Does sound sounds better in a room full of furniture and stuff or without ? - Page 3

In a word, "Yes". But I"m confident dragonfyr will be along shortly with a 2000 word confirmation manifesto.

Yes, and I addressed that long ago in Post 23. In most cases it's not true reverb, but that's what the software calls it, and the graphs display as RT##. Further, smallish rooms can have real reverb when empty. My two-car garage is an amazing reverb room, even with one car in there. The size is 22 feet by 24-1/2 feet, with an 8-foot ceiling, so it's technically a small room. Other than one car my garage is mostly empty, and (amazingly) there's no flutter echo. Although it has bass modes, the reverb quality for mids and highs is superb - you can whistle any musical note and they all sustain for the same time with no emphasis. I never measured the decay time but it's at least a few seconds. And it's definitely true reverb.

--Ethan

LOL: How to spot an online fibber

Yes, proper measurements evaluating both the modal and specular behavior of a room are in order. And the use of these can very definitely facilitate both a more accurate assessment of exactly what is happening in a room acoustically, and can aide in the more appropriate selection and application of appropriate treatments in order to effect the achievement of a desired acoustical response.

But I would not recommend one invest in a limited system such as OmniMic or XTZ for which ones pays a premium price and in turn receives a product the is unfortunately constrained in its abilities in the time domain.

If one wants to spend any money in addition to a free single channel FFT product such as RoomEQWizard, I would very highly recommend one invest the79 Euros or ~$100 in ARTA and obtain a 2 channel FFT with MUCH more extensive capabilities than any of the smaller platforms.

And it would be fascinating to hear Ethan - the self proclaimed "acoustical treatment expert" - explain how, in his small rectangular garage, the parallel surfaces support no flutter echo, but instead promulgate a true reverberant field at all frequencies where an ETC is unable to resolve the various specular reflections into any direction as the are so randomized that the arrival of the energy is equally probable from every direction simultaneously; effectively disproving everyone from Manfred Schroeder to Heinrich Kuttruff. I guess another work featuring acoustic idioms is in order - that is, once the self proclaimed "acoustical treatment expert" finally works through the small personal issue of finally 'deciding' what reverberation really 'is'. It will be the first acoustics text where one can legitimately cite a former president as a reference source.One might suggest a title of "Acoustics as Slang".
Edited by dragonfyr - 6/19/12 at 11:29am

or..."How to make a bundle without confusing people (including oneself) with facts, measurements or acoustics"

It would be a lot more fun to discuss this stuff if you weren't so condescending.

The only difference between "true reverb" and individual reflections is the number of reflections. If you disagree with that then you disagree with Floyd Toole (page 43), Philip Newell (page 82), and Everest (page 134). Every one of those authoritative books describes reverb as the decay of a bunch of reflections. Same for the definition of reverb at Wikipedia:


I don't know why you feel you need to split hairs, to prove that I'm ignorant and you're an expert. Sheesh.

If you insult me or my company again I will not engage you further, and I'll report you to the mods.

--Ethan

This is laughable. And quite frankly I expect more, as you are quite capable of contributing better...

So if we have a room with LOTS of flutter echo, it is reverberant. There is no necessary randomization of specular reflections, no establishment of a statistically homogeneous diffuse soundfield.

And we can glibly dismiss the co-requisite requirement that it in a reverberant soundfield that the arrival of energy is necessarily equally probable from every direction simultaneously.

It is simply a matter of having "lots" of reflections. And if all of the 'many reflections' are oriented from a particular surface, say a concave proscenium or surface, that is "reverb"?

As now in Kutruff's Room Acoustics we can now expect to read that a reverberant soundfield is now properly defined as "allot of reflections".

Again, what we have here is the self proclaimed "acoustical treatment expert" trying to dumb down the formal definitions that define the pre-requisite conditions necessary for RTxx calculations to be applicable into colloquial slang.

And rather than use the proper terminology and the proper measurements that can accurately image and evaluate a LESS than statistically homogenous reverberant soundfield, we persist in attempting to redefine the terminology of acoustics into slang as he rationalizes his original confused misuse of the term and description of the actual behavior.

You see, if he obfuscates the issue any further, even this discussion will become 'reverberant' in its effective confusion of facts, misinformation and slang.

So, at what point does a non-reverberant field become sufficiently non-reverberant so that if we squint we can imagine it to be 'satisfactorily non-reverberant but close enough to reverberant' so that we can begin to apply the term 'reverberant' as slang to something it does not properly describe?

And for the next trick he will explain how a dog is really a cat if we just look at the problem from the right perspective...for you see, we 'hair splitters' are just being far too unreasonable in our insistence that SOME words do have meaning and our failure to treat the English language and the language of acoustical physics as a 'living language' subject to evolution and the arbitrary morphing by anyone for any vested purpose.

Congressman William Hungate from Missouri, who observed during the Watergate hearings, that if they had trotted in an elephant for examination that it would have been redefined as "a mouse with a glandular condition" would properly be aghast!

In acoustics, the term "reverberant" has a special strictly defined meaning. And the statistical tools used to describe this behavior are reserved for circumstances where these criterion are met. The fact that some people persist in using a proper term as slang to describe a statistically non-reverberant sound field does not change the definition nor the proper use.

The sad fact is that a self-proclaimed "acoustical ... expert" continues to obfuscate an issue as they persist in trying to rationalize the continued confusion caused by their (and others) continued inappropriate use of the term as slang has confused allot of people who now think that using the statistical RTxx calculations are now acceptable in any circumstance is a mockery of acoustics and an embarrassment of those who rationalize that doing so is acceptable.


The fact is that in a small acoustical space we do not have a true statistically reverberant soundfield. And as a result we use tools that properly evaluate the locally variable soundfield in what may be a rather diffuse but non-reverberant space.

Its time to move on from what General Honore referred to as being "stuck on stupid" and refer to acoustic behavior accurately and in doing so employ the proper terminology and the proper tools.
Edited by dragonfyr - 6/19/12 at 12:32pm

Thanks for the reply, Dragonfyr. I was actually looking into the XTZ measurement system. What would you recommend as an alternative? What is ARTA and also what is a 2 channel FFT? Can you explain a little bit about what those are and how someone like myself could possibly utilize them?

REW is a great free single channel FFT.
It evaluates the one input signal.

ARTA is a very powerful (with MANY additional features - that may seem intimidating at first simply because of all the choices - but don't let it be) 2 channel platform available for literally a 'steal'. (The joke in the industry for the past ~10 years has been when the Croats figure out the value in the competitive world market that they will add $1000 to the price to make it competitive with the other full featured platforms available...)
So, please check it out! (Pricing and manual downloads here - it is VERY well documented!)

For basic use, all you need is a separate mic preamp (the ART Dual USB Pre available from BandHPhoto.com for $69 delivered is a good bargain), a calibrated Dayton EMM-6 omni mic from Cross Spectrum for $70, and 2-4 cables for the mic, short loopback cable,output of the pre, and an adapter and an RCA cable for the avr from Monoprice.com, and you are set.

The simple advantage of 2 channel FFT over single channel is to compare the two inputs.
This 'simple' ability that has myriad applications ranging from a simple 'dif' (differential/'difference') comparison of the two inputs to more complex studies such as binaural inter-aural cross correlation (IACC) studies (where for a give signal you can compare what each ear 'experiences') or it can be as simple as aquiring two measurements at different locations simulatneously...

At first you may not imagine the need for such capabilities, but trust me, after playing with it for awhile you will begin to realize how this capability expands the possibilities exponentially, and while you may not use all of them immediately, I think you will find that you can use the 2channel capability routinely - as once you have it, it will be hard to imagine trying to do things without it.

Download the ARTA User Manual and take a look. Don't let the myriad options intimidate you - they are the IF you need them. Just focus on the ones you need at first - but know that there is help and the option utilize them you grow into them!

But in any case, I would suggest starting with REW - its worth it simply to fight with their more complicated than necessary documentation and to try to decypher how to connect it! Its free and offers a good array of tools, including the time based ETC that can actually be calibrated in time (as opposed to the canned expensive platforms like Omnimic - augh!).

And then, I might suggest considering ARTA if you have the extra $100 burning a hole in your pocket. But be sure you are going to use it before spending the money - REW will do the job if it is just for the basic but COMPLETE 'necessary' tools you need for evaluating and tuning a small room.

If after looking at them , if you have questions - as I am sure you will! - fell free to PM me and I will be glad to chat and explain more in detail and try to answer any questions you might have.

Oh, and I should note that ARTA is available for use as a fully functional demo - except that you cannot save the results, so its definitely worth playing around with.
Edited by dragonfyr - 6/19/12 at 1:24pm

It is wonderful to see you acknowledge that bandwidth limiting may be necessary in order to extract correct information from ETC. Sadly, I don't recall that aspect being mentioned as people are quickly schooled to go and run a non-band-limited analysis in post after post.

That aside, it would be nice if all one had to do was to use band-limiting and good data just jumped out at you. Alas, the filter there is just a tool and doesn't do that. You need to know what to limit and how to interpret what is being shown in the context of how we hear. Let's look at a great example of this from AES paper on the design of the reference listening room at Fraunhofer Institute (FHG). Just in case you don't know who FHG is, they are the German government funded research group that brought us MP3 compression and contributed to AAC audio codec. They wrote a nice paper on the design of the room where they were trying to do everything by the book. Let's see a key section on the floor effect/treatment:

Vision and Technique behind the New Studios and Listening Rooms of the Fraunhofer IIS Audio Laboratory
Andreas Silzle, Stefan Geyersberger, Gerd Brohasga1, Dieter Weninger and Michael Leistner
Fraunhofer Institute for Integrated Circuits IIS, Am Wolfsmantel 33, D-91058 Erlangen, Germany
Innovationszentrum für Telekommunikationstechnik GmbH IZT, Am Weichselgarten 5, D-91058 Erlangen, Germany
Fraunhofer Institute for Building Physics IBP, Nobelstraße 12, D-70569 Stuttgart, Germany

Figure 11 shows sections of the first 30 ms of two impulse response measurements, band pass filtered from 1 kHz to 8 kHz, both from the center loudspeaker to the reference listening position. The graphs are almost identical with each other, except around 13 ms. The first 10 ms with zero signals represent the direct sound travel time, the strong peaks at 10 ms originate from the direct sound. In the measurement of the untreated room with the regular carpet covered floor (black graph), a second peak at 13 ms with 0.6 amplitude relative to the direct sound, which is equivalent to 4.5 dB attenuation, represents the reflection caused by the floor. The time delay of 3 ms corresponds exactly to the 1.7 m path length difference between direct and reflected sound. This strong reflection violates the requirements of the standard. It can however be attenuated to a negligible level with little effort by placing a piece of moderately absorbing material (e.g. 50 mm thick porous material, about 1.5 m x 1 m) on the floor at the mirror point, i.e. the point, where the sound is reflected on the floor. The measurement result for this case is presented with the red graph in Figure 11. None of the reflections exceeds 0.3 amplitude relative to the direct sound, which is equivalent to more than 10 dB attenuation."

This is Fig. 11:



By now folks following the school of running these time domain tools such as ETC and doing away with these reflections would be giving each other high fives . Sadly for them but fortunately for goodness of audio science, FHG folks did one other important test: audibility. Coming from background of audio compression and science of audio in general, folks at FHG know that what a meter shows is not the same as what the ears capture and the brain interprets. It is that science that allows us to compress an audio file to 128 kbps at 9% of its original file size yet have it sound so close to the CD. And how if you set it to 384 Kpbs you get "near lossless" fidelity yet 75% of the bits are still thrown out! But we digress . Back to our FHG researchers, we read this further in the report:

"3.1.4. Subjective room assessment"

[....]

Regarding the floor reflection, the audible influence by removing this with absorbers around the listener is negative – unnatural sounding. No normal room has an absorbent floor. The human brain seems to be used to this."

Houston, we have a problem!!! The reflection from the room, shown on a band-limited ETC display was sharply reduced using an absorber yet the subjective impact was negative. The authors rightly speculate that as humans we have evolved to take into account that reflections are there. The brain actually puts them to good use, taking into account the differing arrival time and frequency spectrum that I noted in my last reply. It uses this information to aid in intelligibility as noted by Dr. Toole in the same post. Here he is in first person:

"9.1.2 Binaural Hearing, Adaptation, and Comb Filtering
Room reflections arrive from directions different from the direct sound, delivering different sounds at different delays to each of the ears. This means that the details of comb filtering will be different in each ear....because the head is a substantial acoustical obstacle between the ears, sounds arriving from angles away from the forward direction will be different in spectrum at the two ears....The effect of acoustical shadowing by the head is that sounds that arrive at the more distant ear will have reduced amplitude at high frequencies, meaning that the comb filter acoustical interference will also be reduced.

Interestingly, humans seem to cope with these situations very well because the spectrum we perceive is a combination of those existing at both ears. It is a “central spectrum” that is decided at a higher level of brain function. A microphone, at best, can give us a crude estimate of the sound that one ear might hear (crude because a microphone does not have the directivity pattern of human ears—described by the HRTFs). The fact that the perceived spectrum is the result of a central (brain) summation of the slightly different spectra at the two ears significantly attenuates the potential coloration from lateral reflections (Bilsen, 1977; Zurek, 1979). Krumbholtz et al. (2004) confirmed and extended Zurek’s work, showing that the central spectral average of different sound events at the two ears can be mimicked by adding the stimuli from the two ears and presenting the sum identically to both ears.

If many reflections from many directions the coloration may disappear altogether (Barron, 1971; Case, 2001; Moulton, 1995), a conclusion we can all verify through our experiences listening in those elaborate comb filters called concert halls.

This remarkable finding helps further explain why sound in rooms is so pleasant and adds weight to the refl ected sounds shown in the measurements of Figure 9.5."

So unless you are calling how we hear "bogus," then I say the findings here are quite substantial and backed by significant amount of science and listening tests. If you are an expert speaker designer and acoustic engineer, and know how to band limit and interpret the ETC results, there is data that can be gathered from it. But for someone wanting to learn the basics of how to optimize their room, it has no place in their toolbox. It can only lead to incorrect conclusions regarding what to do, and what not to do.

And importantly, reflections are NOT the problematic areas. You don't want an overly live room to be sure so some amount of absorption is necessary. But don't go chasing them with these ETC type tools. There are better ways to get good sound.

Sincerely,
Mr. Bob.

Whatever.

Death by a million cut and pastes, but zero understanding. Your example makes my scanner look like a genius.

And the only thing I call bogus is your newly discovered old research of which most of the rest of the acoustics world has been familiar for many years. But I applaud your enthusiasm, if not your histrionics and incomplete understanding and misrepresentation of others research and awareness.

Sorry friend, but it is YOU who are ignorant of the fact that a bandwidth limited ETC was included in the basic capabilities introduced by Dick Heyser from inception. And I am sorry that you are ignorant of the fact that the broadband behavior was a prerequisite condition from inception as well.

Only if the surface is not broadband in its behavior is such a diagnosis useful, and then it is used to re-mediate the surface so that it is broadband. And since all utilized appropriate treatments are effectively broadband, the need for subsequent bandwidth limited ETC as precluded.

Unfortunately you and your crowd apparently are stuck in the realm of non-broadband boundaries and non-broadband treatments necessitating your languishing in the realm of bandwidth limited ETCs verifying that your treatments are in fact effectively EQing the indirect signal and resulting in coloration of the direct sound.

But it speaks volumes that are your focus, that they are apparently dominated by mistakes in the form of non-broadband treatments necessitating the continued use of bandwidth limited ETCs. My suggestion is to get with the program (and catch up to where the rest of us were 30+ years ago and use better broadband treatments. There is absolutely no reason to remain, as General Honore observed: "Stuck on stupid."

So take a bow. Those more familiar with proper design and the proper use of appropriate treatment long ago solved the imagined hysteria associated with the imagined need for continued bandwidth limited ETCs that merely confirm that what you have done to that point has been flawed.

May I suggest that you go 'back to the future' and progress to where the rest of the acoustics world was more than 30 years ago and realize the value of implementing an environment where the boundary and treatment response is indeed broadband effectively remove your continuing dilemma a the source, instead of lecturing about the need to apply diagnostic band-aids in your newly discovered tool of which you neither understand now apply appropriately.

Amazing how those old farts were way ahead of where you are now. And why when Toole made his momentous discovery that boundaries and treatments can indeed be (improperly) non-broadband circa 1987, that the rest of the acoustics world merely yawned as such startling info was old news and long since resolved through the use of broadband boundaries and treatments.

To repeat Don Davis' oft cited observation: Isn't it funny hoe the ancients keep stealing all of our inventions!

Watch out or you may find yourself catching up to where the world was 30+ years ago!
In other words, your observations were old news then..and only further reinforce just how out of touch you are now.

So I guess we should all stand by breathlessly awaiting your forthcoming revelations. I wonder, with they extend as far as to dare encompass such concepts as the Earth being round? ...or that there are indeed little invisible 'bugs' in water? One can hardly wait.

And then how about explaining why the ultimate early reflection environment created at Blackbird has been effectively converted to a response closer to the LEDE by the covering of the side diffusors with quilts, effectively rendering them absorbers, as the engineers in the facility prefer a more tightly defined image as opposed to your amorphous 'large' image so longingly preferred by those who listen only to surround and worry about the sense of envelopment during FX and exploding suns.

But as has been the case for many months now, this fact will conveniently be ignored as it does not mesh well with his one dimensional cut and paste campaign based upon limited understanding and experience that eschews the reality that there are multiple response models and that not all - especially in the trained ear community - prefer Toole's amorphous indistinct 'big' image based upon lowest common denominator polling data.

Ironically, the exact same 'large' response created by the Bose direct-reflecting speakers dating back to the 1970's that was rejected by many for exactly the same reasons. Its deja vu all over again!

And let's not forget, statistically, the majority of graduating senior cannot find Chicago on a map and believe we never really landed on the moon (I wonder if they can find that?) and believe in ghosts. Great company and high standards there... And the polls prove it!

So while many of Toole's observations that you merely parrot duly constitute what are merely footnotes in a larger body of knowledge, as opposed to the 'one and only definitive correct' response option, the fact is that while allot of folks may like the Bose sound, NOT ALL DO! ...As easily evidenced specifically at Blackbird Studio.

So, now you get to go back and continue your Google and document word searches desperately trying to find a quote in lieu of being able to explain the concepts yourself that will finally convince us that you are desperately trying to find something to justify your cut and paste position.
Edited by dragonfyr - 6/19/12 at 2:29pm

With this you absolutely prove that you have no concept of why measurements are used nor how to employ or to interpret them!But you can cut and paste with aplomb! Too bad some of that effort was not expended in trying to understand the concepts which you can only copy as you saliently try to build a platform for re-marketing Toole's concept. But like most marketing brochures, it is long on claims and hyperbole and short on insight and understanding.

Measurement tools are used at inception in an existing room to evaluate the behavior in that room. They provide a baseline behavior which one can evaluate in light of possible responses that can be achieved in the space, or in light of an already determined response to be achieved.

Assuming one is versed in the basic concept and how these concepts and behaviors can be accomplished, measurement tools do NOT tell anyone what to do. They provide the raw detailed data that one then uses to make their decisions, as this also assumes that the designer knows about that which they are working. the tool is NOT a substituent for this understanding - which readily exposes why you are having such problems with the concept of measurements, seeing as you have no concept of what they indicate nor of the acoustical and psycho-acoustical principles to which they correspond.

Sadly, cutting, pasting and memorizing what someone else has written is not an adequate substitute for actually studying and learning acoustics. Sorry.

Bandwidth limited ETC are primarily useful in two places, after which, if things are done correctly, they are not needed at all. And this apparently confuses you as the majority of persons who use them are already well cognizant of the pre-requisite broadband conditions to be observed and hence do not need to run back to the beginning and take care of that which should have already been addressed.

Their use is in evaluating the basic boundaries and evaluating the comparative nature of the direct and reflected energy. This is usually only necessary in an existing structure with boundaries consisting of unknown composition. In structures where the wall composition and structure is known, experienced acousticians already know the nature of the acoustical impedance of the boundary and how they perform spectrally. Additionally, knowledgeable designers specify boundary construction that exhibits the acoustical impedance and thus the spectral behavior they desire. (And ironically many more knowledgeable designers specify inner boundaries that are reflective to the broadband specular energy while selectively passing the low frequency modal energy.)

And if a structure is identified that violates the broadband behavior, it is corrected and verified before any further steps are taken.

The other primary use of band limited ETCs is in the evaluation and testing of treatments used for absorption and diffusion.

But you will note that this process is all preliminary to the act of specifying and designing the actual space to be created.

Skilled and knowledgeable designers do not need to install and THEN start checking to see if the boundaries and treatments are broadband in nature. That is a MISTAKE. And knowledgeable practitioners avoid mistakes.

If the boundaries are sufficiently broadband in nature, then there is no further use of band limited ETCs.

And in such a room, skilled knowledgeable designers only employ sufficiently broadband treatments, thus there is no effective reason for further use of band limited ETC - at that point the use of full range ETC is quite appropriate.

But that applies to folks both knowledgeable and skilled in this subject area.

On the other hand, in your case, it must shock and confuse you that the ETCs you see posted in texts are employed by those knowledgeable enough to know this and thus use such concepts appropriately and such measurements correctly?

Obviously it confuses you, as you remain dismayed at this use by folks who have employed them since inception over 30 years ago due to your ignorance of acoustical best practices.

But I can easily understand how you and others at the same point in the learning curve regarding the nature of boundary and treatment acoustical impedance and behavior would feel the need to make band width limited ETC measurements after their installation, seeing as how so many are quick to rationalize the use of too thin porous treatments as is so common on this forum, or avoid the use of more effective alternative designs which effectively achieve the performance desired.

In your case with your understanding of both the evaluation of existing boundaries and of the design and application of various treatments without understanding what is required to achieve effective broadband behavior, you SHOULD be using the band width limited ETCs if for no other reason than to warn unsuspecting folks who might erroneously assume that you know what you are doing.

Dragon just explained how you can spend $100 on that software to better your audio experience. I will make you a deal that hopefully you can't refuse. Spend $44 on this book from Dr. Toole: http://www.amazon.com/Sound-Reproduction-Acoustics-Psychoacoustics-Loudspeakers/dp/0240520092/ref=sr_1_1?ie=UTF8&qid=1340154218&sr=8-1&keywords=floyd+toole

Take a few days that it would take to learn to use the tool that Dragon mentioned to read the above book. If at the end of that you did not learn a ton more about audio in general, and acoustics in specific, send me a PM and I will buy the book from you and refund your $44. No questions asked! If Dragon is willing to offer you the same deal for the cost of said software, do both and then please report back on what was more useful. You will have nothing to lose and lots of potential to gain.

Note: I have zero financial interest in you purchasing the above book. So I would be out the $44 since I already have the book.

Note he does not tell you how to get a copy of the book for free....which can indeed be done!

But I am glad to tell you where to get RoomEQWizard for free (go to the HomeTheater Forum)or a fully functional (sans ability to save) copy of ARTA for free.

What he refuses, or more accurately, cannot do is provide an answer as to why the trained ears have opted to cover the lateral mega diffusors in Blackbird with quilts, effectively converting the diffusors into absorbers and tossing out Toole's 'big' amorphous imaging response out in favor of a quasi- LEDE response featuring a much more defined and accurate image. ...So much for preference polls and 'big' amorphous imprecise imaging!...But MSBob is silent on this, as facts confuse him.

And note that after we have made a shambles of his nonsensical objections to ETCs about which he has no knowledge except what he has cut and pasted, he still refuses to actually address ANY of the acoustical issues with anything other than the equivalent of cut and pastes ignoring the greater body of material where Toole actually agrees with the large concepts.

But ask him why he apparently persists in utilizing boundaries and treatments that are not broadband and thus necessitate the use of band-limited ETC in order to expose mistakes that could have easily been avoided if proper treatment broadband treatments and best practice boundary topologies had been followed from the beginning... Just don't expect a direct answer - as he will simply respond with more cut and pastes of the SAME tired material.

Ever the marketeer, never the acoustician.
Edited by dragonfyr - 6/19/12 at 7:53pm

Dragonfyr- I sat down and read the first section of the ARTA manual in order to get some sort of general understanding of how to use it. Seems like it is a hell of a tool that can do lots of different things. As far as the hardware to get setup and using the ARTA software, I will look into the ART Dual USB PRE, as you suggested. I was also wondering if I could use my existing Behringer ECM-8000 mic as I already have it here with me? I am about to send you a pm in just a few minutes as well.

The Behringer ECM-8000 can be used, but I would seriously suggest also using it in conjunction with it's calibration file. Due to the large individual variation in the mics, ideally you will have a calibration file for it such as what Cross Spectrum supplies...

OK, now to this bit. You are correct that research has been on going to try to use this IACC meteric (a value between 0 and 1 showing how close two signals are) and what it means for our preference for a signal. Despite your statement otherwise, Dr. Toole extensively examines the prior science in his book and attempts to bring it all together in a comprehensive yet (relatively) easy to understand manner. Here is the start of that section:

"Many reflections have a positive contribution to listener preferences, but some refl ections are more desirable than others, depending on timing and direction. The deciding factor is which of them are most effective at generating a sense of spaciousness that, in turn, follows from low interaural cross-correlation (IACC). The greater the proportion of sound arriving from the sides, the greater the differences in sounds at the two ears, and the lower the IACC. Start to think in terms of “preference,” “spaciousness,” “low interaural cross-correlation (IACC),” and “lateral refl ections” as positively correlated with each other. IACC is a statistical measure, and although it is a correlate of the desirable perception of spaciousness and envelopment, it needs to be remembered that correlation does not explain causality. Griesinger (1997), for example, is inclined to talk about fluctuations in interaural intensity differences (IID) and interaural time differences (ITD) as a better way of explaining certain spatial perceptions. These quantities are responsible for our ability to localize sounds in space, and when they are randomly varying, fl uctuating, we are logically unable to do that."

Skipping a number of pages of graphs and research references, he summarizes this way:

"A persistent ambiguity exists in terminology relating to the perceptions of spatial effects. It is tempting to apply a single descriptor to it all, like “spaciousness” or “spatial impression,” and just be done with it. That is what Barron did, but who knows what aspects of the complex spatial picture his individual listeners were attending to and what manner of scaling they applied to their judgments? It matters.

All that really can be said with certainty is that refl ections that reduce the interaural cross-correlation increase something called “preference,” and it is difficult to have negative thoughts about that. When pushed to describe what led to the preference, listeners described various kinds of spatial effects relating to the apparent lateral spread, a broadening, of the sound source or to the impression of being immersed in a large reflective space."

And this key punchline:

"There were no complaints about “comb filtering,” the universal justification for absorbing strong reflections, and nothing could be more starkly displayed than a single simulated reflection in an anechoic chamber. It is as though these listeners were in a topsy-turvy world, responding with praise to combinations of sounds that, in a stereo world, have long been thought of as flaws; this gives us all the more reason to press on and fi nd out what is really happening."

If you ask people why reflections should be eliminated, they quickly point to graphs that show that when the direct and indirect sound of the speakers are combined, they generate "comb filtering." Those graphs sure create concern in people looking for flat line response and hence, are a great tool to send people shopping for all manner of acoustic products as to prettify said graphs. Yet listening tests such as those covered here show that listeners are oblivious to these transformations (at higher frequencies). The explanation of why has to do with psychoacoustics and a concept called equiv. rectangular bandwidth (itself driven from equal loudness graphs and auditory filters) which deals with discrimination ability of the ear as the frequencies rise. I will cover more of this in a future post. For now I leave you with this quote from Dr. Toole:

"9.1 THE AUDIBILITY OF ACOUSTICAL INTERFERENCE—COMB FILTERING
The term comb filtering just in itself sounds ugly. And its physical appearance, a succession of deep notches, looks ugly. And ugly is bad, so comb filtering must be bad. But if this is the prosecution’s argument, they lose! The defense can call witnesses, many who will have impressive academic credentials, and many, many more who are just ordinary listeners but can attest to the audible innocence of this phenomenon. Many of them will claim that, in some situations, comb filtering sounds good—and under oath, too!"

[shows what comb filtering is]

"This is one reason why acousticians, for decades, have preferred smoothed versions of room curves: They look better and, typically, we are unable to hear the “grass”—the undulations at high frequencies. The explanation lies in the inability of the ear to separate spectral features that fall within a critical bandwidth or its present-day variant, the equivalent rectangular bandwidth (ERB; Moore, 2003, Figure 3.5). Figure 9.4a repeats 9.3c and shows the critical bandwidths and ERBs at different frequencies.

For the 10 ms delayed refl ection illustrated in (a), spectral bumps and dips are separated by 100 Hz. Obviously, when the ear is unable to perceptually separate such events, the details in a measurement are of little value. The implication of Figure 9.4 is that above 100 Hz, if one adheres to the traditional critical bands, or above about 500 Hz by the new ERB criterion, the ear cannot hear evidence of this comb filtering, and at lower frequencies the effects will be much less severe than the visual presentation suggests. As the delay increases, as would be the case for many small room reflections, the spacing between adjacent peaks and dips is reduced, more of them fall within the critical/ERB bandwidth, and the potential for audible effects is further lessened."

So no, you are not correct that Dr. Toole doesn't cover this topic. Nor that he ignore prior research. The truth is exactly opposite. There are 17 pages of references in Dr. Toole's book. By my quick estimation, that amounts to some 270 external references that he examines in concluding what he does in his book! No one who has read his research and book can assume he is not covering prior research or concepts like IACC. This research, when interpreted by someone with 40 years of professional experience in acoustics and psychacoustics leads Dr. Toole to very different conclusions than commonly believed in forums. Imagine that: the Internet chatter being wrong about something .

Congrats on your ability to do more word searches and cutting and pasting!

And for the record, you are allegations are bogus as usual. No one said Toole did not mention IACC. What was said was that all of your histrionics based upon his vast(sic) references scattered over a WHOLE 17 pages that come some 30 years of extensive research on the topic about which you were previously totally unaware only to have you reduce iot to a number between 0 and 1 is a mockery of the subject, in addition to your not understanding the takeaway of the issue in a properly designed symmetrical environment!

And after all of your histrionics, what does Toole say about the differences in IACC correlation in a symmetrical environment?

Does he say they are reason for substantial concern? And aren't you the one saying that such comb filtering effects in such an environment are not of concern?

Which brings us back to where? Oh, you mean in a well designed symmetrical environment there is no need to perform additional Interaural testing, as a symmetrical response is easily accounted for with a single ETC as both 'sides' are essentially equivalent? Oh my!

And all your hysteria and frantic word searches wasted!

But congratulations on finding the subject mentioned in his book. Now perhaps you will want to spend a bit more time reading up on the subject and actually finding out what it all indicates.

Oh yes, that the effects of inter-aural cross correlation are not significant and not worth chasing the additional complexity of binaural measurements requiring a 2 channel FFT when a symmetrical environment is utilized.

And if an asymmetrical response is the result, we already know the effects that such uneven response will have on localization, now don't we. Well, let me qualify that, don't folks familiar with acoustics already know both the cause and the nature of the result in terms of altered localization and imaging? The simple answer, we avoid asymmetrical designs that result in asymmetrical inter-aural responses! Again, best practices...best practices.

But to correct you in one more detailed respect - binaural inter-aural responses are not limited to IACC correlation "reduced to a number between 0 and 1". But I guess that is the extent of what you word search tells you, as you are obviously not being familiar with the subject to any greater degree than what is mentioned in Toole's book.

Below is a display of the binaural interaural ETCs showing a much more complete binaural response taken with a Knowles in-the-ear microphones showing the actual response including the effects of the pinnae (outer ear) . Sorry we didn't reduce the detailed results to simply a number between 0 and 1 for you.



Binaural Interaural PolarETCs.D'Antonio.png 71k .png file


But I am curious as your silence is deafening in response to why the trained ears at Blackbird have covered the lateral mega 'diffusors to end all diffusors', with packing blankets in order to render them absorbers and 'suffer' interminably accurate precise imaging rather than relish the amorphous imprecise imaging and spaciousness of your reflection rich environment advocated by Toole and your dictatorial tyranny of surveys????
Hmmmmmmm????

We're waiting......
Edited by dragonfyr - 6/20/12 at 9:37am

How?

Thanks . So when you said and I quote, "as it was not mentioned in the single book he read" which book did you mean if it were not Dr. Toole's?

What I got from all that is that this whole thing is about the use of the word "echo" vs the use of the word "reverberation".

Most of us don't have the time to get a PhD in acoustic engineering. And most of us can't afford somebody who did. But if it makes you feel any better, I didn't use a mirror as a tool, and I didn't put absorbers everywhere. Instead, I followed a few general guidelines from reading from other forum members and professionals who care to share their insight. My entire front wall is dead - treated with 40mm (1.5625") mineral wool. On the sides, I have 3 600mm x 1000mm x60mm panels back to the sofa, with the rest of the sides untreated. And one large 8" thick or so absorber stuffed in the large back window in an attempt to attenuate (or whatever highly technical term you use to define the phrase "make the sound quieter before it gets outside") the sound enough so that when it gets outside, it's not loud enough to make the neighbors mad. Oh, and I have a carpet which covers almost the entire floor.

I did use REQ. It does list EDT, T20 and T30. Strangely enough, this is under a tab labeled "RT60", so maybe you should complain to the people who made REQ and have them re-label it since apparently RT60 is not what is being measured. But as far as EDT is concerned, it is about .2 all the way down to about 90Hz or so. That's why I plan to get some thicker panels to put in the corners to see if I can get that last octave a bit lower than the current .68.

Are you asking me who is buried in Grant's tomb again? Here is the graph from FHG room's:



There is no frequency information in that graph. We have amplitude vs time. Much of what we know in psychoacoustics comes from the frequency domain. Let's go back to mother of such research with Fletcher-Munson equal loudness graphs (circa 1930s):

The graph shows ears sensitivity to level and how it varies substantially based on frequency. Each one of those lines represents the same perceived volume. This concept is used in audio compression for example where we can increase the distortion/noise level of low and high frequencies and still have the distortion be inaudible (or relatively so compared to the most sensitive mid-frequencies). An amplitude measurement is devoid of frequency components so you can't perform such analysis and take action appropriately. The FHG graph for example showed an impulse reflection from the carpet prior to addition of the absorber. The carpet clearly filtered the sound before reflecting what was left back. A reflection that was not filtered would have had a different audible impact when combined with the direct sound.

The above is the reason perceptual (psychoacoustic based) audio codecs are almost always "transform based." The audio samples are converted into the frequency domain using an overlapped transform (so that we can invert it fully) and then we perform the analysis there. Once done, we transform back to time domain to get our original time varying audio samples. We would not bother with any of this if we could perform the same analysis in time domain using the music amplitude as our source of insight.

Now, using band limiting can tailor the ETC data somewhat to focus it on the area that we may be interested in. But just throwing numbers in there doesn't do that as FHG researchers discovered. As I noted, I don't recall you all schooling people in how to use ETC with band limiting. Feel free to post threads where you have instructed people this way and the motivation for what values to be used.

Now contrast how much we can learn from frequency domain by looking at a graph from my recent Widescreen Review Article on low frequency optimization:



We quickly see the question that was asked of you two earlier that you punted relative to seat to seat variations mattering in home theater applications where we have multiple seats. We see how changing from seat to seat changes the low frequency response massively (and hence the reason you want to pay attention to that, before you worry about all of this talk about reflections). Your ETC tool is useless in low frequency region as I am sure you agree since we don't even have reflections in the classical sense to detect with it. All of this talk about reflections therefore ignores the massive frequency response errors in low frequencies which are quite audible. The spotlight, pun intended, is incorrectly put in the upper region of this graph whereas the first thing one should do is to tame the low frequency region below transition frequency.

Importantly, in low frequency region we no longer have to concern ourselves with psychoacoustics. What you see is what you get. If you have huge peaks and valleys as any system that is not designed to deal with them has, then you are going to hear that bloated bass. And because of ringing that it represents in time domain, the detail in your music also gets masked by those low frequency overhangs. So start there. Don't waste your time messing with reflections before you have your house in order in low frequencies. My article touches on the top tools there and of course Dr, Toole and AES papers referenced in the article take you the rest of the way there. But really, you don't need to know the theory to know what to do. The steps are all concretely defined and easy to follow.
No need because he does use ETC in his book, just not on the manner you prescribe. As I mentioned, in the hands of an expert it can have value. If you actually buy and read his book you would see all of this. Not too much to ask when you are so critical of his work, is it? After all, he would throw a rock at you for suggesting to cover topics in his book that are already there .
Seems like you didn't read the part about the absorber was place on top of the the carpet that was already there. Therefor the addition of the 50 mmm panel actually made the carpet more broadband but somehow didn't have positive results. Don't look but you are realizing that ETC displays are indeed spectrum blind and lack the data to back your analysis regarding the filtering that the acoustic products perform.

A much better approach and one that saves you a ton of time and hassle is to understand which reflections are beneficial/net-neutral regardless of amount and not worry about them (side reflections are one such thing). And if you need absorption, take it out of other angles of reflection (relative to you, not the speaker). Substitute your gut feeling for this type of research and you will be making a graph pretty and not what your ear and brain like to hear.
That is no what I showed you. I showed you how ETC tool directed researchers with substantial knowledge of audio to treat the room in a way that was detrimental. That conventional forum wisdom of treating such reflections can be dead wrong. Just because you removed a spike, it doesn't mean you made things better. And that the axiom of if a little bit is good, a lot is better is fallacious when it comes to acoustics. The carpet in this instance was doing a fine job and there was no reason to throw a blanket there no matter what the graph told you.
That's right. Carpet is an excellent tool to bring adequate amount of absorption to an empty room. It covers a large space and hence is quite effective that way in making the room less live. And of course has high spouse acceptance factor.
Careful buster. You are talking to Mr. Bob here. I mean Mr. scanner. I wonder what office product Dragon will name me next.

P.S. You and have been around the block on many of these topics. So please forgive me if I don't repeat them all over again. Should you point out something that I can use to expand on a topic relative to this thread as I did here, I might respond. Otherwise, I will probably not.

Please don't worry about me for a moment. I am trying to understand how a person can be so critical of Dr. Toole's position on acoustics, yet not having read his textbook or frankly, much of his AES and ASA papers given your commentary about him. We have a very good data point here where you claimed that Dr. Toole had not covered this topic whereas he does for multiple pages. There were references in the index so it is not like it is hard to find at all. Can you please explain how we can conclude anything other than you not having read this work?
I will come to Blackbird bit later. Trust me, I will .

Thank you for the laugh! And here we have a grand display that demonstrates without a doubt that he is without a clue as to the proper application of the tool! For you see, the ETC is optimal for use in evaluating SPECULAR behavior. And what do we have below the specular region in the "low frequency region"?

OK... now everyone together! Let's help ol' MSBob out!: That's right, The MODAL region!

And what tools are optimal for displaying such behavior, everyone: The waterfall and /or cumulative spectral decay responses.

Its always nice when we are afforded the opportunity to push the distant limits of acoustical understanding just far enough so that the MSBob is able to remove his foot from his mouth.


But to movie on to still more comedic ramblings of the marketeer...

Again, you fail to read for meaning and seek only to push that which you wish to market.

YOU, and to a degree Toole, present Toole's rich reflection model (along with the rest of the REQUIRED speakers, treatment and room correction software (that only costs, was it, ~$16,000!!!) as the ONLY acceptable response model for surround sound.

While that is great as far for marketing purposes in pushing the Harmon/JBL line of speakers (of which you also fail to tell folks feature the ONLY ACCEPTABLE power response that MUST be employed with the Room topology that you say MUST be employed along with Harmon's ROOM CORRECTION SOFTWARE!!! - don't forget to tell others that they cannot properly utilize their plain old speakers with uneven power responses and uncontrolled Q!!!

So, while you attempt to bully others into accepting that your parroted approach, of which YOU understand little of what Toole refers, via the tyranny of surveys and polls which you imagine are sufficient to render other options invalid, you might want to also refer to other statements of Toole's where he acknowledges that other approaches are not only possible and a "matter of taste" and NOT some hallucinated objective fact, but also describes what others just might feel is an advantage that other approaches have.

With reference to surround sound environments he does not here advocate reflection rich environments, but instead favors a more dead environment preferring to allow the various direct surround sources to provide the sense of envelopment:

The Acoustical Design of Home Theaters
byFloyd E. Toole, Ph.D.

"Attempting to bring a little more spatial “flavor” to the reproduction process, loudspeakers are
available in directivities ranging from conventional front-firing, through bipole (bidirectional in phase),
dipole (bidirectional out-of-phase), predominantly-reflecting, to omnidirectional. These present listeners
with very different combinations of direct and reflected sounds, and in most of them the room is a major
determinant. Stereo, therefore, is not really a system at all but, rather, a basis for individual
experimentation.

Multichannel systems offer a partial solution in that there are more channels, and therefore more
directions from which sounds can be made to appear. This also brings a certain independence from room
acoustics, since there are more “real”, as opposed to “reflected” sources. Still, multi-directional, including
“dipole”, speakers rely on reflections, so there are instances in which one must attend to the details.
The standardization of at least some of the more obvious variables by the film industry has helped enormously in achieving some consistency in multichannel film sound tracks. Consequently, what one hears in a cinema is similar to what was heard in the dubbing stage where the sound track was assembled.
Good engineering practices and the Home THX program attempt to continue that relationship into the home.

It will be interesting to see how multichannel music develops."

Indeed!!!!

Yup, one must attend to the issue of room reflections - especially in the instance when anything but diffuse "di-poles" are used. Otherwise the multiple direct sources themselves provide "a certain independence from room acoustics, since there are more “real”, as opposed to “reflected” sources."
And in instance where reflections are desired, isnt it ironic that MSBob eschews entirely the ETC, the one tool capable of displaying the reflected soundfield in complete spatial and temporal characteristics critical to the design and establishment of a well behaved indirect soundfield.


Oh wait, but in the same breadth in which he eschews the ETC, he does advocate band limited ETCs, apparently in order to expose and call attention to his flawed use of deviant acoustical boundaries and specular treatments that are not properly broadband in nature. One might suggest that he, and Toole, walk their talk and actually employ porous and diffusive treatments that are sufficiently large as to be sufficiently broadband in nature. But one must admit, those puny throw rugs, thin porous absorbers and anything but sufficiently large and deep diffusors are cute as obects d'art, if not as effective broadband treatments effectively EQing and coloring the sound!

So.... which way to Oz?, Scarecrow?

And as far as which approach is "correct", you see there are many acoustical responses, and depending upon one's application and preference, regardless of what you or I might prefer, these options are still valid approaches available for others to decide upon.

"The Acoustical Design of Home Theaters
By Floyd E. Toole, Ph.D.

The real solution, for professionals as well as consumers, is loudspeakers that deliver similarly good timbral accuracy in the direct, early reflected and reverberant sound fields. This can be described as a loudspeaker with a flattish, smooth, axial frequency response, with constant directivity (which together result in flattish, smooth, sound power).

Then it becomes an option, whether the room is acoustically damped, or not. If reflected sounds are absorbed, the listener is placed in a predominantly direct sound field, making the experience more intimate, and the imaging tighter and more precise. If the reflections are allowed to add their complexity, the overall illusion is altogether more spacious and open, to many listeners, more realistic. In part, this is a matter of taste. In either case, a room-friendly loudspeaker will yield timbral accuracy. So, at middle and high frequencies, the proper solution to getting good sound quality, is to choose good loudspeakers to begin with."(p.4)

"....A matter of taste"??? But I thought that it is what we MUST prefer. That such a response is a foregone conclusion based upon the tyranny of a bunch of amorphous numbers in multiple surveys!

And now he breaks ranks and dares to admit that it is but a matter of TASTE???????????

So much for the required lockstep and a forced march to a Toole room!

You see, folks DO have valid response choices depending upon their application, be it 2 channel, hybrid music and home theater, and dedicated home theater use. But MSBob is determined that you will not have the opportunity to exercise that choice should he have HIS way.

I know, this simply chafes as he attempts to assemble a marketing approach intent on marketing the "Toole rooms" while he has so valiantly attempted to dictate what others WILL (read "MUST") prefer based upon a tyranny of cut and paste surveys and polls that no individual is dared allowed to differ. Everyone, you have heard MSBob tell you what you WILL prefer: So exercise your free will and "Get in line!"

And fortunately the book is thick enough that if he cannot simply intimidate folks into accepting his declarations of what folks WILL prefer, that he can presumably use it as a bludgeon to correct their thinking and to coerce them into preferring his foregone conclusion dependent upon "taste", to use Toole's word.

The bottom line, Toole's approach is but one of many. Some may like it. Others who prefer more accurate localization and more precise imaging and improved speech intelligibility may not.

Deal with it.

Folks have choices, both in terms of equipment and in terms of the room's response.

And the time of MSBob's marketing tyranny where he tells others what they WILL and MUST prefer is effectively over.
Edited by dragonfyr - 6/20/12 at 3:03pm