About Us

"watch king" wrote in message
...
Actually Ben I've had the "luck" of hearing a few things reproduced
so absolutely realisticly that they were able to totally fool people
into the belief of reality. The sets of these experiences can only be
counted on one hand and 2 fingers would still be left over, but they
never sound "boringly real". I'm not sure that everyone would care to
go through all the trouble and the incredible expense needed to
reproduce these absolute reality experiences from recordings but they
are possible.

First I need to preface this by saying that the possibility of
achieving true musical or vocal realism in a recorded situation has
nothing to do with the parts of this thread dealing with live vs
recorded music per se. If I liked listening to Uzbek folk music,
symphonic music utilizing orchestras of never less than 100 players,
acoustic rock by long deceased performers and secret forbidden
religious ceremonial chants then likely I would most often need to
listen to prerecorded music. There are hundreds of other reasons why
live music isn't a reasonable possibility for many people not the
least of which are the problems that people in some places would have
even trying to get to a place that performed works like The Symphony
of Thousand or Verdi's Requiem. So while there may be nothing that
beats a live and un-sound-reinforced concert by the 3 Tenors from a
front row seat in some grand opera house, hearing this live just
isn't a reasonable possibility for millions of people. But it
actually is possible to make sound equipment and recordings of almost
anything (including your mother's voice) that can sound so realistic,
it can fool anyone into believing that the sound they are hearing is
in fact real and even the directional components are correct. This
experience is not at all "boringly real" but it is a great shock to
the system and gave me goosebumps the size of walnuts each time I was
lucky enough to experience the event.

As well, we aren't considering the fact that some artists want to add
special effects to recordings for the sake of their art. This is
their privledge and with these kinds of recordings there is no way to
reproduce the recording so that it can ever sound like something
real, live or natural. We have to get our definitions straight here.
So let's use a clear example. This example would suppose that you
attend a school event where a group of 5 or 6 students including a
close relative (child, brother/sister, cousin etc.) is singing on one
side of, but very near the front of a stage, say stage left from
where you sit rather towards the center of the first row of seats and
on stage right we have some incredible singing artist like Linda
Ronstadt who has rehearsed the children and is singing with them, but
importantly as well in the rear of this concert venue on the balcony
Harry Connick Jr. is playing piano accompaniment with vocal back-up.
The moment is beautiful, everyone is keeping perfect time, all the
voices come together perfectly and lucky person that you are, you
have a very good video and audio recording of the event.

The video is actually unimportant except that without it you might
need to close your eyes while listening to the audio recording,
because we will assume you hired professional recording engineers
familiar with the special recording techniques and comb-filter
processing required to make these "All Reality Recordings" (ARR is
the name I will use for the process I can briefly describe later).
Since the first demonstrations of this type of recording makeable at
will (in other words any time anywhere) (that I heard) utilized
headphones for the demos, we will assume that later when you play
this recording back you will wear headphones while watching the video
on as large a projection system as possible (to promote visual
realism). I assure you it is possible for you to be totally fooled
into being certain that what you are listening to is the same live
performance you heard the first time in person. In fact the acoustic
envirnment you WERE in would seem to be exactly there again.
Considering this example we can already see that this recreation of
total reality sound recording is limited in many ways but not always
the way you'd think.

It is possible to produce these kinds of recordings of any source
that will not have any further information added to its content in
post production. Electronic instruments are not really more difficult
to record and reproduce this way except that for the most part you
end up listening to what came out of the loudspeaker in the artist's
guitar cabinets or organ box. There is no editing allowed and no
after-the-fact special effects. Some performers can do this kind of
thing and most can't. And there are dozens of reasons why nobody
bothers to make these kinds of recordings or playback systems,
although expense isn't the biggest factor preventing them from being
made, it is considerable.

From 1973 until 1979 I worked for many audio companies and eventually
specialized in the loudspeaker end of the business. When I worked for
Disney as an Imagineer developing loudspeaker systems for EPCOT and
especially for the French Theater at EPCOT, no expense was spared to
try to figure out how to make recorded and played-back sound as
absolutely realistic as possible. Almost by accident a few recordings
were made that seemed to point in the right direction and a few
loudspeaker systems were developed that seemed to have total reality
playback potential. When anyone hears this kind of playback it
staggers you. And it is often the little sounds in life that are the
shockers giving you big goosebumps. The wind blowing through the
trees is an example. There is so much high frequency energy and so
many phase and time relationships that have to be kept controlled to
reproduce such sound that just showing the "specifications" of the
content of the recorded material would require a dozen books. And
always when we few lucky Imagineers heard little snippets of this
sound it was a revelation. As well the recording environment always
seemed to be completely recreated as well so the sound of the room
you were in seemed to disappear. Another very interesting effect in
either mono or stereo was that if you walked towards the sound source
the voice or instrument just seemed to sound like it was nearer, not
really louder, until at a certain point you felt sure that there was
just some invisible barrier right in front of you and if you only put
your arm through it you might actually touch the person or instrument
making the words or music.

The problem was that while we could make these kinds of recordings
once in a while on very special digital recorders sampling up in the
80-100khz range, using special microphones made by Bruel & Kjaer, and
then played back through special test loudspeaker systems that might
take hours to tweak correctly, the results were inconsistant. In an
area of perhaps 30 square feet, on one recording, in one room, this
sense of total reality could be achievable. It wasn't just imaging,
we joked that it was "acoustical miraging". And since EPCOT had to be
built, we workers got back to building all the sound systems needed
for all the shows in EPCOT. While the French Theater at EPCOT has one
of the most realistic large scale sound systems ever made for movie
playback, it wasn't a tenth as realistic as some of the playback the
Imagineers heard back in the lab. The French theater movie audio
track is a recording of the live sounds of the French countryside
recorded from a very soundless balloon. This was why as much reality
as possible was a goal for this venue's sound system. I had the good
fortune to be the Imigineer who designed the sound system for the
French theater and the Imagineer who developed the various
loudpseaker systems that all the Imagineers used in their own venues.
Thus the French theater sound system design could be tailored by me
as much as any major sound system could be, for exactly the job it
had to do.

At nearly the same time as I was an Imagineer working on EPCOT and
Tokyo Disneyland a man named Gary Georgi was making recordings that
would recreate this special reality every time they were played back
through headphones or a very few loudspeakers. He was very interested
in my loudspeaker work for Disney and he attended the AES
presentations I made on the topic. He did a demonstration for me
which has become legend amongsyt audio engineers who heard it is the
early 80s. He had a recording made of a person talking while
wandering around a room mostly in front of the "test subject
listener" but once in a while walking all the way around the
"listener" (actually the recording position of the "listener' was
recreated by processing). There was an interesting acoustic source
used as part of the recording. It was the striking and flaring of a
match. This sound is so unique and soft that any acoustic smearing
will make it sound totally unrealistic. This sound was used near the
end of recording to capture the listener's attention and then came
the shocker as the recorded performer had moved to a position, which
when played back sounded exactly like he was whispering to the
listener from an inch or two away from, and behind, the demonstration
listener's right ear. As all the listeners had done before me I
jumped because of the natural reaction to having someone move so
obviously and unexpectedly inside "my space" where I was unprotected.

Gary wanted to have loudspeakers made that would reproduce the same
effect as headphones could produce (or nearly so) so that the
recording technique and processing he had figured out could be used
in demos without headphones. You see there is a peculiar phenomena
whereby most of the sound impinging on the eardrum has been
reflected....by the ear itself. Our ears are asymmetrical. The comb
filter effect of our ear cavity surfaces and lobes has been well
documented now and so even with single speaker headphones not only is
left/right information easy to determine but front/back and
above/below information can be clearly represented on a single
recording so that it is possible to produce an acoustic image that is
so realistic that it begins to cross the line between reality and
obvious recording. The best binaural recordings using dummy heads
with dummy ears shaped like real ears showed what the different comb
filter effects looked like from all directions. I had been able to
quantify further many of the "other" critieria for the entire
throughput system to be able to recreate a realistic signal including
the dynamic range of the amplifier and speaker, response curve of the
speaker in all directions, bandpass of each component, noise floor,
amp current delivery capability and slew rate, square wave response
(phase), nearfield listening with long delay farfield reverb
components and a dozen other criteria in order of importance.
Eventually Gary got the speakers he needed and he made some
demonstrations with those speakers.

I heard many demos on these loudspeakers using non-processed program
materials and I can definitely say that they had as fine of image
recreation (imaging) as any loudspeakers I have ever heard. There
were some very interesting features about the speakers. The cabinets
were almost totally inert because the secondary sound emissions from
loudspeaker cabinets can easily ruin the "mirage". The speakers also
had a very high ratio of driven (speaker) surface to non-driven
(cabinet, frame, connector etc.) surface (seems the driven surfaces
need to be about 12%+ of the total surface area of the loudspeaker
enclosure/system, assuming the total of all cabinet acoustic energy
emissions in all directions, is down at least 17 db from the output
of the moving drivers) . The front and back of the drivers was firmly
mounted and if possible the low frequency speakers were used in
matched pairs in each cabinet (back to back) so that the movement of
the cabinet from what is known as reactive-opposite force is minimal
if not totally eliminated. As much as it wouldn't seem possible, a
bass loudspeaker cone moving back and forth can easily move a speaker
cabinet "forth and back", thus moving the sound source and muddling
the sound (notwithstanding tip-toes or other partial restraints). Al
Bodine of Bodine Soundrive a vibrational expert of mythical
proportions has commented that these loudspeaker enclosures had the
highest modulus frequency of any he had ever tested (up into the
80-120khz range) thus any of the sound energy the cabinet transferred
into the air was so far outside the audible range that it didn't
affect the acoustic image at all. There were so many atypical and
unusual design factors incorporated into these loudspeakers that most
manufacturers turned down the chance to make them under license.

Gary Georgi was also well known at the time as one of the founders of
Mobile Fidelity Sound Labs recordings, and as the distributor for
Satin Moving coil phono cartridges and step up silver wire
transformers along with many other products. He was reputable and
respected. The recent Stereophile discussion about MFSL recordings of
acoustic and electronic music (like Pink Floyd) point out that Gary's
passion was absolute realism in sound recording and reproduction.
Imperfect soul that he was, he was losing his high frequency acuity
by the time I met him and unfortunately he eventually suffered a
heart attack about 2 years after I first met him in the 80s. He never
really had the chance to popularize the method of processing
recordings so that the original sound of the instrument and the
recording room could be brought to listeners in it's purest form.
Even if he had (because Bruel & Kjaer did eventually produce papers
on the comb filtering and low distortions levels needed) 99% of all
loudspeakers ever made would just trash the recordings anyway. Yes,
loudspeakers can be highly distorted devices (up to 10% combined
distortions at 90db average level is considered quite transparent by
our ear/brain combo) but it is often the phase, moving image source,
rise time, dynamic range and the interference of other spurious but
closely related cabinet noises that ruin the reality of an acoustic
image. Those who are interested can ask further questions in this
thread but some of the smallest details about what I've written here,
require volumes to properly cover. But to again answer Ben's
question, yes I've heard some program materials through a very few
loudspeakers and headphones that sound like absolute reality,
although it could never be described as boring. WK

We don't get enough sound in our glass

(Ben Hoadley) wrote in message
...
I have been listenig to a lot of high end equipment lately and I have
come to the opinion that a lot of it is fundamentally flawed. When I
hear things they sound very nice but they dont sound realistic. What I
mean is often a real instrument or someone singing in a room sounds
kinda boring and hi-fi manufacturers make things sound nicer than they
really are.
For example go and record a single voice with a good mic without
compression,eq at a distance that avoids the proximty effect. then
play it back on your beautiful hifi. It will probably sound good but
it won't sound exacty like you. I think with instruments its even more
obvious. things take on a lush full quality rather than the raw sound
of some guy playing over there.
I know the manufacturers give out all those specs about distortion etc
but I think we all take them with a grain of salt deep down.
Maybe its good that things sound better than real. I'd be interested
to hear other peoples opinions on this. Has anyone heard some
equipment that sounds "boringly real"?

Very interesting account. Thank you.

You're comments about the wind in particular caught my attention. While I
use mostly music for evaluating potential new or different audio gear, I
also use as a test disk a sampler disk put out by Crown for their SASS
microphone. In addition to musical excerpts, the disk contains excerpts
including wind a pine forest, a babbling brook, surf crashing on a rocky
shore, a butane torch being lit, a bicycle pump being used, a bowling ball
hit pins (mic above), a fireworks display, a community swimming pool, a dirt
bike pass-by, and an Indy 500 pit-row recording. Most of these clips
readily show up high frequency anonamalies...I found them particularly
revealing in showing differences in transparency and high frequency
correctness. And among them, the wind in the trees, the babbling brook, and
the surf crashing on shore, all of which seem to have a non-coherent
high-frequency energy level, proved the most difficult. It takes accurate
reproduction and extreme transparency for them to sound "real".

Thanks again for the story.

Thread Tools
Show Printable Version
Display Modes
Switch to Linear Mode Switch to Hybrid Mode Threaded Mode

Menu

About Us