Whats are the different theories on room acoustics to create a great sounding room?

[KlausR.]

When the system behaves itself the captured acoustic of the performance takes over, as far as the ear/brain is concerned, rendering the acoustic of the listening room largely irrelevant.

This is one of the questions to which I did not find an answer: what exactly happens when the acoustics of the performance venue, as present on the recording, and the acoustics of the playback room are superposed. Is there a difference between 2-channel and multi-channel? All I know is that with 2-channel I do not have the feeling of being there, regardless of the size of the performance venue.

I have read your post with interest and would like you to sum up for us... What's your take on Room Acoustics? We all know we can't replicate Live... We all also seem to think we get close when the room is well treated. So what is your take on room treatment and which one do you favor and why?

Having read most parts of Toole’s book, and also many of the literature he refers to, I draw the following conclusions, which in part are the same as Toole’s.

1. Reverberation time
When entering a room you immediately know whether or not the room is too reverberant, no need to measure. If too high, speech intelligibility goes down, a fatiguing experience. I could not find any basis for the often recommended optimum reveberation times for small listening rooms, and certainly no psychoacoustic research to that respect, so if the room sounds ok in terms of RT60, forget it, if not, use absorption, and the normal stuff of life will do, specialized absorbers are not necessary

2. Early reflections
I could not find evidence that early reflections are detrimental as a matter of principle, in each and every case. There is some research that looks very promising, but the 2-channel, multiple reflections case has not yet been investigated. There are cases where reflections might benefit from being treated, such as strong left-right asymmetry, bad off-axis response of the loudspeakers, very low reverberation time. In any case, each case should be looked at individually.

3. Room modes
The excitation of room modes depends on the position and orientation of the source with respect to the individual mode. You cannot say that a room has a certain number of calculated modes which inevitably will be excited, regardless of circumstances.

You further cannot measure a room impulse response or such and say, there’s a problem that needs being addressed. Measurements do not take into acount the mechanisms of human hearing. Measurements are useful once you have detected an audible problem, not the other way round. Don’t panic if a measurement looks bad, it’s easy to excite room modes with broadband signals, but you listen to music, don’t you? So wait until the modes show their ugly face when playing your favorite music, then come into action. In 10 years time I have found only 3 tracks where the 72 Hz width mode is excited, but then, your favorite artists are maybe heavily using synthesizers.

My take is that the influence of the room is generally heavily overrated, the room is the last link in the chain but not by definition the weakest. Yet, this seems to be the general consensus, and when people ask for advice they almost always get to hear: treat the bass, treat the modes, treat reflections. And this without having reported an audible problem and without the person replying having all necessary data. The room and the loudspeaker/room interaction may cause problems, yes, but not necessarily.

If you also prefer multi-channel that is fine ... I hasten to say , that I am on the fence for MC as I haven't ..yet found it entirely satisfying ...and its logistics are for me cumbersome

The furniture arrangement in our living doesn’t allow correct placement of multi-channel speakers, that’s why I did not look into this technology and related acoustics/psychoacoustics at all, haven’t even heard a system.

The best thing about the Toole book is that he presents his opinions and the opinions of others that do not agree entirely with him, enabling a critical mind from the reader. As you say he shows a preference for multichannel, but accepts that stereo can also do the "trick".

I read many of the papers Toole refers to and some that Toole does not mention. When looking at the original papers in some cases the conclusions I draw are somewhat different from Toole’s and in one instance I even think that Toole made an error. I hence conclude that it is always a good idea to look at the evidence yourself and draw your own conclusions, that’s why I posted a list of literature relevant to small room acoustics (http://www.whatsbestforum.com/showt...em-and-acoustics&p=30479&viewfull=1#post30479) offering to mail the pdf’s to those interested.

Considering your comment " I never had even remotely the impression of being there when listening to live recordings" it is difficult to define exactly what is the impression of being there, as many members will happily explain you, as it is a subjective notion.

Obviously recordings can only transmit the auditory aspects of live events. A symphony orchestra is say 10-15 m wide, it is 10-50 m away, the hall has 10.000+ cbm volume, all this is not possible to convey using 2-channel. Tom Danley has some demo tracks on his website (http://www.danleysoundlabs.com/technical downloads.html), in the one with the Harley there is a plane flying over. You hear it, but it’s not 200 m up in the sky. I’v never heard good multi-channel recordings on good multi-channel systems, I’ve never heard wave field synthesis recordings and systems, perhaps that those are capable of conveying the real thing.


Klaus

 

[fas42]

Measurements do not take into acount the mechanisms of human hearing. Measurements are useful once you have detected an audible problem, not the other way round. Don’t panic if a measurement looks bad, it’s easy to excite room modes with broadband signals, but you listen to music, don’t you? So wait until the modes show their ugly face when playing your favorite music, then come into action. In 10 years time I have found only 3 tracks where the 72 Hz width mode is excited, but then, your favorite artists are maybe heavily using synthesizers.

This to me is key to the whole debate: the ear/brain can certainly perform beyond simplistic processing of the sound waves that normal measuring equipment responds to, and provides numbers for. But overload this human mechanism with too much irrelevant information in the form of added distortion generated internally within the playback componentry, prior to reaching the speaker drivers, and then the balance is tipped in favour of the room acoustics dominanting. For me the whole room acoustics discussion is always lacking, because I see it always failing to mention the aspect of the quality of the sound reproduction.

Rephrased simply, the listening room acoustics are always perfect, there is zero distortion of their contribution because it is a "live" component of the sound that the listener hears. The reproduced acoustic, on the other hand requires the playback system to work at a high level to faithfully recreate the the signature sounds of that recorded acoustic, in all its fine detail. If the playback system is relatively weak in conveying the recorded acoustic then the listening room acoustic will dominate; conversely high performance audio will reproduce the musical event's acoustics convincingly, these will be in sympathy with, correlate perfectly with the direct sound of the instruments, and so then the ear/brain will register that acoustic as being the one most meaningful, and the influence of the listening room will be largely filtered out of the equation by the brain.

Frank

 

[Orb]

Klaus,
whats your take on ambisonics compared to the multichannel that is available these days (SACD/etc)?
I just wonder if the ideal setup was ambisonics and critically the process on how ambisonics is recorded, was this before its time and became a missed opportunity....
Ironically I understand that quite a bit of the algorithms have found their way into other decoding techniques-products that audio listeners purchase these days, so sort of lives on but IMO not the same without the recording process.

Thanks
Orb

 

[Kal Rubinson]

When entering a room you immediately know whether or not the room is too reverberant, no need to measure. If too high, speech intelligibility goes down, a fatiguing experience. I could not find any basis for the often recommended optimum reveberation times for small listening rooms, and certainly no psychoacoustic research to that respect, so if the room sounds ok in terms of RT60, forget it, if not, use absorption, and the normal stuff of life will do, specialized absorbers are not necessary.

My take is that the influence of the room is generally heavily overrated, the room is the last link in the chain but not by definition the weakest. Yet, this seems to be the general consensus, and when people ask for advice they almost always get to hear: treat the bass, treat the modes, treat reflections. And this without having reported an audible problem and without the person replying having all necessary data. The room and the loudspeaker/room interaction may cause problems, yes, but not necessarily.

I have singled out these two statements because I think they should be apposed. I agree with the first statement but not with the implications of the second because, all to often, the acoustics of a domestic multipurpose room are all to obviously detrimental upon entering the room and listening to footfalls and/or conversation. Clearly, one cannot say that the owner's "normal stuff of life will do" in these instances and specialized treatments are necessary.

How typical are such rooms? I do not have any data but, from anecdotal personal experience, it is very common.

Kal

 

[FrantzM]

This to me is key to the whole debate: the ear/brain can certainly perform beyond simplistic processing of the sound waves that normal measuring equipment responds to, and provides numbers for. But overload this human mechanism with too much irrelevant information in the form of added distortion generated internally within the playback componentry, prior to reaching the speaker drivers, and then the balance is tipped in favour of the room acoustics dominanting. For me the whole room acoustics discussion is always lacking, because I see it always failing to mention the aspect of the quality of the sound reproduction.

Rephrased simply, the listening room acoustics are always perfect, there is zero distortion of their contribution because it is a "live" component of the sound that the listener hears. The reproduced acoustic, on the other hand requires the playback system to work at a high level to faithfully recreate the the signature sounds of that recorded acoustic, in all its fine detail. If the playback system is relatively weak in conveying the recorded acoustic then the listening room acoustic will dominate; conversely high performance audio will reproduce the musical event's acoustics convincingly, these will be in sympathy with, correlate perfectly with the direct sound of the instruments, and so then the ear/brain will register that acoustic as being the one most meaningful, and the influence of the listening room will be largely filtered out of the equation by the brain.

Frank

Hi

From the get go I disagree with you .. Most rooms distorts what comes from the speakers .. The goal of room/acoustics treatment is to give a chance to the reproduction from the speaker to reach your ears ... Truly such extreme point of view seem to have two goals :
To confuse
To induce discussions for the sake of discussing ...

I apologize for being so blunt but that is what I get from this post ... How can one state that the acoustics of a room is perfect.. IT is not even an opinion but a fact that the best transducer in a room that distorts its output completely is of no value.. The goal of Acoustics/.Room Treatments is not to eliminate the room entirely rather it is to reduce/mitigate some of its most deleterious effects or characteristics...
If we are about absolute then nothing will ever replicate the live ideal ... It just can't be .. but we can get pretty darn close , even at times with two channels and to me that is the crux of the mater we are looking for the best approach (given the constraints of space and budget and maybe aesthetics) that would allow one to treat a room in such a way that the speakers waveform is not too distorted by the time it reaches your ears. What's you take on that ? and Please! Let's assume the gear all of them are at least as good as your modified HTIB .... which if I understand is your current reference

 

[JackD201]

Sorry Frank but I have to go with Frantz on this one. The brain is great at many things but filtering everything except the direct sound is definitely not one of them. If a room can make it difficult to understand what someone in there with you is saying it sure as heck will obscure even the most distortion free system devoid of any second, third or billionth order effects.

 

[Nyal Mellor]

1. Reverberation time
When entering a room you immediately know whether or not the room is too reverberant, no need to measure. If too high, speech intelligibility goes down, a fatiguing experience. I could not find any basis for the often recommended optimum reveberation times for small listening rooms, and certainly no psychoacoustic research to that respect, so if the room sounds ok in terms of RT60, forget it, if not, use absorption, and the normal stuff of life will do, specialized absorbers are not necessary

Hi Klaus. I agree that it is quite obvious when you walk into a room to observe whether it is too live or too dead. I think though that for a high performance environment it is necessary to have absorbers that approach a constant absorption throughout the spectrum (at least down to 300Hz), such that any sound waves or packets hitting those absorbers do not end up being selectively absorbed. From a practical perspective I worked in a room with a lot of thin drapes that were used to reduce single figure RT60 to a reasonable level (the room didn't sound too live). However use of these thin drapes was causing different decay times at high frequencies compared to midrange frequencies. Ripping out the drapes and using the right surface area of 3" absorbers rebalanced the decay times across the spectrum and audibly improved sound quality.

 

[Nyal Mellor]

2. Early reflections
I could not find evidence that early reflections are detrimental as a matter of principle, in each and every case. There is some research that looks very promising, but the 2-channel, multiple reflections case has not yet been investigated. There are cases where reflections might benefit from being treated, such as strong left-right asymmetry, bad off-axis response of the loudspeakers, very low reverberation time. In any case, each case should be looked at individually.
Klaus

All depends on where the reflections are coming from. In all cases I have had practical experience with, eliminating the ceiling reflection notably improves clarity and reduces listener fatigue. The back wall reflection is also one that does nothing beneficial. I could agree with you on the sidewall reflections though. Once you've done enough rooms you can draw a practical correlation between the location of a reflection and its effect on sound quality. It is also something you can experiment easily with yourself in your own room.

 

[fas42]

Most rooms distorts what comes from the speakers .. ... How can one state that the acoustics of a room is perfect.. IT is not even an opinion but a fact that the best transducer in a room that distorts its output completely is of no value.. The goal of Acoustics/.Room Treatments is not to eliminate the room entirely rather it is to reduce/mitigate some of its most deleterious effects or characteristics...
If we are about absolute then nothing will ever replicate the live ideal ... It just can't be .. but we can get pretty darn close , even at times with two channels and to me that is the crux of the mater we are looking for the best approach (given the constraints of space and budget and maybe aesthetics) that would allow one to treat a room in such a way that the speakers waveform is not too distorted by the time it reaches your ears. What's you take on that ? and Please! Let's assume the gear all of them are at least as good as your modified HTIB .... which if I understand is your current reference

I had a feeling that some people might misunderstand what I was trying to say so I will restate: what I am saying is that the acoustics of the listening room are perfect, irrespective of what they actually are, echoing ferociously or dead as a door nail, because they are the real thing, the acoustics of that room are a live experience. If you recorded what the sound was in your listening room with a high quality microphone setup, and then replayed that on some system of a certain quality in a room with completely different acoustics from your listening room, would you be able to easily hear the "signature" of your listening room in that other room? Does that make sense?

I agree that the room should not distort the sound emerging from the speaker drivers, but the sound emerging from the speaker drivers should not be distorted in the first place!! That is why I have this strange obsession with sticking my ear right next to the driver while it's working: I'm treating the speaker as a headphone transducer because if it doesn't perform at least as well as that then it has no chance of producing good sound in the rest of the room! Here's an interesting experiment that I've tried and I suggest others do: get the best headphones you can get hold of working at the highest volume still with optimum sound, take them off and put them down with the headband stretched out so the transducers become a set of very closely spaced stereo speakers at the appropriate height. Walk back from them, and you should be amazed at how big a soundstage they produce, even at this relatively low volume.

I will add, as a variation of what I have mentioned in other posts, that a system working well will sound the same irrespective of how close you are to the speaker. When you are very close the direct sound of the drivers obviously dominates, the room is out of the equation. As you move further and further away from the speakers room acoustics will, if they are going to have an effect, "distort" the sound because of room reflections. The mark of the system working well is that that doesn't happen, the sound character of the playback doesn't change.

Frank

 

[terryj]

edited

 

[FrantzM]

Frank

In using a headphones you remove the room from the equation, an understatment ... My point is that the discussion assumes the equipment are good enough and frankly nowadays they really are very good .. If I am to take account your own experience even an HTIB is (to you) good enough to reproduce .. a lot of things ... When such systems are considered the room becomes dominant and in most rooms that are untreated the results are usually not good ...

I will add, as a variation of what I have mentioned in other posts, that a system working well will sound the same irrespective of how close you are to the speaker. When you are very close the direct sound of the drivers obviously dominates, the room is out of the equation. As you move further and further away from the speakers room acoustics will, if they are going to have an effect, "distort" the sound because of room reflections. The mark of the system working well is that that doesn't happen, the sound character of the playback doesn't change

I am not sure what you mean by "sound the same" ... I can tell you that in a large multi way speaker you need a certain distance for all the drivers to blend.. simple law of physics if you are close to one of the driver its sound thus frequency band dominates and what you hear is likely not "the same' as what you would have heard at the distance where all the driver blends which often is sufficient for the room to dominate as well .... The speaker will still be operating well if you are too close toit you will not hear it well though .. So your point is not made.
Now if you advocate near field listening .. fine some speakers works well with that some don't , many don't .. A quad ESL 63 works wonderfully well in nearfield.. Large speaker such as an Arakis, Altair VR9 or toher multi drivers would not work well in headphones territory proximity .

 

[fas42]

In using a headphones you remove the room from the equation, an understatment ... My point is that the discussion assumes the equipment are good enough and frankly nowadays they really are very good .. If I am to take account your own experience even an HTIB is (to you) good enough to reproduce .. a lot of things ... When such systems are considered the room becomes dominant and in most rooms that are untreated the results are usually not good ...

Yes, a headphones removes the room, but the room being removed is not necessary for the experience. Some years ago, as I mentioned in another thread, an audio acquaintance lent me a pair of Sennheiser HD650s, which he was enamoured with, for some time. My setup at that time was by the standards of WBF of a decent order, modified top of the line Yamaha battleship CD player, and Perreaux 200 watt amp, better in some ways to my current HT arrangement, worse in others. Yet listening to the HD650s directly plugged into the player didn't improve the acoustic experience at all, it just felt a bit claustophobic if anything. I quickly tired of the sensation and went back to room sound. Ah, much nicer, much "bigger".

I am not sure what you mean by "sound the same" ... I can tell you that in a large multi way speaker you need a certain distance for all the drivers to blend.. simple law of physics if you are close to one of the driver its sound thus frequency band dominates and what you hear is likely not "the same' as what you would have heard at the distance where all the driver blends which often is sufficient for the room to dominate as well

I know what you mean about speakers being driven normally. A couple of years ago I heard many expensive setups using Wilson speakers, and the drivers on these were yelling at the top of their voices, look at me, look at me! This is the problem with these systems, you have to get way away from the raw drivers for the sound to start to become reasonable. The point I keep making is that the system has to be operating overall at a certain level for the ear/brain mechanism to do its magic of integrating the room sound, no matter where you are in the room. Acoustic science is not going to give you all the answers, because it doesn't take enough notice of what's going on inside your brain.

Frank

 

[FrantzM]

fas42

So you favor close-field listening ? Fine .. It requires speaker that works well in the near field and most speakers, especially multi-way, multi-driver speakers don't but please don't state that the room has "perfect" acoustics .. That was my point of contention. I am not a fan of Wilson but properly set-up in a well treated room many Wilson speakers will project an utterly realistic depiction of an original music event ...

 

[fas42]

So you favor close-field listening ? Fine .. It requires speaker that works well in the near field and most speakers, especially multi-way, multi-driver speakers don't but please don't state that the room has "perfect" acoustics .. That was my point of contention. I am not a fan of Wilson but properly set-up in a well treated room many Wilson speakers will project an utterly realistic depiction of an original music event ...

I'm sorry Frantz, if you think I'm saying the room has "perfect" acoustics: this is something I am most certainly not saying! I am also not fussed about where I am listening from: to the side of the speakers, in the next room, down the hall; all of these should still give me the good experience. If you ran any of the tests that the trained people use, on my listening environment, the results would be dire!! What I am saying is that the poor room acoustics don't matter IF the system is working particularly well! If a group of real jazz musicians started playing in a big bathroom would you say that it doesn't sound real, that the bathroom has to be rid of all the echoing for it to be tolerable or musical to listen to?

Yes, the Wilson speakers will sound excellent if all the right things are done, if all the right procedures are followed, if you sit in exactly the right position. But I am aiming to get a system working beyond that ...

Frank

 

[FrantzM]

fas42

We are losing ourselves here, what again do you mean by if the system is working well? Again coming back to one of my ealrier post assuming perfect speaker, perfect set up in bad room equates to bad reproduction very far to the original .. Room acoustics matter and a great deal .. I am sure you know it but are pulling some legs ... Let's drop it I know you know what the truth is

 

[fas42]

We are losing ourselves here, what again do you mean by if the system is working well? Again coming back to one of my ealrier post assuming perfect speaker, perfect set up in bad room equates to bad reproduction very far to the original .. Room acoustics matter and a great deal .. I am sure you know it but are pulling some legs ... Let's drop it I know you know what the truth is

OK, we'll drop it from this point on. My last comment is that Roger's, Vince's and my system can all work "well", we at least know what the "effect" is. The key thing is that what matters is not only the "perfect" speaker and "perfect" setup, but "perfectly" sorted out electronics -- see the new thread just starting: http://www.whatsbestforum.com/showthread.php?3232-The-Combined-Effect-of-Tweaks-Can-Be-Impressive!

Frank

 

[KlausR.]

Whats your take on ambisonics compared to the multichannel that is available these days (SACD/etc)?

Orb,

I have never looked at anything else than 2-channel, so I don’t know anything at all about ambisonics, sorry.

Klaus

 

[KlausR.]

When entering a room you immediately know whether or not the room is too reverberant, no need to measure. If too high, speech intelligibility goes down, a fatiguing experience. I could not find any basis for the often recommended optimum reverberation times for small listening rooms, and certainly no psychoacoustic research to that respect, so if the room sounds ok in terms of RT60, forget it, if not, use absorption, and the normal stuff of life will do, specialized absorbers are not necessary.

My take is that the influence of the room is generally heavily overrated, the room is the last link in the chain but not by definition the weakest. Yet, this seems to be the general consensus, and when people ask for advice they almost always get to hear: treat the bass, treat the modes, treat reflections. And this without having reported an audible problem and without the person replying having all necessary data. The room and the loudspeaker/room interaction may cause problems, yes, but not necessarily.

I have singled out these two statements because I think they should be apposed. I agree with the first statement but not with the implications of the second because, all too often, the acoustics of a domestic multipurpose room are all too obviously detrimental upon entering the room and listening to footfalls and/or conversation. Clearly, one cannot say that the owner's "normal stuff of life will do" in these instances and specialized treatments are necessary.

How typical are such rooms? I do not have any data but, from anecdotal personal experience, it is very common.

Kal,

What you seem to be describing are rooms with too much reverberation. Of course, in these cases the stuff present in that room doesn’t do the job. What I intended to say was that non-specialized absorbers such as upholstered furniture, drapes, carpet, rugs, cushions, plush toys, are sufficient. Why? Because specialized treatments are targeting specific frequencies/frequency ranges, and since there is no psychoacoustic research (that I know of) relating to how the frequency dependence of RT60 should look like, there is no need to use surch “targeted drugs” for tailoring the curve RT60 vs frequency. Obviously, in cases where the normal stuff of life is not an option, specialized treatments might be the solution.

Klaus

 

[KlausR.]

Nyal,

I think though that for a high performance environment it is necessary to have absorbers that approach a constant absorption throughout the spectrum (at least down to 300Hz), such that any sound waves or packets hitting those absorbers do not end up being selectively absorbed. From a practical perspective I worked in a room with a lot of thin drapes that were used to reduce single figure RT60 to a reasonable level (the room didn't sound too live). However use of these thin drapes was causing different decay times at high frequencies compared to midrange frequencies. Ripping out the drapes and using the right surface area of 3" absorbers rebalanced the decay times across the spectrum and audibly improved sound quality.

You can measure different decay times for different frequencies, and using 3" absorbers still leaves the decay times unbalanced. The question is, is an unbalanced RT60 really disturbing given the fact that the sound field in the case of directional loudspeakers in a small room is highly directional?

All depends on where the reflections are coming from. In all cases I have had practical experience with, eliminating the ceiling reflection notably improves clarity and reduces listener fatigue. The back wall reflection is also one that does nothing beneficial. I could agree with you on the sidewall reflections though. Once you've done enough rooms you can draw a practical correlation between the location of a reflection and its effect on sound quality. It is also something you can experiment easily with yourself in your own room.

According to research by Bech with speech and noise it is the floor reflection that has the greatest impact. According to Guski the ceiling reflection may cause confusion when being the only reflection remaining. Was the ceiling reflection the only or merely one of several reflections to be treated? You say that the back wall reflection does not do anything beneficial, but is it disturbing? With music Wrightson found that a rear wall reflection did not change the image width significantly.

Since off-axis response is responsible for how the reflection will look like spectrally it would be interesting to see how the off-axis responses of the systems you’ve treated looked like. The cases you are mentioning may have presented reasons that made reflection treatment have beneficial effects. Were the systems 2-channel or multi-channel? I’ve prepared a 23 page literature review (88 references) on this issue of early reflections (2-channel only), if interested, send me a note.


Klaus

 

[Kal Rubinson]

Kal,

What you seem to be describing are rooms with too much reverberation. Of course, in these cases the stuff present in that room doesn’t do the job. What I intended to say was that non-specialized absorbers such as upholstered furniture, drapes, carpet, rugs, cushions, plush toys, are sufficient. Why? Because specialized treatments are targeting specific frequencies/frequency ranges, and since there is no psychoacoustic research (that I know of) relating to how the frequency dependence of RT60 should look like, there is no need to use surch “targeted drugs” for tailoring the curve RT60 vs frequency. Obviously, in cases where the normal stuff of life is not an option, specialized treatments might be the solution.

Klaus

Yes, I am describing rooms with too much reverberation and reflections because that is what, anecdotally, I have found to be common.