SO ... HOW do we measure soundstage???

[DVDdoug]

Amplifiers don't affect soundstage unless there is some processing. If an amplifier makes a difference it's probably imaginary and the effect will disappear (or become unrelated to the amplifier) in a blind listening test. Apparently EQ can have side-effects on the perception of soundstage.

I think sound bouncing around the room adds to the perception of width. So maybe omnidirectional speakers (or the old Bose 901 speakers) in big room with hard surfaces?

Of course moving the speakers farther apart will widen the soundstage.

And is that effect comparable to apply a small percentage of “stereo widening” effect on the music production side?

"Tricks" can be done with delays & phase. If you reverse the connection to one of your speakers it will be out-of-phase with the other and you'll get some weird spacey effects. (or you can invert one channel digitally in Audacity, etc.). (That will also cause the bass soundwaves to mostly-cancel.)

There is a lot of gray ground when it comes to hifi in my opinion.

Sound perception in an acoustic environment is complicated but the electronic & digital side of things is well defined and straight forward. See Audiophoolery and What is a blind ABX test?

 

[tmtomh]

"Soundstage" is often a concept that is confused. There's a difference between the pattern of dispersion of the sound coming out of the speaker, which is what I think about when I use the word soundstage, and spatial differentiation of elements in a mix, which I might call imaging. Different things affect both of these.

Amplifiers don't have a soundstage. The idea that you could compare two amplifiers and say one creates a bigger or smaller soundstage is silly, because amplifiers don't produce sound waves. They produce voltages that move drivers back and forth.

If the amplifier has less distortion and higher dynamic range, and its power and current handling are adequate for the load you're presenting it, it can help the system reproduce more details that might improve our perception of spatial cues in the resulting audio. Same if the amplifier is controlled i.e. using its damping factor to effectively slow down and stop the movement of the driver. If the amplifier output is accurately matched across both channels, it will help the stereo image feel more coherent because you aren't getting mismatched arrival times/arrival levels of the sound to your ears (which is what creates our perception of spatial cues.)

If you're talking about dispersion, it is hugely dependent on speaker/driver and cabinet design.

This thread is huge and may be of help. https://www.audiosciencereview.com/...creates-a-large-and-precise-soundstage.48542/

Yes, I would agree with you that soundstage makes sense as the one-word shorthand for the perceived size of the musical presentation, while imaging makes sense as the one-word shorthand for the perceived precision/sharpness of location of elements within that perceived presentation.

In practice, though, it seems the most clear and concise we can get in online discussions among groups of folks with different knowledge bases and word-choice habits is "soundstage size" and "soundstage imaging".

 

[Mikig]

I don't give a technical definition. but when I listen I could define a situation that I think you can share, and which I could perhaps mistakenly refer to the sound stage: when the speakers "disappear" the music can be perceived as coming from the space between the wall and the speakers themselves.

This is the only contribution I can make, but I have no idea if it is measurable.
I only know that this phenomenon can only be changed by moving the speakers.
If positioned incorrectly, the music can be perceived as coming from the speakers themselves, losing "field".

The only component I could bring into play is the amplifier.
I have tried poorly performing amplifiers, and probably not suitable for the speakers and room which, with the same speaker positioning, gave the system a less "wide" sound and more focused on the drivers...

 

[Anton D]

Soundstage/imaging, is a sum of how we interpret sounds arriving at our ear/brain machine and translate that into a three dimensional impression.

I don't there is any one thing to measure, it's everything put together. I'd even call it it one of the top goals of putting together our Hi Fi systems, as ephemeral as it is.!

So, I think your ears are the most honestly objective measurement of soundstage and imaging. (Knowing what I just said is anathema to SINADologists.)

No controvesry intended.

 

[kemmler3D]

Sound stage is hard to measure because it's hard to define, despite all the discussion we don't even seem to have a solid definition of something that could be rated on a 1-10 scale, which would at least be a start toward quantifying it.

"How convincing is the stereo illusion" might be it?

Or

"How far apart are the apparent lateral limits of the stereo image"?

Or something else entirely more subtle about the perception of an illusory space?

We at least need to pin down a single perceptual phenomenon / dimension of preference before we start correlating it to measurements.

To put this in perspective and extend @dasdoing 's comment a little - in the studio there are many "tricks" that can be used to create/enhance something you would call soundstage. Recording in stereo, with varying levels of reverb in the recording space... using mics to record the room vs. direct... adding (various styles of) reverb after the fact... doubling/panning a track and tweaking pitch/phase/EQ, adding stereo chorus, etc. All of them have to do with partially decorrelating the left and right signals.

For "depth" (front/back perception) you use filtering/EQ, dry/wet reverb ratios, and of course simple volume controls.

So if I had a piece of electronics that noticeably enhanced soundstage, I would assume something like @dasdoing said - it's messing with phase or adding some kind of semi-randomness on a per-channel basis. Distortion, in other words.

So, bottom line... high fidelity gear shouldn't enhance soundstage.

However, the question remains whether there is some kind of distortion that would be especially likely to diminish soundstage that exists in the recording... that's something that I think is interesting, worth looking into via measurements. But first we have to have a good idea of what perception of "soundstage" really is in practice.

 

[Anton D]

Well done, my good fellows!

 

[Blumlein 88]

Q-sound can manipulate where things come from and the soundstage as well. So whatever they do is an example of what effects where we hear things. Some of the simpler things are known. The problem with any DAC is measuring it is a waste of time. No good DAC will alter anything enough for us to hear a difference. This is an on going discussion with a new member about this. He pointed to a Neural Network model of human hearing at MIT. Again a DAC isn't changing anything you could hear as soundstage.

You could try this test if you want.

Online LEDR™ Sound Test | Listening Environment Diagnostic Recording Test

Audio Test Tones : check your stereo system for correct imaging

www.audiocheck.net

 

[Mikig]

Sound stage is hard to measure because it's hard to define, despite all the discussion we don't even seem to have a solid definition of something that could be rated on a 1-10 scale, which would at least be a start toward quantifying it.

"How convincing is the stereo illusion" might be it?

Or

"How far apart are the apparent lateral limits of the stereo image"?

Or something else entirely more subtle about the perception of an illusory space?

We at least need to pin down a single perceptual phenomenon / dimension of preference before we start correlating it to measurements.

To put this in perspective and extend @dasdoing 's comment a little - in the studio there are many "tricks" that can be used to create/enhance something you would call soundstage. Recording in stereo, with varying levels of reverb in the recording space... using mics to record the room vs. direct... adding (various styles of) reverb after the fact... doubling/panning a track and tweaking pitch/phase/EQ, adding stereo chorus, etc. All of them have to do with partially decorrelating the left and right signals.

For "depth" (front/back perception) you use filtering/EQ, dry/wet reverb ratios, and of course simple volume controls.

So if I had a piece of electronics that noticeably enhanced soundstage, I would assume something like @dasdoing said - it's messing with phase or adding some kind of semi-randomness on a per-channel basis. Distortion, in other words.

So, bottom line... high fidelity gear shouldn't enhance soundstage.

However, the question remains whether there is some kind of distortion that would be especially likely to diminish soundstage that exists in the recording... that's something that I think is interesting, worth looking into via measurements. But first we have to have a good idea of what perception of "soundstage" really is in practice.

The good "Amused to Death" by Roger Waters comes to mind as an example. sounds perceived in the room coming from everywhere, even in two stereo channels

 

[pablolie]

You don't measure "sound stage". It is included in the FR of a recording that attempts to showcase it. The FR reflects the time delays etc that sound stage is encoded within.

It's kind of asking "How do I measure how well an Ibanez guitar is recorded ?"... all you can do is record the original information in a way that hopefully captures everything relevant in a standardized format - and then you measure the accuracy when you play it back, as in how true it remains to the original signal. Stage is there if it was captured.

 

[pablolie]

Soundstage is either encoded in the recording or it isn't. All the equipment has to do is reproduce the recording accurately. Linear, neutral equipment will provide sound stage if (a) it measures as delivering accuracy to the original signal and (b) if it is set up appropriately in a suitable environment.

There is nothing special about sound stage. If it is in the recording and the equipment is accurate, it will be highlighted accordingly. But no equipment will focus on better "soundstage" delivery, all it can do is deliver the original recording with accuracy.

Outside of that basic stuff, I find the key to sound stage is speaker positioning (provided the chain is accurate).

 

[BDWoody]

Don't worry, you haven't entered "The Twilight Zone", I just merged a couple of related threads.

 

[Blumlein 88]

The good "Amused to Death" by Roger Waters comes to mind as an example. sounds perceived in the room coming from everywhere, even in two stereo channels

Yes it was encoded with Q-sound.

 

[Chester]

Q-sound can manipulate where things come from and the soundstage as well.

Yes, anyone who wants to hear what a wide soundstage sounds like, take a listen to Madonna - The Immaculate Collection. Sound as wide as your listening room, thanks to Q-Sound.

 

[Anton D]

Yes, anyone who wants to hear what a wide soundstage sounds like, take a listen to Madonna - The Immaculate Collection. Sound as wide as your listening room, thanks to Q-Sound.

Great example, as is the Roger Waters recording!

 

[Anton D]

Soundstage is either encoded in the recording or it isn't. All the equipment has to do is reproduce the recording accurately. Linear, neutral equipment will provide sound stage if (a) it measures as delivering accuracy to the original signal and (b) if it is set up appropriately in a suitable environment.

There is nothing special about sound stage. If it is in the recording and the equipment is accurate, it will be highlighted accordingly. But no equipment will focus on better "soundstage" delivery, all it can do is deliver the original recording with accuracy.

Outside of that basic stuff, I find the key to sound stage is speaker positioning (provided the chain is accurate).

I find mono recordings actually give me great info regarding soundstage.

A good set up will place all the sound in what can seem like a very good central image/soundstage, with no wandering outside this central image. It's not in the recording at all, it's in the playback!

Warning, trigger word ahead, SINADologists might be disturbed by the word....stop now if you are a delicate flower...

Stereophile's John Atkinson uses monophonic white noise as an indicator of a system's soundstaging/imaging.

He will play that and look for how central and how narrow the soundstage/imaging is. There seems to be a correlation between how well a system performs on this 'test' and it seems to match well with subsequent stereo listening and image/soundstage quality.

 

[Mikig]

that I know of, stereophony and so Spatial listening presupposes recording with at least two microphones and diffusion with at least two loudspeakers. This gives the possibility, since we hear with two ears, to perceive the space and the origin of the direction.

 

[pablolie]

that I know of, stereophony and so Spatial listening presupposes recording with at least two microphones and diffusion with at least two loudspeakers. This gives the possibility, since we hear with two ears, to perceive the space and the origin of the direction.

It really is as simple as that.

 

[pablolie]

...

Stereophile's John Atkinson uses monophonic white noise as an indicator of a system's soundstaging/imaging.
...

Well, we hear a lot of weird things in audio, but if that's true it deserves some sort of audiophile nonsense award.

 

[Anton D]

Well, we hear a lot of weird things in audio, but if that's true it deserves some sort of audiophile nonsense award.

It actually makes sense if you think about it!

It should yield a narrow and specific "image."

 

[kemmler3D]

He will play that and look for how central and how narrow the soundstage/imaging is.

That actually makes good sense, I think? I'm assuming he listens to both speakers with the same noise signal.

If nothing else it will give you a good sense of channel matching between two speakers. I *think* it would also give an idea of directivity being even or not, since if some frequencies sound "wider" than others, it would indicate they're getting more or less than their fair share of reflections.

Using white noise instead of pink will tend to overemphasize treble, but that would probably just make the above issues easier to hear.

It's a similar concept to doing a null test... listen in mono and see if any stereo effects creep out anyway. A truly good stereo setup should reproduce mono that sounds perfectly centered.

Well, we hear a lot of weird things in audio, but if that's true it deserves some sort of audiophile nonsense award.

Respectfully disagree per the above points. I don't think this is a full and definitive test of stereo / soundstage / imaging capabilities, but I like it because it seems like it would reveal flaws quickly. For me, that's the utility of any material used for a listening test. I don't like to sit there judging nuances of sound quality subjectively, but rather go down a list eliminating flaws via stress tests, first.