• 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!

What makes the sensation of "sound stage"?

richard12511

Major Contributor
Forum Donor
Joined
Jan 23, 2020
Messages
4,335
Likes
6,700
2 dimensions would be 2d :) 3d with time as a dimension is also in the mix. Most anything in our audio experience is an illusion, tho....

Are you agreeing with me or disagreeing with me? I honestly can't tell :D.

My thinking is:

2 points = 1D
3 points = 2D
4 points = 3D
2x3D points = 4D

In actual speaker config, though(no subs), I'm thinking:

2.0.0 = 1D
5.00 = 2D
5.0.2/4 = 3D

There was a great youtube video that really made understanding multiple dimensions all click for me, but I can't seem to find it :(. This one is pretty close to what I remember, though. Sorry for the off topic, but it's a cool concept.
 

Chrispy

Master Contributor
Forum Donor
Joined
Feb 7, 2020
Messages
7,751
Likes
5,910
Location
PNW
Are you agreeing with me or disagreeing with me? I honestly can't tell :D.

My thinking is:

2 points = 1D
3 points = 2D
4 points = 3D

In actual speaker config, though(no subs), I'm thinking:

2.0.0 = 1D
5.00 = 2D
5.0.2/4 = 3D

There was a great youtube video that really made understanding multiple dimensions all click for me, but I can't seem to find it :(. This one is pretty close to what I remember, though. Sorry for the off topic, but it's a cool concept.

How would one point be defined then? No dimensions? 2 points would define two dimensions at least in my schooling....
 
OP
D

Destination: Moon

Senior Member
Joined
Feb 28, 2021
Messages
478
Likes
314
Location
Western USA
Are you agreeing with me or disagreeing with me? I honestly can't tell :D.

My thinking is:

2 points = 1D
3 points = 2D
4 points = 3D
2x3D points = 4D

In actual speaker config, though(no subs), I'm thinking:

2.0.0 = 1D
5.00 = 2D
5.0.2/4 = 3D

There was a great youtube video that really made understanding multiple dimensions all click for me, but I can't seem to find it :(. This one is pretty close to what I remember, though. Sorry for the off topic, but it's a cool concept.


It's sound, like the light we can sense, it exists in 3 dimensions. We sense light with 2 eyes that present the illusion of objects as they exist in 3D space (let's not confuse things by adding time on top). The same occurs with sound. We have 2 ears that sense sound that exists in 3 dimensions. If you only had 1 eye you can still see shadows so some 3d info is still there but greatly diminished. If we only had 1 ear?
 

richard12511

Major Contributor
Forum Donor
Joined
Jan 23, 2020
Messages
4,335
Likes
6,700
It's sound, like the light we can sense, it exists in 3 dimensions. We sense light with 2 eyes that present the illusion of real objects as they exist in 3D space (let's not confuse things by adding time on top). The same occurs with sound. We have 2 ears that sense sound that exists in 3 dimensions. If you only had 1 eye you can still see shadows so some 3d info is still there but greatly diminished. If we only had 1 ear?

I don't disagree with what you say, but I think we're talking about two completely different things.
 

audiofooled

Addicted to Fun and Learning
Joined
Apr 1, 2021
Messages
530
Likes
588
And then there's this:


Edit: I see this has already been published on the forum, as well as other stuff from Geoff Martin.
 
Last edited:

fmplayer

Member
Joined
Sep 3, 2017
Messages
96
Likes
72
That said, I do agree that it can seem 3D at times, which is really awesome, but it's just a clever illusion. A testament to the cleverness of mix engineers, I suppose.
Exactly. Mixing is very a challenging activity in which the engineer must use every psychoacoustical trick to fool the brain which is all but able to do its usual job in sound differentiation and localization
I don't have great experience with multichannel music, but what I have heard does sound a little better, but it's not a huge difference(at least what I've heard). What I've heard has only been 2D, though(5.1/7.1). I do have height channels, but I don't own any 3D music yet.
Additional channels are maybe more oriented towards a greater perception of the acoustics of the recording space. It's more of a "you being there" than "they being here" strategy
 

audiofooled

Addicted to Fun and Learning
Joined
Apr 1, 2021
Messages
530
Likes
588
Not about the Beolab 90's but the description of how we perceive imaging and the effects of loudspeaker directivity at different frequencies and room reflections on imaging. What is closely related to our topic here would be around 13:30 but to better understand the concept you may want to watch from the beginning:

 

audiofooled

Addicted to Fun and Learning
Joined
Apr 1, 2021
Messages
530
Likes
588
Please note that nothing behind the speakers plan is shown ...

Yes, this is not about the imaging depth, but to show how inter-channel amplitude difference we hear differently than inter-channel time difference. And how it is manipulated in the mix.
 

fmplayer

Member
Joined
Sep 3, 2017
Messages
96
Likes
72
Beolab 90 is one of those DSP powered speakers aiming at minimizing room influence, alongside Dutch&Dutch 8c and Bruno Putzeys' Kii Three which I cited earlier.
 

audiofooled

Addicted to Fun and Learning
Joined
Apr 1, 2021
Messages
530
Likes
588
Beolab 90 is one of those DSP powered speakers aiming at minimizing room influence, alongside Dutch&Dutch 8c and Bruno Putzeys' Kii Three which I cited earlier.

Yes, there's plenty on here on Beolab's but regardless of those in this presentation Geoff Martin I think provides a good explanation about very sound stage and imaging different speakers provide, as an observation of OP. They just use DSP and many drivers to mimic these (many different speakers) all in one. If we disregard Beolab 90's and just think of different speaker characteristics, we can see the difference of speaker-room interaction.

I think that most of us love big sound stage and precise imaging and depth all together, because it is required if we want to "declutter" and unfold the mixed cues and perceive them as layered, focused and orderly for critical listening. These are different things, and may be a lot to ask of just one pair of speakers, at least for far field listening. I am glad that there is so much research and finally modern ways to find "good ones", objectively measure and predict their in room response (read ASR ;)). I wish I had access to any, let alone modern equipment when I designed mine.

So, it turns out that if we want correct and focused imaging (true to the recording), we need better, more uniform directivity across the frequency spectrum. The more frequency dependent directivity is (i.e if we let lower frequencies be wider and higher frequencies narrower), the less focus we get. As given example, vocal occupies more places at the same time, which in return confuses the brain by blurring the perception. Further more, lower, i.e wider ranges of frequency reflect of the wall and give away where the speakers are located. Even more confusion. How can a listener compute directional cues if they are telling contradictory things? Not going to happen.

When we are engaged in near field listening, we just point any speakers on axis, et voila! But far field is apparently a different story. I think a lot is explained in the video if we look carefully.
 
Last edited:

fmplayer

Member
Joined
Sep 3, 2017
Messages
96
Likes
72
So, it turns out that if we want correct and focused imaging (true to the recording), we need better, more uniform directivity across the frequency spectrum. The more frequency dependent directivity is (i.e if we let lower frequencies be wider and higher frequencies narrower), the less focus we get. As given example, vocal occupies more places at the same time, which in return confuses the brain by blurring the perception. Further more, lower, i.e wider ranges of frequency reflect of the wall and give away where the speakers are located. Even more confusion. How can a listener compute directional cues if they are telling contradictory things? Not going to happen.

When we are engaged in near field listening, we just point any speakers on axis, et voila! But far field is apparently a different story. I think a lot is explained in the video if we look carefully.
If we want true to the recording listening, we must listen like in the studio, from lack of listening in the studio, and as you pointed out, nearfield listening is the easiest way to achieve the goal, provided your speakers are far from the walls.
It's not a matter of directivity, it's a matter of sound reflection. For us, mere audiophiles, directivity is so important because tonal balance of the speaker may be altered due to the reflections, and amir's measurements show that perfectly.
In the studio, they get rid of the reflections by positioning the speakers far from the walls or by soffit mounting them (and by acoustic treatment and construction of course).
Good directivity characteristics are here to allow more people (the band, the producer) be in the sweet spot at the same time.
Lack of reflections allow the sound engineer to be in good conditions to make the processing necessary to mix and make the illusion of depth, because then, he can rely only on his monitors, which are a tool to tailor the sound.

There are lots of good articles on Sound On Sound where all this stuff is explained for those interested. Few audiophiles know how the records are produced and have expectations that usually can't be fulfilled in our amateur environment
 

jonfitch

Senior Member
Joined
Sep 2, 2019
Messages
479
Likes
532
I think with headphones its mostly delayed reflections, the ringing that Amir has shown in the 1-3K range, is what people call soundstage. A lot of headphones with "big drivers" that envelope a listener are described as having a small soundstage, so soundstage is not necessarily the size of the sound field, but rather something that occurs within it.

With speakers I think it has to do with some kind of acoustical dip in the high frequency range, is BBC dip is pretty famous. There are a lot of speakers with wide dispersion that "envelope" a listener that people don't seem to associate with a large soundstage. I think soundstage is mostly a psychoacoustical effect in the high frequency range and may require a dip or recessed output here. My theory is we naturally think high frequency = narrow directivity. So a speaker that's simultaneously neutral in-room with wide dispersion in the high frequencies will end up sounding forward and boxed-in because frequencies we think should be narrow in directivity in a room are suddenly wide, which psycho-acoustically implies the object must be very close to the listener. So the listener will simultaneously think the speaker sounds "big" but also has "no soundstage"--because there is no directivity separation between instruments--every frequency sounds wide when we don't expect them to be.
 

audiofooled

Addicted to Fun and Learning
Joined
Apr 1, 2021
Messages
530
Likes
588
If we want true to the recording listening, we must listen like in the studio, from lack of listening in the studio, and as you pointed out, nearfield listening is the easiest way to achieve the goal, provided your speakers are far from the walls.
It's not a matter of directivity, it's a matter of sound reflection. For us, mere audiophiles, directivity is so important because tonal balance of the speaker may be altered due to the reflections, and amir's measurements show that perfectly.
In the studio, they get rid of the reflections by positioning the speakers far from the walls or by soffit mounting them (and by acoustic treatment and construction of course).
Good directivity characteristics are here to allow more people (the band, the producer) be in the sweet spot at the same time.
Lack of reflections allow the sound engineer to be in good conditions to make the processing necessary to mix and make the illusion of depth, because then, he can rely only on his monitors, which are a tool to tailor the sound.

There are lots of good articles on Sound On Sound where all this stuff is explained for those interested. Few audiophiles know how the records are produced and have expectations that usually can't be fulfilled in our amateur environment

Let us not forget Dr Toole's book, chapters 6-9. Here we may find extensive research on the subject of reflections and get familiar with the terms which very well describe listener's preference qualities of a loudspeaker, as well objective measurement results which would allow for more gratifying reproduction of upmixed recordings.
 

kristiansen

Member
Joined
Mar 12, 2021
Messages
81
Likes
29
Location
Hillerød Danmark
What is interesting about beolab 90 is not only the beam control, but also that it uses a large part of its speakers to mute the sound behind and around the speaker so that reflections from side and front walls are avoided, the principle has been used for many years in recording Studies and is called LEDE (Live end, Dead end) In studies it realized with acoustic materials, You can do the same if you're allowed by your wife.

I have LEDE and no wife anymore, I have made a drawing below, which shows the development of my soundstages, the closer you get to the optimal the more the speakers disappear ,you could say the core lines are being straightened out, the lines also indicate the system will play with the same soundstage out at one speaker as in the middle.
My experience is that amplifier etc. Plays a major role in the soundstage / imaging and its ability to vary in wide depth and height depending on what is on the source file.
perspektiv-engelsk.png
 
Last edited:

fmplayer

Member
Joined
Sep 3, 2017
Messages
96
Likes
72
I think with headphones its mostly delayed reflections, the ringing that Amir has shown in the 1-3K range, is what people call soundstage. A lot of headphones with "big drivers" that envelope a listener are described as having a small soundstage, so soundstage is not necessarily the size of the sound field, but rather something that occurs within it.

With speakers I think it has to do with some kind of acoustical dip in the high frequency range, is BBC dip is pretty famous. There are a lot of speakers with wide dispersion that "envelope" a listener that people don't seem to associate with a large soundstage. I think soundstage is mostly a psychoacoustical effect in the high frequency range and may require a dip or recessed output here. My theory is we naturally think high frequency = narrow directivity. So a speaker that's simultaneously neutral in-room with wide dispersion in the high frequencies will end up sounding forward and boxed-in because frequencies we think should be narrow in directivity in a room are suddenly wide, which psycho-acoustically implies the object must be very close to the listener. So the listener will simultaneously think the speaker sounds "big" but also has "no soundstage"--because there is no directivity separation between instruments--every frequency sounds wide when we don't expect them to be.
BBC dip was designed because of a subjective peak at the Xover frequency of their 2 way monitors. Google for "BBC research reports" about the speakers designed by the research department.
Some known facts:
High frequency EQUALS directivity. A driver becomes directional starting from the frequency where wavelength equates the driver diameter. It's jsut a fact of acoustics.
Box sound is rather related to the enclosure, and bad designed bass reflex is a known cause of boxy sound.
Big sound is related to energy in the 100-200Hz, and depends closely on driver membrane area. The more the better. EQing can help a little, but nothing replaces membrane area.
Please read also read my previous posts in this thread. It's a summary of things learned during the last years. No point to repeat. Do some mixing and mastering by yourself if you don't believe me.
Headphones: wanna know what soundstage is? try some artificial head recording. Nothing in common with frequency response which must be coherent with the harman curve.
 
Top Bottom