Paul's Posts — 26 July 2012
By Paul McGowan
Yesterday I modeled what would happen if you moved a perfect reproduction of a live venue to your home. It wouldn’t sound live because of the change of venue. We are quite capable of acoustically defining the space we’re in and it’s hard to fool us.
I remember once I visited an anechoic chamber and it was an eery experience to say the least. When the door to the chamber closed it felt as if all the life had been sucked out of the room. When I closed my eyes, not only could I not tell what size the room was but all I could hear was the blood pumping in my ears. For all I knew the room could have been the size of a football field. This is about the worst situation you could have for a live group or a stereo system because there are no reflections and we depend on those reflections as audible cues. Yet, this chamber changes the apparent size of the room we’re in and is a key to solving our puzzle.
And so with that in mind, the answer to yesterday’s thought problem is to change the room – with technology.
Imagine if we could simulate the room the original combo was playing in. How would that feel? I am quite convinced it would feel like you were there. But to be convincing you’d have to be able to sit in your listening position and close your eyes and believe you are in a a different sized room and hear the space around you. If you coughed or shuffled in your seat, all that you hear would have to be convincing enough to fool you.
Haven’t we heard of “concert hall” sound and tricks that fool you into believing the music’s playing in a larger venue than it is? Sure we have. Many of you are probably familiar with prior attempts at adding room characteristics into the stereo system – some Yamaha receivers did this and it was moronic. Really, you can’t add the room into the music, you have to add the room into the room. In other words, you have to create the room environment in the room and independently of the stereo system.
How would one do that? With another loudspeaker system surrounding the listener and a microphone or two placed appropriately in the room. Imagine 4 loudspeakers, one in the middle of each of the rooms 4 walls – and a microphone setup either near the loudspeakers or near the listener. Through tricky electronics and known DSP techniques, this setup could manage the reflections from the walls and provide a convincing room to anyone in it – and of course the room dimensions could be adjusted electronically as well.
Easy to do? Practical? No, but it is another view of how to overcome this age old problem.
Like they say, if you can’t bring a horse to the water, bring the water to the horse.
Forward to a friend and help us engage more readers
Related Articles
Share
About Author
(4) Readers Comments
Leave a Reply
You must be logged in to post a comment.
Soundminded
Now we are coming to what for me is by far the most interesting part of this challenge, thinking about the nature of sound itself. How it is propagated into space, how rooms affect it, how it arrives at us, and how we perceive it. This for me is where the real problem lies. It’s also where the most opportunity to make improvement is found because there’s been so little effort expended on it and what has been tried has largely failed. Yet this is where the overwhelming preponderance of what we hear when we listen to music live at a public concert venue is to be found. How could it not be very important? If this were easy it would have been solved already. It’s very hard which may be why so many in this business choose to ignore it or even pretend it doesn’t exist.
An anechoic chamber really is an eerie experience. There are no echoes of anything. But most people have experienced anechoic or nearly anechoic environments even if they didn’t think about it. Out on a golf course, in a park away from trees or any structures, out in an open field such as on a farm, an outdoor athletic field with no stadium around it, even in the desert. Every sound you hear passes you once and you will hear little or nothing of it again. Life experience teaches us to associate what we hear with what we see. When we go indoors whether it’s a normal house or office, a log cabin, a tool shed we hear pretty much what we see and have learned to expect. Some are unusual and a little surprising if we’re not accustomed to them. A concrete vault, the space under an overpass over a roadway especially an arched one made of concrete (my favorite) or a tunnel, even inside a cave. Then there are places we hear live music performed, a high school auditorium, a church, or if we’re lucky a cathedral, concert hall, or opera house. With every one of these experiences we learn to associate what we see and what we hear. As hahax pointed out yesterday, even ambient noise such as from air conditioning or heating fans, shutting a door, outside noises from cars and trucks that enter and are heard in the space creates in our minds an auditory impression of that space and an association with what we see immediately. It reinforces that connection we’ve come to expect.
And then there are spaces where this association breaks down. To our eyes an anechoic chamber looks like an ordinary sized room although it has no windows and strange looking floors, ceilings, and walls. It’s usually a fairly small room because they are so expensive to build. But they don’t sound like any other inside room. There is a psychological disconnect that is immediately obvious and it is for me a very unpleasant place to be. It’s certainly no place to listen to music live or recorded. It is however a useful place to make certain acoustic measurements because it isolates what we want to measure from the acoustics of normal places. I’ve considered that this experience is like Gulliver visiting Lilliput. His whole acoustic world has shrunk to nothing, it’s all inside his body. The only echoes are those inside his own ears. Everything outside is small. (This is what Einstein said would happen if you reach the speed of light, the universe is reduced to a single point.) It strikes me that when audiophiles install lots of sound absorbing material in their rooms this is where they’re headed. But what about the other extreme, Gulliver traveling to Brobdingnag? This is the concert hall, opera house, cathedral, cave amphitheater. Now everything is very big and it is we who are small by comparison. The sounds are powerful even from far away. The echoes are coming from everywhere, even the sound of a pin dropping can be huge. It surrounds us, sometimes to the point where the echoes are so strong we can’t even tell where the original sound is coming from. And they are persistent, they last a long time as they bounce back and forth on the hard surfaces traveling long distances. (a sound of 90 db in a typical concert hall will travel nearly half a mile before it becomes inaudible. In a cathedral it can be a mile or more.) Now what if you could recreate this convincingly in a normal sized room with an electronic sound system without turning the room into an anechoic chamber (building such a system inside an anechoic chamber would make the problem much more expensive to solve although potentially much more controllable and the equipment far more complex)? Imagine that disconnect. What a surprise. So does the sound of the room’s own acoustics affecting locally generated sounds like a cough, talking, outside noises entering, or especially the first arriving sounds from the loudspeakers where the source of sound (as opposed to the reflections) is being reproduced? My experience of 36 years with it is that it does not, at least not appreciably. What a lucky result. The effect of the reproduced space overwhelms it.
What happens when you make a close up recording of something big like a symphony orchestra and play it back in a small room with small room acoustics? You’ve started the voyage from Brobdingnag back to Lilliput. You’re well on your way. Small wonder some audiophiles play their recordings very loud. But no matter how loud they play them, even to the point of deafening themselves, they can’t break through the walls of that small room with its small room acoustics. The missing dimensions of time and space remain missing. Can you put a number, a factor by which the impact of sound is different between Brobdingnag and your Lilliput room on it? By my method of calculation based on matching the loudness and bass level but taking into consideration the difference in perceived distance to the source, the perceived difference in the volume of the room, and the persistence of the echoes, for a symphony orchestra heard in a typical concert hall it’s about 43 db, for a pipe organ or choir heard in a cathedral it’s about 50 db. Those are factors of 20,000 and 100,000 respectively. What a huge difference. What an enormous opportunity even if you get only part of it.
hahax
I believe there’s a second reason for playing loud, compensation for dynamics. Lots of speakers and systems are smooth these days but the are non-linear dynamically and the ea4r/brain recognizes this compression as not being real. Overloading the ears hides this to a degree.
And always remember to get flat bass you usually need to play loudly to eliminate Fletcher/Munson unless the recording compensates in some way.
Soundminded
At a live concert in a good concert hall, a symphony orchestra sounds like a very big source of sound. It can play very softly or at tremendous power levels without any blaring, blasting, strain, or other irritating effects. It is clear, mellow, powerful. The sound comes from the orchestra itself but it also seems to surround you too. When there is a rest or pause at the end of a phrase or note you can hear the reverberation die out and sense the immensity of the hall. The bass fiddles cellos, tubas, and trombones are deep and strong. You can feel the plucked strings of the basses and cellos. The violins shimmer and are never strident, always sweet. They are not muffled or muted either. The brass have bite but aren’t shrill. You can feel the power of the bass drum and the kettle drums. Every sound is clear and distinct. Each instrument’s sound is also well integrated with the other sounds. These characteristics are part of the music itself, they are what the composer intended, what the musicians and conductor try to create.
These qualities are largely inadequate or completely absent from even the best recordings heard on the best sound reproducing systems. The reason is clear, the technology isn’t up to it. It’s not about a speaker or amplifier not being good enough, it’s the whole idea of how to do it that doesn’t work well enough. It does not take a golden eared audiophile or concert goer to hear this difference. I think anyone with normal hearing even listening casually would know the difference immediately. Let him sit blinfolded in the hall and then bring him blindfolded to another room where a hi fi is playing a recording of the same music, tell him it’s another performance in another hall and I don’t think he’ll believe it for even one second.
acuvox
I have heard several good approaches to solving this problem.
Michael Gerzon had a magic box that translated B format Ambisonic recordings into a headphone signal with pots that controlled the altitude, azimuth, distance from source and separation angle of a virtual near-coincident pair. It felt like flying; but of course headphones all have timbral problems.
Ralph Glasgal has a remarkable room where you are literally surrounded by 18 large SoundLabs electrostatic panels in a circle, so you can’t see the walls at ear height. Each speaker has a dedicated DSP channel to process existing recordings and simulate acoustic measurements of real concert halls. This was a remarkable illusion, but only for one listener in the center.
James Johnston developed a proprietary six channel recording format at AT&T Labs. That demo was the best at reproducing a recorded space over a good portion of the floor. I can still remember hearing the sound travelling down and back the large captured space, auditioning in a small hotel room.
There is one good sounding system offered as a product to start with a relatively dead room and create synthetic acoustics up to the largest halls. It is “Concertino” from SH Acoustics, and consists of ceiling microphone array and a large number of full range speaker in the walls (16 or more), driven by an equal number of DSP channels and amplifiers. It works symmetrically for live and reproduced sources.
The large number of sources is necessary because the pinnae are directional phase encoders. Level differences (pan pots) do not produce spatiality nor does timing difference from two speakers suffice due to cross-coupling. If you want an echo to come from angle x, there needs to be a speaker there.