Acoustic treatment for my living room - Page 8

amirm, you don't know when to stop, do you.

Let's say you want to build that rotating 3 sided (or 2 sided for simpler construction) jig for DBT of treated vs untreated room. How would you come up with the gap behind it so that it can turn? Build another layer of wall in front of existing wall? How about this, saw cut the existing wall and install that pivot in order to preserve the look of your current listening room? Do you know how much it costs to do such construction? Many thousands at least.

As they say, don't dig a hole when you are already in one.

Please be cordial if you like me to answer you. It should not much to ask if your interest is for me to respond to you.


This is not about building a "room" but testing a single reflection point. You can do this to all the walls you want but then it is a larger project. As it is, folks are not even evaluating the effectiveness of a single panel so getting there is a huge help.


It is a room within a room. Or in this case, the room wall you see is a false wall shared with another room behind it. If you are curious how it is done, download the AES paper: "A New Reference Listening Room For Consumer, Professional and Automotive Audio Research" by Sean E. Olive


Why would you build a lab in your listening room? The suggestion is for people in the industry who are objecting to testing done by Harman in this manner. I am asking them to produce their own unbiased data on the efficacy of what they advocate as a person's perception is not the same as a single microphone and a graph. If you are an individual and still want to do this, you can but that was not the point of the argument.


That's right. That would be quite affordable relative to the cost of doing business. I have a $15,000 meter I use just to calibrate video. I have an audio precision analyzer that cost $25,000. Sometimes it costs money to do something and if you want data, you need to spend it. Science doesn't owe a cheap answer .

While I have not cost how much it is to go to an anechoic chamber, that is also another option where you can simulate the same with two or more speaker and you don't have to convince your wife to let you do it .


Again, I am going to ask for you to please be professional. I have seen these fixtures, I have read the papers, and I know what is involved in making them. You don't have the benefit of any of this so I would be cautious about gloating too much .

Hi Terry. I was hoping that Localhost would answer since he post the picture without attribution. I don't know which PowerPoint that graphic came from so can't provide insight as to what they meant by it. But I do have the reference paper(s) used to generate it.

While the paper is by Dr. Toole, the data for that specific graph was generated by one of his researchers, Sean Olive who presented it a year earlier (both worked at National Research Council of Canada (NRC) at the time and not Harman). The paper included extensive research from many sources to address common points in question such as the effect of reflections. There is an updated version of the same paper in 2003 which is basically a much denser version of what you read in Dr. Toole's book on room/loudspeaker acoustics.

Anyway, Sean's research was a listening test conducted using the anechoic chamber method I mentioned earlier. A speaker provided direct sound and then two others provided reflected sound at different angles. Electronic means was used to delay/change level/transform the sound going to the other speakers, simulating on demand what the reflections are doing as perceived by the listeners. A switcher allowed instant change to what was being simulated. They also did tests in a real room but that is not related to the graph below.

I will do my best to summarize the information here but this is a very detailed paper and look at the problem that goes on for a good bit.

With that introduction, let's look at the graph again:


As the graph indicates, this is a test of speech as the source material. It is thought that if we have trouble understanding speech, then likely music and movie sound suffer just the same. For this test, only a single reflection is looked at by simulating in the anechoic chamber per above.

The graph has a level on the vertical axis and a time delay on the horizontal. These are the two dials they could manipulate and then determine audibility and character of what was heard. The delay simulates distance from the reflection and the level is how strong the reflection is relative to the direct sound.

The curves have these meanings from bottom up:

1. Absolute Threshold. This is the level/delay at which the listeners detected any change “in the nature of the sound itself or its sound field.” In other words, anything below this curve is not of perceptual interest as listeners could not tell the reflections were there at all. Reading from the graph above, where the blue vertical arrow is at 20ms of delay, looks like the level needs to be stronger than -15 dB relative to the direct sound for anything to be detected.

2. Image-shift threshold. This is the level of level/delay at which “there was a just-discernible shift in the location or size of the principal auditory image.” In other words, this is the level at which listeners could tell that the image had shifted (usually toward the reflection point) or widened. Using the same blue line as an example, that point at 20 msec of delay is around -7 dB.

3. Separate Source. This is the level/delay at which the listeners perceived listeners report hearing “a second sound source or image, simultaneously coexisting with the original one.” This threshold came from Lochner and Burger’s work so the testers and test conditions may not have been exactly the same. Nevertheless we see the interesting effect that the reflections actually have to be louder than the main source for us to perceive this at many delay levels! Needless to say, that is not going to happen with natural reflections in smaller rooms.

As you look at the graphs, keep in mind that sound travels around 1 foot/msec. In a typical home listening room therefore, the numbers past 15 msec or so are probably not applicable (remember, these are first reflection points).

Now you have the literal explanation. The paper by Dr. Toole (especially the 2003 version) analyzes this and many others to arrive at what it all means. And it is what I/Dr. Toole have been saying. That there is nothing “wrong” with having reflections. Starting at the top, the term “image-shifting” is not the scary thing that it seems to sound. It simply says that the sound didn’t originate from that single point where the speaker was. Perceptually as I noted earlier from Harman research, this is considered a positive feature, not negative. Think of dialog in a center channel of a movie. Do you really want that to only come from that single point no matter who is talking on the screen? And keep in mind that the graph is at the threshold of when this shift occurs, detectable in quick AB tests. You may not realize that this is occurring when you are just listening and not comparing.

There is a ton more analysis past this point in the paper including what happens with multiple reflections and tests in real rooms as opposed to anechoic chamber. This is how Dr. Toole nets it out:

” 5 EFFECTS OF REFLECTIONS—A SUMMARY
5.1 Speech
Readers who have been keeping score will have noted a distinct absence of negative effects from reflections on any aspect of speech perception we have looked at. In fact, the effects range from neutral to positive. No single reflection has been shown to be a problem for speech reproduction in small rooms (see Table 1). Multiple early reflections contribute even more to intelligibility.” Here is the table he shows:



Bottom line, “these are not the droids you are looking for.” Yes, scary terms are used but that is all they are. In that regard, I am unclear why Localhost is putting forward this graph to bolster his case. It would be odd for Dr. Toole to contradict himself and the paper reads just like his book.

amirm, you respond when trying to shill av products or cover up your mistakes no matter how coarse the replies are (the term "rookie mistake" got you all riled up but you responded for many pages). You ignore questions when exposing of your shilling or mistakes no matter how cordially asked.


Testing a single reflection point and not DBT of treated vs not treated room? "redirect", your dance move #2.


Really? I thought:

As for your audibility claims about HDMI vs S/PDIF connections:

Sometimes meaning that when it's your turn, you are exempted?


So what? You haven't mentioned a thing about actual example of room/s capable of double blind testing acoustically treated vs untreated which requires switching out acoustic treatments in seconds at the push of a button. You've only cited some hardware (pivots).

You have the benefit of what? Not much relating to DBT of treated vs untreated room.

the diagram was posted to further attempt to communicate that not all reflections are the same, as you continually and erroneously lump them all together. even after the chart was presented, you still refer to them as "having reflections". the point is, their arrival time and gain clearly dictate the psycho-acoustic response.

and im sorry, but harman is most certainly NOT in a position to dictate to anyone what their personal design requirements are - as if one were to prefer an accurate, critical listening space that features precise imaging and localization, they would certainly NOT want any high-gain early arriving "image shifting" indirect reflections - regardless of whether harman considers them a positive feature or negative.

if one wants a wider and fuzzier (less precise) image, then maybe that is their preference. but it is NOT for harman to decide.


hmm, so before you seemed to imply that toole never references the ETC - therefore, you deemed it must not be important. but for this scenario where we are discussing very different types of psycho-acoustic responses based on Gain and Time arrival of the indirect reflection - then it seems ETC is the right tool for the job. hmm, pretty much what i've been saying all along regarding "reflections" and measuring via the ETC in the time-domain --- NOT the frequency-response in the frequency-domain. but quite frankly, i don't think you've ever even utilized the ETC NOR have any actual acoustical measurement experience with respect to the context of this conversation.



psycho-acoustics 101.

it simply means that the low-gain reflections are below human detection threshold for ear-brain; but it does NOT mean that they are not heard - just not processed (effectively anechoic != anechoic). the low-gain reflections do not trigger localization, imaging, or any other keyed response. the ISD-gap in LEDE/RFZ, for example, is to be an effectively anechoic time period - so that all processing in the ISD Haas interval is keyed to the early arriving direct signal. and there-after, you have the ambient spaciousness, the re-inforcement of the localization based on the early arriving direct signal, and the liveliness of the room - if such is a design goal.

and what is even more comical, as you have continued to push my apparent "Cookbook to Kill All Reflections" and mock my "buzz-words" - the fact of the matter is, one should be measuring via the ETC to determine the ACTUAL room's acoustic response (gain and time arrival of the reflections) before making decisions on what to do based on THEIR choice of specular response. it is continually stated that if the user chooses an accurate listening space and wishes to attenuate high-gain early arriving destructive indirect specular reflections, then one should surgically place broadband absorption ONLY at areas incident of high-gain reflections. so, as the chart lists above, low-gain reflections that are below the human detection threshold are not a worry - thus, broadband treatment is NOT required at those reflection points - LIMITING the amount of broadband absorption placed within the room to ONLY those points and boundaries incident of the HIGH-GAIN reflections.



it is for any specular energy, not simply first reflection points - as in small acoustical spaces we can and typically do have later-arriving high-gain reflections that still maintain their vector, magnitude, and time arrival components. hell, even flutter echo across the listening space is easily identifiable within the ETC...

and if the listening position is ~ 7ft from their rear wall, then that first reflection point would be approx 15ms after the direct signal.

you should also be aware the first significant reflections determine the perceived acoustical size of the space. eg, if the room's physical dimensions induce a 7ms high-gain reflection, then the brain will key on that for perceived acoustical size of the room. while one generally cannot modify the physical dimensions to create a larger room - attenuating the first order high-gain reflections which delay in time when the first significant reflection impedes the listening position will INCREASE the perceived acoustical size of the room.

a quick demonstration of this is blindfolding yourself and speaking in your bathroom. you will be able to determine and perceive how large of a space you are in based on the time arrival and gain of the reflections. now, go into a large room like a movie theater while still blindfolded, and your brain will be perceived into being in a very large space.


the gain and arrival time of indirect specular reflections very-much-so determine the psycho-acoustic response, as toole indicates. but maybe we should just return to the 2D Flatland that is the frequency-domain and start attempting to 'view' reflections in the frequency response instead...

OK, whatever


There was no talk of "room." Here is what Arny said: "Doing a DBT related to room treatments is very difficult because of the physical scale of the changes that are involved."

He says related to room *treatment*. No one is selling you a whole room as to require DBT of that. They are selling individual acoustic products which lack any perceptual studies. They tend to scare you with graphs and ask you for your money. Sounds you like they have taken some of your seeing how you are concerned about me asking for proof. Is that the case?


You thought what? That Ethan is not in the business of selling acoustic products? You don't know what he does? His site shows acoustic products that his company sells and they are devoid of double blind tests that show efficacy.


Sounds like you are exempting companies selling acoustic products from showing listening tests. I don't exempt them on that basis.


I showed how they can switch out speakers in a few seconds. If you don't have the imagination to realize you can present different material that way, then that is your issue, not mine.


See above


See above .

Where did it come from Local? Can you please provide a reference? I assume you didn't create it. Please post a link.


There is no "me" in that response. You quoted research from Dr. Toole. I provided word for word explanation of what the axis meant. You were here but didn't provide that so I assume you didn't know what the graph meant or where it came from.

I then quoted the conclusions of that section of the paper which was published in the *Journal* of Audio Engineering Society meaning it was "peer reviewed." Not once but twice. The author had this conclusion about everything there:



Are you in agreement with the author of the paper? Or is your position that you know more than he does about the research his own team conducted at NRC, written in such clear, and non-ambiguous terms?

welcome to psycho-acoustics , amirm - im glad you finally were able to discover the measuring tool that has been imperative to the development and understanding of psycho-acoustics of indirect reflections for the past 30+ years.

keep your hands inside the vehicle at all times -

and remember, guys ---harman will now dictate to YOU what is positive feature and negative feature. and apparently, precise imaging is NOT a positive feature -

"Starting at the top, the term “image-shifting” is not the scary thing that it seems to sound. It simply says that the sound didn’t originate from that single point where the speaker was. Perceptually as I noted earlier from Harman research, this is considered a positive feature, not negative." ---

The world according to amirm:

Double blind taste tests have shown that people prefer Pepsi (untreated listening rooms) to Coke (listening rooms treated with broadband absorption and diffusion). The scientific evidence demonstrates that the reason for such preference is that Pepsi is sweeter (has more reflections) than Coke. Therefore, everyone must drink Pepsi (listen to music in untreated listening rooms). Preferring Coke (a listening room with "a predominantly direct sound field, making the experience more intimate, and the imaging tighter and more precise") is wrong.

Sometimes you just want to quote something and stand back!

Wonder if the guy selling you power cables can justify his product that way. "Sir, it doesn't mean anything that double blind tests didn't show an improvement. The listeners still heard the difference but didn't process it!"

I ask again. Can you please provide a link to where the powerpoint presentation came from and whether you agree with the conclusions of the paper.

you really put forth the image that you do not have ANY basic understanding of psycho-acoustics...
the copy-paste efforts does not imply an understanding of the topic.

do you not understand the difference between anechoic and effectively anechoic?

this is the difficulty of trying to converse with a flatlander who lives and views everything from the frequency-domain --- he is simply unable to comprehend what the time-domain data means, even though it is from toole. but hey, maybe now he understands that reflections of different time arrival and differing gain can INDEED have different psycho-acoustic effects.

Heaven help us all if figuring out how to create a high fidelity experience in our listening room is a matter of taste and not science. In that regard, we better not say a thing to anyone asking what treatment they should use. The answer must be that "I don't know what you like so do as you please. And oh, go conduct a bunch of tests to see what works for you. No, I am not paying for that, you need to do that!"

Look, if you all want to put aside research and published writing from one the top people in the industry, that's cool. Just don't position it as my views alone with silly analogies.

I didn't put forward the image. You did! I am providing the back up as requested with respect to how the data came about. Do you have the link yet for us to read what that author said? And do you agree with Dr. Toole's own conclusion?


So you created that graphic and didn't cut and paste it?


Why don't you ask Arny or Ethan if they agree with the exactly what I quoted from you. I will wait for their replies.

before a few hours ago, you didnt even understand the different psycho-acoustic responses of reflections based on their gain and time arrivals.

the problem is, you see acoustic 'treatment' as a product to add to the experience. acoustic treatment (be it a broadband porous absorber, a diffuser, or simply a LARGE FLAT PIECE OF WOOD) is merely a modification to the room to achieve the desired acoustical response at the listening position - both in the frequency-domain as well as the time-domain.

this fundamental misunderstanding leads to these types of conversations...



more deflections - im not talking to ethan or arny - im talking to YOU.
what is the difference between anechoic and effectively anechoic?

here's my statement again: read it a bit more careful this time:


psycho-acoustics 101.

it simply means that the low-gain reflections are below human detection threshold for ear-brain; but it does NOT mean that they are not heard - just not processed (effectively anechoic != anechoic). the low-gain reflections do not trigger localization, imaging, or any other keyed response. the ISD-gap in LEDE/RFZ, for example, is to be an effectively anechoic time period - so that all processing in the ISD Haas interval is keyed to the early arriving direct signal. and there-after, you have the ambient spaciousness, the re-inforcement of the localization based on the early arriving direct signal, and the liveliness of the room - if such is a design goal.

You spent tens, if not hundreds, of thousands of dollars creating your reference theater, including engaging Keith Yates to do 3D modeling and perform computational fluid dynamics to tell you where to place your subwoofers, speakers, absorption panels, diffusion panels and other acoustic treatments. Tell me, was his model that of "the prototype IEC 268-13 (1985) listening room, which in essential respects was intended to be representative of typical domestic listening spaces" which Olive and Toole - and you - advocate as the preferred listening environment.

You omitted the letter "s" from "treatments". Shame on you.

It was about "Doing a DBT related to room treatments" but somehow you narrowed it down to one thing, "testing a single reflection point". Making things up as you go, your dance move #4.


Willful ignorance or problem with English language.

Is "I am asking them to produce their own unbiased data on the efficacy of what they advocate" for "this is something you need to do if you are going to say there are improvements relative to the products you offer for sale" or "for people in the industry who are objecting to testing done by Harman in this manner."? Can you narrow that down to one?


"deflect", your dance move #1.


Trying to lump speaker switching with acoustic treatments switching into one category? Only a rookie would do that. Why do I say so? See questions in post #206

Amir, I am confident that if you sized a typical sound absorber or diffuser to fit into this little gizmo's rather limited capabilities, the results would be totally non-representative of a typical installation of effective room treatments.

Now if you built a room that had enough of these to cover the walls, floor and ceiling...you'd have something that worked and you would also be back in line with my budget estimates.

I am reminded of an earlier post on this thread by someone who chided me for thinking of room treatments as if they were like other audio components. It was of course very wrong-headed as applied to me because I am very aware of what it takes to impmlement effective room acoustics treatments.

Agreed. Just tells me that while some people show neat-looking pictures of acoustically treated rooms , they have no actual clue about what it takes to materially change the acoustics of a typical listening room in the real world.

This looks a little more real-world to me:



This picture shows maybe 1/6 of a typical listening room. Swapping the absorbers shown on just the walls would take at least 6 of the Harman speaker-swappers that Amir mentioned. So, now we are talking 36 units with the walls and the ceiling yet to go...

How is that? Ethan sells 2 by 4 feet traps. You are saying I can't put that bit of fiberglass and frame on a rotating fixture at first reflection point and test whether it does anything good or bad?

Does its function fall below the threshold detectability?

Interesting watching your posts writhe and curl in the winds of reason, Amir.

You keep changing the question to avoid exposure of what you do and don't know. For one thing you keep talking in terms of bad versus good, while there are many changes that are neither bad nor good, just different.

I said "non-representative of a typical installation", and you've gratuitously changed the criteria to threshold of detectability.

At the threshold of detectibility, the differences are far too subtle to relate to bad or good. I bet you didn't know that!

I frankly don't know whether a single 2' x 4' absorber in a reasonable-sized listening room will make a reliably detectable audible difference. When I last tried something like that with my listening room, adding a single absorber that size was like spitting into a hurricane.

When I've worked with absorbers like that in my listening room, I wasn't interested in a minor change on the order of the threshold of hearing. I wanted an readily and unmistakably audible difference.

At one point I had moved at least 6 of them into and around the room, and the results were far less than what I hoped for.

OTOH, I know that 8 4 x 4 pieces of 2" Sonex foam placed ideally can do quite a bit. That's the comparison of 48 square feet of absorber versus 128 square feet of absorber.

Rule of thumb is that the effectiveness of absorbing material relates to the area in the boundaries of the room, versus the area of the absorbing material. Obviously the thickness and type of material also relates.

My main listening room is about 24 x 16 x 9, so as they say do the arithmetic.

"into and around the room"
how did you determine where to place the absorbers in your room?





what do you mean by "can do quite a bit" regarding the thin porous foam absorber?

and im confused by your "rule of thumb" as well - why would you be concerned with area coverage in a small acoustical space that is dominated by focused reflections?


gas-flow-resistivity is the key value.

Thanks for the observation Arny. That is very useful. Here is what Ethan said earlier:


Bolding mine. Here is the tag line below that graph on his site: "Figure 1: Response comparison with and without MicroTraps at the first reflection points." The spec for MicroTrap says: "MicroTraps are 2 by 4 feet by 1-1/4 inch thick, and weigh 8 pounds. "

You are saying that the transformation that he shows on the graph is inaudible then? Don't you think having objective data that says that in the form of a non-sighted test would help settle a lot of arguments? And that graph is highly misleading if you are even close to being right?

If you could read well enough to usefully participate in detailed conversations like this Amir, you would notice that he said: "Response comparison with and without MicroTraps"

IOW, he was talking about a number of Microtraps that was greater than one. You asked me about just one.

And my opinion was: "I frankly don't know whether a single 2' x 4' absorber in a reasonable-sized listening room will make a reliably detectable audible difference."

What is unclear about "I frankly don't know"?

This is the picture that is shown right above that measurement:



You are saying that even though he is showing one unit there, and he talks about the mirror method of finding first reflections and covering them in his videos and such, that he still used multiple units? If that is the case, you think he has made it clear enough to the buyers that they would not be getting that transformation with one unit? Again, the image above is right above that measurement.


Oh, if you mean you have no idea, then you are wasting our time .

I don't know why it is so hard to get you to accept that if a product is sold with unknown impact on our audio system, it should not get a free pass for a listening test.

Earlier I explained the tests that NRC had run on reflections. They used an anechoic chamber. You have a set of excuses for that too?

amirm, as for the feasibility of doing a DBT related to room treatments, have you given up on such "not difficult to build at all" idea since even you couldn't think up a "readily doable" way?

why are these guys discussing the changes of a single sidewall absorber? the proposed arrangement is moot because it disrupts symmetry. why bother?

Thanks Amir (and mark)

appreciate the time you took to explain it, the missing explanations/captions certainly helped.

I've kept out of this after the last one, seems a pathological trait of this forum that everything degenerates.

I understand a lot of where Amir is coming from. Firstly, it is true that normal living rooms can be very satisfying, I am not sure that point of harmons is really in dispute is it?

Equally, that is a far cry from saying that normal rooms with normal furnishings (the wording alone is a hint, very woolly) cannot be improved upon. Else Amir would not have his theatre designed and done professionaly.

I think the following is perhaps the most important point...it IS a very confusing area to delve into, with a lot of contrariness even between advocates.

Still, WE are here because we want to go further and as such I think discussions are important as long as the person can learn from those discussions. No different in 'content/purpose' in learning about different speakers to improve what we have, just a different field.

That is my problem with amirs point, tho I can understand it. Still, as he says, it is a steep learning curve and yes many would roll their eyes before grasping what is needed to make sense of it. But that is no reason to not discuss it for the rest of us.

Another point of amirs is the cost to 'do it at home and find out for yourself'. For a lot of people yes. This is however also a diy forum, and for many that cost does not need to be as horrendous as sometimes represented. Ethan (laudably) has lots and lots of diy info on his website as well as but one example.

I just finished relaying some floors, and as a result now have half a bag of f/glass insulation left over. It is a trivial matter and absurdly cheap to start experimenting with those to begin to put this stuff into context for yourself.

What I do not want to see is any throwing up of hands as it were (which I think is why I am 'against amir' on this) and not find out.

Look, twas relatively recently that things like multiple subs, heck even eq on bass, was looked at in audio. Ten years if that? Don't reckon there was a lot of discussion and how to's (even now tho it is more mature and well known) in that area too? It is more and more accpeted that subs are not the evil the purists thought it was, the techniques needed and knowledge required is getting more and more widely known, and even then there is enough person to person variation on how THEY do it that mimics this treatment discussion.

The point being that to get this wider acceptance and knowledge that topic too had to go thru these forum disputes, it is part and parcel of the growing up of the field if you will, and I see this topic of room treatment going thru the same maturing curve.

The worst thing to do would be to stifle it and follow set formulas.

I will go thru the explanation again amir, buit a few points popped out on first reading.

This graph was generated in an anechoic chamber essentially, I can see that it would be very precise in investigating these threshold levels etc, but even WITH first reflection points etc I find it hatd to beleive that it would mimic a normal listeneing room.

Just by the by, has anyone listened in an anechoic chamber with additional speakers etc, does it (the additional speakers) 'bring it back to real world sound'?? Anyone had a demo like that?

Also, I acknowledge you did not bring the graph up, and it seems harmons tests were different.

I am also sure the green line is the important one in this discussion, it does clearly show that early not very attenuated reflected sounds do affect perception, and using the cheap method I espouse is not hard to test.



My room is not 'normal' (17 foot ceiling for starters), and I used treatment for (what turned out to be a very different reason), slap echo. But the essential point is that it is not hard nor expensive to test this stuff.

I did not find that first reflection points helped (nor did they hinder btw) so I don't have it. Then again my layout kinda does not 'have' first reflection points so moot in my case.

I would not be without the treatment, but horses for courses, so each needs to find out for themselves.

Amir, you seem to be unaware of the potential ambiguities involved with looking at plan view drawings. The actual number of absorbers is unknown since they could be stacked vertically.


His text says what it says, and what it says is plural. BTW, that is more than one!


Since I said what I said several interchanges back, it is not me that is wasting our time but you, Amir.


There's no talk here about free passes for listening tests, except from you Amir.


Amir, if you can't figure out the wisdom of testing reflections in an otherwise reflection-free environment...

Amir is on his own in any discussion changes related to the use of a single small absorber.

I guess that he is trying to make an argument that since a single small absorber's effect may be inaudible or subtle, the use of any number of them is also subtle.

This contrasts with Amir's previous advocacy of any number of things which remain inaudible, even when used in large numbers.