In article ,
"Gary Eickmeier" wrote:
Audio_Empire wrote:
In article ,
"Gary Eickmeier" wrote:
Oh boy. A lot here to respond to. Is it OK if I edit a little and cut to the
chase?
It might survive Dick Pierce, but it doesn't survive Audio_Empire.
Presently, the very best way of doing what you postulate, above is via
a system devised by a North Carolina computer company called Zenph.
They take a (commercial) recording of an artist and run it through a
computer program which registers each piano note played, analyzes it
for pitch, amplitude duration, and probably a number of other
parameters. The computer then turns that information into piano
keystrokes that mimic the information gleaned from the original
performance. They have then built a device that can exactly replicate
those keystrokes when placed in front of a piano (sort of like the old
vorsetzer from the late 19th, early 20th Century). They took this
player replicator to Toronto Canada to the very studio where the
recording they used was made by Gould in 1955. Not only was the studio
exactly the same, but the piano they used was the same piano as well.
They then had the piano tuned to exactly the same state of tune as
Gould's was on the recording dates. Then they invited people from the
Gould fan club to come and hear the "re-performance" of their hero's
most famous recording, the 'Goldberg Variations' by Bach. After the
performance (which was hailed by the assembled acolytes) they recorded
the performance using modern mikes for Sony SACD. I have that SACD and
I also have the original 1955 mono recording of that performance and
I have to say that, ignoring the sonic attributes of either, I'd have
to say while the performances are very close, they are not exactly the
same.
Yes, that is what I was talking about. I have the Oscar Peterson CD of the
recording of his performance.
I don't agree with Amar Bose and others here (I have always thought
Dr. Bose was a bit of a quack and snake-oil salesman, and his products
have always been a triumph of marketing over substance). Bose might
have a point if recordings were designed to bring a performance into
one's listening room. They're not. At best, they are designed to
transport us, the listeners, to the venue where the performance took
place, and at worse they are designed just to allow the listener to
merely hear all the instruments without any thought of a soundstage or
any venue ambience. The microphone, correctly used, captures the space
that the instrument or ensemble occupies. If all you want is the
instrument sound. you might as well Frap (use contact mikes) all the
instruments and be done with it. I've heard that done... I don't want
to hear it again!
This is a gross misread of the Bose research and everything I have said
about the philosophy of what it is we are doing with the process. It was
William Snow, one of those Bell Labs researchers, who said that the object
of the binaural system is to transport the listener to the original space,
and of the stereophonic system to transport the sound to the listener's
room.
I use the terms interchangeably, I guess, and I shouldn't do that. To me
a stereo system is supposed to open a window onto a live performance.
You asked in your original post in the previous thread, why doesn't it
sound like they are right there in your room? My statement is that it is
indeed possible for it to sound like they are right there in front of you,
while at the same time getting your room to sound more like the recorded
space to give it more of the "flavor" of the recorded acoustic. That is a
big pill to swallow, but I'm afraid it is the best we can do, because of the
central recording problem, and because we simply cannot make a room sound
larger by playing a recording of a larger room within it.
First of all the system has to sound like music. With Bose 901s, you
have one strike against you right there. 901s sound terrible (to me)
they have no highs, the bass is muddy and slow (with the equalizer.
Without the equalizer they have NO bass) and that reflected sound off
the wall is, to me incredibly annoying and unrealistic.
Is there something in this allegory that can be applied to the
general situation, and maybe show a path to some improvement in our
playback systems? We saw that if we could model the playback spatial
qualities after the real acoustic event, it would sound more real,
and that if we paid no attention to those characteristics or didn't
know about them or beleive them, then it would emphatically sound
different from the original.
Your conclusion is faulty because it's based on facts not in evidence.
We do not see any evidence that proves to us that if we could model
the playback spatial qualities of the instruments after the real
acoustic event that it would result in more realistic sound from our
stereo systems. It's quite a leap in logic to go from an example of
someone like Oscar Peterson playing on a Disklavier and then playing
that Disklavier on the same piano in the same space and sounding
identical to the real performance, and then somehow transferring that
(mostly) truism to a performance in a living room playing a commercial
recording. IOW, your Oscar Peterson example proves only that Oscar
Peterson sounds like Oscar Peterson when playing the same piano that
he played before, Whether he plays in person or via an accurate player
piano device. It does not prove or even infer that this experience is
transferrable to your ideas about sound reproduction.
If you do not agree that spatial characteristics of sound are audible, then
I am at a loss how to proceed.
I'm not saying that they aren't audible. I'm saying that recreating them
artificially is not possible.
Even if we could prove that what you postulate is true, it's simply
not doable. Every recording is different and each brings it's own set
of acoustic parameters to the table. Short of using some technology
that embeds digital information about the acoustic signature of the
recording being played in that recording. And having done that,
playing said recording and data back through some system that can,
somehow, physically rearrange the room and speakers to conform to
those embedded parameters, I don't see how it could even be
implemented.
So what is your solution to all that? You have the same problem I do, but
have not any answer. You certainly can't reproduce those qualities by
ignoring the problem and playing everything with a direct field.
You listen to the stereo system that you like and be happy that you can
come THAT close. Frankly, I'm pretty happy with what we CAN do. I wish
the industry would catch-up with me, but they have gone in another
direction entirely! Properly recorded stereo sounds magical to me, I
can turn out the lights (with a proper recording) and point to each and
every instrument in the ensemble with pin-point accuracy. I can hear the
highest highs (that my old ears can respond to) and the lowest lows. The
midrange is very realistic and distortion-free. I'm content with that
because I know what's possible and what's impossible.
Gary, all of that research has been done. In the 1930's Bell Labs did
experiments with up to THIRTY channels of sound (they couldn't record
30 channels in those days, so they had musicians in a soundproof room
with 30 microphones feeding 30 speakers in another room). They'd move
speakers and change the room around a thousand different ways trying
to get what we would call a holographic sonic image of the musicians
playing in that other room. Out of these experiments came the final
realization (after continuing to remove mikes and speakers one at a
time) that two channels and two microphones and two speakers were
actually ideal for stereo.
No, not "ideal," just the most practical approximation for commercial use.
And yes, as I have mentioned, the pioneers knew some things about stereo
that we have since forgotten.
Gary, don't you think that If the Bose's concept of "direct-reflecting
Sound" had any real merit beyond its marketing appeal, that it would
have gained some traction in the industry? By that I mean don't you
think that some others company would have, in the ensuing years,
copied Bose's principle to make similar, competing products? Yet in
the almost 45 years since the 901's first broke cover, no one has ever
tried to replicate Dr. Bose's product or use any of his concepts in
other products.
BTW have you ever heard the MBL Radialstahler 101 or the X-Treme?
These speakers (as closely as possible with current technology) mimic
the physics ideal of the perfect loudspeaker being a pulsating sphere.
The MBLs are (mostly) omnidirectional and yet they throw pinpoint
images. They are among the best loudspeakers I''ve ever heard. But
then so are the latest Wilson Alexandrias as well and they are just a
set of cones in boxes. And the M-L CLX is the most transparent
loudspeaker I've ever heard. There are lots of paths to audio nirvana
it seems. All of them out of my reach...... 8^)
Audio_Empire
OK, first you poo poo the Bose concept of a combination of direct and
reflected sound, then you point out the best reproduction you have heard,
from the omnidirectional MBL! Why do you think the MBL sounds the way it
does? And no, it has nothing to do with "perfect pulsating spheres." The
Bose research project disproved that one, and I don't even know where that
concept came from, but it is bunk with no research behind it.
Actually it is true the perfect loudspeaker would be a pulsating (truly
omnidirectional) sphere and it would also (here's the difficult part) be
infinitely small. This is a mathematical model, and it can't be
researched as such except on a theoretical level.
I do not understand your saying that you think Wilson speakers sound like
MBLs.
Well, I can understand your confusion since I never said that. I said
that both the MBLs and the Wilson Alexandria XLF were AMONG the best
speakers I've ever heard (obviously, for different reasons). I also said
that the Martin-Logan CLXs are the most transparent speakers I've ever
heard and those are the one's I'd choose if money were no object. No
speaker does everything well, so this speaker might impress me in one
way, and that speaker might impress me another way.
BTW, MBL 101s only work in giant rooms. In the average 14 X 18 living
room they don't work at all.
The Martin Logans are, again, dipolar multidirectional speakers that
"splash sound all around" much like 901s and MBLs and have an equal
radiation to the rear as to the front. Those factors are what makes speakers
sound the way they do, so I propose that in essence you are agreeing with
me!
Gary Eickmeier
---
news://freenews.netfront.net/ - complaints:
---