The Stereo Soundstage Page 2

There is a large demand today for processors that can make simulated stereo out of a mono source. The money that is driving this technology comes not from the music industry but from Turner TV and Hollywood, where producers who are colorizing old black-and-white movies for video sale would also like to enhance their mono soundtracks so they will sound good in a home video theater. Of course film purists kvetch about colorization, and audio critics have always complained about fake stereo, especially since most early examples of each were awful. But some colorized movies actually are more satisfying on TV than the black-and-white versions, because most video equipment has insufficient visual dynamic range to properly reproduce the chiaroscuro contrasts that made certain great movies look so fine on theater screens. Colorization and stereoization are unstoppable anyway, since video sales and rentals are now a vastly larger business than either movie theaters or records. Such processing may even be a good thing, at least for new audiences who have no opportunity to see classic films in a theater. It also can restore the faded and distorted colors of movies that were filmed in color during the '50s using unstable dyes that have changed in the vaults.

And as cable-TV systems increase from 30 to 50 or 100 channels, every added channel represents an increased demand for 24 hours of different programming, 365 days a year. A 50-channel cable system needs 1200 hours of programs every day, 8400 hours every week, to keep the pipeline filled. For outlets such as Nickelodeon, the Family Channel, the Comedy Channel, and the Sci-Fi Channel, old TV shows from the '50s and '60s are being dusted off and recycled. Even the classic "M*A*S*H" series from the '70s, which has been running continuously in syndication for nearly 20 years, is being re-released in fresh video prints with enhanced sound and much better video transfers from the original film negatives.

The huge size of the video market is driving rapid improvements in both computer-based video colorization and simulated-stereo processing. In the past, most attempts to make pseudo-stereo out of mono have produced obvious side effects (phasiness and comb filtering), but recently a few processors have done it fairly well. During the Las Vegas CES I heard one, in prototype form, whose artifacts were still audible but not seriously offensive. It's not good enough yet to make a genuinely satisfying stereo image from a mono source. But if a computational algorithm can be used at full strength to widen a mono signal and create fake stereo, it can also be applied at reduced strength to widen all of the individual images in a stereo mix. Thus, a sound that already contains the elements of a true stereo image can be usefully widened and still sound pretty good. If the processing is applied conservatively, its artifacts may be subtle enough to remain unheard. This was the case with the prototype processor.

In a paper presented at last fall's AES convention (AES Preprint 3423, "Signal Processing for Simulating Realistic Stereo Images"), Michael Gerzon emphasized that one of the benefits of wide-stage stereo is "directional unmasking." When two similar sounds arrive at the listener from the same direction, they tend to be heard as a single composite sound whose character is dominated by the louder component. But if the sounds are separated by even a small angle (5 degrees or so), they are easily resolved by the ear as individual sources.

To mention an obvious example, the contrast between stereo and mono is not merely the introduction of left-right ping-pong separation, with violins on the left and bass fiddles on the right. More importantly, stereo provides a basic improvement in the quality of the sound, giving you more to hear---making it easy, for instance, to resolve the individual sounds of two similar instruments playing in the same frequency range (eg, a duet between clarinet and English horn in the middle of the orchestra). It also allows you to resolve the reflections of these sounds as they bounce off the walls of the stage, giving you a sense of the width or depth of that stage. All this is part of what some writers describe as a stereo system's ability to "throw" a wide, deep soundstage. And since improved resolution of the soundstage lets you hear everything with less mental effort, you may experience less listening fatigue. (Recall what GL wrote.)

As Gerzon pointed out, only a small change in stereo spread is needed to increase directional unmasking and produce a noticeable improvement in subjective quality. Thus I hear an obvious change in soundstage clarity and resolution when I move my head just 18" forward, and GL experienced a dramatic transformation when he moved 4' closer to his speakers. Simple experiments with the placement of your speakers or your chair could pay off with big dividends. Next year, don't be surprised if you see a control knob called "distance" or "image width" on a preamp or processor.


Doctor Fine's picture

I have my listening chair positioned so that I can close the "hole in the middle gap" by leaning back or increase the spacing by leaning forward.