Room Reflections & Human Adaptation for Small Room Acoustics

[Gary Eickmeier]

Floyd - what you and everyone else fail to consider is how to USE the reflected sound as part of the speaker system and thus eliminate it as a nuisance variable. You also never think of the possibility of using MORE reflected than direct sound in the radiation pattern in order to engineer that image shift into the total frontal soundstage. Imagine if you will a speaker like this placed in a rectangular room with specular reflectivity at the speaker end in order to ensure that the first reflections have the same freq response as the actual speakers. You will place them 1/4 of the room width in from the side walls and the same amount out from the front wall. What you now have is a matrix of direct and reflected (virtual) speakers that is twice as wide and deep as the actual speakers and positioned perfectly evenly from each other. The soundstage is now formed by a summing localization between the actual and the virtual speakers that forms behind the plane of the speakers and as wide as the room itself. The perception is of a huge, spacious, deep soundstage that can actually decode the direct and reflected sounds that were recorded and that has no extra early reflections because you are using them as part of the whole model. I am doing this at my home and the results are sensational. Most aspects of this design are just the opposite of historical practice, so I offer it as something that no engineers would ever think of on their own.

Outside the Box,
Gary Eickmeier

 

[DD66000]

Floyd - what you and everyone else fail to consider is how to USE the reflected sound as part of the speaker system and thus eliminate it as a nuisance variable. You also never think of the possibility of using MORE reflected than direct sound in the radiation pattern in order to engineer that image shift into the total frontal soundstage. Imagine if you will a speaker like this placed in a rectangular room with specular reflectivity at the speaker end in order to ensure that the first reflections have the same freq response as the actual speakers. You will place them 1/4 of the room width in from the side walls and the same amount out from the front wall. What you now have is a matrix of direct and reflected (virtual) speakers that is twice as wide and deep as the actual speakers and positioned perfectly evenly from each other. The soundstage is now formed by a summing localization between the actual and the virtual speakers that forms behind the plane of the speakers and as wide as the room itself. The perception is of a huge, spacious, deep soundstage that can actually decode the direct and reflected sounds that were recorded and that has no extra early reflections because you are using them as part of the whole model. I am doing this at my home and the results are sensational. Most aspects of this design are just the opposite of historical practice, so I offer it as something that no engineers would ever think of on their own.

Outside the Box,
Gary Eickmeier

Not everyone. The room I designed/built in the '90's had such a layout... the room was 17.5' x 20'.
I had the mains set 57" in from the side walls from center of driver arrays. Or approximately 1/4. They were >48" out from the end wall. I'm thinking it was around 52".
Depending on the recording being played, that soundstage would sound even wider than the room. And in all cases a huge soundstage, in width and depth.

As one reviewer said many years ago, "speakers need room inside the box, and outside the box."
The few exceptions would be horn loaded speakers designed to be placed in the corners or near the side walls, as in the case of the Everest series.

 

[Floyd Toole]

Hi Gary, glad to see that you are still promoting your philosophy. I really don't disagree with you. As I say at the end of the piece: "For optimum stereo listening if your music tastes are as eclectic as mine, one really needs adjustable acoustics and, possibly, variable-directivity loudspeakers, but we know that won’t happen."
The article started with my old "classical" listening room that was outfitted with horizontally omnidirectional loudspeakers - lots of reflections - and I was listening 20ish feet away - not a lot of direct sound in proportion. So I think I did not "fail to consider how to USE the reflected sound". My head is not in the sand (it sounds terrible . For classical and some kinds of pop music it was a splendid experience. But it required a large room. Your approach is a better fit for a small room.

However, if you consider the "circle of confusion" - namely that if we want to hear the art as created, both the mixers and the listeners should have closely similar listening situations. You have found something that suits your preferences and listening habits. Good. If you can find a market for it, even better. The 901 had the basic idea - it just didn't sound great.

The fundamental problem is stereo itself - it is a spatially and directionally deprived medium. We need multichannel music so we don't have to play silly games.

Inside the room,
Floyd

 

[Gary Eickmeier]

Hi Floyd -

I will have to disagree in several areas, if you will bear with me.

Number 1 and most importantly, there are not different speakers for different musical tastes. I can play all genres on my system and there are no recordings that would sound better on some other directional scheme. Nor is it a matter of philosophy, but rather science. That is why I called my IMT (Image Model Theory) a whole new theory for stereophonic sound. I know you have studied the spatial nature of live sound, how the early reflected sound comes from a much broader and very different directions from the direct sound, and yet historical stereo theory has us reproducing ALL of it from those two points in space.

Number 2, how to use the reflected sound. I have said that there should be a greater amount of radiation in the reflecting direction and the speakers need to be positioned for the proper use of that reflected sound. Most of the illustrations in your book and also above in your post have the speakers too close to the walls and corners of your lab rooms. Everyone just thinks that the reflected sound is an unavoidable but undesirable artifact of speakers. I am trying to tell everyone that it is an essential part of reconstructing the spatial characteristics in the recording.

Number 3, the "circle of confusion" - most of this confusion lies at the lap of speaker design and stereo theory, which, if it were correct from the outset would have given us a much greater insight into all aspects of recording and reproduction. In my universe there are not some speakers for recording engineers and some for audiophile enjoyment. There is but one correct approach and it is the one that takes into account all sound fields present in the live situation. Most of the engineers and audiophiles think that the idea is to get the sound that went into the microphones straight to your ears with a quality called "accuracy" with no distortions, nonlinearities, or room effects interfering, and then the stereo will form in your brain. What should have been the standard is the image model of the playback acoustical situation to that of the live sound and how to reconstruct sound fields in rooms, not how to get waveforms to ears. This is so basic, so fundamental, and the loss of it so tragic I could scream.

Number 4, no, the 901 did not sound bad, it started a revolution in the industry that just didn't take, because Bose did not include speaker positioning as part of the deal, probably to lead us to think that we could put the speakers just anywhere that decor dictated. Well, I wrote to them a long letter, visited the factory, talked with Dr. Bose, to no avail. But it was Joe Veranth, his chief engineer, who put me onto the technique of image modeling to illustrate the reflected sound - how to show where it is coming from with the speakers placed in various positions. So I specify my 1/4 rule in the theory because if you place a multidirectional speaker too close to the walls you get a clustering of images that pull the sound toward them instead of spreading it evenly across the soundstage. This causes a stretching of soloists, sometimes called a hole in the middle or wandering images.

Number last - thanks for staying the course - I have seen a lot of crazy ideas on how to build speakers and how to do spatial audio in some papers I have read lately, and why stereo is so failed a medium, none of which even scratch the surface without an understanding of IMT. We are not recording and reproducing the perception of sound, we are recording and reproducing the sound itself, reconstructing sound fields within our rooms, not signals for the ears. The function of a loudspeaker is not to be a direct radiator but rather to be an Image Model Projector.

No longer a theory, doing it in my home, paper at next NY convention
Gary Eickmeier

 

[Floyd Toole]

Hi Gary,

I am not your foe. I have been criticized for excessively "liking" reflections, hence my little article.

Your approach goes farther, requiring very specific loudspeakers and room setup. You say that it is a universal solution, but the hoards of recording engineers who gravitate to essentially dead rooms, direct sound only, to create their art would disagree.

I don't know, but suspect, that you listen to a lot of classical repertoire when you say: "There is but one correct approach and it is the one that takes into account all sound fields present in the live situation." For most of the recordings created the only "live" event is the final mix in front of a pair of loudspeakers in a substantially reflection free room. Listening to pop, today and yesterday - it hasn't changed - one hears mono left, mono right, double-mono phantom center, and not much else. A bit of "reverb" sweetens the mix and it is done. Adding reflections during playback of such tracks is not a bad thing. And, as I emphasized in my article, I employed a large reflective space and omnidirectional loudspeakers to advantage for classics.

At a time when video displays are approaching the "peel and stick" (on a wall) configuration, people are looking for similar loudspeaker solutions. Few people are willing to devote the necessary square footage of floor space needed for a truly "reflective" solution, and the prominence of a pair of speakers puts this more into the category of a dedicated listening space. But that is a marketing issue, not a technical one.

I guess that is why I continue to have faith in multichannel solutions. They have the huge advantage of delivering the reflected sounds of the live experience with something resembling the large delays of the performance space - tens, even hundreds, of milliseconds from different directions. A while back I heard two musical selections played on Auro 3D, one of the "immersive" formats, in this case created by a music lover, not specifically for over-the-top blockbuster movies, as Atmos was. The symphonic performance in a concert hall and the organ concerto in a cathedral, were spine-chillingly realistic, and the illusions held up as one walked around the room. This is a truly dedicated listening room experience, but with elaborate home theaters still being constructed, it is an option for the well-heeled audiophile.

How many channels? More than two, even with reflections. The point of diminishing returns sets in at a tolerable channel count.

The Bose 901? I am looking at my anechoic measurements on it right now. It sounded bad because it used transducers that had gross audible problems. It is asking a lot for a single small paper cone to do 20 Hz to 20 kHz. Today's transducers would sound better, no doubt. When we reviewed it for one of the Canadian magazines we experimented with listening to it much farther out into the room - even then sensing the recommended placement was too close to the boundaries (marketing again, I suspect). The spatial aspect improved, but the sound quality did not.

The future in multidirectional loudspeakers is - or should be - in active arrays. They exist. Then sounds can be steered where you want them, and more importantly, the pattern can be changed. I still don't think one size fits all

Cheers,
Floyd

 

[Gary Eickmeier]

Floyd -

No, you are not my foe and I hope I didn't get too rambunctious in my enthusiasm for this stuff. What you are is one of the most knowledgeable and influential audio luminaries, I have corresponded with you several times in the past, read your book(s) and hope become friends. So to me talking to you is like a chance to speak to the industry at large and express some of my concerns.

I wish I could have heard JJ's Auro 3D - I have read about that before, and would like to see exactly what he is doing with his 7 channel microphone and especially the playback end. I can't get anything out of him.

I am in central Florida, Lakeland, if you ever get over to this side of the country.

Gary Eickmeier

 

[Floyd Toole]

Hi Gary,

Auro-3D is the brainchild of Wilfrid Van Baalen, a Belgian, and is owned or has shared ownership with Barco. Wilfrid owns Galaxy studios, near Antwerp. He is dedicated to reproducing music in all its spatial splendor, and from the small sample I heard, he has succeeded remarkably well. Yes, it does immersive movies too, but that is the easy part.

If you mean JJ (Johnson) you are talking about a different product. I know nothing about it.

Cheers,

Floyd

 

[Gary Eickmeier]

Organs must be a popular subject for surround sound researchers.

Gary

 

[Danzilla31]

Thanks nathan_h. Audio purists have long adhered to the belief that music was the "perfectionist" content, and movies, and home theaters, were somehow inherently inferior. Rubbish! It was movies that motivated the development of stereo, and then multichannel. The music world would have been content with mono. I am old enough to remember when stereo was introduced that a lot of music lovers said, basically, "who needs it?". It is clear in my last AES paper that movies can stand some improvement, but they are much more uniform than music, and they are multichannel. Movie mixers have long ago learned how to work with multichannel and some excerpts of music in movies are quite impressively good. In contrast, some of the efforts of the music side have been, to me, colossal failures because they were mixed by "stereo" engineers. When a multichannel music recording appears and there is no center channel . . . gimme a break! It has happened on some "audiophile" recordings. Play the test signals and all channels are present. Play the music and nothing is coming from the center speaker. Unfortunately this also happens in some otherwise entertaining music videos.

Sound is sound, good sound is good sound, and then there is content: I would not argue that the latest superhero movie is equivalent to the great classical, jazz or pop music - they are just different. Each one benefits from respectful treatment in the playback process. However, the fact that the music domain has been stuck in the stereo groove for decades is, to me, evidence that direction and space in music has lost out to rhythm, melody and timbre, all of which are audible in mono, or "headphone stereo" which is not real stereo. Too bad . . .

These days it is possible to have a multichannel home theater that can do it all, and do it at levels of excellence that would satisfy the fussiest of listeners. Just replace the crummy surround speakers with good ones (not dipoles, please!) and try it. All loudspeakers in a surround system need to be comparably good from a timbral perspective, and, if it is affordable, all the same.

I totally agree with you and am so grateful that you've brought up the point of using good quality surround speakers in the listening environment.

This has been my experience both in my old multi channel theater/music room in my former home that was set up in the living room and the new one that is in my new home set up in a dedicated theater room area.

I just became convinced through trial and error that in both rooms very capable surrounds and if possible identical speakers in each position were beneficial to both my music and theater experiences. I believe long gone are the days where surrounds and the content placed in the mixing tracks for these channels can be decided as superficial.

In fact I don't know if I'm crazy but my ears actually prefer when the surrounds are capable enough to have all channel stereo when when listening to music in both rooms. Although I still haven't figured out why this is.

Thank you for such a wonderful insightful article I really gleaned a lot of useful information from it.

 

[DBlank]

Having dealt with lots of different materials for absorption of reflections, it's important to understand how the absorption coefficient between 125hz and 500hz is critical and how most common materials like compressed fiberglass, mineral wool tend to over absorb in that region. It's also a good thing to make aware that not only are those materials not designed for music and speech acoustic treatment, nor are many of them a good healthy product to have exposed to the people in the room because they can have tiny particles that become airborne and since it's harmful to one's health, these types of materials should only be used as their intended usage which is contained within a wall and used for insulation against hot or cold instead of acoustic treatment. Yes, I have found an acoustic foam that is specifically designed for acoustic treatment with regards to music and speech and this is by far the best I've found to date. It can be purchased in either raw panels or it can be be placed into a wooden frame and covered with fabric if one so desires.

Another issue is with low frequency absorption. There are only three types of devices that can really absorb low frequencies. When I talk about low frequencies, I'm talking about 100hz and below. These frequencies have to be absorbed differently than mid's and high's. Traditional limp mass, fiberglass, mineral wool, acoustic foam, simply isn't going to be effective. so one needs to use either a Diaphragmatic device, Membrane (similar to diaphramatic, yet not as efficient or effective) or Helmholtz resonator. Helmholtz resonators are not typically used because one typically needs many of these of different sizes, but they have to be carefully designed and built and they take up a lot of room which most of us do not have. Yes, some are using subwoofers and trying to use those, but you needs many of them and they don't necessarily provide enough coverage and may not be as effective as a high quality diaphragmatic. Low frequencies go everywhere in the room and putting subwoofers may not be enough coverage and they take a short period of time to react to the incoming signal where they simply might not be as effective. One also should measure the room to figure out what frequencies are giving the biggest problem, where within the room they are located and getting properly designed tuned cabinets to go after those frequencies and place them in the room where those modes exist. Also, it may be required to have additional diaphragmatic devices that are broadband to place in between tuned cabinets should there be a large problem. Typically speaking, it can take at least 10 or more of these diaphragmatic units around the perimeter of the room to be effective, but if you use the right product and place them in the right places, the results are truly amazing.

Then there is diffusion. Obviously quadratic is the best and one can get them in either horizontal or vertical or both and they can be purchased to handle different frequency ranges, but you have to figure out which ones would be appropriate for your room, and there should be a fair amount of distance between the listener and the diffusors so they are effective. But you just can't stick them wherever you want, careful placement has to be considered.

One thing that bothers me is that people that are recommending products should have experience using many different products on the market. I don't believe either of the people in this video have done this and I think they should. using traditional materials such as compressed fiberglass is a mistake and I've made that mistake before and will not make that same mistake twice. Mineral wool, rock wool are also on my list of materials to not use either. These MMVF's shouldn't be considered if they are exposed inside the room, they can release small tiny particles that become airborne, inhaled and they can cause lung diseases. So, careful with recommending these materials, and covering them with fabric isn't going to necessarily going to fix the issues with them.

 

[DBlank]

Other things to consider are the size of the room, the size and number of speakers used within the room. It's common for people to have rooms that can't handle large speakers or many speakers because the room dimensions/volume. Also ceiling height. The lower the ceiling the more problems one is going to have, especially in the low frequency area and with soundstage. If you have 8 ft ceilings, which most people do, it's not conducive for a larger stereo or surround sound system. Even 9 ft ceilings are going to have a problem.

I know there is a big push for people to buy Atmos systems, but the problem is with the room size. Is your room big enough to handle the extra full range sound sources? One should consider the room's usage. A home theater, 2 channel listening room, control room for a studio, etc. have different things to consider and treatment might change depending on the type of usage of the room. Also, what works in one room size doesn't necessarily work in another room of a different size, so be careful not to walk into that trap either.

 

[Ole]

Hi Floyd,

did you ever listen to Apple HomePods (stereo pair) and if so: What is your opinion on that approach? I would probably call it "computational audio" in analogy to "computational photography" that is happening right now.

Cheers,
Ole