting different information. Their distance is the same, however the impulse they are receiving is reversed. When the top mic has the head moving away from the mic, the bottom mic simultaneously has the head moving towards it. Most mix engineers will reverse the polarity on the bottom mic in this case, although not always, it depends on what sounds better.

----

Anytime you have two instances of the same thing there is a phase and timing relationship. Sometimes this relationship is stable, such as bass guitar being recorded both with a DI and miced bass cab. The two takes on this single source will have slightly different timings, and the mix engineer will make a decision about how they want it to sound. The decision they could make is one of those sources being mostly "out of phase" (shudder shudder :eek:;)) with the other one, as that could actually be what fits the mix the best. And guess what?! You would never know; as it is a stable timing relationship, and you have no reference.

For what it is worth this phase relationship often corresponds with frequency; so being out of phase at a certain frequency can create a dip there (rather than summation) it might make the instrument fit in frequency wise without even touching the EQ, which is pretty cool!

----

When you have two or more mics in a room that relationship is often not stable, as they are often picking up more than one source.

----

In processing land this can be further manipulated, for example a source can be artificially widened in the left/right domain. If you take a source, such as two mics on a mono guitar cab, hard pan the mics, and delay one side 8-13ms this will "widen" things out. Now is this decision creative and artistic OR simply mucking about and causing phase problems? That probably depends.

A musical example of this would be this album, the mix feels spacious on headphones, but can feel artificially wide on many systems, while I cannot state with certainty, I would guess there is time manipulation going on (probably a decision made in the recent release of it for better or worse) or at minimum that the mics were placed quite far apart in the original recording.

Also, interestingly, if you took the same source and doubled it, hard panned them, and did this, the one that arrived first would be perceived as louder and closer, even though the volume is the same. Even a 1-2ms is quite perceivable for a timing relationship between sources.

Another interesting example of phase manipulation, would from Roger Waters album Amused to Death, specifically the track "Perfect Sense, Part 1," listen to when the piano comes in at 40 seconds, where do you hear it coming from, and where does it move to? On a reasonably good system, in decent room it is jaw-dropping, and while I don't know what wizardry was committed in the mixing process, I would guess it has to do with the phase relationship between the L/R speakers on this particular source.
 

Tks

Major Contributor
Joined
Apr 1, 2019
Messages
3,221
Likes
5,494
Of course soundstage is an illusion, whether picked up by the microphones or created by the engineer or both. Is it worth pursuing? Maybe for some:

According to a paper written by Wolfgang Klippel and cited by Floyd Toole in his book (third edition, pages 185-186), the “feeling of space” makes a 50% contribution to "naturalness” (realism and accuracy), and a 70%(!) contribution to "pleasantness" (general satisfaction or preference). Here is how Toole sums it up (page 186):

"Sensations of sound quality and spaciousness contribute equally to impressions of "naturalness", and spatial quality dominated the impression of "pleasantness". Therefore whether one is a picky purist or a relaxed recreational listener, the impression of space is a significant factor."



It still doesn't make sense even after you've experienced it!

A couple of decades ago I bought a CD by a local musician. On one song his voice seemed to come from about where the speaker met the floor. We were friends so the next time I saw him, I asked him about that.

Turns out the album was recorded in one day, just before he left for a European tour, because his manager decided at the last minute that he needed to have a "new album" to go along with the tour. On that particular song they were so rushed the microphone in the sound booth wasn't lowered into position so it was ABOVE this head somewhat. Apparently somehow that resulted in the listener's perspective corresponding with where the microphone was in relation to his mouth. I don't even have a theory for how that could happen. And no it doesn't make sense.



Well I claim that "soundstage can be a thing", but I'm not sure just what you are asking for a demonstration of. Could you clarify?

Soundstage isn't an illusion though, I don't take post processing effects to be an illusion, all they are, is edits to already recorded material that aren't illusions as the term is usually used. Especially not an illusion if it's recorded (like a highly spacious and reflective recording setting, vs an anechoic chamber). There's no illusions there, one had more reflectivity, which contributed to spaciousness, while the other.. you get the idea.

As for "pursuing", sure it's worth pursuing, but it depends on what you're pursuing. I'm currently only interested in pursuing a definition, otherwise I have no clue what I'm looking for. That paper you referenced and quoted is utterly useless in this regard because it only compounds the problem, why? Instead of bringing more clarity as to what soundstage is being considered, it introduced more terms outside the audio dictionary. Words like "naturalness" aren't informative if they're not accompanied with technical and very direct explanations as to what that means on paper in audio science. It's fine if one replies "well I don't know what it means". But then that's a problem that needs to be rectified first before the notion of a more expansive concept like soundstage could even be entertained.

As for soundstage not making sense. It makes 100% to me, because my definition is coherent to me. That being: Recording setting, channel panning, and post processing effects like reverb and similar concepts that allow for some multi-channel emulation. The listening apparatus has zero notion of soundstage in my view (meaning the driver, as long as the driver can reproduce the range of frequencies, and doesn't experience noise/distortion artifacts, that to me is the extent of what a driver does by definition, anything else just seems insane to me). Now that's not to say all listening devices sound the same or produce identical soundstage, because if you can throw more microphones at a recording, and also follow that with an identical number of speakers per microphone, then sure, you're going to feel more of that recorded soundstage than you would with stereo pairs. Likewise listening to a nice pair of speakers properly powered, won't sound identical to a pair of headphones (even if we forcefully disregard something like a chest thump from heavy bass, or vocal cord rattle from a high frequency noise coming out of a speaker).

What I personally don't understand is when people talk about what speaker has better soundstage. That question seems like a category error for me until the entailment's of soundstage are explained by the person asking the question. And just to be more clear, I don't want the discussion to devolve into "well this speaker has a higher max SPL before distortion sets in, so the soundstage is great with this speaker if you like to really crank your system up". No.. If we're comparing soundstage between two devices, all other factors must be made equal as much as possible, so we can't have things like different frequency response, that must be equalized before such comparisons are allowed to proceed if one is to demonstrate soundstage as a distinct concept.

So when you ask for clarification of what sort of demonstration of soundstage I'm asking for. I'm asking for things that can be toggled or put on a slider that increase or decrease soundstage, using a concept or concepts that are as distinct as possible from already existing concepts. I don't want a demo of a Spatial Audio remastered album, because that type of soundstage I understand already, that's wholly a post-processing toolbag being used. What I want to understand are those people who make claims about devices having more soundstage than the other, yet don't use already existing concepts as the primary account for the phenomena as I have. If you hold the similar defintion of soundstage that I do, then you wont be doing this, you'll admit all soundstage is comprised of, are the things I mentioned (or if the untouched recording is concerned, it's simply reflections and time delay aspects inherent to the setting where it was recorded). But if you think there's soundstage inherent in a driver or something to that effect, please, show me what I would need to tweak on paper technically speaking in order to increase my driver's soundstage, or decrease it to nothing (witty replies like "low pass most of the sound" doesn't count for obvious reasons).

But before demonstrations of soundstage are made, I'd more than anything would like an actual definition. Ideally with recordings to where you can difinitively say X songs have more/less soundstage than Y songs. Just so I can know what's actually being tracked as a phenomena.
 

tomchris

Active Member
Joined
May 14, 2021
Messages
210
Likes
415
Location
Denmark
Not going too much into theory, we have the following framework for stereo sound reproduction:

- We have two ears (sound localization due to ear shape and time & level differences between ears)
- Stereo loudspeakers (sound wave reproduction due to frequency response, directionality, blend effect and time & level difference between speakers.
- Room interaction (interaction between direct and reflected sound due to loudspeaker radiation pattern and the listener)
- Source (number/type/placement of microphones, mastering covering level and stereo placement virtualization techniques)

We can influence a) loudspeakers, b) room interaction and c) the recorded source, enhancing depth and width cues.Unfortunately, people’s hearing is subjected to their beliefs, thus complicating and conflating matters regarding views on sound stage and depth, what we wish to achieve and how to achieve it.
 
Last edited:
Top Bottom