• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required as is 20 years of participation in forums (not all true). Come here to have fun, be ready to be teased and not take online life too seriously. We now measure and review equipment for free! Click here for details.

Technical Article: Listening Room "Modes" (Frequency Response Changes)

Joined
Jun 20, 2018
Messages
442
Likes
234
#21
I have tried listening music in a true anechoic room (we had one at university) and it sounded awful. Even a people voices sounded awful. I would definitely not want to have that kind of sound in my room.
A good fully anechoic room doesn't have a sound. At all.
A lot of things can be responsible for your experience. Maybe the room wasn't anechoic enough, many small anechoic rooms for research are only anechoic in the mids and treble, not the bass. This can make the sound very unbalanced which can be horrible (pressure on ears / clogged ears feeling as soon as you even walk into the room let alone play music in it).
Second thing is that in an anechoic room you will hear every fault in a speaker. If the speakers weren't very good you'll hear it very prominently.
And thirdly if the music isn't mixed well you'll hear it too, can sound quite awfull.
And lastly there's the prominent treble thing I mentioned. Long story to explain why this is so, but usually one uses an EQ curve to make the sound balanced (as mentioned I'll be using a different method myself this time which I won't elaborate on now).
Once all of the above is done right I can assure you there's nothing that comes even remotely close in extreme audio quality and wow factor.
 

Krunok

Major Contributor
Joined
Mar 25, 2018
Messages
4,600
Likes
2,807
Location
Zg, Cro
#22
A good fully anechoic room doesn't have a sound. At all.
A lot of things can be responsible for your experience. Maybe the room wasn't anechoic enough, many small anechoic rooms for research are only anechoic in the mids and treble, not the bass. This can make the sound very unbalanced which can be horrible (pressure on ears / clogged ears feeling as soon as you even walk into the room let alone play music in it).
Second thing is that in an anechoic room you will hear every fault in a speaker. If the speakers weren't very good you'll hear it very prominently.
And thirdly if the music isn't mixed well you'll hear it too, can sound quite awfull.
And lastly there's the prominent treble thing I mentioned. Long story to explain why this is so, but usually one uses an EQ curve to make the sound balanced (as mentioned I'll be using a different method myself this time which I won't elaborate on now).
Once all of the above is done right I can assure you there's nothing that comes even remotely close in extreme audio quality and wow factor.
I think you missed the part where I said it was a true anechoic room for audio experiments at electrotechincal university.

Everything sounded awful in that room. After few minutes you could hear your heart beating like a thunder and a strange noise which I was told was coming from the blood running through my veins.

Btw, where exactly did you find an idea that a perfect listening room is one without any reflections? Can you post a link to a research claiming that?
 

Cosmik

Major Contributor
Joined
Apr 24, 2016
Messages
3,075
Likes
2,000
Location
UK
#23
I think two things are at play here: a truly anechoic room AND extreme quiet. Maybe you need the system to maintain background noise at all times when music isn't playing...
 

Krunok

Major Contributor
Joined
Mar 25, 2018
Messages
4,600
Likes
2,807
Location
Zg, Cro
#24
I think two things are at play here: a truly anechoic room AND extreme quiet. Maybe you need the system to maintain background noise at all times when music isn't playing...
Sure. But never mind, my question still stands: where did he got the idea that a perfect listening room is the one without any reflections?

It simply isn't true, some rflections are desirable for a good sound and speakers need to have way to operate in a space with reflections, the way it is described in Spinorama measurements theory. It is nonsense to state that rooms need to be treated to kill all reflection - that would result in an incredibly "dull" sound.
 

Cosmik

Major Contributor
Joined
Apr 24, 2016
Messages
3,075
Likes
2,000
Location
UK
#25
Sure. But never mind, my question still stands: where did he got the idea that a perfect listening room is the one without any reflections?

It simply isn't true, some rflections are desirable for a good sound and speakers need to have way to operate in a space with reflections, the way it is described in Spinorama measurements theory. It is nonsense to state that rooms need to be treated to kill all reflection - that would result in an incredibly "dull" sound.
I think it can be justified as an idea. What you get is the pure recording - conveyed to you with HRTF and acoustic cross-coupling from the speakers (unlike headphones). I won't personally be going down that route myself, though.
 

Krunok

Major Contributor
Joined
Mar 25, 2018
Messages
4,600
Likes
2,807
Location
Zg, Cro
#26
I think it can be justified as an idea. What you get is the pure recording - conveyed to you with HRTF and acoustic cross-coupling from the speakers (unlike headphones). I won't personally be going down that route myself, though.
I don't think so. A room without reflections is simply not natural habitat for sound and everyything sounds "weird" in it. I tried to play guitar in it and it sounded also "weird".

Btw, I don't think when you go for a professional room treatment that the target is to kill all reflections.
 
Last edited:

Cosmik

Major Contributor
Joined
Apr 24, 2016
Messages
3,075
Likes
2,000
Location
UK
#27
I don't think so. A room without reflections is simply not natural habitat for sound and everyything sounds "weird" in it. I tried to play guitar in it and it sounded also "weird".
But that's my point about the two-way nature of listening in ordinary rooms. If you are prepared to forego being able to 'interact' with the performance, an anechoic room makes logical sense. I suppose being outdoors can be close to anechoic, and yes, it does begin to sound a bit weird when your voice dies as soon as you say anything. And live music played outdoors can sound very dry. But recordings include their own 'ambience'.
 

Krunok

Major Contributor
Joined
Mar 25, 2018
Messages
4,600
Likes
2,807
Location
Zg, Cro
#28
But that's my point about the two-way nature of listening in ordinary rooms. If you are prepared to forego being able to 'interact' with the performance, an anechoic room makes logical sense. I suppose being outdoors can be close to anechoic, and yes, it does begin to sound a bit weird when your voice dies as soon as you say anything. And live music played outdoors can sound very dry. But recordings include their own 'ambience'.
Cosmic, you and all of us, are living day by day in "ordinary" rooms. We are accustomed how our voices, music and other things we hear sound in them. For that reason everything sounds dull and weird in a room without reflections. I'm pretty sure the recording engineers are aware of that and are producing music to sound well in "ordinary" rooms. I don't remember reading anywhere that a goal of good room treatment is to kill all reflections - in fact, I remember reading that that exactly is NOT a goal. Anyway, I'm too lazy to try to find reference to that right now so let's wait for pros to pop in and shed some additional light on this.
 
Last edited:

LTig

Major Contributor
Forum Donor
Joined
Feb 27, 2019
Messages
1,918
Likes
2,767
Location
Europe
#29
Sure. But never mind, my question still stands: where did he got the idea that a perfect listening room is the one without any reflections?

It simply isn't true, some rflections are desirable for a good sound and speakers need to have way to operate in a space with reflections, the way it is described in Spinorama measurements theory. It is nonsense to state that rooms need to be treated to kill all reflection - that would result in an incredibly "dull" sound.
@Floyd Toole deals with this in his book. If I recall correctly only professional mixing engineers preferred a room with low reflections for work. For recreational listening they preferred rooms with reflections, as did all others, even professional mastering engineers during work.

I once was in a true anechoic room and it was a very strange experience. Although it was totally quiet I felt a kind of pressure in my ears, similar to the feeling when fast loosing altitude.
 

Krunok

Major Contributor
Joined
Mar 25, 2018
Messages
4,600
Likes
2,807
Location
Zg, Cro
#30
@Floyd Toole deals with this in his book. If I recall correctly only professional mixing engineers preferred a room with low reflections for work. For recreational listening they preferred rooms with reflections, as did all others, even professional mastering engineers during work.
That is exactly what I meant when I said I read it somewhere, but I couldn't find the article to quote it.
 
Last edited:

Hipper

Senior Member
Joined
Jun 16, 2019
Messages
300
Likes
227
Location
Herts., England
#31
As it happens, sound waves in a room also interact, affecting the amplitude of sound you hear at different frequencies. Sound waves from more than one source (for example, two speakers) will meet and interact. Sometimes they add, sometimes they subtract (cancel), but there will be interactions in any room. “Ah, but what about absorption?” you might ask. Ah, but remember I said “more than one source”. If I put two speakers in a perfect anechoic chamber, absorbing all sound when it interacts with any surface (walls, floor, ceiling), there are still those two sources to consider. If I sit between them, waves from one interact with waves from the other, despite having no other reflections to contend with. See Figure 1; sound waves from the two speakers spread out and interact at the listener’s location.


Figure 1. Sound wave interaction.
Let's talk about these two sources - speakers - in an anechoic chamber. I'm assuming that they are set up as we would in a good listening room, equidistant to the ears. I am also assuming that the signal from each speaker is identical.

If we are talking about frequencies that will cause room modes (those below the transition frequency, with relatively large wavelengths) how do these two signals interact in such a way that we will hear a difference?

Are you talking interaural crosstalk (XTC) right ear hears left speaker etc.) which might introduce interference but at very low frequencies with really large wavelengths, would that be significant? Would these large wavelengths be bent round the head, nose etc. like higher frequencies? On another thread about the processor BACCH it was said that below 94Hz there was thought to be no effect- post 117:

BACCH
 

Hipper

Senior Member
Joined
Jun 16, 2019
Messages
300
Likes
227
Location
Herts., England
#32
Toole says that in listening tests most people preferred side wall reflections.

Indeed in his third edition he discusses the matter at length in chapter 7, page 157 onwards. It's not a simple explanation as it involves early and late reflections, psychoacoustics, loudspeaker directivity, and the development of the phantom image. Indeed he suggests that side wall reflections can reduce the damage done to the phantom image by comb filtering from interaural crosstalk.

To me, with no scientific back up, it seems logical to stop side wall reflections. This allows all the information (above the transition frequency) in the recording to be heard via the direct sound from the speakers and not added to by early side wall reflections or directionally muddied by later ones, and therefore achieving its most accurate rendition (ignoring variables of recording techniques).

I tried without reflections for a few months, then introduced absorbent panels immediately to the outside of each speaker (i.e. not on the walls) and found I got a much sharper central image which I like and fits in with the above theory.
 
OP
DonH56

DonH56

Technical Expert
Technical Expert
Forum Donor
Joined
Mar 15, 2016
Messages
4,181
Likes
6,367
Location
Monument, CO
Thread Starter #33
Let's talk about these two sources - speakers - in an anechoic chamber. I'm assuming that they are set up as we would in a good listening room, equidistant to the ears. I am also assuming that the signal from each speaker is identical.

If we are talking about frequencies that will cause room modes (those below the transition frequency, with relatively large wavelengths) how do these two signals interact in such a way that we will hear a difference?

Are you talking interaural crosstalk (XTC) right ear hears left speaker etc.) which might introduce interference but at very low frequencies with really large wavelengths, would that be significant? Would these large wavelengths be bent round the head, nose etc. like higher frequencies? On another thread about the processor BACCH it was said that below 94Hz there was thought to be no effect- post 117:

BACCH
Let's start by me stating upfront that acoustics is not my area of expertise let alone details of how we perceive sound.

A perfect anechoic chamber would absorb all sound waves from the emitters. If we were not in the room, the sound waves from the two speakers would still interact related to the source signal (frequency and phase), relative positions of the speakers, and where you observe ("see", "hear", measure) the wavefronts. The ripples in a pond analogy works, or sticking two fingers in a pan of water, just imagine the pond is infinite or the pan has no edges and extends forever. Ripples from the two sources (perturbations) will interact to create peaks and valleys, often sort of a crosshatch or checkerboard pattern.

If the ripples (waves) were close together, like HF sounds, then there would be a lot of little perturbations (ripples, cross ripples, ripples on ripples). A point-source measurement (or observer) would only see the ripples as they occur right at the observation point. An observer, even a "small" one, might span several ripples. If they were very long, then locally it would be hard to see the effect of interactions, so LF waves are less observable if the observer is "small" with respect to the wavelength( distance between ripples).

So based on the hand-waving analogy, which is admittedly all wet :) , then yes higher-frequency effects are more observable (hearable). The big counter to that model is that standing waves create peaks and valleys (nulls) related to the position of speakers, room dimensions, and listener placement. Those steady-state standing waves are quite noticeable, creating places where the bass is "boomy" and other places *nulls) where it seems to disappear.
  • Room modes occur at specific points in the room for specific frequencies. For example, there will always be a null in the middle of a sealed room at a frequency related to the length of the room (see first post for the math). The depth of the null is related to the stiffness and reflectivity of the walls; very hard untreated walls (like concrete) lead to deeper nulls than flexible walls (e.g. thin wallboard on widely-spaced studs) that have absorption or diffusion on them. I have measured nulls 20~40 dB deep in a normal room. The usual solutions to nulls are either to move the MLP (main listening position) of the null or to add drivers (like more subs) to "drive" the null point so it is actively canceled at the MLP.
  • Reflections from walls and other surfaces (a coffee table between couch and TV is a common one), or speaker-boundary interference response (SBIR. see e.g. http://tripp.com.au/sbir.htm ), are generally at higher frequencies, and lead to comb filtering and similar effects. The first reflections from surfaces have the most energy, are closest in time to the direct sound, and are the ones most often discussed. Other reflections are more attenuated and take longer to arrive so do not impact the initial sound as much. Absorption or diffusion is often used to control the first reflections from walls, ceiling, and floor near the speakers. Ray diagrams are useful to show how the waves move from the speaker, to the wall (or whatever), and reach the MLP. That is how I set up the absorbers in my listening room to reduce interaction among early reflections. If you draw a line from the speaker to the wall, then the angle it makes coming in is the same angle the wave will take on the other side going out. A drawing program makes it easy; just use triangles and vary their angles to cross the listening position, then calculate the path length for the reflected vs. direct sound to see what frequencies are affected. It is straightforward but messy; I use Mathcad or Matlab programs to trace out a room.
Dr. Toole's book has much more, and much better presented, information about all of this. My short version is that room modes affect bass while reflections generally affect midrange on up. Too many reflections, especially first reflections that are large and close, can degrade the stereo image. Too few reflections make the room sound "dead" and all ambience is from the source. I personally tend toward a more "dead" room than most folk but it is all about preference. Most folk prefer a somewhat live room but with first reflections reduced to provide a better image. YMMV.

HTH - Don
 
Joined
Jun 20, 2018
Messages
442
Likes
234
#34
@Krunok There is a very logical reason to think an anechoic chamber has many ideals for sound reproduction. Rooms colour the sound to a very very great deal, moreso than speakers or any other link in the reproduction chain. If your ideal is to hear exactly what's on the recording, not less not more then an anechoic chamber is the starting point for this.
Perhaps you've indeed been in a good anechoic chamber, though almost all anechoic chambers are not anechoic in the bass, it is the size and treatment that determine at which frequency the chamber stops being fully anechoic. In most anechoic chambers loudspeaker placement still matters a great deal in the deep bass.
And as mentioned, there's a big difference between a quiet room (giving the effects you describe) and an anechoic room.
But assuming you've listened in a good anechoic room, I gave many other points to take into account to explain your negative experience.
I can tell from personal experience that if all those point are taken into account and done well that the soundquality and experience is second to none.

As for research. There is truly a ton of research which shows the detrimental effects of room reflections and modes. And hell one needs to only use a microphone at listening position in a room to show the severe effects. It is wrong to think that because we are used to rooms that we can "filter out" these effects or something along those lines. Again there's a ton of research which shows the opposite.
As for Toole. This is not where things stop. Many people have taken research much further. One example is again Northward Acoustics. He takes into account the "state of the ears" and makes a room where the ears still hear a room regarding self-noise, but not regarding sound coming from the speakers to put the ears into a relaxed state. Some people will need this more than others, myself I'm not bothered by an anechoic room, spent almost a year in one daily.
And yes, many many pro's use near-anechoic rooms / non environment rooms. Those who can afford one and have done their homework visiting many types of rooms often end up wanting exactly such a room.
As for lower budget studio standard, it is standard to at least treat all direct reflections and then try to tackle modes as best as possible. This makes a big difference already but a (near) anechoic room is yet another step up.
As I stated before, with a fully anechoic room though (and to some degree near anechoic room) there is some issue with treble. Most use an EQ curve to tame this. I have done original research into this and think I have a new and better solution. With this solution fully anechoic rooms become truly ideal in my opinion. Will share how it works later.
 

Cosmik

Major Contributor
Joined
Apr 24, 2016
Messages
3,075
Likes
2,000
Location
UK
#35
It is wrong to think that because we are used to rooms that we can "filter out" these effects or something along those lines. Again there's a ton of research which shows the opposite.
How does the research reveal this phenomenon?
 

Krunok

Major Contributor
Joined
Mar 25, 2018
Messages
4,600
Likes
2,807
Location
Zg, Cro
#36
As for research. There is truly a ton of research which shows the detrimental effects of room reflections and modes.
Sure, excessive reflections and room modes are hurting the sound, but I was asking for a research that proves via listening tests that ideal room should be without any reflections/anechoic.

Btw, I understand what you are claiming but I would like to see some research where listening tests were conducted and majority of listeners voted in favor of anechoic room.
 

Krunok

Major Contributor
Joined
Mar 25, 2018
Messages
4,600
Likes
2,807
Location
Zg, Cro
#37
As for Toole. This is not where things stop. Many people have taken research much further. One example is again Northward Acoustics. He takes into account the "state of the ears" and makes a room where the ears still hear a room regarding self-noise, but not regarding sound coming from the speakers to put the ears into a relaxed state.
Well, I certainly hope audio research doesn't stop with Toole but I do hope future researchers will be at least serious as he was. But it was not he who favored Spinorama, instead he conducted a serious of numerous tests with various subjects and only when vast majority of them voted in favor of that model he accepted it. The point is not about your guy from Northward Acoustic and you who think that anechoic room sounds good or me who think it doesn't, but about serious research with multiple subjects which is conducted according to scientific principles to confirm or reject that thesys.
 
OP
DonH56

DonH56

Technical Expert
Technical Expert
Forum Donor
Joined
Mar 15, 2016
Messages
4,181
Likes
6,367
Location
Monument, CO
Thread Starter #38
Here's a chart showing the effect of first reflections on the sound level at the MLP (main listening position). These are not room modes. Only one speaker is shown and is spaced either about 1' (Mag_1) or 2' (Mag_2) from the side wall. In both cases the MLP is 8' from the speaker. The sound travels direct from the speaker to the MLP, while the reflected wave travels to the side wall then bounces to the MLP (taking a longer path). The longer path means it arrives at a different phase than the direct signal. The phase depends upon the frequency so sometimes the interference is additive, other frequencies subtractive, resulting in the peaks and dips seen in the plot. Also notice the distance from the wall changes the frequencies at which peaks and valleys occur. For this plot the reflected wave is the same magnitude as the direct wave just to make it easier to see the effect, and the magnitude is normalized to 1. The phase shift is more significant as frequency goes up thus the peaks and valleys get closer together. The plots look somewhat like the tines on a comb, thus the term "comb filter". Hopefully this picture helps a bit. - Don

1566148218069.png
 

Cosmik

Major Contributor
Joined
Apr 24, 2016
Messages
3,075
Likes
2,000
Location
UK
#39
Here's a chart showing the effect of first reflections on the sound level at the MLP (main listening position). These are not room modes. Only one speaker is shown and is spaced either about 1' (Mag_1) or 2' (Mag_2) from the side wall. In both cases the MLP is 8' from the speaker. The sound travels direct from the speaker to the MLP, while the reflected wave travels to the side wall then bounces to the MLP (taking a longer path). The longer path means it arrives at a different phase than the direct signal. The phase depends upon the frequency so sometimes the interference is additive, other frequencies subtractive, resulting in the peaks and dips seen in the plot. Also notice the distance from the wall changes the frequencies at which peaks and valleys occur. For this plot the reflected wave is the same magnitude as the direct wave just to make it easier to see the effect, and the magnitude is normalized to 1. The phase shift is more significant as frequency goes up thus the peaks and valleys get closer together. The plots look somewhat like the tines on a comb, thus the term "comb filter". Hopefully this picture helps a bit. - Don
Not an issue - it has got to be like that. Constant tones will exhibit comb filtering and musical transients will be modified by reflections. But here's a thing: an algorithm could be designed to work back to the source even if it knew nothing about the room. Physically moving an array of two microphones around would help to, effectively, infer a model the room, and allow it to be removed from the signal with DSP. The room sound could be kept as a separate stream.

The question is, do humans do something similar, and if they do (or not) how would you demonstrate it, seeing as it is all within the listener's mind?

The typical experimenter would immediately start to reach for speakers and microphones and comb filtering DSP, etc., but this would be missing the point that the listener has to have access to the complex sound fields in order to do the separation of room and source. Merely adding comb filtering to some music coming out of a speaker is not a valid test.

And 'difference' alone is not evidence that the listener isn't doing this separate streaming: the listener may hear a difference if they choose to focus on it, but it would be in the stream of room sound *not* their perception of the source.

A clue that this idea is true: record a concert with a cassette recorder from the back of the room and play it back when you get home. It will sound nothing like the concert (much more hollow, echoey) because you no longer have the complex sound fields available that are letting you separate the source from the room. It's why most recordings are made close-mic'ed, and why an experiment that applies comb filtering to the sound of a speaker is not a simulation of a real reflection.
 
OP
DonH56

DonH56

Technical Expert
Technical Expert
Forum Donor
Joined
Mar 15, 2016
Messages
4,181
Likes
6,367
Location
Monument, CO
Thread Starter #40
Psychoacoustics and room correction algorithms are far outside the scope of this thread.

This was just to show the effect, which is quite real and measurable as well as audible, to give folk a visual reference. I am not sure what you mean by "typical experimenter" but this was in no way meant to delve into complex sound fields and such. It was just meant to provide a pictorial example of comb filtering. I have had to deal with this many times, partly because I lived with dipole speakers for many years, but also in applications such as hanging choir mics to provide the best frequency response for the congregation. You can clearly hear the dips and valleys in the response when the choir is singing and the impact of moving the mics or adding treatment to minimize the effect.

I am not sure where you want to go with this but it sounds like a great topic for another thread exploring the interaction with musical or more complex sources. Like most of these threads, they are meant to be introductory and help beginners understand various concepts. More advanced readers will find them boring and very limited in scope and variables (too simplified).
 
Top Bottom