SAA7715 Datasheet, PDF (22/36 Pages) NXP Semiconductors – Digital Signal Processor

Philips Semiconductors
Digital Signal Processor
Preliminary specification
SAA7715H
10.1.1.5 Attack time
Attack time controls the rate at which gain-reduction is
engaged following the detection of the input signal
exceeding the threshold level. Typical values are in the
range of 0 to 100 ms. Fast attack times tend to smooth out
abrupt transients thereby helping to ensure the output
level remains fairly consistent; however, at the same time
fast attack times can easily destroy much of the dynamic
character of sources having very distinguished attack
transients (such as a piano or an acoustic guitar). Slow
attack times, on the other hand, allow the sources attack
transients to pass through virtually unaltered, thereby
retaining most of the dynamic signature of the source. The
danger here, however, is the possibility of clipping the
output, or overloading one or more downstream
components.
The present implementation of the compressor does not
provide user access to attack time.
10.1.1.6 Release time
Complementing the attack time control, release time
controls the speed at which the compressor disengages
after the input level falls back below the threshold. Typical
values here range from around 100 ms to several
seconds.
The present implementation of the compressor does not
provide user access to release time.
10.1.1.7 Output or ‘make-up gain’
In order to make maximum use of the available bit
resolution, it becomes necessary to boost the
compressors output in order to ensure the signal swings
close to the maximum excursions allowed by the digital
output. Notice in Fig.7 how the output level can be
dramatically reduced, particularly at low threshold levels
and high compression ratios. In the present
implementation, this rescaling is managed automatically
according to the current threshold and ratio settings.
10.2 Audio enhancer
10.2.1 THEORY OF OPERATION
The enhancer uses non-linear processing to generate
extra harmonics, which are added to the audio to improve
high frequency detail. It is particularly useful with
streaming audio from the Internet, which is typically
compressed to the extent that the original high frequency
content is lost.
The enhancer is also a very effective means of improving
the sound of CD-quality audio, by restoring the presence
and brilliance of the original acoustic performance.
10.2.1.1 Control parameters
The enhancer has a single mix control, which determines
the amount of generated harmonics to be added to the
signal. High settings will result in a brighter effect with
greater depth. For particularly dull audio, such as is often
received over the Internet, a high mix level will have a
pleasing effect. Intermediate settings are appropriate for
CD-quality audio, although classical music listeners may
prefer to use the enhancer sparingly.
10.3 Equalizer
10.3.1 GENERAL DESCRIPTION
• 2-channels
• 5-bands
• Control range: 20 Hz to 20 kHz.
10.3.2 OVERVIEW
The fundamental ideal for any high-fidelity audio rendering
system is to reproduce the aural experience present at the
time and place the original audio material was recorded.
Unfortunately, practically all systems fall short of this ideal
to some degree for a number of reasons. While
environmental acoustics can play a significant role, in
many cases performance deficiencies associated with the
loudspeakers cause most of the ‘distortion’. This happens
when the loudspeakers cannot deliver a uniform frequency
response over the entire audio range (20 Hz to 20 kHz).
Equalizers were invented to deal with frequency response
problems by boosting or cutting selected frequency bands
in the signal. Used in the right manner, a properly adjusted
equalizer can effectively compensate for loudspeaker
performance deficiencies, or any other frequency
dependent amplitude variations in the system.
Additionally, equalization can be used to create a
customized frequency response which is better suited for
a particular listener or a particular style of music, for
instance.
The type of equalizer provided with this system is of the
parametric variety. Parametric equalizers differ from
graphic equalizers by giving the user more control over the
filters that actually effect the boost or cut of a particular
band. More specifically, for each band, users can control
the band’s centre frequency, and also the width of the
band of frequencies that are affected.
2001 May 07
22