Carl Valle wrote:

I took some advice from Howard.
Rather than make a long winded post, I'll compress here...
Howard said I could get better imaging with a center channel.
I tried this several ways and have not been able to get a better image. I
was able to raise the sound pressure, but the image just collapsed into
mono.

Whether this works with two-channel source material depends
upon the quality of the steering circuitry. Generally, you
will get decent results with Dolby Surround by backing the
center level off about 3 dB below the Dolby reference level.
If you do not back off the level the soundstage will often
collapse toward the center. This is not the case with
advanced steering systems that outfits like Lexicon offer
with their processors, nor will it occur with a well set up
Dolby Pro Logic II (music) arrangement.

Also, it nearly always works at its best when the left and
right speakers are somewhat further apart than typical.

Finally, it will vary from recording to recording
considerably.

The sweet spot did indeed disappear, and from the sweet spot position, the
center channel was clearly defined. This even when it was operating at -20db
(the lowest relative level that can be set on my receiver).

At -20dB down the center channel would essentially not be
audible at all from a typical listening location. In that
case, the center feed would be missing and all you would
hear would be the left and right side of the soundstage,
with any part of the L+R signals pulled out of them. The
effect would be weird.

From off axis
positions, the closest speaker became predominate.

I wrote about this kind of situation once for The Sensible
Sound.
Here is an edited excerpt of several sections from that
column:

To make my evaluations, I used both musical source material
and a series of fixed-image, pink-noise panning signals
available on the Delos "Surround Spectacular" test disc set,
DE-3179. This is a two-disc set with stereo and matrixed
surround test signals on one of them and musical excerpts
recorded by John Eargle on the other. The test sequences
were engineered by David Ranada.

There are actually two versions of this test on the disc,
with one being configured for checking out the center
steering with Dolby matrixed program material and the other
configured for checking out the soundstage spread,
smoothness, and imaging with conventional two-channel
recordings. The balance for each is somewhat different, but
each test offers a sequence of white- and pink-noise signals
at five locations: HARD-LEFT, HALF-LEFT, CENTER, HALF-RIGHT,
and HARD-RIGHT. They do this continuously back and forth
across the soundstage. They also offer continuous, smooth
sweeps in addition to a five-position sequence.

The two-channel, conventional-stereo version is what we are
interested in here, since that will be the most relevant to
what we get when playing regular two-channel recordings.
With standard, two-speaker stereo playback the levels at
each of those five positions, even at half-left and
half-right, should be at about the same levels - provided
the listener occupies the sweet spot. The noise bursts at
each location should also be well focussed, with them being
uniformly strong at each of the five positions: strong /
strong / strong / strong / strong. You should also
experience this uniformity with good musical program
material.

Unfortunately, if the listener moves some distance away from
sweet-spot listening position, the smoothness of the
transitions go somewhat to pot, with the sound shifting
unevenly toward the closer main speaker. For example, if I
move a modest distance to the right of center, the staging
effect will came across as strong / very weak / weak /
strong / very strong. This same kind of two-speaker
soundstage skewing can also be heard with most musical
source material, although it will not usually be as emphatic
as it is with those test-tone noise bursts. Some speakers
are designed to counteract this tendencies (the dbx
Soundfield One system of a number of years ago comes to
mind), and it can also be counteracted somewhat by toeing
conventional speakers inward.

For my evaluation of the Dolby Pro Logic II (music) and DTS
Neo:6 (music) steering systems with two-channel music source
material I primarily used two speaker/processor
combinations. These were my own Allison IC-20 main speakers
and a custom center speaker (plus Allison and RDL
surrounds), controlled by the Yamaha RX-Z1 receiver, and
then a bit later on three NHT M6 satellites and Hsu VTF-3
subwoofer (with additional NHT and RDL speakers as
surrounds). The latter combination was controlled by the
Sunfire Theater Grand III processor/tuner, with three
Sherbourn 1/300MB monoblocks powering the front satellite
speakers and an assortment of other, smaller amps powering
the surrounds.

Utilizing the center channel with either Yamaha
Classical/Opera or standard Dolby Pro Logic processing (and
even with the movie/cinema versions of DPL II and Neo:6),
the balance at the sweet-spot listening position with this
series of noise burst test signals was always: strong / weak
/ strong / weak / strong. In addition to being subdued in
level, the half-left and half-right images were often vague
in terms of stability and focus, with only hard-right,
center, and hard-left sounds firmly stabilized and strong. I
found that if the Yamaha's front-enhancement "effect"
channels were turned on the Classical/Opera mode added a
somewhat spacious characteristic up front. However, the
soundstage imaging focus remained unchanged.

When listening from off axis, things got somewhat better.
For example, if I sidled toward the right a couple of feet
the spread was always: strong / weak / strong / strong /
strong. The center steering stabilized the imaging much more
effectively than what I got with only two channels.

Unfortunately, when listening from the sweet spot with a lot
of complex musical material the soundstage often will
partially collapse towards the center, and sometimes the
collapse will be rather dramatic. The steering systems often
do this with musical sources, even though they may not do it
with test signals. Consequently, to make both old-style DPL
and Classical/Opera steering work effectively up front it is
nearly always a good idea to back the center level off about
3 dB below the normal Dolby set-up level.

I have been doing this as a matter of policy when checking
out recordings in my record-review column, and it works
quite well. Of course, with the Delos disc's noise-sequence
test, the sweet-spot imaging will then show up as: strong /
weak / weak / weak / strong. Fortunately, this far-left and
far-right weighting does not cause problems with most music
material. Indeed, in nearly all cases the soundstage is
improved dramatically, with a smooth left-right blend,
particularly when listening from off axis. At such
locations, the image remains nearly as stable as what we
have with the center set at the Dolby calibration level.

However, both DPL II (Music) and Neo:6 (Music) are in a
position to be superior to some of those other modes when it
comes to soundstage focus, spread, and stability. In
addition, with the Surround Spectacular test sequence, the
same results are obtained from the sweet spot as when using
two speakers: strong / strong / strong / strong / strong.
There was no lack of focus or image shifting at all with the
Allison/Yamaha combination, nor did I detect any with the
NHT/Sunfire combination. Musical signals are equally well
stabilized and spread out.

Now, to get these results with DPL II (Music), it is
necessary to work with an adjustment parameter called
"center width." At the lowest setting, the steering is
similar to what we have with both standard DPL and DPL II
(Movie) and also with the Yamaha steering modes. At the
highest number, the center speaker is bypassed and you have
a phantom center.

However, at an in-between setting the steered center is
solidly blended with the phantom, and you get a mix that
stabilizes the center, while at the same time offering up
stabilized half-left and half-right images, too. I got the
best results with the Dolby center width set at number 3.
Neo:6 (Music) cannot be adjusted as to center width, but as
best I can tell it has a fixed setting that is similar to
that one.

When listening from off axis with the lowest setting of DPL
II (Music) or the setting is adjusted for a modest
phantom/center blend, the results are easily as good as what
we get with the other steered modes, and similar results are
obtained with Neo:6 (Music). If we shift the listening
location to the right again we get: strong / weak / strong /
strong / strong. With music, the slight weakness at the
half-left location is nearly always inconsequential.

While it is debatable whether these new DPL II and Neo:6
music technologies are always superior in terms of
soundstaging to what Yamaha offers with their
Classical/Opera mode (with the center level reduced 3 dB to
minimize center collapse), it is likely that with some
musical sources they will be. I now use all three, and have
decided to optimize the DPL II center-width setting at
number 2, with Neo:6 being the alternate if I want a still
more blended soundstage. It is not necessary to back off the
center level at all with those functions selected. If I want
a more lively hall acoustic, I generally opt for the Yamaha
Classical/Opera mode.

These new music-oriented Dolby and DTS technologies are also
remarkable when it comes to what they can do with the
surround ambiance they extract from a recording.

With the ambiance clicks on the Surround Spectacular disc,
DPL II (Music) generated an almost out-of-phase
characteristic from the surround channels. It was unlike
anything else I have heard using those test-signal clicks.
The effect is almost mysterious and when musical signals
were played, the result added an uncanny
three-dimensionality to the sense of large-room space in my
home-listening room. The result was particularly effective
with my IC-20 main speakers, because they generate a lot of
side-wall reflections anyway, and those blended seamlessly
with the DPL II (Music) ambiance further back into the room.

Unlike with the front soundstage, where the two technologies
are fairly similar, with the surround channels Neo:6 (Music)
was somewhat different from DPL II (Music). The surround
ambiance had a wide feel to it, but it was more coherent and
less diffuse sounding. As with the movie version, Neo:6
(Music) also makes use of the center-rear channel (if you
have it hooked up), and that feed was probably responsible
for the somewhat less diffuse sound from the surrounds. By
now, you are probably aware that Neo:6 technology involves
simulating six channels from a two-channel input.

With music, I felt that the sense of space with DPL II
(Music) was a bit superior to what I detected with Neo:6.
However, much will depend on the source material, the size
and shape of the room, the speaker arrangement, and the
listening position. From off axis, Neo:6 might have been a
bit more stable in terms of soundfield consistency, but it
was really hard to judge, even with the ambiance clicks.

Of course, Yamaha is noted for its soundfield enhancements,
and in this case the Dolby and DTS music-mode advantages
were anything but cut and dry. In most cases, the Yamaha
processing came across as a bit superior, particularly with
some of its concert-hall and jazz-club modes, but also with
the Classical/Opera function. DPL II and Neo:6 lack the
flexibility of the Yamaha hall-ambiance palette.

To a lesser extent, the same goes for the Sunfire TG III,
which has a user-adjustable Jazz mode that incorporates
center steering and exhibits a remarkable ability to
simulate a number of different-sized listening spaces,
including a concert hall. The Sunfire also has the ability
to drive a pair of up-front, side-wall-mounted effects
speakers (called "side-axis" speakers), and those also
imposed a degree of spaciousness and depth to the sound -
and could do so even with the new Dolby and DTS modes in
operation. In that respect, the Sunfire has a leg up on the
Yamaha.

Speaking of adjustments, the DPL II version offers up three
adjustment options that allow the user to adjust or engage
several parameters to accommodate room size and shape, the
listening position, and most importantly, taste.

One, called "dimension," allows the user to shift the
emitting soundfield further towards the front or rear of the
room. A second, the aforementioned "center width"
adjustment, allows one to vary the center imaging from a
phantom mode to a solidly center-focussed mode, with a
number of combinations in between.
The most radical of these adjustment options is "panorama,"
which partially incorporates the surround channels into the
soundstaging. The result is a wraparound effect that is not
exactly my cup of tea. However, if you have been thrilled
with some of the 5.1-channel musical presentations featuring
"sound in the round" from all channels, the panorama mode
might appeal to you. It did not appeal to me, however.

To summarize. Prior to DPL II and Neo:6, the only way an
enthusiast could enjoy the benefits of high-tech,
surround-sound signal enhancements with their two-channel
source material was to spring for some pretty expensive DSP
ambiance-synthesis hardware.
All that has changed. For the most part, Dolby Pro Logic II
(Music) and DTS Neo:6 (Music) are important advances in the
art of simulating a live musical performance from
two-channel source material. DPL II (Music) goes Neo:6
(Music) one better, because it does allow for more
adjustment parameters, although Neo:6 works terrifically
right out of the box, and it also includes a center-back
feed. The best thing about both is that they are now
available in a large number of audio/video receivers and
processors, and some of those are reasonably priced.

Indeed, this is the primarily reason why they are so
revolutionary and should be of interest to every sensible
enthusiast.

End of excerpts from the article.

When I switched to 5.1 it got worse.

I am not sure what this means. You cannot use the 5.1
function with two-channel source material, because 5.1 is a
discrete-channel recording process and the extraction of
surround material and the deriving of center-channel
material from two-channel sources involves steering
circuitry.

Part of the problem is that you have to
deal with time delays on my receiver for the back channels.

Yep, the surround speakers cannot let the ambiance signals
get to the listener before the primary, first-arrival
signals from up front. This requires adjustments to either
the speaker distances or some kind of adjusted-in electrical
delays.

Thanks for your suggestions Howard!

Glad you finally got satisfactory results. Yes, it does
require some experimentation to get ambiance and center
feeds properly adjusted, but what the heck, I thought that
one reason audio was supposed to be fun was because it
allowed one to experiment. You cannot begin to experiment as
much with two-channel audio as you can with surround sound.

Enjoy.

Howard Ferstler