[Gary Eickmeier]

THE PROBLEM WITH STEREO



I want to take you on a time travel trip to an alternate present, a present
in which even more mistakes are made than have been made now, in 2016. What
mistakes? Ride with me:



Suppose that architectural acousticians were even slower than audio
engineers in catching on to what causes good sound. We go into a new concert
hall that they are designing and observe them treating the walls for great
sound. They reason that what we need to hear is just the direct sound from
the instruments because anything else bouncing off the walls all over the
place would muddy things up and dilute the imaging and everything else. They
line the front wall, the one behind the players, with Sonex or similar, to
keep the reflected sound from coming back to the audience. Similarly with
the side walls, sound absorbing materials all over the place - especially
those first reflection points, so that it will, after all, sound just as
good as our stereo recordings of them.



TROUBLEMAKERS



Things would have stayed that way except that a few troublemakers had gone
to Europe and listened in some halls that had not been treated. The sound
had a certain "spaciousness" or width to it, seeming to come from much wider
than the orchestra itself. And the tonality of the instruments! They didn't
realize that the violins and cellos had such a warmth and musicality to
them. And the percussion! It sounded a lot more important there, with all
of those reflections. The team came back and reported to the American
acousticians that maybe we DO need to hear all of those reflections that we
have been dampening and controlling. Reluctantly, the acousticians try it,
taking down all of the sound killing materials until we once again get back
to the way we know it today.



In the concert halls, anyway.



STEREO



I get it now, said one acoustician. What we should be doing is building
sound fields within the concert hall, not just the direct sound from the
instruments. If we could label these fields, we could call them the direct
sound, the early reflections, and the full reverberant field. This is all
really quite important and just the opposite of what we had been doing. Now
we can hear the full sound power put out by the instruments in all
directions, and the sound doesn't trail off so abruptly as you go back in
the hall. This is what we should have been doing all along - building sound
fields, rather than just the direct sound for the audience's ears.



"But what about stereo reproduction?" one of them asked. Our current
practice is just the direct sound from the speakers, with all reflected
energy dampened away with Sonex or clever room shapes. We have been told to
make "reflection free zones" for the primary direct sound and don't let
anything get past those first reflection points. "But this is just the
opposite from what we have learned about the concert halls" blurts one
acoustician.



Maybe we should think about this.



Yes - this principle has proven very important for the production of music,
but what about the reproduction? Why would it be any different?



Oh, that's easy. According to time honored principles, we have already
recorded those qualities of the concert hall that make good sound. Now all
we need to do is play it back and we will hear it.



Are you saying that our ears work differently for stereo than they do for
live music? The huge, complex set of direct and reflected sound fields for
live music sounds the same as the two high direct fields from two points in
space for stereo?



Well, yes, at least for the area between the speakers.



That's the best thinking for stereo reproduction today? We have just
discovered that the most important factor in the concert hall is the
building of the various sound fields within the room, but in stereo we're
going to use just the direct sound from two speakers? These two could not
sound the same.



Maybe someone from our discipline should get with the audio engineers and
explain about the spatial nature of sound - that it is very audible and must
be addressed in the reproduction just as it is for live sound or it will
sound different. That is, if they are really interested in reproducing all
characteristics of live sound.



Gary Eickmeier