Soundstage, what is it exactly?

[Suono]

You can have a large soundstage but filled with only blurred images

For me, the first thing that should be looked for is focus, starting from the center and continuing to look for it as much as possible on everything, the width of the acoustic scene is secondary, it must be sought without losing focus.

 

[olieb]

You can have a large soundstage but filled with only blurred images.

As someone mainly interested in classical (concert hall) music, I could live with “blurred“ images '(not too much of course), after all it is rare to have pinpoint location of instruments even in a good seat in a concert hall.
But I found “blurred images“ in reproduction to be strongly connected to “blurred sound“ of the instruments. What I mean is that the sound is somehow non-realistic, like a broad-stroke, posterised version of itself, whereas a violin in the concert hall still sounds like a violin even when the location is a bit imprecise.

I've found that having a fairly long time gap between the first-arrival sound and the strong onset of reflections can help.

That seems like the silver bullet, but not so easy to have in a “small room“.
And there is that school of thought (Toole) which says that (significant and specular) early reflections are not (that) detrimental, can even be beneficial.

 

[thcdru2k]

I imagine toe-in in relation to room geometry / listening positions are important. Choosing the correct tradeoffs. Ceiling material even (smooth plaster vs textured vs ceiling treatment). How much reflection do you want or not want.

In an open-plan asymmetrical room, some of the usual assumptions about first reflections may not apply because the room itself is already modifying reflection timing and strength.

 

[MattHooper]

As someone mainly interested in classical (concert hall) music, I could live with “blurred“ images '(not too much of course), after all it is rare to have pinpoint location of instruments even in a good seat in a concert hall.
But I found “blurred images“ in reproduction to be strongly connected to “blurred sound“ of the instruments. What I mean is that the sound is somehow non-realistic, like a broad-stroke, posterised version of itself, whereas a violin in the concert hall still sounds like a violin even when the location is a bit imprecise.


That seems like the silver bullet, but not so easy to have in a “small room“.
And there is that school of thought (Toole) which says that (significant and specular) early reflections are not (that) detrimental, can even be beneficial.

I’ve mentioned this before, but I actually found a simple “hack” for combining spaciousness with image precision and density.

So I very much like the sensation of wide open acoustic space, and focused sonic images happening within that space. So in my 15’ x 13’ room, as is often the case, if I sit closer to my loudspeakers and/or do not toe them in a lot, that tends to open up the sound stage, at the expense somewhat of imaging precision and density.

I found that introducing a single curved diffuser placed between and just behind my loudspeakers seems to focus images and give them more presence and density… similar to more toe in speakers or if I move back from the speakers.

In this way, I’m able to enjoy the benefits of sitting closer to my speakers and having them angled so they are shooting somewhat “past” my ears a bit (no doubt, engaging some sidewall reflections), which allows for a spacious sound stage and larger images. But within that stage the diffuser keeps images focussed, dense and palpable. So it feels like the best of both worlds.

Here’s a photo of what I’m talking about.
You can see my left channel speaker in the distance and I place the diffuser facing up sitting on top of my large centre channel underneath my projection screen:

(I’ve brightened up this picture where the lighting is normally dark, to see it from the listening position)

Btw: I can also place the diffuser on the floor, leaning back against the centre channel , with the curved part facing me.
That increases the sense of punch and density…. But at the expense that the sound starts to get a bit more congealed around the diffuser and less spacious. It seems with the curved part facing up towards the ceiling adds the density but maintains the spacious separation of the images. In fact, I can even dial in the specific aspects of the imaging in 3-D presentation by shifting the curved diffuser forwards or backwards.

 

[goat76]

As someone mainly interested in classical (concert hall) music, I could live with “blurred“ images '(not too much of course), after all it is rare to have pinpoint location of instruments even in a good seat in a concert hall.
But I found “blurred images“ in reproduction to be strongly connected to “blurred sound“ of the instruments. What I mean is that the sound is somehow non-realistic, like a broad-stroke, posterised version of itself, whereas a violin in the concert hall still sounds like a violin even when the location is a bit imprecise.

That's why you should use controlled test recordings when setting up your loudspeakers' positioning in your room, and not just rely on your favorite choice of classical (concert hall) music, or any other random recordings that you have no idea if they were mixed in the "non-realistic" way as you describe.

It's better to use something way more "boring", like the simple sound mix I made (link below) with 3 different types of sound objects, where each object is panned hard left, mid-left, center, mid-right, and hard right. These sound elements are the same no matter where they are panned in the stereo field, and if the loudspeakers are set up in a correct way, the sound objects should sound pretty much the same, at least when it comes to their perceived width and distinctiveness/punch in all of the panned positions. (The tonality may change a little for the phantom-panned sounds versus the hard-panned sounds, but that is another subject.)

The single most important thing to get right is the center image, which I see as the "anchor" or the "compass" of the whole stereo image. If you get the center image to sound almost as distinct-sounding as if there were a physical center speaker in place, pretty much everything else in the stereo image and soundstage will geometrically fall in the right places, too.
If the center channel sound "smeared out", it may indicate that you have your speakers too wide apart (the equilateral listening triangle may be too large for your choice of loudspeakers), and if the center image lacks focus, some changes to the toe-in setting may be the solution.

Here is the stereo panning test file: https://www.dropbox.com/scl/fi/t02v...ey=68b1lt6m902odkada7zwcb80k&st=b37tdqd0&dl=0

 

[pablolie]

Very good discussion and many good points.

I think we all know a LOT of recorded music these days has completely artificial staging - most recordings these days do NOT have all musicians playing live along each other in the studio as if it was a live concert. So how relevant is artificial staging? I know it's the "intent", but I personally don't obsess over it. I do like clarity, though. And spacing between instruments supports that.

I also completely agree with @olieb when he points out even in the best seat in a classical concert you cannot pinpoint the exact location of every instrument (the goal in classical is always that "the whole is greater than the sum of its parts". Putting razor sharp separation in that -if even possible- could be distracting from the "performance". But of course we want the clear overall location info that classical is most often arranged into.

Jazz -especially trios- are what really screams for staging, since with three instruments you *do* hear very clear "staging" if you're listening to them live. And I find it interesting to listen to the different recording techniques to capture that "atmospheric" magic.

I also use test tracks (Sheffield Labs has a few great CDs) for setup, but that has to tested with my fav test tracks, which are well recorded classical and jazz.

Years ago I started a thread about the "30ft wide piano" that touches this subject as well.

 

[Suono]

But I found “blurred images“ in reproduction to be strongly connected to “blurred sound“ of the instruments. What I mean is that the sound is somehow non-realistic

but this is high fidelity, the more focused the violin is, the closer it gets to the fidelity of stereo microphone recording.

 

[Grimoire]

Very good discussion and many good points.

I think we all know a LOT of recorded music these days has completely artificial staging - most recordings these days do NOT have all musicians playing live along each other in the studio as if it was a live concert. So how relevant is artificial staging? I know it's the "intent", but I personally don't obsess over it. I do like clarity, though. And spacing between instruments supports that.

Whether the staging and imaging be artificial or "organic", if it is in the recording then it is capable of being reproduced.

 

[Duke]

That [a fairly long time gap between the first-arrival sound and the strong onset of reflections] seems like the silver bullet, but not so easy to have in a “small room“.

Imo to get there in a small room, good radiation pattern control is crucial. Yes I know there are approaches to room treatment that successfully push the strong onset of reflections back in time, but ime the ones that also preserve the spectral balance of those reflections tend to be quite expensive.

And there is that school of thought (Toole) which says that (significant and specular) early reflections are not (that) detrimental, can even be beneficial.

Yes, it's a tradeoff. By pushing the strong onset of reflections back later in time, in practice you lose the increase in apparent source width from having strong early same-side-wall reflections. Imo you gain increased clarity and image precision, and an increased sense of being within the acoustic space on the recording.

 

[pablolie]

Whether the staging and imaging be artificial or "organic", if it is in the recording then it is capable of being reproduced.

My point is just that it is utterly artificial. I have nothing against artificial mixes to achieve the illusion of staging, I never said I can't enjoy it. I just don't obsess over it... unless they start doing stupid crap like panning individual instruments from left to right and back for absolutely no good reason...
But there's a certain synergy when they really record all live together that -to me- is undeniable.

 

[olieb]

Imo you gain increased clarity and image precision, and an increased sense of being within the acoustic space on the recording.

I totally agree.

Yes, it's a tradeoff. By pushing the strong onset of reflections back later in time, in practice you lose the increase in apparent source width from having strong early same-side-wall reflections.

I never understood what the advantage of this apparent source width is supposed to be. Maybe for electronic or close-miked instruments that have no natural ambience in the track and where listening room reflections are a surrogate and at the same time create a “they are here“ impression. But for well-recorded classical music?
I cannot see the advantage and I do not hear it either, quite the contrary. (It is the same in the hall, where one doesn't want to sit next to the wall.)

 

[Grimoire]

My point is just that it is utterly artificial. I have nothing against artificial mixes to achieve the illusion of staging, I never said I can't enjoy it. I just don't obsess over it... unless they start doing stupid crap like panning individual instruments from left to right and back for absolutely no good reason...
But there's a certain synergy when they really record all live together that -to me- is undeniable.

Who cares whether it's artificial or not? If it's there, I want to hear it. "Synergy" (whatever that may be) and all.

 

[goat76]

Very good discussion and many good points.

I think we all know a LOT of recorded music these days has completely artificial staging - most recordings these days do NOT have all musicians playing live along each other in the studio as if it was a live concert. So how relevant is artificial staging? I know it's the "intent", but I personally don't obsess over it. I do like clarity, though. And spacing between instruments supports that.

Artificial staging of a controlled test recording, like the one I posted, is at least the best possible tool for setting up the positioning for your loudspeakers. And by doing that, you can also be more certain that you, at least, got your loudspeakers in a correct position, also for your non-artificial recordings.

I also completely agree with @olieb when he points out even in the best seat in a classical concert you cannot pinpoint the exact location of every instrument (the goal in classical is always that "the whole is greater than the sum of its parts". Putting razor sharp separation in that -if even possible- could be distracting from the "performance". But of course we want the clear overall location info that classical is most often arranged into.

The same goes for this.
If you get a sharp pinpoint imaging of the location of the instrument in most audio productions containing artificial instrumental staging, you can also be more certain that you will get a more accurate representation out of your classical recordings, as well.

 

[pablolie]

Who cares whether it's artificial or not? If it's there, I want to hear it. "Synergy" (whatever that may be) and all.

Like I said, *I* do. We are entitled to personal preferences. I didn't word it in any way that denigrated artificial staging, all I stated is I don't overly obsess over it when I know that's the case. And that includes several recordings I very much enjoy, otherwise I probably wouldn't keep buying a lot of new recordings.

 

[pablolie]

Artificial staging of a controlled test recording, like the one I posted, is at least the best possible tool for setting up the positioning for your loudspeakers. And by doing that, you can also be more certain that you, at least, got your loudspeakers in a correct position, also for your non-artificial recordings.

You miss the fact I clearly stated I use test tracks for setup.

 

[Duke]

I never understood what the advantage of this apparent source width is supposed to be.

I think it's a "personal preference" thing.

My recollection is that Erin really likes wide-pattern speakers that "throw" a wide soundstage extending laterally well beyond the speakers. I agree that that's fun, but imo it comes at a price if it's something we're trying to synthesize for every single recording (via strong early ipsilateral reflections), rather than letting it be part of the presentation when the recording calls for it and not when it doesn't.

 

[MattHooper]

I never understood what the advantage of this apparent source width is supposed to be. Maybe for electronic or close-miked instruments that have no natural ambience in the track and where listening room reflections are a surrogate and at the same time create a “they are here“ impression. But for well-recorded classical music?
I cannot see the advantage and I do not hear it either, quite the contrary. (It is the same in the hall, where one doesn't want to sit next to the wall.)

I could try to explain from my goals.

I’m not currently using dipoles or Omnis, just conventional forward, radiating floor standing speakers.

I like the sensation of voices and instrument occurring in real acoustic space. And I love a sense of development and the sense that I’m sort of entering the space of the recording. Or that it takes over my room somewhat.

In many systems I find there’s a sense that the sound stage and the recorded acoustic appears mostly between the loudspeakers, in a way that makes it feel like you’re in one room, but peeing through a window into the recorded space.

Whereas I prefer the impression that the speakers just disappear and the room beyond around the speakers becomes the recorded acoustic (whether artificial or naturally recorded).

And then I get a better sensation of sonic images occurring in real acoustic space.
(so for me that often means I’m not doing heavy toe in, and there’s likely some engagement with side wall reflections).

The other aspect is that I often find that “too focused imaging” sounds artificial to me.
In fact, I generally tend to find recordings on both systems to sound reductive in that way - it’s like voices and instruments, having squeezed down to smaller points than they sounded naturally. Whenever I listen to a live instrument, my first impression is how much bigger and richer and more expensive it is than the squeezed, reductive points in space that I experience on many sound systems.

So I try and get a little bit of that back with some tow out of the speakers and my position and with room acoustics… so that the images expand in size a bit more, and also that the outlines are not sounding artificially crisp to me. Very often, I use a well recorded centred vocal to start that process off - where if I have too much toe in it can sound like the singer has been squeezed to small and sharp, but with the right amount of speaker angling the image becomes less squeezed less hard edged, less mechanical and artificial sounding and I can “ believe” the voice in front of me more. Same with any number of instruments.

The trade-off has usually been that highly focused images do something right in regards to sounding a bit more real - the images take on more density…. And density is one of the things you lose vs the real thing because of the inherent limitations of stereo. With stereo images you get more of a phasey see through sensation that does not seem as corporal as the real thing. Focussed imaging at least increases the sense of corporality. But again to my ears at the expense of harder edges and squeezing the images too tight.

So that’s why I have always sought the best balance between these two trade-offs that I can manage - trying to expand sound stage and image sizes with the least trade-off in terms of image density.

And that as I mentioned has been helped quite a bit when I started using the central diffuser. With that I maintain the large spacious imaging and immersion that I’m looking for, and the sense of images floating in real acoustic space, and also the image being slightly more large, larger and realistic… but the diffuser adds the Sonic density into the picture as well.

I find this deeply satisfying, not just for acoustic music but for all music.

For instance, a couple nights ago I was listening to the new Boards Of Canada “Inferno” album which is highly artificial electronic music, but which contains wide and deep spacious and airy sound design.
The sensation of the rest of my room just melting away into what sometimes felt like a limitless depth, and to some extent width, with very palpable sonic images, was extremely satisfying. It felt like sort of entering the sonic creation somewhat rather than just observing through a port hole from another room. (So I can certainly understand why Floyd Toole points out that listeners find spaciousness and immersion to be very appealing). (I get immersion from my surround system, but not the same level of coherent three-dimensional palpable imaging as I get from my two channel set up) z

 

[KSTR]

What approach(es) have you found to be helpful?

Near-field listening in a not too small and well-damped room, mostly. But that can be too intimate and dry with some recordings, therefore I sometimes take the liberty to apply some reverb and early reflections, the wet signal passing though the "blow-it-up-to-3D" mechanism mentioned earlier (slight amount of speaker crosstalk cancelling applied on a slightly narrowed stereo width). And I find proper "digital room correction" a must, in almost all situations.

Trinaural-matrixed playback also can work well for increasing the separation between discrete and diffuse image sources (again, best with room influence reduced), but it really is a different rendering mechanism with some advantages and some drawbacks. The need for an (ideally identical) center speaker and the +-45° opening angle for the L/R speakers is probably the biggest. It also has difficulties to render HRTF-encoded sound ("Q-sound", etc) properly, and it can have a touch of phasiness that some listeners find distracting.

 

[Mort]

What approach(es) have you found to be helpful?

Near-field listening in a not too small and well-damped room, mostly.

Agree that nearfield (usually with narrow-ish directivity coaxial) provides pinpoint imaging and solid envelopment. It's the way I listen to 90% of my music, which is bass-heavy rock, pop, metal. Since this practically requires a separate space and usually accommodates only a single listener, it's a luxury many can't afford. But it is a solution.

Alternatively, I also have a Farfield, wide directivity system which we strongly prefer for movie content. Wide directivity smears the soundstage a little bit, and we find that more appealing for movie content. It also sounds good on acoustic, particularly philharmonic content because it creates a wall of sound and adds a little extra reverb for envelopment.

 

[audiofooled]

What we (or I) want is pinpoint 2D imaging of the discrete sources plus large and 3D-ish imaging of the diffuse sources (ambience, reverb tails, early reflections).

Sorry for being late for the discussion but I completely agree.

What approach(es) have you found to be helpful?

@Duke

Time - intensity trading, yes, near constant directivity if possible in the most critical range 700-7kHz for pinpoint imaging and wide coverage or sweet area, rather than sweet spot, if you will. This also helps the perception of depth (distance and proximity). However, there's also perception of space in the low frequencies and the perception of what would sound like the diffuse field. For that you really need very wide coverage but the kind that would work with the setup.

What I've learned so far experimenting with what would hopefully be a natural perception (or rather acceptable illusion) of recorded space, or spatial quality of what was recorded, is that time domain performance is critical, the more close to full audible range, the better. Tight inter-channel tolerances are also important.

Also, I don't want vertical reflections in order to more effectively manipulate direct to reverberant sound ratio. When it comes to side reflections, in terms of energy, the right spectral balance and time of arrival of lateral and contralateral reflections IMO is required both for early and late spatial impressions. So again very, very wide coverage helps if you want to manipulate the side energy, but with the right amount of attenuation of what radiates to the sides.

To illustrate, here are some graphs:

T60M (L-yellow, R-blue, L+R-red, vector average of L/R-orange:

Some Airy CWT's:

L:

R:

L+R:

Vector average of L/R:

Left/Right A/B:

The same, but R is inverted (important for out of phase signals):

Morlet CWT revealing problematic frequencies on the first arrival (TV and coffee table, SBIR), the kind of restrictions we normally find in living rooms:

The amount of attenuation throughout the full range for one channel inverted and what the room has to say about it:

Bottom line, for me at least, the concept of sound stage is entirely a brain construct (what you hear may be different from what I hear), so much that it escapes definition, especially if we introduce the dimension of height. Imaging however we can work with in terms of how well defined or pinpoint if you will. It's all in the recording.

EDIT:

Perhaps I should note that the measurements should represent rather complex intra and inter-channel interactions of what we would normally and perceptibly call "phantom center", at 3.5m distance. Array is now focused about half a meter in front of the listening position.

The last graph should be compared to the fifth one in the row to show relative attenuation and room behavior. In this kind of setup and when it comes to low frequencies, center sub/bass module actually runs 13 dB hotter than the mains array to get this amount of attenuation and in room direct, early and late energy response behavior relative to the rest of the spectrum. Pressurization down to DC is not shown on the graphs.