GIK Acoustics 244 Sound Panel Review

[gene]

Do you have a system with a parametric EQ, and a way to measure and display ringing? If so, I'd love to see some before / after graphs showing ringing being reduced by a meaningful amount. But I'd also want to see what happens a few inches away.

I do have a system in a relatively bad room which opens to my whole downstairs floor. I am using a Velodyne SMS-1 to bass manage my sub and reduce modal peaks in the primary listening area. Of course it does next to nothing for all other seats unfortunately, which is why I am going to be adding another sub (wife permitting) to that system.

Please check our review of the SMS-1 and you will see some measurements I pulled using LMS. We also did a detailed write up on how to interpret measurement anomalies that you may find interesting.

Velodyne SMS-1 Review

I plan on revisiting this issue soon once I add the second sub. I will also do higher resolution plots so you can see whats really going on.

 

[ScottMayo]

I do have a system in a relatively bad room which opens to my whole downstairs floor. I am using a Velodyne SMS-1 to bass manage my sub and reduce modal peaks in the primary listening area. Of course it does next to nothing for all other seats unfortunately, which is why I am going to be adding another sub (wife permitting) to that system.

Multiple subs can give a smoother response, but if you continue to use a managed EQ for bass, you will continue to have problems elsewhere in the room. Adding more subs doesn't fix that; the EQ will continue to adjust output for the point the microphone is at, and other points will continue to be wrong. Multiple subs can make things worse as well as better.

Ultimately, getting well controlled bass at multiple points in a room requires diffusion, epic amounts of absorption, and good, well behaved drivers. None of that is simple, and it rarely comes cheap. I think EQ is great for the single listener in a nearfield arrangement. It can be wonderful in a room that is already very well behaved. As the "big cure", though, it's a mistake. Make it the *last* thing you apply.

 

[Audioholics]

I think EQ is great for the single listener in a nearfield arrangement. It can be wonderful in a room that is already very well behaved. As the "big cure", though, it's a mistake. Make it the *last* thing you apply.

First of all, I dont recall anyone, at least not Gene, saying that EQ was his "big cure" or anything other than the last thing - so we appear to be in agreement on that one point. Secondly, adding additional subs in almost every test and measurement series we have done (and that I have seen) has broken up modes and resulted in better overall bass response for multiple seated positions. This also jives with most of the Harmon research and with recommendations from both THX and the HAA. Thirdly, adding "diffusion, epic amounts of absorption" is - as you state - not simple or cheap... and it's certainly not a recommendation for anyone who isn't a bachelor!

Far be it from us to recommedn ANY EQ as a "magic bullet". I think it is more accurate for us to caution readers that neither is randomly filling your room with 4" panels. The review we did was for the effectiveness of these panels in a room - not to endorse them as the "big cure" or end-all, be all to room acoustics problems.

Readers can reference this article (I think some others appear to have already made up their minds) for what we're talking about when we say to use multiple siubwoofers:

http://www.audioholics.com/techtips/setup/loudspeakers/SubwooferplacementP1.php

In our testing, two well-placed $1000 subs will outperform a single $2000 sub almost every time - if you care about more than 1 seat. This is just one trick you shoudl have in your arsenal to achieve better room response in multiple listening positions...

 

[HAA Acoustics]

It is usually NOT possible to sit the listeners equidistant from all speakers. Don't you agree that time alignment in the processor takes care of this problem?

No, it's geometrically impossible to time align all listeners. Time alignment simply minimizes the error. The complexity of the sound field as provided by reflections is what masks any comb filtering resulting form the error.

As for spatial averaging in the primary listening positions to look for system pathologies, Do you apply any weighing to those averages? For example, you emphasize more importance for 2 or 3 of the 6 seats for the owner who still wants their money seat?

Interestingly, a properly designed mike array for spatial averaging is a randomized placement. You know it's truly random if you rerandomize the placement and receive a similar response despite new positions. As one draws in the radius of the mikes the response becomes more and more seat specific and would potentially allow the weighting you suggest. It all depends on what the customers needs are; a one person show or a for a group. This why we stress placing listeners in acoustically similar areas. This relieves us of some of the concern for good versus bad seat measurements. If we make one area sound good (bass wise) the others should follow suit. This also reminds me of a comment from an earlier post, "move 4 " and get 15 db difference at 100 Hz". This simply points out the futility of single point measurements. we hear both the 15 db peak and the flat response at 100 Hz simultaneously because we hear/perceive over a space not a point.

On another note, how do you deal with asymmetric "L" shaped rooms where the front L speaker is near a wall, while the front R speaker is open to the L portion of the room?

Punt. I'd renegotiate the direction of the system or advise my client of the drawbacks. Without acoustical balance there are many sins that are not negotiable.

HAA

 

[HAA Acoustics]

Multiple subs can give a smoother response, but if you continue to use a managed EQ for bass, you will continue to have problems elsewhere in the room. Adding more subs doesn't fix that; the EQ will continue to adjust output for the point the microphone is at, and other points will continue to be wrong. Multiple subs can make things worse as well as better.

This is only true if one doesn't understand the nature of the problem; room modes. I can't imagine why one would place multiple listeners in acoustically dissimilar locations. I would think its easier to talk someone into a little EQ and moving their easy chair than the major league dose of fiberglass such treatment would require. Also the empirical and laboratory evidence tends to support Floyd Tooles contention that the bass response is nicely moderated over many usable locations. Also , note my comments on single point miking above.

Ultimately, getting well controlled bass at multiple points in a room requires diffusion, epic amounts of absorption, and good, well behaved drivers. None of that is simple, and it rarely comes cheap. I think EQ is great for the single listener in a nearfield arrangement. It can be wonderful in a room that is already very well behaved. As the "big cure", though, it's a mistake. Make it the *last* thing you apply.

A bass diffuser or absorber of 2 to 4 feet depth perhaps. can be effective but I've only seen one installation like that. I'd avoid epic amounts of absorption due to the epic amounts of power it would consume. At the end of the day it depends who is our customer; a hobbyist or somebody's Mom and Dad.

HAA

 

[Doug917]

I care about only one seat in my HT room, mine! Everyone else is on their own. I built my HT room for me as I have no need to impress others. I am in to HT because I love it, and it is my hobby. My wife could really care less about improvements in sound. Even when someone is in any seat in my HT room (I only have 5) they still tell me how great it sounds. No one else cares as much as I do about how things sound. It has probably been more than a year without anyone besides the wife and I watching a flick in my room.

I can tell everyone that my very reflective room with paneling has benefited more from the GIK panels than any other single improvement I have made, be it a subwoofer, processor, amp, etc. I have eleven full panels now and the room sounds great. Slap echo is gone! Channel separation is excellent. Yeah, my wife hates how it looks, but it's the one room I have total control over...so tough. Oddly enough, when it's dark, it really doesn't bother her .

I use EQ as well, but feel the panels have made a much larger difference than the EQ's (SMS-1 & Audio Control Bijou). I am very happy with the results and as I said above, that is all I am concerned with.

Some more pics. Sorry, the room is still a mess.

 

[ScottMayo]

This is only true if one doesn't understand the nature of the problem; room modes. I can't imagine why one would place multiple listeners in acoustically dissimilar locations.

Me either, but I see it fairly often. People with odd shaped rooms often hope that just slinging in another sub will solve their problems. $1000 later, they have unnatural sounding bass (less peaks and valleys, but it "sounds wrong"), an irritated wife, and a for sale sign up on audiogon. I get to explain to them that no matter how many people say that phase and direction don't matter in low frequecy sounds, bass drums just don't sound right when two subs at different distances, with complex resonance patterns, pretend to be a drum.

I would think its easier to talk someone into a little EQ and moving their easy chair than the major league dose of fiberglass such treatment would require.

Frequently! A $200 component that adjusts itself is an easier sell than improved room acoustics, which takes more work, more planning and more cash. It's not for everyone, definitely.

A bass diffuser or absorber of 2 to 4 feet depth perhaps. can be effective but I've only seen one installation like that. I'd avoid epic amounts of absorption due to the epic amounts of power it would consume.
HAA

My room is something like that - I have 2'+ of absorption in several places. Absorption (and lots of it, well placed) doesn't make a stereo seem quieter; in fact if it's done *right*, the increase in clarity and detail makes for a perceived increase in the music. People stop cranking the volume to get at the details, only to find they are feeding the room problems that mask what they are after. In terms of measurement it probably cost me a few db of energy on average, but it sure doesn't sound that way. And since a decent stereo is capable of going far louder than anyone wants in any case, I'm happy to lose a little actual energy to get the increase in detail.

As you say, it depends on the customer. My room is not a traditional HT, it's one obsessive audiophile's vision of "I will get this right". I don't employ any EQ, and I don't need it. (Which doesn't mean I won't fiddle with some someday.) Using EQ would have been cheaper overall - I spent hundreds on materials - but I get results in this room that I don't think anyone's EQ could give me.

For a family HT where the goal is to get the best shake-the-walls rumble while keeping the dialog clear, my approach is clearly overkill. EQ (properly applied - as mentioned, the single-mike stuff audiophiles play with is just not useful for large listening areas) is the way to go in such settings. But if you want great HT and world class music reproduction (most of my customers to date have been muscians), EQ should be a small part of the equation. Taking a sledgehammer to the signal in the hopes of counterbalancing the sledgehammer effect of untrammeled room resonance, shouldn't be any perfectionist's final answer.

 

[ScottMayo]

...
Some more pics. Sorry, the room is still a mess.

Can't comment on the mess, but I have to say, the music/movie collection looks very respectable to me.

 

[HAA Acoustics]

bass drums just don't sound right when two subs at different distances, with complex resonance patterns, pretend to be a drum.

Again, for any consumers out there (thats why I'm here); if properly calibrated, the bass generated from multiple subs should integrate with the stereo images generated by the main speakers. Proper calibration of a sub includes smooth response, proper level and a smooth transition through the crossover. This integration is possible despite odd sub locations if it's done right even pleasing the most demanding audiophile. Doing it right though is the key requiring flexibility on speaker/listener position by the client and technical knowledge and instrumentation by the calibrator.

In terms of measurement it probably cost me a few db of energy on average

3 dB of lost sound energy is easily replaced by doubling the amount of power in the amp. For the typical music lover this is not a problem since so much of what we listen to is dynamic contrast (transients) versus dynamic range. However, for home theater and even audiophile systems designed to recreate live sound levels 3 dB is a very high price to pay (100 watts needs to be 200 watts and for a sub 1000 watts needs to be 2000 watts). It's important to note that with speakers of average sensitivity (86 to 90 dB), even without treatment, a power level of 500 watts is required to hit 0 dBFS (max theatrical level). While not everyone needs this capability, a well calibrated system is truly breathtaking and quite palitable when called upon to reach these heights; no sense of harsh loudness just impact. Audiophile quality systems can and many do sound great at low levels without this capability. They depend on dynamic contrast to create impact but then become very nonlinear and poor sounding if driven to a sustained higher level (as is common with much music and most movies). The cost factor of reaching true reference level dynamics with smooth response is the cost of admission. Losing 3 dB (typically much more than that for tons of absorption) provides a smoother bass response but at a sizable cost to dynamics, particularily if live sound levels or movie reference sound is desired.

EQ should be a small part of the equation. Taking a sledgehammer to the signal in the hopes of counterbalancing the sledgehammer effect of untrammeled room resonance, shouldn't be any perfectionist's final answer.

EQ is the last and in many cases the least effective means to smooth response. A properly designed system will employ more elegant means to the smooth response end before employing EQ. Many very expensive rooms have been designed over the years in quest of smoothing response with out resorting to EQ. Much of the work was focused on studio design and systems where there was a generous budget. Obviously the object is to make a recording nuetral of the monitoring enviroment. RPG acoustics is a splendid example of this science as well. For the typical consumer though, the concepts of design flexibility and WAF do not go together. Add to that the breathtaking advancements in DSP parametric equalization (no I'm not a big fan of auto EQ) and many of the problems and failures associated with EQ are gone. As a long term audiophile who has physically thrown many EQs out the door. I am very pleased with a host of EQ products for solving many (not all) of the everymans theater issues without resorting to the mega-acoustics solution... especially if the customer won't buy a second sub or allow it out of the corner.

HAA

 

[Ethan Winer]

Doug,

> my very reflective room with paneling has benefited more from the GIK panels than any other single improvement I have made, be it a subwoofer, processor, amp, etc. I have eleven full panels now and the room sounds great. Slap echo is gone! Channel separation is excellent ... I use EQ as well, but feel the panels have made a much larger difference <

Exactly.

So where's Glenn - he should be basking in this praise.

--Ethan

 

[Ethan Winer]

Folks,

A few more random comments and points:

> the loss of extra reflections reduces the complexity of the sound field and exposes audible comb filtering ... This "comb filtering" is irrevocable because any path difference between two or more sources causes it; there's no cure ... Reflections do create comb filtering but they also add complexity to the sound field and act to mask the more egregious comb filtering of the multiple speakers <

I understand what you're saying about comb filtering due to path length and arrival time differences. But this is mostly mitigated by the fact that each ear hears a very different response, so the sum of what's heard is not as skewed as you'd think looking at the response at each ear individually. And I can say absolutely for certain that imaging and response problems caused by early reflection comb filtering is far more damaging. I cannot imagine a situation where adding absorption at the first reflection points will not make things much better, let alone make things worse.

> I do have a system in a relatively bad room which opens to my whole downstairs floor. I am using a Velodyne SMS-1 to bass manage my sub and reduce modal peaks in the primary listening area. <

But do you have a way to display ringing, and be able to show the results of EQ reducing that ringing? To me this is the crux of the matter.

> Far be it from us to recommend ANY EQ as a "magic bullet". I think it is more accurate for us to caution readers that neither is randomly filling your room with 4" panels. <

Glenn and I are not suggesting that consumers should randomly fill their room with panels! Any more than you are suggesting they should insert an EQ into the chain and "randomly" set the knobs. There are specific guildelines for properly treating a room, and any vendor who knows what they're doing will advise people where to place them.

> This also reminds me of a comment from an earlier post, "move 4 " and get 15 db difference at 100 Hz". This simply points out the futility of single point measurements. we hear both the 15 db peak and the flat response at 100 Hz simultaneously because we hear/perceive over a space not a point. <

Yes, the fact that our ears are 6-7 inches apart means that one ear often hears what the other is missing. As I just explained a few paragraphs above. My comment was in response to Gene's statement, "where wavelengths are very long (about 22 feet at 50 Hz), a few inches make little difference unless one is near a narrow interference dip or null."

> the empirical and laboratory evidence tends to support Floyd Tooles contention that the bass response is nicely moderated over many usable locations <

Yes, but ringing is still present, and ringing is at least as damaging as a skewed response.

> I'd avoid epic amounts of absorption due to the epic amounts of power it would consume <

This is mostly a myth, at least for low frequencies. In most cases adding bass traps to a room increases the perceived level of bass. This is because so much energy is lost to nulls from destructive interference. But I agree that at higher frequecies adding a lot of absorption reduces the overall volume somewhat.

> EQ is the last and in many cases the least effective means to smooth response. <

That's all I've been trying to say here all along!

--Ethan

 

[Glenn Kuras]

Doug,

> my very reflective room with paneling has benefited more from the GIK panels than any other single improvement I have made, be it a subwoofer, processor, amp, etc. I have eleven full panels now and the room sounds great. Slap echo is gone! Channel separation is excellent ... I use EQ as well, but feel the panels have made a much larger difference <

Exactly.

So where's Glenn - he should be basking in this praise.

--Ethan

AAAAAAAAAAAAAAAAAAAAAH I need a personal assistant like NOW!!! The move is driving me crazy. Sorry guys I have been gone for a few days but need the day to be 30 hours to get everything done.
Doug thanks for posting the pictures. You rock sir.


Glenn

 

[Doug917]

Yes, but ringing is still present, and ringing is at least as damaging as a skewed response.

This is where I think this thread has gone really wrong and I am in agreement with Ethan here. To me, as long as each speaker sounds similar to one another, the system will sound very good and pans and so forth will be quite smooth. The speakers do not all have to have a totally flat frequency response.

I had health issues and had to recently get a hearing test. Not a school-nurse type, but a close you in an acoustically treated room and give you a series of tones across the bottom end of the spectrum ~20-3000Hz. After seeing the results of this test I was in awe. My hearing when graphed looks like mountain range with differences of -15 to -20db in places. I know other people would have similar results especially as we get older.

My point; if all speakers are adjusted to a totally flat response using RTAs and testing equipment I still won't be hearing what the director intended due to the fluctuations in my own hearing. Sure, I could adjust things to my hearing, but then when someone else comes into the room, it may sound wrong to them because their hearing is different.

Skewed response is fine as long as it is skewed across all speakers. Ringing, slap echos, and reflections need adressing as they actually distort the sound a person is hearing. Even if the speakers all have the same flat response, but are prone to reflections the sound will be muddied. The acoustic panels treat these issues quite well.

I am not an acoustics expert, just a practical HT enthusiast. This is my opinion. If you are in doubt of what I am saying go see an ENT and get a professional hearing test and see if your hearing has a flat response.

 

[HAA Acoustics]

that imaging and response problems caused by early reflection comb filtering is far more damaging.

Yes that is what I used to think, but I was always haunted by high-end two channel systems I installed without side wall treatment that sounded pretty darn good. I also had been fascinated by AES articles relating how speech intelligibility could actually be improved by these reflections. Yikes!

National Research Council in Canada recently released a multi-year study showing how lateral reflections operate and based upon this and many other academic studies of listener perception of reflections (yes Haas included) Dr. Floyd Toole (including several of his own) has officially asserted the dependence of successful mirror point treatment on the off-axis response of speakers. Translation: Mirror point absorption improves the sound of speakers with poor off-axis response but reduces sound quality and envelopment (soundstage width detrimentally) for speakers with good off-axis characteristics. It's 2006 we should only be buying these good speakers. The difference between a good speaker of yesteryear and today in many cases is in it's off axis characteristics. If one only uses older speakers than you should hide those reflections, otherwise trade up.

Reflections also, most assuredly reduce the audibility of comb filtering due to time alignment errors. Listening to two speakers then moving in and out of the sweet spot is all the evidence you need of the audibility of this comb filtering. All speakers sound much better in a room with reflections than in an anechoic room or even out of doors if you don't have one handy. Certain reflections can be a very good thing; lateral ones especially. The key is understanding what are good versus bad reflections (another time perhaps; many researchers are still decoding that). How many folks prefer their speakers outside with no reflections... raise your hand.

At the HAA we conduct classes in calibration. During one of the exercises a team of calibrators experiments with adding then removing side wall treatment (we used to just tell them to put it there). Many are dumbfounded by the fact that for a good pair of speakers they prefer the walls blank and a little better with diffusion (but not too much).

Floyd's AES paper is due out soon, he references the dozens of studies from which he draws his conclusions. Hopefully, we can put together a synopsis of the discussion since AES will likely not want it released directly. For those who are AES members I recommend reviewing the scientific literature on the subject it's quite interesting. I'll be teaching the Level I HAA Seminar at RPG Acoustics next month, I've asked Peter D'Antonio to say a few words about this exciting new research. It means that as acoustical advisor's, we have many more options in a regular room.

But do you have a way to display ringing, and be able to show the results of EQ reducing that ringing? To me this is the crux of the matter.

This is not up for debate. I haven't read a reputable scientific discussion of room acoustics in some time that does not acknowledge that a properly tuned parametric EQ reduces the resonance (ringing) of modes in a small room.

Yes, but ringing is still present, and ringing is at least as damaging as a skewed response.

Most studies use impulse response measurements which are energy versus time; thus including ringing. The basis of Dr. Tooles's research is on reducing ringing. The use of spatial averaging is designed to expose ringing which is indeed very audible but as a response distortion. Ringing causes skewed response. No one is suggesting that acoustical treatment is not a vital part of well designed system. It's simply important that we recognize the progression of science and our understanding of human sonic perception.

> I'd avoid epic amounts of absorption due to the epic amounts of power it would consume <

This is mostly a myth, at least for low frequencies. In most cases adding bass traps to a room "increases" the perceived level of bass. This is because so much energy is lost to nulls from destructive interference. But I agree that at higher frequencies adding a lot of absorption reduces the overall volume somewhat.

I have the happy experience of dealing with many different rooms on the job. There is no mistaking an over damped room to the degree of several dB at midrange frequencies. As to bass, I prefer a tuned trap (Helmholtz, diaphragmatic absorber etc.)which focuses on specific frequencies more than a broad band absorber that lowers bass level indiscriminately. In a high end theater every bit of power is precious. The perception of increased bass caused by absorption is because the treatment lowers the level of midrange energy making the bass (we are less sensitive to) appear louder. This is a well documented phenomena we discuss in class.

> EQ is the last and in many cases the least effective means to smooth response. <

That's all I've been trying to say here all along!

I left that out there assuming that someone would jump on it out of context. I believe as industry professionals we should keep up to date on the world of science. 10 years ago I would have been right there with you discussing EQ and bass traps. We owe it to our clients to agree on certain things that have been shown as more cost effective and practical solutions. Bass traps are a very good thing but they shouldn't be sold as the only thing. I'd hope once a scientist had proven something in the forum of scientific debate the discussion would benefit from it. Lets work together to make better sound. After all, many of the HAA Dealers would be good resellers of a well made product such as this.

HAA

 

[HAA Acoustics]

This is where I think this thread has gone really wrong and I am in agreement with Ethan here. To me, as long as each speaker sounds similar to one another, the system will sound very good and pans and so forth will be quite smooth. The speakers do not all have to have a totally flat frequency response.

Your right! Flat frequency response is not the issue. A smooth response is paramount and a truly flat response would not be to most folks liking. The real truth in what you say is that each speaker sounds similar to one another. I'll add to that, that the off axis response of each speaker (that's what we hear from reflections mostly) is also similar.

HAA

 

[MarkS]

RE: Low Frequency Effectiveness of 4" Absorption Panels



Beranek, Keele, et. al. show that modeling a surface (umm, let's say a 4" thick panel such as that discussed) as a collection of co-planar, massless plates (upon which sound impinges), each connected to a mass (freestanding) that is, in turn, connected to a damper is a means by which
to model the acoustic energy-dissipating behavior of said panel. Back of the envelope, 1st order approximation kind of stuff.

The high frequency (where resistive absorption occurs) performance of panels, such as that discussed in this thread, is generally well known, intuitively easy to grasp and frequently documented, whereas the low frequency (LF) (reactive) performance of the same panel is often ignored or otherwise buried in the charts & graphs. Conventional thinking seems to focus on the resistive component of a panel's boundary impedance, ignoring the reactive component.

For diverging spherical sound waves, sound sources located at a distance greater than the wavelength (λ) of interest, the acoustic impedance, Z, is largely resistive (real) and for sources located a distance much smaller than λ, Z is complex and largely reactive (i.e. the acoustic pressure & particle velocity are not always in phase). Referring again to the panel model mentioned in the first paragraph, at high frequencies, the mass is essentially immobile and the energy is absorbed by the damper. At low frequencies, the entire mechanism moves, providing a mass reactance to the impinging wave.

Having said that, the greater the match (at any frequency of interest) between boundary mechanical impedance and the specific acoustic impedance of the incoming sound wave, the more effective said boundary is in reducing the reflection coefficient.

So, can you get LF (resistive) absorption out of a 4" panel? No, but you can get mass reactance! How much LF-taming you can get in practice owing to mass reactance depends on a number of factors or variables. In practical terms, the results are typically minimal to vanishingly small (when compared to the high frequency absorption capabilities of the panel in question) and often difficult to predict. Another words, a 4” panel really isn’t a very effective device for the dissipation of low frequency acoustic energy - even factoring in its mass reactance. This is In contrast to, say, a Helmholtz resonator, an old, well-proven technology that behaves in very predictable ways when dealing with LF acoustic energy

To illustrate an approach that fully exploits mechanical HF resistance & LF reactance, I've attached a photo of an anechoic chamber (Bruel & Kjaer) referred to as an "acoustic jungle" that features not the usual wedges, but a huge collection of contrivances that are purpose built to maximize the matching of boundary and acoustic impedance at all frequencies. Effectively exploiting both resistive and reactive behavior, there is no where to be seen the usual monster-size foam wedges. Dr. techn. Per V. Brüel written quite a bit about the effectiveness of the "acoustic jungle" approach, something that's more popular in Europe (particularly in Germany) than here.

If you're interested in reading more, try these resources:

J. Wang and B. Cai, "Calculation of Free-Field Deviation in an Anechoic Chamber," J.
Acoust. Soc. Am., vol. 85(3), pp. 1206-1202 (1989 Mar.).

D. B. Keele, Jr., "Low-Frequency Loudspeaker Assessment by Nearfield Sound-
Pressure Measurement," J. Audio Eng. Soc., vol. 22, pp. 154 - 162 (1974 Apr.).

Industrial Acoustics Company, "Anechoic Rooms," Bulletin AN-33RI, (Bronx, NY,
USA, 1962).

L. E. Kinsler and A. R. Frey, Fundamentals of Acoustics (Wiley, New York, 1962).

W. Koidan and G. R. Hruska, "Acoustical Properties of the National Bureau of
Standards Anechoic Chamber, "J. Acoust. Soc. Am., vol. 64(2), pp. 508-516 (1978 Aug.).

E. Skudrzyk, The Foundations of Acoustics (Springer-Verlag, New York, 1971).

P. M. Morse and K. U. Ingard, Theoretical Acoustics (Princeton University Press,
New Jersey, 1968).

L. L. Beranek, Acoustics (American Institute of Physics, New York, 1986).

D. B. Keele, Jr., "Anechoic Chamber Walls: Should They be Resistive or Reactive at Low Frequencies?" Audio Eng. Soc., Preprint 3572 (G2-2) (March 1993) Berlin
Germany

G. Rasmussen, "Anechoic Sound Chambers," Bruel & Kjaer design document (1972
Sept.).

C. Fog, "Anechoic Chamber at Bruel & Kjaer," Proceedings of the Nordic Acoustical
Meeting at Aelborg, Denmark, pp. 285-288 (1986 Aug.).

 

[Savant]

I realize I'm jumping in here late. Gene also asked me to chime in and I wanted to thank him for the invite! This post caught my eye and I thought I'd comment on it and some of the other stuff that's been discussed:

The only way I know to measure absorption reliably below 100 Hz is with software like ETF, by comparing the ringing in a room with and without traps in place.

If the purpose is to quantify how much a device absorbs - in a general sense - then there is at least one standardized test. It is possible to use the ASTM C384 test, provided the impendence "tube" (really a chamber when measuring for LF) is large enough. A drawback to this test is that it will only measure normal incidence absorption as opposed to the random incidence absorption that is commonly published for acoustical devices. What I believe is a modified version of this technique has been used by RPG in their research at the Fraunhofer Institute. The way I understand it, they excite a reverb chamber with the 1/3-octave center frequencies at low frequencies and measure differences in decay at various points around the room. I couldn't find documentation of the technique on RPG's site, but it's another approach more manufacturers could take to help quantify general performance of a device at low frequencies. FWIW, testing an absorber with ETF in a bedroom will only reveal what that one device does in that situation. YMMV and all that. With the extant objective, quantitative approaches, as well as the new approaches being developed by folks like RPG, there should be no reason for "guessing." As Gerry pointed out - it's 2006.

Very few acoustic labs are certified to measure below 100 Hz, and I make that clear on the RealTraps web site. Even at 125 Hz, where acoustic labs are certified, they can vary as much as 50 percent.

What varies by 50%? If the same sample is measured the same way in the same lab, I doubt you'd see that much variation. On the other hand, that much variation is not out of the question when the same samples are measured in different labs, possible arranged differently (or not). But 50% of what? If the sample yields 0.10 in one lab, a 50% difference would mean either 0.05 or 0.15 in the other lab. None of the three values (at 125 Hz) are what would be considered "high." On the other other hand, if the sample in one lab measures 0.90 and either 0.45 or 1.35 in the other lab, that would be noteworthy. If this comment could be put into some context, it would be more useful, perhaps. Any data available to help clarify?

Part of the reason is because labs use a microphone that is always moving during the tests (for averaging). While this works well at midrange frequencies, it is inadequate at low frequencies where the lab's own room modes begin to dominate. This is why even at 125 Hz labs vary a lot.

No, quite the opposite. The moving microphone technique is in the standard precisely to help minimize the effects of the room resonances. It is mostly because of the moving mic that low statistical deviations are possible for frequencies like 125 Hz. If the microphone were stationary, the results would not be statistically meaningful for low frequencies. With the moving microphone, uncertainty is typically low for the entire frequency range of the test (typically 100-5000 Hz). In fact, the measurement uncertainty is low enough that (a) the lab has been certified to test at that frequency to begin with and (b) the results are trustworthy. I will grant that different labs can produce different result, particularly for the low bands. This is unfortunate, but that's where I believe other methods could come into play to help quantify LF behavior for everyone equally well.

Why don't they just place a number of static mics, say in a grid, and average them for the mid frequencies, but use static results for the low frequencies?

I realize that you'd been misled into asking this question. However, I will mention that the ASTM C423 standard allows for either a moving microphone, or a random array of a minimum number of stationary microphones. Either way, many measurements must be made to ensure measurement uncertainty is minimized.

More recently...

I had health issues and had to recently get a hearing test. Not a school-nurse type, but a close you in an acoustically treated room and give you a series of tones across the bottom end of the spectrum ~20-3000Hz. After seeing the results of this test I was in awe. My hearing when graphed looks like mountain range with differences of -15 to -20db in places. I know other people would have similar results especially as we get older.

I don't think this should alarm you all that much. I've had at least half a dozen hearing screenings in my life and without exception, they all used pure tones. Thus, the huge differences you're seeing are for single frequencies. I suspect (thought I'm not an audiologist) that this is part of the reason why "dips" of up to 15-20 dB are considered within the "normal" range for adults. It's when you start to see the HF roll-off (usually on the order of 30+ dB at 2 kHz and above) that you need to start thinking about hearing aids.

That said, I do see you're point. Individual hearing differences are often overlooked in things like HT design.

************

To close for tonight (it's late here), I will say that I pretty much agree with what Gerry has said about HT design below 80 Hz. I have had the wonderful opportunity in my life to acoustically treat several rooms down into the 20-30 Hz range. It's fun...and expensive...and space-consuming. This is simply not a reality for many people. This makes the recent advances in "EQ" (I use quotes since some of the systems aren't really equalization in the true sense) all the more exciting...

As for multiple subs, did anyone else notice that "modal variation is virtually eliminated" with 5000 subs?! That's always been my favorite part of that paper. I'm sure Ethan and Glen can show that 5000 bass traps also virtually eliminates modal variation!

 

[HAA Acoustics]

Great Post! I haven't seen the Bruel & Kjaer stuff. This is a proper post including references. Bravo!

HAA

 

[Buckle-meister]

...testing an absorber with ETF in a bedroom will only reveal what that one device does in that situation.

Not sure if you're referring to my results in the adjacent thread. Assuming you are, I understand the above but thought it'd be helpful at worst to have included untreated/treated room responses since readers would have a real world example of the difference RealTraps make instead of just the results from a reverberation testing chamber. Your Mileage May Vary, but I'd counter that with More Product Data Is Always Beneficial (made that up myself ).

Oh, and it's my living room.

 

[Savant]

Buckle,

Not sure if you're referring to my results in the adjacent thread. Assuming you are, I understand the above but thought it'd be helpful at worst to have included untreated/treated room responses since readers would have a real world example of the difference RealTraps make instead of just the results from a reverberation testing chamber. Your Mileage May Vary, but I'd counter that with More Product Data Is Always Beneficial (made that up myself ).

Oh, and it's my living room.

I wasn't referring to any room/person/test in particular. I am delighted that folks are taking charge of their rooms and understanding - from a subjective and objective standpoint - what is happening to sound that is produced or reproduced in them. However, I have to disagree with the "More Product Data" comment. I am of the opinion that too much data - even for something like acoustical treatment products...especially for something like acoustical treatment products - is not a good thing.

Data, like pumpkin pie, should be consumed in moderation to appreciate its goodness.