6  Sound Settings

The sound settings menu offers a selection of sound settings you may change to customise your listening experience.

6.1  Volume

This setting adjusts the volume of your music. Like most professional audio gear and many consumer audio products, Rockbox uses a decibel scale where 0 dB is a reference that indicates the maximum volume that the player can produce without possible distortion (clipping). All values lower than this reference will be negative and yield a progressively softer volume. Values higher than 0 dB are available and can be used to raise the volume more than would otherwise be possible. These volume levels will ordinarily lead to distorted sound, but might work nicely for music that has an otherwise low volume level. The volume can be adjusted from a minimum of -74 dB to a maximum of +6 dB.

6.2  Bass

This setting emphasises or suppresses the lower (bass) frequencies in the sound. A value of 0 dB means that bass sounds are unaltered (flat response). The minimum setting is -24 dB and the maximum is 24 dB.

6.3  Treble

This setting emphasises or suppresses the higher (treble) frequencies in the sound. A value of 0 dB means that treble sounds are unaltered (flat response). The minimum setting is -24 dB and the maximum is 24 dB.

6.4  Balance

This setting controls the balance between the left and right channels. The default, 0, means that the left and right outputs are equal in volume. Negative numbers increase the volume of the left channel relative to the right, positive numbers increase the volume of the right channel relative to the left.

6.5  Channels

A stereo audio signal consists of two channels, left and right. The Channels setting determines if these channels are to be combined in any way, and if so, in what manner they will be combined. Available options are:

Stereo.

Leave the audio signal unmodified.

Mono.

Combine both channels and send the resulting signal to both stereo channels, resulting in a monophonic output.

Custom.

Allows you to manually specify a stereo width with the Stereo Width setting described later in this chapter.

Mono Left.

Plays the left channel in both stereo channels.

Mono Right.

Plays the right channel in both stereo channels.

Karaoke.

Removes all sound that is common to both channels. Since most music is recorded with vocals being equally present in both channels to make the singer sound centrally placed, this often (but not always) has the effect of removing the voice track from a song. This setting also very often has other undesirable effects on the sound.

6.6  Stereo Width

Stereo width allows you to manually specify the effect that is applied when the Channels setting is set to “custom”. All values below 100% will progressively mix the contents of one channel into the other. This has the effect of gradually centering the stereo image, until you have monophonic sound at 0%. Values above 100% will progressively remove components in one channel that is also present in the other. This has the effect of widening the stereo field. A value of 100% will leave the stereo field unaltered.

6.7  Crossfeed

Crossfeed attempts to make the experience of listening to music on headphones more similar to listening to music with stereo speakers. When you listen to music through speakers, each ear will hear sound originating from both speakers. However, the sound from the left speaker reaches your right ear slightly later than it does your left ear, and vice versa.

The human ear and brain together are very good at interpreting the time differences between direct sounds and reflected sounds and using that information to identify the direction that the sound is coming from. On the other hand, when listening to headphones, each ear hears only the stereo channel corresponding to it. The left ear hears only the left channel and the right ear hears only the right channel. The result is that sound from headphones does not provide the same spatial cues to your ear and brain as speakers, and might for that reason sound unnatural to some listeners.

The crossfeed function uses an algorithm to feed a delayed and filtered portion of the signal from the right channel into the left channel and vice versa in order to simulate the spatial cues that the ear and brain receive when listening to a set of loudspeakers placed in front of the listener. The result is a more natural stereo image that can be especially appreciated in older rock and jazz records, where one instrument is often hard-panned to just one of the speakers. Many people will find such records tiring to listen to using earphones and no crossfeed effect.

Crossfeed has the following settings:

Crossfeed.

Selects whether the crossfeed effect is to be enabled or not.

Direct Gain.

How much the level of the audio that travels the direct path from a speaker to the corresponding ear is supposed to be decreased.

Cross Gain.

How much the level of the audio that travels the cross path from a speaker to the opposite ear is to be decreased.

High-Frequency Attenuation.

How much the upper frequencies of the cross path audio will be dampened. Note that the total level of the higher frequencies will be a combination of both this setting and the Cross Gain setting.

High-Frequency Cutoff.

Decides at which frequency the cross path audio will start to be cut by the amount described by the High-Frequency Attenuation setting.

Most users will find the default settings to yield satisfactory results, but for the more adventurous user the settings can be fine-tuned to provide a virtual speaker placement suited to ones preference. Beware that the crossfeed function is capable of making the audio distort if you choose settings which result in a too high output level.

6.8  Equalizer

Rockbox features a parametric equalizer (EQ). As the name suggests, a parametric EQ lets you control several different parameters for each band of the EQ. In some ways the EQ is similar to the Bass and Treble settings described earlier, but the EQ allows you to control the sound much more carefully.

Rockbox’s parametric EQ is composed of five different bands:

Band 0: Low shelf filter.

The low shelf filter boosts or lowers all frequencies below a certain frequency limit, much as the “bass” control found on ordinary stereo systems does. Adjust the “cutoff” frequency parameter to decide where the shelving starts to take effect. For example, a cutoff frequency of 50 Hz will adjust only very low frequencies. A cutoff frequency of 200 Hz, on the other hand, will adjust a much wider range of bass frequencies. The “gain” parameter controls how much the loudness of the band is adjusted. Positive numbers make the EQ band louder, while negative numbers make that EQ band quieter. The “Q” parameter should always be set to 0.7 for the shelving filters. Higher values will add a small boost around the cutoff frequency that is almost always undesirable.

Bands 1-3: Peaking filters.

Peaking EQ filters boost or lower a frequency range centered at the centre frequency chosen. Graphic equalizers in home stereos are usually peaking filters. The peaking filters in Rockbox’s EQ lets you adjust three different parameters for EQ bands 1 through 3. The “centre” parameter controls the centre frequency of the frequency range that is affected as described above. The “gain” parameter controls how much each band is adjusted, and works as for the low shelf filter. Finally, the “Q” parameter controls how wide or narrow the affected frequency range is. Higher Q values will affect a narrower band of frequencies, while lower Q values will affect a wider band of frequencies.

Band 4: High shelf filter.

A high shelf filter boosts or lowers all frequencies above a certain frequency limit, much as the “treble” control found on ordinary stereo systems does. The high shelf filter is adjusted the same way as the low shelf filter, except that it works on the high end of the frequency spectrum rather than the low end.

As a general guide, EQ band 0 should be used for low frequencies, EQ bands 1 through 3 should be used for mids, and EQ band 4 should be used for highs.

Enable EQ.

This option controls whether the EQ is on or off.

Graphical EQ.

This option brings up a graphic EQ screen, which allows adjustment of each of the three parameters described above (gain, centre frequency, and Q) for each of the five EQ bands.

Key	Action
Right	Raises the highlighted parameter.
Left	Lowers the highlighted parameter.
Play	Moves to the previous EQ band.
Submenu	Moves to the next EQ band.
Select	Toggles the cursor among the three parameters (gain, centre frequency, Q) for the selected EQ band
Power	Exits the graphic EQ screen.

Pre-cut.

If too much positive gain is added through the graphical EQ, your music may distort. The Precut setting allows you to apply a global negative gain to decoded audio, cancelling out positive gain from the EQ. This will prevent distortion when boosting certain frequency ranges, at the expense of making audio quieter.

Alternatively, precut can be used with a flat EQ curve to implement a volume cap. For example, on a player that allows overdriving the headphone output to +6dB, maximum volume can be capped to +0dB by applying 6dB of precut. Note that precut is not applied if EQ is disabled.

Simple EQ.

This option provides an easier alternative for those who are daunted by all of the parameters that can be adjusted using the graphical EQ. With the Simple EQ, the only parameter that can be adjusted is the gain.

Advanced EQ.

This sub menu provides options for adjusting the same parameters as the Graphical EQ. The only difference is that the parameters are adjusted through textual menus rather than through a graphic interface.

Save EQ Preset.

This option saves the current EQ configuration in a .cfg file.

Browse EQ Presets.

This menu displays a list of EQ presets, as well as any EQ configurations saved using the Save EQ Preset option. Users unfamiliar with the operation of a parametric EQ may wish to use the presets instead of trying to configure the EQ, or use the presets for designing their own custom EQ settings.

6.9  Dithering

This setting controls the dithering and noise shaping functionality of Rockbox.

Most of Rockbox’ audio file decoders work at a higher bit depth than the 16 bits used for output on the player’s audio connectors. The simplest way to convert from one bit depth to another is simply discarding all the surplus bits. This is the default behaviour, and adds distortion to the signal that will vary in character along with the desired sound.

Dithering adds low-level noise to the signal prior to throwing away the surplus bits, which gives the resulting signal a uniform noise floor which is independent of the signal. Most people find this noise preferable to the time-varying noise heard when not performing dithering.

After dithering, noise shaping is performed. This basically just pushes the dithering noise to the parts of the frequency spectrum humans cannot hear so easily. In Rockbox’ case, some of the noise is pushed up to above 10 kHz.

This setting will be put to its best use when listening to dynamic music with frequently occuring quiet parts, classical music being a typical example. It is worth noting that the effects of dithering and noise shaping are very subtle, and not easily noticable.

Rockbox uses highpass triangular distribution noise as the dithering noise source, and a third order noise shaper.

6.10  Timestretch

Enabling Timestretch allows you to change the playback speed without it affecting the pitch of the recording. After enabling this feature and rebooting, you can access this via the Pitch Screen. This function is intended for speech playback and may significantly dilute your listening experience with more complex audio. See section 4.3.3 for more details about how to use the feature.

6.11  Compressor

The Compressor reduces, or compresses, the dynamic range of the audio signal. This makes the quieter and louder sections closer to the same volume level by progressively reducing the gain of louder signals. When subsequently amplified, this has the effect of making the quieter sections louder while keeping the louder sections from clipping. This allows listening to the quiet sections of dynamic material in noisy environments while preventing sudden loud sections from being overbearing.

There are several settings associated with the compressor. The first, and most important, is the Threshold. The threshold is the audio input level at which the compressor begins to act. Any level louder than the threshold will be compressed to some extent. The maximum amount of compression, or the quietest level at which the compressor will operate, is -24 dB. The default of Off disables the compressor.

The Makeup Gain setting has two options: Off and Auto. Off means that the compressed audio will not be amplified after compression. The default of Auto will amplify the signal so that the loudest possible signal after compression will be just under the clipping limit. This is desirable because the compressed signal without makeup gain is quieter than the input signal. Makeup Gain in Auto restores the signal to the maximum possible level and brings the quieter audio up with it. This is what makes it possible to hear the quieter audio in noisy environments.

The Ratio setting determines how aggressively the compressor reduces gain above the threshold. For example, the 2:1 setting means that for each two decibels of input signal above the threshold, the compressor will only allow the output to appear as one decibel. The higher the ratio, the harder the signal is compressed. The ratio setting of Limit means essentially a ratio of infinity to one. In this case, the output signal is not allowed to exceed the threshold at all.

The Knee setting determines how abrupt the transition is from a non-compressed signal to a compressed signal. Hard Knee means that the transition occurs precisely at the threshold. The Soft Knee setting smoothes the transition from plus or minus three decibels around the threshold.

The Release Time setting sets the recovery time after the signal is compressed. Once the compressor determines that compression is necessary, the input signal is reduced appropriately, but the gain isn’t allowed to immediately return to normal levels. This is necessary to reduce artifacts such as “pumping.” Instead, the gain is allowed to return to normal at the chosen rate. Release Time is the time for the gain to recover by 10 dB.