This page contains settings for mixing control.
 Mixing matrix
The main element of this page is mixing matrix. It is represented in a form of matrix multiplication:
Simpler it means a rule for every cell: we mix the channel given by the column into the channel given by the row with gain given by the cell. Therefore, if we want to route the left channel into the right speaker we should set the value of cell on the cross of ’L’ column and ’R’ row as 1.
Matrix row defines the contents of an output channel (speaker) given by this row. Therefore, the first row of the matrix above indicates sound, which will be heard from the left speaker:
L’ = 1 * L + 0.7071 * C + 1 * SL + 1 * LFE
i.e. from the left speaker we will hear the left channel, part of the central channel, the left rear and the low frequency effects channel. Thus, mixing from 6 original channels into 2 stereo channels is performed (this process is called downmixing).
One important property of matrix is that when swapping contents of two matrix rows we can ’interchange’ appropriate speakers. Therefore, swapping of the first and the third rows results in ’interchange’ between the left and right speakers. This property can be used to rearrange speakers without their physical switching.
Values of the matrix column show the target of the channel mixing. Therefore, the central channel is routed both to the left and right speakers with the 0.7 factor (otherwise the center outgoing from two speakers, will sound louder than if we would have the real third speaker). Thus, we will hear the central channel in the middle of the left and right speakers. Changing factor of the left or right channel you can ’move’ the center. Thus, setting the value of cell on the cross of ’C’ column and ’L’ row as 1 will result in louder sound of the center from the left channel and center will seem to be moved to the left.
In most cases there's no need to change the matrix manually: when the AutoMatrix option is on the mixing matrix is calculated automatically based on the following parameters: channel configuration, Voice, Surround, LFE levels, Normalize matrix, Voice control and Expand stereo options. Thus, changing the Voice level will result in changes of matrix coefficients at column ’C’.
 Output format
The same as output format at Main page. It is placed here because channel configuration also affects the mixing matrix. So you're able to see the effect of output format change without jumping around pages.
 Matrix preset
Here you can save or load the matrix you have made.
To save a matrix:
- Enter a name into the box
- Press 'Enter' or click 'Save' button.
To load a previously saved matrix just select it from the drop-down list.
- Gain for the center channel.
- If the input signal contains real central channel, then this level affects only on this channel.
- If there is no central channel but Voice control flag is set, this level controls the level of ’virtual’ center.
- This level is a matrix parameter, so it makes sense only when Auto matrix option is on. When changing level the matrix is recalculated.
- Rear gain level.
- If rear channels present at the input, then this level affects only on them.
- If there are no rear channels in source but Expand stereo flag is set, then this level controls the level of ’virtual’ rear.
- This level is a matrix parameter, so it makes sense only when the Auto matrix option is on. When changing level the matrix is recalculated.
- LFE channel level. This level is a matrix parameter, so it makes sense only when the Auto matrix option is on. The matrix is recalculated when changing the level.
- Auto gain control
- This option reduces the current gain level to avoid an overflow.
- When this option is off, an extremely strong sound will be cut resulting in distortions, which are unpleasant to hear.
- If the option is on then in case of overflow the sound becomes quieter thereby excluding overflow. System behavior after overflow is defined by One-pass norm option:
- When One-pass norm is disabled, the gain will return to Master level. At that, this level specifies an average gain level. If there are many locations with overflows such behavior may result in unpleasant leaps of loudness. Therefore you should not set Master level too high. It is recommended to use this mode when overflow is not expected or when it appear relatively rarely.
- When One-pass norm is enabled, the gain does not recover after overflow. Thus, every overflow reduces the loudness along with the probability and ’intensity’ of next overflows, Therefore, after some time the gain is automatically adjusted to a certain value that does not produce overflows. As a rule, level becomes stable after ~10 minutes of film playing. This mode is used to reach the highest possible loudness with minimum artifacts. In this case, the Master level acts as an initial gain value for the one-pass normalization operation. As such, it should be adjusted to the maximum gain desired.
- One-pass norm
- Enable one-pass normalization. Refer to Auto gain control.
- Auto matrix
- Automatic calculation of the mixing matrix.
- When this option is on you cannot edit the matrix directly. LFE, Voice and Surround gain levels are parameters for matrix calculation; when changing them matrix is changed immediately.
- When this option is off the following control elements loose their effect: LFE, Voice and Surround levels, Normalize matrix, Expand stereo, Voice control options. Ability of direct editing of the mixing matrix becomes available.
- Normalize matrix
- When this option is active, the mixing matrix is normalized to ensure that no overflows occur in output channels (the sum of the channels <= 1). But at the same time, changing of LFE, Voice and Surround gain levels may lead to loudness redistribution (for instance, when increasing LFE loudness, loudness of other channels will decrease) that is not always convenient. Besides, in most cases matrix normalization strongly reduces total loudness. Switching this option off allows to avoid those effects, however possibility of overflow appears at the same time.
- Expand stereo
- Create 'virtual' surround. Takes effect only when there are no rear channels at input. In this case filter creates virtual surround channel, gains it with Surround level and mixes it into the output.
- If output configuration has rear channels, virtual surround channel is mixed there. I.e. upmixing takes place. It is useful to utilize rear speakers even when a movie has only a stereo sound.
- If output configuration does has no rear channels, then virtual surround is mixed back to front channels. The purpose of this mode is to enhance surround sounds in stereo track.
- Voice control
- Create 'virtual' center. Takes effect only if there is no center channel at input. In this case filter creates virtual center channel, gains it with Voice level and mixes it into the output.
- If output configuration has center channel, virtual center channel is mixed there. I.e. upmixing takes place. It is useful to utilize center speakers even when a movie has only a stereo sound.
- If output configuration does has no center channel, then virtual surround is mixed back to front channels. The purpose of this mode is to enhance dialogs in stereo track.