Our perception of audio

[Cosmik]

Speakers with a downward sloping power response don't require voicing. The aim is always flat axial response and a downward sloping power response. Or am I misunderstanding what you mean by voicing?

Also, CD (including omni) do sound different depending on the room they are in. In all cases (omni, CD, downward-sloping, etc), the room will absorb and reflect differently at different frequencies. Speakers will interact differently and sound different in different rooms. This is true regardless of the dispersion pattern, although of course the narrower the directivity the less the extent of interaction (due to a higher ratio of direct to reflected sound).

I refer you to my earlier posts! We hear the direct sound from the speaker, and we hear the room. The combination will only sound natural if the room corresponds naturally with the direct sound: every reflection accounted for with the necessary frequency peak or trough; every frequency-selective attenuation decaying at the right rate at the right time, etc. Otherwise we hear the artificiality of it. Only a constant directivity speaker can do this.

If we keep our CD speaker but change the room, we hear the same direct sound but a different room. It will still sound right, even though we may prefer the sound of one room over another.

 

[svart-hvitt]

I refer you to my earlier posts! We hear the direct sound from the speaker, and we hear the room. The combination will only sound natural if the room corresponds naturally with the direct sound: every reflection accounted for with the necessary frequency peak or trough; every frequency-selective attenuation decaying at the right rate at the right time, etc. Otherwise we hear the artificiality of it. Only a constant directivity speaker can do this.

If we keep our CD speaker but change the room, we hear the same direct sound but a different room. It will still sound right, even though we may prefer the sound of one room over another.

Is your argument «combination will only sound natural if the room corresponds naturally with the direct sound» correct if the speaker is not point source?

 

[Cosmik]

Is your argument «combination will only sound natural if the room corresponds naturally with the direct sound» correct if the speaker is not point source?

Ideally, I think it needs to behave as a point source.

 

[andreasmaaan]

I refer you to my earlier posts! We hear the direct sound from the speaker, and we hear the room. The combination will only sound natural if the room corresponds naturally with the direct sound: every reflection accounted for with the necessary frequency peak or trough; every frequency-selective attenuation decaying at the right rate at the right time, etc. Otherwise we hear the artificiality of it. Only a constant directivity speaker can do this.

If we keep our CD speaker but change the room, we hear the same direct sound but a different room. It will still sound right, even though we may prefer the sound of one room over another.

This argument doesn't make sense to me, or perhaps I'm misunderstanding what you mean by "natural".

Real sounds are rarely (never??) constant directivity. Do these not sound natural in real rooms?

Or what do you mean by "sound natural"?

 

[svart-hvitt]

Ideally, I think it needs to behave as a point source.

«As if».

Sounds like an economist...



Even if point source is an ideal, we have different ways of measuring if the speaker behaves as a point source.

Any suggestions what qualifies or disqualifies as a point source, or what is acceptably point source?

 

[Cosmik]

This argument doesn't make sense to me, or perhaps I'm misunderstanding what you mean by "natural".

Real sounds are rarely (never??) constant directivity. Do these not sound natural in real rooms?

Or what do you mean by "sound natural"?

This came up in a different thread. We are not aiming to recreate each instrument as though in a fairground organ (which contains a physical violin, etc.) but to reproduce the sound of that instrument combined with others as they sounded in the original venue - and how their individual dispersion characteristics affected the sound in that venue (give or take the caveats to do with how recordings are made with multiple mics, etc.)

We have a choice of beaming the recording directly to our ears, 'dry' - various ways to do it e.g. sit very close to the speakers, use an anechoic chamber or super-beamy speakers, headphones, head tracking, etc.

Or we can allow our room to add a little of its own ambience and hopefully achieve a more relaxing experience that perhaps resembles having the musicians in our own living room or transforming our living room into a window onto the live venue. It's a two-way thing: our own voices blend with the recording in our room and it feels very natural.

Directivity allows us to dial in the ratio of room ambience that we prefer and to some extent how much we can move around. It doesn't affect our perceived frequency response because we still hear the direct sound separate from the room.

The system only works if the transducer is constant directivity. Otherwise there is a difference in colour between the direct and room sounds. The timing and frequency response cues that should all point back to the direct sound instead point at a different, coloured source. Maybe if this non-uniform dispersion characteristic was applied to only a single dry source e.g. the natural sound source you mention above, it would sound plausible like a fairground organ. But if the recording comprises several sound sources and the venue acoustics, then passing the whole lot through a single, fixed non-neutral transducer will expose the characteristic of the transducer - this is true for many other forms of distortion, too.

The discrepancy cannot be corrected with EQ.

Or that's the way I see it. I think it's one of those ideas that's either going to be totally right, or totally wrong. The evidence, I tentatively suggest, is that the latest technological developments are showing it to be right . (Again, the quote from the D&D designer about voicing being unnecessary for a speaker with constant directivity being Exhibit A).

 

[svart-hvitt]

This came up in a different thread. We are not aiming to recreate each instrument as though in a fairground organ (which contains a physical violin, etc.) but to reproduce the sound of that instrument combined with others as they sounded in the original venue - and how their individual dispersion characteristics affected the sound in that venue (give or take the caveats to do with how recordings are made with multiple mics, etc.)

We have a choice of beaming the recording directly to our ears, 'dry' - various ways to do it e.g. sit very close to the speakers, use an anechoic chamber or super-beamy speakers, headphones, head tracking, etc.

Or we can allow our room to add a little of its own ambience and hopefully achieve a more relaxing experience that perhaps resembles having the musicians in our own living room or transforming our living room into a window onto the live venue. It's a two-way thing: our own voices blend with the recording in our room and it feels very natural.

Directivity allows us to dial in the ratio of room ambience that we prefer and to some extent how much we can move around. It doesn't affect our perceived frequency response because we still hear the direct sound separate from the room.

The system only works if the transducer is constant directivity. Otherwise there is a difference in colour between the direct and room sounds. The timing and frequency response cues that should all point back to the direct sound instead point at a different, coloured source. Maybe if this non-uniform dispersion characteristic was applied to only a single dry source e.g. the natural sound source you mention above, it would sound plausible like a fairground organ. But if the recording comprises several sound sources and the venue acoustics, then passing the whole lot through a single, fixed non-neutral transducer will expose the characteristic of the transducer - this is true for many other forms of distortion, too.

The discrepancy cannot be corrected with EQ.

Or that's the way I see it. I think it's one of those ideas that's either going to be totally right, or totally wrong. The evidence, I tentatively suggest, is that the latest technological developments are showing it to be right . (Again, the quote from the D&D designer about voicing being unnecessary for a speaker with constant directivity being Exhibit A).

Did the D&D engineer say something among these lines?

https://www.genelec.com/directivity-control-waveguide-dcw-technology

(I am baffled again and again at how old technology is presented as novel. First cardioid radiation and now constant directivity).

 

[Floyd Toole]

Ideally, I think it needs to behave as a point source.

A point source is a theoretical concept. A practical implementation would have all sound radiating from a source that is small compared to the shortest wavelength being radiated (17 mm at 20kHz). Good luck with that! Since the dominant reflecting surfaces in rooms are vertical or horizontal the most important (strongest) reflections occur in those planes. This is why, in 1983, I began full-circle measurements using polar and equatorial orbits. There is no requirement for loudspeakers to be perfectly omnidirectional, but there are preferred performance criteria on horizontal and vertical axes. Because the ears are in the horizontal plane, the lateral reflections have the greatest influence on perception.

BTW, I just noticed your "motto". A "sloping in-room response" IS a measurement - a steady-state one. Humans prefer flat direct sound, which requires loudspeakers with flat on-axis and listening window frequency responses. Typical forward-firing loudspeakers will then exhibit downward sloping steady-state in-room response because of the frequency-dependent directivity, combined with some air absorption at very high frequencies. It is an idea worth repeating, but it is not correct as written. Sorry.

 

[andreasmaaan]

@Cosmik this was a great post, exactly hitting on those points of yours that I was struggling to understand.

We are not aiming to recreate each instrument as though in a fairground organ (which contains a physical violin, etc.) but to reproduce the sound of that instrument combined with others as they sounded in the original venue - and how their individual dispersion characteristics affected the sound in that venue (give or take the caveats to do with how recordings are made with multiple mics, etc.)

Ok, I essentially agree with this as you know.

The system only works if the transducer is constant directivity. Otherwise there is a difference in colour between the direct and room sounds. The timing and frequency response cues that should all point back to the direct sound instead point at a different, coloured source of direct sound. Maybe if this non-uniform dispersion characteristic was applied to only a single dry source e.g. the natural sound source you mention above, it would sound plausible like a fairground organ. But if the recording comprises several sound sources and the venue acoustics, then passing the whole lot through a single, fixed non-neutral transducer will expose the characteristic of the transducer

But how can something that wasn't ever even present in the recording (i.e. a reflection in the room) be either accurate or "coloured"? It just can't be to my mind. There is no correct reference to which it can be compared.

It seems to me that it's this basic difference in how we see things that's led to our different ideas about this topic. I do appreciate the elegance of your approach to it, however.

(Again, the quote from the D&D designer about voicing being unnecessary for a speaker with constant directivity being Exhibit A).

Well, voicing isn't necessary for any speaker that aims to be neutral, so I'm not particularly moved by that comment. It's got nothing particularly to do with whether a speaker is constant directivity or not.

"Voicing" is to smooth the power response of speakers with erratic polar response, or for tinkerers who don't believe in accuracy. Little to do with whether or not a speaker is CD.

Re: the point source discussion @svart-hvitt and @Cosmik (and now @Floyd Toole!), I've always felt somewhat conflicted about it.

I can clearly see why it's preferable to lobing exhibit by non-coincident radiators in the upper mids and highs, so no argument from me there, and in my own speakers I am at pains to reduce this lobing as much as possible.

But when it comes to the low mid-range and bass (the region most affected by destructive first reflections and the modal region, respectively), in most cases having a distributed sound source will tend to smooth out the steady state response. My view is that there's at least a plausible argument why this is more important than point source behaviour in these frequencies.

Whereas in the rest of the spectrum we tend to clearly distinguish the direct sound from the reflected sound in terms of perceiving tonal balance, my interpretation of the studies (including Dr Toole's) is that, in-room, below a certain frequency, our ability to distinguish the direct sound from the steady state response collapses. In this frequency range I would argue that point source behaviour is in fact a disadvantage, as it will tend to have a negative impact on the steady state response as compared to a distributed sound source.

 

[Cosmik]

@Cosmik this was a great post, exactly hitting on those points of yours that I was struggling to understand.



Ok, I essentially agree with this as you know.



But how can something that wasn't ever even present in the recording (i.e. a reflection in the room) be either accurate or "coloured"? It just can't be to my mind. There is no correct reference to which it can be compared.

It seems to me that it's this basic difference in how we see things that's led to our different ideas about this topic. I do appreciate the elegance of your approach to it, however.



Well, voicing isn't necessary for any speaker that aims to be neutral, so I'm not particularly moved by that comment. It's got nothing particularly to do with whether a speaker is constant directivity or not.

"Voicing" is to smooth the power response of speakers with erratic polar response, or for tinkerers who don't believe in accuracy. Little to do with whether or not a speaker is CD.

Re: the point source discussion @svart-hvitt and @Cosmik (and now @Floyd Toole!), I've always felt somewhat conflicted about it.

I can clearly see why it's preferable to lobing exhibit by non-coincident radiators in the upper mids and highs, so no argument from me there, and in my own speakers I am at pains to reduce this lobing as much as possible.

But when it comes to the low mid-range and bass (the region most affected by destructive first reflections and the modal region, respectively), in most cases having a distributed sound source will tend to smooth out the steady state response. My view is that there's at least a plausible argument why this is more important than point source behaviour in these frequencies.

Whereas in the rest of the spectrum we tend to clearly distinguish the direct sound from the reflected sound in terms of perceiving tonal balance, my interpretation of the studies (including Dr Toole's) is that, in-room, below a certain frequency, our ability to distinguish the direct sound from the steady state response collapses. In this frequency range I would argue that point source behaviour is in fact a disadvantage, as it will tend to have a negative impact on the steady state response as compared to a distributed sound source.

Re. voicing, I am in agreement with the D&D designer who says:

Voicing is required to balance differences between direct and off-axis sound. The 8c has very even dispersion. It is the first loudspeaker I ever designed that did not benefit from voicing. The tonal balance is purely based on anechoic measurements.

He is clearly saying that a constant directivity speaker does not need voicing because its direct and off-axis sounds are the same (would you not say?).

Beyond that, you could apply any EQ curve you like, but then I think you are in that position of passing the whole orchestra (for example) through a coloured transducer which cannot help but show up the EQ curve.

I simply have to stick my guns when it comes to the bass . I will continue to use speakers in a single box (or bolted together as one) i.e. as close to 'point source' as possible and without any particular fiddling with the bass response except for where I roll it off - I will take it as it comes, in the expectation that the whole system interacts with the room at all frequencies in a 'natural' way. I can't relent and go non-purist on just one part of the spectrum! It sounds pretty good to me as it is.

 

[andreasmaaan]

He is clearly saying that a constant directivity speaker does not need voicing because its direct and off-axis sound is the same (would you not say?).

Yes of course, but he’s comparing it to speakers that exhibit polar response discontinuities between drivers at the xo point. There’s no need to “voice” speakers that have a smoothly sloping polar response by virtue of matched directivities between the drivers at the xo points. IMHO in any case.

I simply have to stick my guns when it comes to the bass . I will continue to use speakers in a single box i.e. as close to 'point source' as possible and without any particular fiddling with the bass response except for where I roll it off - I will take it as it comes, in the expectation that the whole system interacts with the room at all frequencies in a 'natural' way. I can't relent and go non-purist on just one part of the spectrum! It sounds pretty good to me as it is.

Very fair

 

[Cosmik]

Yes of course, but he’s comparing it to speakers that exhibit polar response discontinuities between drivers at the xo point. There’s no need to “voice” speakers that have a smoothly sloping polar response by virtue of matched directivities between the drivers at the xo points. IMHO in any case.

Ah, well maybe I'm misunderstanding the meaning of constant directivity. I am meaning (and I think the D&D designer is too..?) a fixed uniform dispersion angle at all frequencies, not just avoidance of discontinuities. For sure it's not possible to achieve this perfectly - just like a literal point source is not possible - but it's a good objective.

 

[andreasmaaan]

Ah, well maybe I'm misunderstanding the meaning of constant directivity. I am meaning (and I think the D&D designer is too..?) a fixed uniform dispersion angle at all frequencies, not just avoidance of discontinuities.

No that’s what I understand it to mean too.

I think the difference in understanding is of the word “voicing”. In my book, the BBC dip is an example of voicing. The tweeter is far wider dispersion than the woofer, so it’s axial response is nudged down just above the xo point to achieve a flatter power response, which is considered to result in a more neutral sound in-room. You trade flat axial response for a smoother power response.

If OTOH you have matched directivities between drivers at the xo point, there’s no need to voice (at least not in my view) - even if the net result is a downward sloping power response.

What do you think of that way of looking at it?

 

[Cosmik]

BTW, I just noticed your "motto". A "sloping in-room response" IS a measurement - a steady-state one. Humans prefer flat direct sound, which requires loudspeakers with flat on-axis and listening window frequency responses. Typical forward-firing loudspeakers will then exhibit downward sloping steady-state in-room response because of the frequency-dependent directivity, combined with some air absorption at very high frequencies. It is an idea worth repeating, but it is not correct as written. Sorry.

Many thanks for the comment Floyd. What I am getting at with my signature is that what the microphone and FFT 'hears' (i.e. displays on the screen) is not the same as the listener hears. A different measurement (e.g. gated) will give a different curve for the same sound, so clearly an in-room measurement can slope independently from what the listener is hearing - and from what would be displayed for the most appropriate measurement type - even with a perfectly neutral transducer.

 

[Cosmik]

Is this a way to look at it?

Have you seen those TVs that have 'ambilight'? They project light reminiscent of the picture colour onto the wall behind the TV.

But supposing that the typical TV didn't have very accurate ambilight colours. So when you were watching a grey image, there was a reddish halo around the picture, and whatever the picture showed, the halo appeared somewhat pinkish in comparison. It would be very annoying and, unlike a good ambilight setup, would draw attention to itself more than if it wasn't there at all. A measurement of the combination using a diffuse spectrometer (I just made that term up) would show a downward sloping colour balance.

You could try to compensate for it. By making the image slightly blue-ish, you could make it so that the measured colour balance was slightly less downwards sloping. You could, in fact, achieve a flat measurement if you wanted to. However, the viewer would not be fooled, seeing the screen image as distinctly blue when viewed directly. There would be no substitute for a TV with a neutral screen colour balance and accurate ambilight colours.

 

[andreasmaaan]

Is this a way to look at it?

Have you seen those TVs that have 'ambilight'? They project light reminiscent of the picture colour onto the wall behind the TV.
View attachment 19251

But supposing that the typical TV didn't have very accurate ambilight colours. So when you were watching a grey image, there was a reddish halo around the picture, and whatever the picture showed, the halo appeared somewhat pinkish in comparison. It would be very annoying and, unlike a good ambilight setup, would draw attention to itself more than if it wasn't there at all. A measurement of the combination using a diffuse spectrometer (I just made that term up) would show a downward sloping colour balance.

You could try to compensate for it. By making the image slightly blue-ish, you could make it so that the measured colour balance was slightly less downwards sloping. You could, in fact, achieve a flat measurement if you wanted to. However, the viewer would not be fooled, seeing the screen image as distinctly blue when viewed directly. There would be no substitute for a TV with a neutral screen colour balance and accurate ambilight colours.

Haha well no

Ambilight is a deliberate effect added to the picture for... well I have to admit I’ve never understood.

Sonic reflections in room are a necessary result of reproducing the recording, so the question is not whether and how but simply how.

A+ for diffuse spectrometer, can we think up an audio thing and use that for it maybe lol?

 

[svart-hvitt]

A point source is a theoretical concept. A practical implementation would have all sound radiating from a source that is small compared to the shortest wavelength being radiated (17 mm at 20kHz). Good luck with that! Since the dominant reflecting surfaces in rooms are vertical or horizontal the most important (strongest) reflections occur in those planes. This is why, in 1983, I began full-circle measurements using polar and equatorial orbits. There is no requirement for loudspeakers to be perfectly omnidirectional, but there are preferred performance criteria on horizontal and vertical axes. Because the ears are in the horizontal plane, the lateral reflections have the greatest influence on perception.

BTW, I just noticed your "motto". A "sloping in-room response" IS a measurement - a steady-state one. Humans prefer flat direct sound, which requires loudspeakers with flat on-axis and listening window frequency responses. Typical forward-firing loudspeakers will then exhibit downward sloping steady-state in-room response because of the frequency-dependent directivity, combined with some air absorption at very high frequencies. It is an idea worth repeating, but it is not correct as written. Sorry.

@Floyd Toole , your position, and many ASR members’, on point source seems to be:

1) Point source is ideal
2) Point source is a theoretical concept that cannot be fully implemented in real life
3) Psychoacoustically, traditionally stacked drivers work just fine, as evidenced by listening tests where researchers used traditionally stacked driver speakers

You didn’t really write (1), but I have never met anyone who hasn’t been in agreement that point source is ideal. So I put (1) there to make this discussion a target for a wider audience.

On (2): Because something, an ideal, is impossible to work out 100 percent in reality, does it mean we should give up the idea of the ideal? We have different ways of engineering things:

a) You can tweak and perfect an old proven design
b) You can come up with a better idea, still unproven, and set out on a journey to realize that idea even if you know that the idea is a theoretical one that can never be realized 100 percent.

I believe I am a (b) person; if someone tells me my idea is an impossibility, I may still spend some time on it because I know option (a) is unchallinging and boring. And I believe that if I realize just a part of a BIG idea, I may still create something which is greater than a perfect realization of the conventional, yet smaller idea.

On (3): All (?) psychoacoustic studies that we normally refer to have been done on conventionally stacked design. How can one use these studies in arguments on the pros et cons of point sources? Isn’t it like running in circles? Shouldn’t point source (attempting) designs be used to evaluate the value and validity of point source design (attempts). My point is, how strong is the psychoacoustic research that we normally refer to? Is it so strong that we can disregard attempts - existing and future - to realize the point source ideal?

Existing point source designs have been proven to have more robust directivity than even a two-way design. So I cannot understand that @Floyd Toole and others think chasing the point source ideal is futile («Good luck with that!»).

Because existing point source designs already have superior directivity off and on axis, I think one should spend more time discussing the drawbacks (and positive sides) of these point source designs - as evidenced by measurements and data - instead of attacking attempts at bridging the gap between idea and real-life engineering.

Let’s not kill the idea of point source without data to support the position that point source is futility.

 

[andreasmaaan]

You didn’t really write (1), but I have never met anyone who hasn’t been in agreement that point source is ideal.

Here’s one! See my previous post.

 

[svart-hvitt]

Here’s one! See my previous post.

Heretic!



So you disagree, then, with Watkinson’s list:

https://www.thebroadcastbridge.com/...echnology-part-15-a-catalogue-of-shortcomings

Maybe I should start a thread on Watkinson’s 16 articles on Broadcastbridge? He seems to be an ideas guy.

PS: Did you refer to your post #296

 

[Cosmik]

@Floyd Toole , your position, and many ASR members’, on point source seems to be:

1) Point source is ideal
2) Point source is a theoretical concept that cannot be fully implemented in real life
3) Psychoacoustically, traditionally stacked drivers work just fine, as evidenced by listening tests where researchers used traditionally stacked driver speakers

You didn’t really write (1), but I have never met anyone who hasn’t been in agreement that point source is ideal. So I put (1) there to make this discussion a target for a wider audience.

On (2): Because something, an ideal, is impossible to work out 100 percent in reality, does it mean we should give up the idea of the ideal? We have different ways of engineering things:

a) You can tweak and perfect an old proven design
b) You can come up with a better idea, still unproven, and set out on a journey to realize that idea even if you know that the idea is a theoretical one that can never be realized 100 percent.

I believe I am a (b) person; if someone tells me my idea is an impossibility, I may still spend some time on it because I know option (a) is unchallinging and boring. And I believe that if I realize just a part of a BIG idea, I may still create something which is greater than a perfect realization of the conventional, yet smaller idea.

On (3): All (?) psychoacoustic studies that we normally refer to have been done on conventionally stacked design. How can one use these studies in arguments on the pros et cons of point sources? Isn’t it like running in circles? Shouldn’t point source (attempting) designs be used to evaluate the value and validity of point source design (attempts). My point is, how strong is the psychoacoustic research that we normally refer to? Is it so strong that we can disregard attempts - existing and future - to realize the point source ideal?

Existing point source designs have been proven to have more robust directivity than even a two-way design. So I cannot understand that @Floyd Toole and others think chasing the point source ideal is futile («Good luck with that!»).

Because existing point source designs already have superior directivity off and on axis, I think one should spend more time discussing the drawbacks (and positive sides) of these point source designs - as evidenced by measurements and data - instead of attacking attempts at bridging the gap between idea and real-life engineering.

Let’s not kill the idea of point source without data to support the position that point source is futility.

As I see it, the conventional vertical line of drivers is a reasonable approximation of a point source because the drivers get smaller and closer together as the wavelengths get shorter.

The main reason for wanting a point source is - I think - the alternative thought experiment: do you want your audio coming from multiple drivers spread around the room, even if they are delayed/phased correctly for the listening position? What will the reverberation point back to? What will it sound like if you move your head?

And then there is the question of directivity. How do you achieve that from a point source? You may end up having to use something that looks bigger than a point source, but behaves like a directional point source. e.g. the Kii system of using extra drivers to cancel out the sound trying to get round the back. Hopefully the cancellation is perfect and the speaker behaves like an acoustically small object that is directional.