An Enticing Marketing Story, Theory Without Measurement?

[Guermantes]

Try to listen to someone sing outdoors versus indoor. You will change your mind.

This reminds me of the time we recorded vocals for a song in the backyard because we didn't like the sound of the recording room. Of course we added artificial reverb during the mix

 

[Guermantes]

The Schroeder frequency, as indicated by Schroeder himself, was created in the context of large performance spaces with significant diffusion, not small, relatively dead listening rooms. However the principle still applies but starting at a higher frequency than the one calculated. See Figure 6.2 in the 3rd edition of my book. It is not a well defined frequency in any room; it really indicates how much overlap there is between and among room resonances. I chose to call it the transition frequency. Below this frequency judicious equalization can be used to address individual prominent room resonances, but it is only functional at the point of measurement - the prime listening location. All other seats will be different. This is the reason why multiple subwoofers are advantageous (Chapter 8), along with the greatly increased efficiency.

At the upper range of this frequency range and above, up to a few hundred Hz, we encounter adjacent boundary effects and these can be addressed by equalization - See Chapter 9.

If I understand correctly, it's also about the best way to mathematically model the acoustics of the room. We can use reasonably simple mathematics to model room resonances below the Schroeder frequency but as we move above it, things become increasingly complex and so stochastic analysis is a better fit. It's almost as if there are two rooms (or at least a split personality): one mostly governed by dimensions and one mostly by materials.

Is it the transition area where use of room EQ is the most contested, partly because of above and partly because it also falls into a important area for musical perception?

 

[Floyd Toole]

If I understand correctly, it's also about the best way to mathematically model the acoustics of the room. We can use reasonably simple mathematics to model room resonances below the Schroeder frequency but as we move above it, things become increasingly complex and so stochastic analysis is a better fit. It's almost as if there are two rooms (or at least a split personality): one mostly governed by dimensions and one mostly by materials.

Is it the transition area where use of room EQ is the most contested, partly because of above and partly because it also falls into a important area for musical perception?

Simple math only works for perfectly rectangular, perfectly reflecting, empty rooms. More elaborate shapes can be analyzed, of course, using heavy math. I discuss some examples in my book in which wood frame construction exhibited measurable errors due to structural inconsistencies. However, being practical one can get useful guidance. When wavelengths shrink to the dimensions of furnishings and surface irregularities in domestic houses the simple math breaks down, resonances are damped and a combination of the increased density in the frequency domain and the lowered Q result in unstructured acoustical interference among modes. This is how I interpret the division between the two domains.

In addition, there are adjacent boundary issues superimposed on all of this, often extending above the transition frequency. Both the individual low frequency resonances and the evidence of adjacent boundary effects are responsive to parametric equalization. Chapters 8 and 9.

 

[Cosmik]

Hmmm. Once I notice it, I cannot but attend to how it changes my perception of the source.

Well this is fairly crucial to this topic. I say that I am 'hearing through the room' to the source, so that whether I am in an anechoic chamber or a church, I recognise voices and where they are coming from. I am aware of the acoustics of course, but I am separating them from the source. To me, then, room correction* would be folly: changing the source but leaving the room the same.

But you say that the room affects your perception of the source. Maybe we need a distinction: is the room affecting your perception of the tonality of the source, or at some other 'meta' level e.g. church-like acoustics imbue the source with calmness and seriousness etc.?

To me, the former seems unlikely. It would suggest that you would not be able to look at a source and determine from its materials, dimensions and so on, how it was going to sound or vice versa. Instead, it would be dependent on the room's acoustics, and even walking around the room, the source would be changing its apparent dimensions and materials as you walked! I perceive no such changes as I walk around a room chatting to someone.

In our house, if I walk from a highly reverberant large room into a carpeted smaller room carrying an iPad playing a podcast, I do notice a huge change in the acoustics, but I perceive no change in the voices, nor the iPad's own 'sound'.


*Not necessarily speaker correction/compensation, however.

 

[JJB70]

Anechoic chambers are awful places. I spent quite a bit of time in one testing certain equipment which was to be installed on board military vessels and found it uncomfortable and disconcerting.

 

[Ultrasonic]

In our house, if I walk from a highly reverberant large room into a carpeted smaller room carrying an iPad playing a podcast, I do notice a huge change in the acoustics, but I perceive no change in the voices, nor the iPad's own 'sound'.

Do you prefer the sound in one environment more than the other though? I have a cheap little battery powered radio that is definitely more enjoyable in my bathroom (where I use it) than in my much less reverberant lounge.

Going back to one of your earlier posts, being able to identify a person, instrument or whatever in different acoustic environments is very different to there being no difference to the perception and enjoyment of the sound. If I play a recording of a guitar in my car or on my main hifi both are recognisable as a guitar but they don't sound the same.

 

[March Audio]

Try to listen to someone sing outdoors versus indoor. You will change your mind.

Funny thing is, not saying anything difinitive here, but I have always found outside rock concerts sound infinitely better than in an enclosed arena. Probably a specific case, but in that instance I think the lack of loud long delayed reflections helps.

I don't think an unamplified orchestra would be good though.

 

[andreasmaaan]

I also prefer the sound of electronic music outdoors, all else equal.

 

[svart-hvitt]

I also prefer the sound of electronic music outdoors, all else equal.

@andreasmaaan and @March Audio , I guess this open air thing is about preferences, though I cannot escape the idea (IDEA as opposed to scientific fact) that open air is correcter than listening to a mix of direct and reverberant sounds.

@amirm , I wonder if your remark on singing in open air sounds strange is somewhat false. It depends on for example how close you are, and if the singer uses a microphone and speakers, doesn’t it?

Reverberations are colour, in my mind. Church builders, I guess even cave men, understood this - and used it to paint a picture that let the listeners get closer to God through music.

However, I cannot escape the idea that the ideal for producing and playing back music us open air, because then the «room» factor has been taken out of the equation. But in practice, we use and will forever (?) use our rooms to colour our music experience. And those rooms will never be the same, neither on production nor playback side.

That brings me back to a listening experience where I turned the DRC (digital room correction) on and off. My friend thought the uncorrected sounded the best. Was he more used to boomy bass than I was? Who is correct; he who enjoys uncorrected sound in a «deficient» room, or he who corrects the sound to match an ideal of flat’n smooth?

 

[Juhazi]

^yep, and controversially a choir or a small acoustic group sounds very bad outdoors (without PA) Many old outdoor veues have a half dome behind the stage to reflect sound to audience.

Tallinna laululava

 

[Cosmik]

Do you prefer the sound in one environment more than the other though? I have a cheap little battery powered radio that is definitely more enjoyable in my bathroom (where I use it) than in my much less reverberant lounge.

I think it would depend on what material I was listening to: I would prefer to hear a choir in a reverberant room, but a political discussion in a fairly 'dry' room.

Going back to one of your earlier posts, being able to identify a person, instrument or whatever in different acoustic environments is very different to there being no difference to the perception and enjoyment of the sound. If I play a recording of a guitar in my car or on my main hifi both are recognisable as a guitar but they don't sound the same.

But in that case, it's the source that is being changed (by the car stereo versus the living room hi-fi). The question would be whether the source sounded the same to you if you were using the same reproduction equipment in the two different environments. And this is further complicated by another variable: if your speaker doesn't have neutral directivity characteristics then (in my view of how this works) the room *will* change your perception of the source.

This is why I think that pure thought is more useful than 'try-it-and-see'. The (my) pure thought version says that a neutral transducer will be perceived as neutral regardless of the room; it *doesn't* say that the overall experience will not be changed by the room. But it also realises that a non-neutral transducer in terms of directivity will be perceived differently dependent on the room.

The distinction allows EQ to be targeted correctly: no EQ change for a truly neutral source; some EQ compensation for a non-neutral source.

 

[Ultrasonic]

I think it would depend on what material I was listening to: I would prefer to hear a choir in a reverberant room, but a political discussion in a fairly 'dry' room.

The point is that you would have preferences.

But in that case, it's the source that is being changed (by the car stereo versus the living room hi-fi). The question would be whether the source sounded the same to you if you were using the same reproduction equipment in the two different environments. And this is further complicated by another variable: if your speaker doesn't have neutral directivity characteristics then (in my view of how this works) the room *will* change your perception of the source.

This is why I think that pure thought is more useful than 'try-it-and-see'. The (my) pure thought version says that a neutral transducer will be perceived as neutral regardless of the room; it *doesn't* say that the overall experience will not be changed by the room. But it also realises that a non-neutral transducer in terms of directivity will be perceived differently dependent on the room.

The distinction allows EQ to be targeted correctly: no EQ change for a truly neutral source; some EQ compensation for a non-neutral source.

Your point earlier appeared to be that floor bounce echoes were obviously irrelevant since you could still identify a person in different locations. My car vs hifi example was to demonstrate that just because you could identify a sound source under different conditions didn't mean that the sounds were equivalent, or that enjoyment would be the same.

You've now clarified your position to say that the 'overall experience' won't necessarily be the same. This I agree with, but means that the question of whether the presence or absence of a floor bounce echo is desirable or not is a perfectly valid question to debate and conduct experiments to investigate.

 

[Cosmik]

Your point earlier appeared to be that floor bounce echoes were obviously irrelevant since you could still identify a person in different locations. My car vs hifi example was to demonstrate that just because you could identify a sound source under different conditions didn't mean that the sounds were equivalent, or that enjoyment would be the same.

You've now clarified your position to say that the 'overall experience' won't necessarily be the same. This I agree with, but means that the question of whether the presence or absence of a floor bounce echo is desirable or not is a perfectly valid question to debate and conduct experiments to investigate.

I think you misunderstand what I'm saying. Although the acoustics changes the experience, it does not follow that you are perceiving the source to be any different. To quote someone, a Steinway piano still sounds like a Steinway piano regardless of the concert hall. If there is floor bounce it is naturally part of the acoustics, and you hear it as 'spaciousness' - *not* frequency response colouration.

*This does not apply to floor bounce recorded in the recording* - which will be heard as colouration. Again, I refer to the experience most of us must have had, of making a far field recording in a room and finding that the recording sounds a lot more coloured than the live experience, whether you play it back over speakers or headphones. Did it never puzzle you?

This point is crucial to understanding whether, where and when, EQ is appropriate in audio.

 

[Cosmik]

You could say that what you record with your cassette recorder is what room correction software 'hears' when it makes its measurements.

It 'hears' that horrible, hollow, echoey sound that you record on your cassette recorder, but instead of recording over it with the chart show on Radio 1, as it should do, it uses it as the basis of changing the performer (the speaker) to invert the hollowness of the room that it measured.

But you, the listener never heard the hollowness in the actual room because your hearing is not a dumb, single-mic FFT; it is an acoustic scene interpreter that models the world from its acoustics. So now the modified speaker sounds like a shrill, peaky torture device at the end of the same room.

This is almost literally true:

Once GLM has served you a flat in-room frequency response on a silver plate, the result might be brighter than you like. While a flat frequency response is necessary as a reference, it may not be perceived as the best option with all types of content and environments, particularly when working long hours...

 

[Ultrasonic]

I think you misunderstand what I'm saying. Although the acoustics changes the experience, it does not follow that you are perceiving the source to be any different. To quote someone, a Steinway piano still sounds like a Steinway piano regardless of the concert hall. If there is floor bounce it is naturally part of the acoustics, and you hear it as 'spaciousness' - *not* frequency response colouration.

All of which still leaves open the discussion of whether floor treatment, or any other adjustment, may result in a preferable listening experience. It was your particular argument for this being an obvious irrelevance that I was disagreeing with.

I do by the way think the argument that for typically close-miked recordings it makes sense not to try to remove the floor bounce is logical. Ultimately I think listening tests are the way to confirm this, although some conflicting results have been posted in this thread.

I have no intention to start trying to attenuate floor-bounce signals in my own system either way though. I'm far more interested in the question of side-wall and ceiling reflection point treatment. I forget how the floor-bounce question even came up in the first place now but all of this is rather off the intended topic of this thread .

 

[Kal Rubinson]

But you say that the room affects your perception of the source. Maybe we need a distinction: is the room affecting your perception of the tonality of the source, or at some other 'meta' level e.g. church-like acoustics imbue the source with calmness and seriousness etc.?

Yes, the perceived tonality of the source changes from one acoustic environment to another.

To me, the former seems unlikely. It would suggest that you would not be able to look at a source and determine from its materials, dimensions and so on, how it was going to sound or vice versa.

Can you? Can you make such a determination upon being presented with an unknown speaker in an unknown room? I can make some presumptions but not a determination.

Instead, it would be dependent on the room's acoustics, and even walking around the room, the source would be changing its apparent dimensions and materials as you walked! I perceive no such changes as I walk around a room chatting to someone.

That is room dependent and of a different (smaller) magnitude than moving from one acoustic space to another.

I like to refer to a common experience of mine: Being hailed by a friend from across the main space of the Grand Central Terminal. First, I hear the address (my name) from afar but I can only recognize who it is by sight. However, as he and I approach each other, I can increasingly identify his characteristic voice but it is highly colored by the influence of that reverberant space. When we do come face-to-face and I hear him almost exclusively by direct sound (my ear brain now easily suppressing attention to the reverberation), he sounds as he ever does when in close proximity. That shift occurs in smaller and less imposing acoustical spaces but to a much lesser degree.

 

[Cosmik]

I forget how the floor-bounce question even came up in the first place now but all of this is rather off the intended topic of this thread .

I know of few things *more* relevant to this thread!

 

[Hugo9000]

^yep, and controversially a choir or a small acoustic group sounds very bad outdoors (without PA) Many old outdoor veues have a half dome behind the stage to reflect sound to audience.

Tallinna laululava

The Hollyrock Bowl was known for great sound!

 

[amirm]

@amirm , I wonder if your remark on singing in open air sounds strange is somewhat false. It depends on for example how close you are, and if the singer uses a microphone and speakers, doesn’t it?

Reverberations are colour, in my mind. Church builders, I guess even cave men, understood this - and used it to paint a picture that let the listeners get closer to God through music.

That is your mind's eye thinking, not how you hear.

Let me ask you this: if i proposed two listening situations to you, and you prefered one all the time, would you opt for the other for any reason?

 

[Cosmik]

I like to refer to a common experience of mine: Being hailed by a friend from across the main space of the Grand Central Terminal. First, I hear the address (my name) from afar but I can only recognize who it is by sight. However, as he and I approach each other, I can increasingly identify his characteristic voice but it is highly colored by the influence of that reverberant space. When we do come face-to-face and I hear him almost exclusively by direct sound (my ear brain now easily suppressing attention to the reverberation), he sounds as he ever does when in close proximity. That shift occurs in smaller and less imposing acoustical spaces but to a much lesser degree.

I am not yet convinced that your friend's *voice* "is highly colored by the influence of the reverberant space". I can see two possibilities:

1. the room colors your friend's voice meaning that the two are inextricably woven together;
2. to your hearing system, the room has its own sound (reverberation) and your friend's voice has its sound; when the reverberant sound is much stronger than your friend's direct voice, you hear only the reverberant sound, coloration and all. It *doesn't* mean that the room has colored your friend's voice; merely that the voice is very quiet in comparison.

I think that another of your comments ("my ear brain now easily suppressing attention to the reverberation") shows that it is option (2).

Which of the options it is is surely crucial to whether 'room correction' is a valid process. If you are able to suppress attention to the reverberation in a normal room, you should not change the source in response to measurements of the reverberation, otherwise you will hear a changed source.

Grand Central Station is merely 'a bad room' in which to listen to voices, or recordings of voices at a great distance. Of course your attention is going to be drawn to the almost totally dominant reverberant sound - it's simply a case of relative levels of two 'sources' (the voice and the room). In any reasonable listening room and normal distance between listener and speaker, you will hear the friend's voice (or recording thereof) at a satisfactory level, and the reverberation will not dominate. The reverberation will add a sense of 'spaciousness' but it won't affect your perception of the voice.

In Grand Central Station, the reverberation wasn't affecting your perception of the voice, it's just that the voice was very quiet; you didn't mistake the reverberation for your friend's voice - you *always* knew it was reverberation.