App. #4
Application Notes
Studio and Production Applications of the MTI-3 TriSonic Imager
Contents:
Surround RecordingsGeneral Overview
The Miles Technology MTI-3 TriSonic&trade Imager is a high-quality signal processor providing important proprietary amplitude and phase functions without introducing nonlinearity or dynamic distortion.
In addition to TriSonic Imaging, which creates an excellent soundstage when reproducing stereo sound through three loudspeakers or loudspeaker clusters (and surround loudspeakers if desired), the MTI-3 also uniquely provides a variety of audio processing functions that are quite useful for production work involving stereo sound recording and remixing.
The MTI-3 performs a mild frequency tailoring in the stereo process, smoothing low-frequency performance below 80 Hz. This provides consistent performance with standard mixes which normally "go mono" in the low bass.
Maintaining highly correlated low bass in the stereo signal affords subjective advantages such as higher bass impact, as well as technical advantages like higher bass efficiency. The result is retention of overall signal integrity and no degradation of sound quality in any way.
The MTI-3 incorporates a crossover concept to apply the stereo process at frequencies above 80 Hz, where it's needed, while allowing the low bass to pass through unmodified. The use of first-order networks ensures that perfect frequency and phase integrity is maintained when the signal is recombined for the output.
Most outboard signal processors supply various delay and reverb functions at the expense of signal quality. In contrast, the MTI-3 offers a unique set of signal functions while maintaining the exceptional signal quality of state-of-the-art analog audio electronics.
These functions include: high-quality implementation of surround recording, creation of an extra-wide soundstage, vocal cancellation, and SpreadSoundÔ, a proprietary Miles Technology stereo synthesis design.
Surround Recordings
Surround recordings are achieved by taking the signal to be panned as surround and putting it into the stereo mix as a difference signal. In other words, the signal is mixed to left, and an equal level is inverted and mixed to right. (The reverse can also be done. It generally does not matter which polarity is used unless you wish to move the source to the front while recording, with continuous panning.)
But the situation is a bit more complicated. First, conventional mono compatibility is lost because the surround signal disappears in a mono left/right mix-a physical result of the basic surround recording process. (If mono compatibility is important, consider using the MTI3's SpreadSound function described later in this document, rather than recording the signal as surround.)
Second, most surround playback systems do not support the lowest frequencies. The majority of them utilize a mono subwoofer and smaller, lower-power loudspeakers for surround. The result is a loss of bass if it is "encoded" as surround.
The MTI-3 inherently solves this problem because output signals will remain in phase at the lowest frequencies (below 80 Hz). The frequency-dependent transition is very smooth and maintains a transparent response upon playback.
To Setup 1: Surround RecordingsIncreasing The Pan Range Of The Mix
A normal stereo mix-made with a typical mixing console-has a soundstage ranging from the left loudspeaker to the right loudspeaker. The pan controls at each input channel route the signal to a place somewhere at or between these two points.
The MTI-3 can be used to expand the range of all the pan controls on the console. Instead of being limited to panning across a line from the left to the right loudspeaker, the MTI-3 allows panning from a place at the left of the left loudspeaker to a place at the right of the right loudspeaker. The span of the soundstage can be electronically "widened" to practically any degree desired.
The console pan controls will still pan normally from left to right, but within the central portion of the control rotation. As the pan control is moved toward the extreme left or right position, the sound source will be panned beyond the loudspeakers.
An extreme pan position causes the signal to appear in the opposite channel of the stereo mix with inverted polarity. Acoustically, this moves the audible sound source beyond the loudspeaker (the same process used in "boom boxes" and stereo television sets to widen sound beyond the closely-spaced loudspeakers). It's an approach that works well for reproduction on any good stereo sound system. In the extreme case, the signal becomes essentially a surround signal.
To Setup 2: Arbitrary Signal PlacementUsing SpreadSound In The Mix
The MTI-3's SpreadSound function is an excellent way to "spread" a sound across the playback soundstage. Rather than having to choose a specific place to locate a sound with a pan control, any desired signal elements of a mix can be routed through the SpreadSound system so that those signals do not localize at a specific point. Instead, they are reproduced with a "big sound" spanning the entire loudspeaker setup.
SpreadSound can be used in the recording process just as well as in the playback process-applied to certain channels in the mix, these channels will offer the desired big sound when played back on any sound system. These enhanced signals are superimposed on the rest of the normal stereo mix, which remains unaffected.
This process is different from using a stereo reverb or "space expander," because the signal remains "clean" and "dry". SpreadSound can be used together with any digital effect to achieve a combined result, such as an all-around-you echo, spread-out flanging, etc.
A unique feature of the SpreadSound process: if left and right output signals are recombined to form a mono signal, the overall level does not change. Therefore, stereo mixes made with SpreadSound have much better mono compatibility than ordinary stereo mixes that typically suffer from "center buildup" when mixed to mono.
To Setup 4: SpreadSound OperationSynthesizing Stereo With SpreadSound
The SpreadSound function is a very effective way to synthesize a stereo signal from a monophonic program or signal source. Unlike other stereo synthesis techniques, it will not alter the frequency response of the signal nor create any audible delay or other type of distortion. It only adds a small amount of phase shift to the signal, which in itself is practically inaudible.
SpreadSound will create a big, spacious, yet totally clean version of a mono input signal. Either of the described connection approaches can provide excellent stereo synthesis. In addition to working perfectly with all stereo or TriSonic playback systems, the SpreadSound channels can be mixed back to mono with perfect results.
To Setup 4: SpreadSound Operation Synthesizing Mono With SpreadSound
To Setup 5: Variable SpreadSound
The idea of synthesizing a mono signal from stereo may seem trivial - you could simply mix the left and right channels together to create a mono sum. However, as mentioned earlier, there are some problems with this approach. The center-panned portion of the stereo mix will be over-emphasized in the mono-sum signal, relative to the left and right components. And, any surround components in the stereo mix will be completely lost due to cancellation in the summing process. Therefore the mono mix is seriously compromised, especially it if is derived from a surround recording.
The MTI-3 solves this problem. Using SpreadSound process with a stereo input signal prevents both center buildup and surround cancellation. The MTI-3 Center Output provides a mono send which includes the original left, center, right, surround, and all other pan locations at equal levels in the mono mix.
There is one caveat: since SpreadSound does not phase-shift at the lowest frequencies below about 100 Hz (because that could cause interference problems in listening rooms) there will still be surround cancellation of these lowest frequencies in the mono output. But since surround components almost never include the lowest frequencies, this is normally not a problem. There may be a slight change in overall frequency balance, which is easily correctable with equalization.
This method is probably the only way to accurately create a mono mix of a surround recording. It is recommended for any application that requires a high-quality mono mix.
To Setup 7: Synthesizing MonoCompare The Sound
The quality of TriSonic and SpreadSound processing speaks for itself, and can be applied to be dramatic or subtle. Both are unique in that they do not "color" the sound to achieve the spatial effects, and both actually enlarge the stereo listening area.
Other processes such as Spatializer, SRS, and QSound, utilize "generic ear response" filter functions, which are never a perfect match for your ears. They also require a centered, forward-facing listening position, and often take on an artificial-process character. The use of elaborate process chains with dynamic, spectral, and temporal modifications not only can be quite expensive, but often cause "mucked up" sound.
Wide panning and SpreadSound allow you to spatially place sounds where you like, without signal-quality penalties. And you can easily use it to spatially separate different instruments, your favorite delay effect, reverb, equalization, or any other type of signal. The end result is a much cleaner mix!
MTI-3 Front Panel
MTI-3 Setup 1: Surround Recording
To utilize the MTI-3 for surround recording, send the surround signal source (typically from a channel or subgroup direct output) to the MTI-3's Left Input, and mix the unit's Left and Right Outputs back into the stereo mix. With the unit's Input Balance control fully left, and the unit's TriSonic Balance control fully clockwise (to "Diff" setting), the Left and Right Outputs provide a stereo difference signal from the original input.
There will be a transition between approximately 60 Hz and 100 Hz, where the stereo placement of the signal will gradually move from surround, at frequencies above that range, to the center at the lowest bass frequencies. This frequency compensation occurs automatically and requires no user intervention or control adjustments.
The unit's Right Input can also be used for a signal-source connection, or, in the case of two signals that are uncorrelated-e.g. separate instruments or effect outputs-use both the Left and Right Inputs with the Input Balance control centered.
MTI-3 Setup 2: Arbitrary Signal Placement
To precisely place a signal anywhere on the 360° soundstage, connect the input signal to the unit's Left and Right Inputs (via a "Y" connector or a stereo feed with the source panned center). With the unit's Input Balance control at full left, the TriSonic Balance control will pan the signal from the rear or surround position ("Diff" or full clockwise), around the left (TriSonic Balance straight up) and then to straight center ("Mono" or full counterclockwise). Moving the Input Balance control to full right and moving the TriSonic Balance control through its range pans the signal around the right side from center to surround.
Another method: Set the TriSonic Balance control to the 3 o'clock position and use the Balance control to pan the mono input signal from left-rear, all the way around center, to right-rear.
With these techniques you can easily achieve precise placement of any particular sound source at any position in your mix. Meanwhile, the lowest bass will always be maintained perfectly at front center so that consistent sound and compatibility with popular playback systems is maintained.
MTI-3 Setup 3: Increasing The Pan Range
To utilize the MTI-3 for pan range expansion, insert the unit into the stereo mixer output signal path using the unit's Left and Right Inputs and its Left and Right Outputs. The MTI-3 Input Balance control should be straight up (12 o'clock). The unit's TriSonic Balance control will adjust the pan range, and hence the achievable soundstage width.
With the TriSonic Balance control at about the 9 o'clock position, normal panning will result. At the 12 o'clock position, significant left-to-right expansion will result. Set it toward the 3 o'clock position for even more expansion where a pan control set to full left and right will yield essentially a surround signal. In the other direction, such as at the 8 o'clock position, the soundstage will be narrowed.
With the TriSonic Balance control at full left or right, the soundstage will collapse to mono or go to difference (with no center), respectively. To preset the pan range, pan a signal to full left or right on the mixer and adjust the TriSonic Balance control for the desired effect in this extreme pan position.
The normal range for the TriSonic Balance control in this application is from 9 o'clock to 3 o'clock. The end result of using the MTI-3 for this application is a great improvement in the capability of every pan control of the mixer for full-range soundstaging.
Meanwhile, standard in-between panning is easily achieved with the pan controls nearer the center of their range.
NOTE: Due to the nature of this process, the level may change slightly as a signal is panned. Therefore, it is best to set the pan positions before precisely adjusting the levels in the mix.
MTI-3 Setup 4: SpreadSound Operation
The approach for standard SpreadSound operation is to use the unit's Discrete Surround input connection and it's Surround 1 and Surround 2 Outputs as the stereo output. Make sure the unit's Surround Bandwidth switch (also on the back panel) is set as desired (20K for full bandwidth).
Route a channel send or a subgroup send to the Discrete Surround input and return the Surround 1 and Surround 2 Outputs back to a stereo return or pair of inputs on the mixing console. Set the unit's Surround control (on the front panel) all the way down (full counterclockwise) to maintain isolation with the main TriSonic circuitry. You can simultaneously use the TriSonic circuitry for TriSonic Imaging, or for another application.
In addition to providing a big, spread-out sound, SpreadSound maintains mono compatibility. If SpreadSound signals are mixed back to mono, the result is a replica of the original signal. SpreadSound is probably the very cleanest way to expand a single sound source, not causing any audible delays or comb filter distortions. It utilizes a gentle phase shift approach which maintains perfectly flat frequency response at the outputs.
If you listen to each of the outputs, you'll find that they sound just like the original signal. Their sum also sounds like the original.
MTI-3 Setup 5: Variable SpreadSound
The MTI-3 actually includes two SpreadSound circuits. They can be used independently with different signals, or together for multichannel recordings or productions.
Using the main signal path in the MTI-3 for SpreadSound provides a variable control over its effect. To do this, use the Left and Right Inputs to the MTI-3 (the input should be monophonic; a Y-connector can be used if necessary) and press the SpreadSound Button on the front panel. The Left and Right Outputs will then provide the SpreadSound signal. The unit's Center Output can also be used where needed.
In this application, the TriSonic Balance control will then adjust the amount of SpreadSound applied. The normal position of this control is straight up (12 o'clock), with reduction of the spreading effect achieved by moving the control counterclockwise and an increase of the spreading effect achieved by moving the control clockwise.
Multichannel SpreadSound also can be achieved by additionally using the MTI-3's Surround 1 and/or Surround 2 Outputs. These outputs offer additional phase shift and are suitable for multichannel productions where it's desired that sound be "spread out" among any number of additional channels.
When more than five channels are employed, alternate the Surround 1 and Surround 2 Outputs on adjacent channels in the system. In this application, the level of Surround 1 and 2 Outputs will be affected by the Surround control on the unit's front panel because they are derived from the unit's Left and Right Inputs.
MTI-3 Setup 6: Vocal Cancellation
The MTI-3 accurately performs vocal cancellation (removal) from existing stereo recordings. This process is based on the principle that a lead vocal signal in a stereo recording is almost always panned to the center. It also works if the vocalist is off center.
Set the MTI-3's TriSonic Balance control to "Diff". Then adjust the unit's Source Balance control until the vocal signal is minimized. With a high-quality signal source, the vocal portion will be almost completely removed from the output. Note that the Left and Right Outputs will have the same signal but they will be inverted from each other. Either can be used as a mono signal.
Note also that the output will not be deprived of bass, as would occur with a straight differencing operation. Due to the frequency contouring, the vocal-removed signal will still have the bass portion of the music even though it is panned center.
For a stereo recording or production of a vocal-removed signal, use the unit's Left and Right Inputs in the procedure described above. However, instead of using the MTI-3's Left Output as is, patch it back into the unit at the Discrete Surround Input. Make certain that the front-panel Surround control is turned all the way down. Then use the Surround 1 and Surround 2 Outputs as the stereo output.
Again, adjust the Source Balance control until the vocal signal is minimized. You will then have a wideband, SpreadSound stereo signal, with plenty of bass and yet with vocals (or other lead instrument) removed! This works well for both live productions and for recording.
MTI-3 Setup 7: Synthesizing Mono
To create a good mono mix from any stereo recording, connect the stereo signal to the MTI-3 Left and Right Inputs. Set Source Gain and Balance as appropriate, switch on SpreadSound and set the TriSonic Balance to the center of its range.
The MTI-3 Center Output then provides a high-quality mono mix of the stereo input, with no center buildup and no surround cancellation. The output mix is a simple sum below 100 Hz, but above 100 Hz it includes all original pan locations at an equal level.