NetNews Usenet Archive 1992 #31/NN_1992_31.iso/spool/rec/audio/17334

Internet Message Format | 1992-12-25 | 7.2 KB

Path: sparky!uunet!zaphod.mps.ohio-state.edu!saimiri.primate.wisc.edu!ames!biosci!ucselx!crash!cmkrnl!jeh
From: [e-mail]
Newsgroups: rec.audio
Subject: Re: Carver and the FAQ
Message-ID: <[e-mail]>
Date: 25 Dec 92 09:50:04 GMT
References: <[e-mail]> <[e-mail]>
Organization: Kernel Mode Consulting, San Diego, CA
Lines: 120

In article <[e-mail]>,
 [e-mail] (Allan Armstrong) writes:
> [...]
> And then there's "Sonic Holography." In college this was considered the
> ultimate piece of gear to have if you were going to take acid and play
> Dark Side of the Moon past the clipping threshold of your amplifier. I
> actually never heard sonic holography. Can anyone report on what it
> really is? How does it actually work?
> [...]

Carver's "Sonic Holography" is an imaging-enhancement mechanism.

Suppose your system reproduces a sound that is supposed to be heard as coming
from only the left speaker.  Your left ear hears it first, and then (because
your right ear is a bit farther from the speaker) your right ear hears it also.
So far this is all demonstrable fact.

The theory behind "Sonic Holography" says that hearing the sound in both ears
muddles the soundstage.

What the Sonic Holography circuit does is to take the left channel signal,
delay it by about the inter-ear time as described above, invert it, and add it
to the right channel.  (And, obviously, vice versa.)  This inverted, delayed
signal reaches your right ear at the same time as the original signal from the
left speaker, and so as far as the right ear is concerned, the sound is
cancelled.  So only the left ear hears the sound, so the imaging is better.
Carver's user manual (at least the one I had, for the C-1 preamp) claims that
imaging will be better localized and that the soundstage will extend beyond the
speakers on either side and will also have obvious depth.

You may also recognize this as the theory behind Polk's "Stereo Dimensional
Array" speakers of a few years ago.  Polk did it more directly, with an extra
midrange driver per side (and in the highest-priced models, an extra tweeter
as well), connected to the "other side" and out of phase.  I'm not sure how (or
if) Polk introduced any delay.

There are several theoretical problems with this approach.

- Cancelling of sounds with alike-but-inverse-polarity sounds doesn't work all
that well if the sound sources (the speakers) are separated from each other,
and/or from the listener, by more than a small fraction of the wavelength of
the sound.  (Try it.  Leave your speakers in their usual positions, reverse
the connections to one of them, and then set your FM tuner for mono.  You will
hear markedly less bass than before but the levels of the mids and highs won't
be much affected.  The closer the speakers are to each other, the higher the
frequencies that will be effectively cancelled.)

- in the example above, note that the left ear will hear not only the original
signal but also (one head-width later) the cancelling signal from the right
speaker.  If hearing the delayed, normal-polarity signal in the "opposite ear"
messes up your localization of the sound, then hearing a delayed, reverse-
polarity signal in the "same ear" should mess it up also, probably more so.

- When we hear sounds in nature, obviously both ears hear each sound, and
there is no "cancelling" signal to help us out.

The standard rebuttal from the Sonic Hologram defenders is that stereo
reproduction is not even close to how we hear sounds in real life anyway.  And
this point is absolutely correct.  It seems to me that the absolute best that a
stereo system can do is to reproduce the sound you would hear in a room
adjacent to the one where live music is being played, with two speaker-sized
cutouts in the common wall.  Would you expect to hear a realistic sound stage
(similar to the one you would hear without the wall) from this arrangement?  Of
course not! It's a lot better than you'd get from just one hole, of course, but
it doesn't come close to the real thing.  So, since stereo is already playing
tricks (like exaggerating the level differences between the channels) to fool
us into thinking there's a soundstage, it should be okay to play a few more
tricks so that we can be "fooled better".

However, my ears and brain must like to be fooled in this way.  I found the
Sonic Hologram circuit to be very effective at doing "something" to the
perceived soundstage.  The effect varied widely with different recordings
(which is only to be expected).  But at no time did it produce what I would
consider a realistic soundstage.

For example, in a simple imaging test where a sound is panned from one speaker
to the other, the Sonic Hologram made the "image" better defined, but it
wandered all over the place when it was supposed to simply move from left to
right and back again.  It's a widely accepted tenet that if a system can't
pass this simple test it isn't going to image worth a damn on real program
material.

But, one might well say, that simple test is nothing like real music; the
instruments and vocalists tend to stay put; we don't need to be able to track a
"panned" source.  True enough.

However, in my experience, the Sonic Hologram circuit had problems with real
music.  What I heard (mostly from recordings of reasonably complex orchestral
pieces, some miked with "minimalist" techniques, others with lots of mikes)
was, in general, an improved soundstage, but very often a particular instrument
would jump out and grab me by the throat:  "Hey, listen to ME!  I'm over HERE!
Isn't that AMAZING!!!"

Well, yes, it's amazing, but it's nothing like real life.  And ultimately it
became prety tiresome, like a once-funny story that you've heard too many
times. There's no question that it will amaze your friends, though, so if
that's what you buy stereo gear for, go for it.

(I had the same reaction to Hughes's Sound Retrieval System.)

One other point:  The Carver units with Sonic Hologram that I've seen all had a
two-position switch to control the level of image enhancement, something like
"normal" and "wide".  As you might expect, the "enhanced" setting made the
"listen to ME!" effect more pronounced.  I've often wondered what this circuit
would sound like if I could set it for somewhat less enhancement than Carver's
"normal" setting.  It might be much better.  Might not:  The Hughes SRS unit I
tried has a continuously-variable control for the degree of image enhancement,
and I couldn't find a setting that wasn't fatiguing to listen to, but this is
of course dependent on my speakers, room, ear lobes, etc., etc., etc.

My conclusion is that these units, and all other units that are attempting to
improve imaging, are very much subject to personal preferences, the other
equipment with which they're used, the room, etc. -- maybe more so than even
speakers. If you're interested in one of these, the obvious thing to do is to
get one on a money-back-if-not-satisfied-in-30-days arrangement, and then
really test the thing in your own system on music you listen to a lot.

    --- Jamie Hanrahan, Kernel Mode Consulting, San Diego CA
Internet:  [e-mail], [e-mail], or [e-mail]
Uucp:  ...{crash,eisner,uunet}!cmkrnl!jeh