Every Mac comes with a Sound control panel for recording system alerts. With
extra software and hardware you can turn your computer in a multi-track
recording studio!
The Mac’s analogue to digital converter (ADC) digitises audio signals by sampling, using a specified sample rate and number of sample bits. These values are chosen for the best compromise between sound quality and the available memory or disk space.
During a recording the Sound control panel or a recording application transfers the sound onto hard disk — it can then be processed using a sound utility. When replayed it’s sent to a digital to analogue converter (DAC) that feeds the Mac’s audio output. The DAC automatically uses the correct sample rate and number of bits.
Digital audio recording and playback sometimes ‘skips’ when virtual memory is enabled — if you have this problem just turn off Virtual Memory in the Memory control panel.
Sound Software
qqqqqqqqqqqq
The Sound Control Panel
wwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwww
  The Sound control panel operates in following areas:-
Menu Item For selecting:-
Alert Sounds Alert sound file and volume
Sound In Sound input hardware
Sound Out Output, sample rate, resolution, mono or stereo
Volumes Overall output volume
The Alert Sounds window usually looks something like this:-
 
In this example Simple Beep has been chosen as an alert sound with an Alert Volume of 1. All the sounds in the list are system sound files in the System file that resides inside the System Folder. The Alert Volume is also influenced by any settings in the Volumes window.
If you select Add… the system presents you with a recording window:-
 
The control buttons, elapsed time meter and sound input indicator work in the same way as a tape recorder. When recording is complete you can Save the result — a new system sound file is then created in the System file
å On a PCI PowerMac the Sound panel is replaced by Monitors & Sound.
å Some AV models can’t always play some alert sounds properly unless Alert Volume
is at maximum. To compensate for this select Volumes in the pop-up menu and
adjust the Built-in volume.
å You can Copy any sound highlighted in the list of sound files — and then Paste it
into the Scrapbook or any other application.
å Some application make alert sounds that can’t be stopped. To prevent such sounds
you can create a short silent recording (unplug the microphone) and name it Silent.
You can then select this in the application in place of the original alert sound.
SimpleText and SimpleSound
wwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwww
  SimpleText can record sounds that are automatically saved inside the current
document. To make a recording select Record in the Sound menu — the standard
recording window will appear. Once the recording is complete you can recall it at
any time by selecting Play in the menu. To get rid of it simply select Erase.
If QuickTime is installed you can use SimpleText to open movie files of Type MooV. These can contain graphics, sounds or music as a MIDI sequence. To make SimpleText open such a file automatically in the Finder you should change its Creator code to ttxt or ttro.
√π See the Music chapter for about MIDI and musical sequences
√π See the Movies chapter for more about QuickTime
  SimpleSound is an application that creates alert sounds in the same way as the
Sound control panel and opens sound files outside of the System file. When a
file is opened a window shows you its resolution, whether its mono or stereo, its
sample rate and duration. To hear the sound just double-click in the window.
Inside a Sound Utility
wwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwww
  A sound utility can extract sounds from various documents, including
generic sound files (see below). For example, SoundHandle can extract snd_
resources from any file, including applications! Such a utility can then process
these extracted sounds and save them in another form.
  The window below is from an utility called SoundEffects — a stereo sound file has
been opened and a segment of one channel is selected. In the top left are details
about the file, including its length in bytes.
The centre boxes give details about the whole of the selected channel whilst those at the top right show the position and size of the selected segment — in this example these are measured in samples.
 
The waveforms shows how the signal level of each channel vary with time — the left-hand channel (channel 1) is at the top and the right-hand channel (channel 2) at the bottom. Below these are buttons to enlarge, reduce or normalise the viewing magnification. To the right of these are scrolling buttons for moving the view through the sample. The usual tape recorder buttons appear below these. The last button lets you play the a the file or a selected segment continuously!
In a typical utility you can process either a selected segment or the entire file. These effects can include:-
Chorus adds replicas of sound slightly spaced in time
Dither adds background noise to improve subjective quality
Downsize reduces sample resolution from 16-bit to 8-bit
Echo repeats the sound fading away with time
Fade In fades volume from zero to maximum
Fade Out fades volume from maximum to zero
Filter emphasises or reduces frequency components
Flange as Chorus but with feedback giving metallic effect
Keyboard for playing sound using a musical keyboard
Mono converts stereo samples to mono
Noise inserts white noise (hiss) in place of sound
Pan moves sound around the stereo stage during sound
Pitch Bend modifies pitch during sound
Resample changes sample rate (or resolution)
Reverb similar to Echo
Reverse same as Backwards
Robotise removes tonal components
Silence inserts silence in place of sound
Smooth removes spiky components
Stereo converts mono samples into stereo for later processing
Upsize increases sample resolution from 8-bit to 16-bit
Waveform inserts fixed frequency tone in place of sound
QuickTime
wwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwww
  Some sound applications gain extra features when QuickTime is installed. They
can then open QuickTime movie soundtracks or audio CD tracks. They can also
import sounds from generic sound documents or from MIDI files and picture files
— and then convert them into a movie file.
√π See the Music chapter for more on MIDI and musical sequences
√π See the Movies chapter for more about QuickTime
Sound Recording Applications
wwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwww
  With a suitable application you can create hi-fi sound recordings on your Mac,
assuming its hardware is capable of high-quality sound (see below).
16-bit recording, sampled at 44.1 kHz, requires around 5 M of hard disk space for every minute of sound — that’s about 75 K per second. For stereo multiply this by two; for a multi-track application (see below) multiply it by the number of tracks!
å For serious audio work you’ll need several gigabytes (G) of disk storage and lots of
memory to store sounds whilst they’re being recorded or edited.
Hard Disks and Audio Recording
  A hard disk used for audio recording can easily become fragmented — data for
each sound file gets scattered all over the disk. During playback it’s crucial that
the mechanism can read audio data from the disk in time — if a file’s badly
fragmented this simply isn’t possible.
Fragmentation can be avoided by regular formatting, which erases the disk, or by disk optimisation. The latter transfers all data via the computer — so a back-up is vital.
√π See the Hard Disk chapter for more about Optimisation
§ The number of sound files on a disk should be kept to a minimum.
§ Avoid using one drive for both audio and non-audio files.
Multi-track Recording Applications
wwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwww
  In a multi-track application layers of sound are built up as with a multi-track tape
machine. Each track is usually recorded, or laid, one at a time — or two at a time
if you only have stereo inputs!
During playback the sounds from each track are mixed together as required. The application can drop into overdub mode at a set time, so allowing extra material to be recorded onto any tracks that are enabled for recording.
Most Macs need a digital audio card, complete with extra audio inputs and outputs. The Audiomedia II card fits into a NuBus slot whilst the Audiomedia III fits into a PCI slot in a PowerMac — the latter can record four tracks simultaneously and provides eight outputs.
AV models and PowerMacs can support multi-track applications directly, subject to the limitations of RAM and drive speed:-
Model Max No of Tracks
Quadra 660AV 6
PowerMac 840AV 8
PowerMac 6100/66 8-10
Faster machines let you use up to 32 tracks!
Deck II is an example of a multi-track application. It can be synchronised with Metro, its companion MIDI sequencer application , or with any QuickTime movie.
√π See the Music chapter for more on MIDI and musical sequences
The Mac’s Sound Hardware
qqqqqqqqqqqqqqqqqqq
Inputs and Outputs
wwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwww
  Most modern Macs use 3.5 mm 3 pole (stereo) mini jacks for their input and output
connections — whether the machine actually works in stereo or not! Older Macs
use the mono version of this connector.
å Don’t put a mono plug into a stereo socket — it shorts out one channel but may
prevent both channels from working.
å The output of 1 volt peak-to-peak from a source impedance of 47 ohms is intended
for Walkman-style headphones. To connect it to the line level input of a hi-fi system
requires an adaptor from a stereo mini jack plug to RCA phono (PIN) plugs.
å The input accepts any suitable microphone fitted with a stereo plug. To connect it
to a line level source from a hi-fi system you’ll need an adaptor with an attenuator.
PlainTalk-compatible inputs don’t need an attenuator (see below).
Some Macs don’t provide a stereo output — on these machines the left- and right-hand channels from a sound file are automatically mixed to create a mono signal.
PlainTalk Microphone Inputs
  Some Macs can use a PlainTalk microphone instead of a simple dynamic
microphone. The PlainTalk version comes with an extended plug incorporating a
fourth contact at the tip that’s used to connect a 5 volt supply to the microphone.
Not all Macs have the necessary extra contact inside their input socket — so they can’t be used with a PlainTalk microphone! The PlainTalk device includes its own amplification that boosts the output to nearly line level. This is matched by the low input sensitivity of a PlainTalk-compatible computer, making such machines unsuitable for use with the old-fashioned dynamic microphones that were supplied with earlier Macs.
Hardware Limitations
wwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwww
  The sound quality of a Mac depends on its ADC and DAC hardware. Older models
are limited to 8-bit operation, sampled at 22 kHz — this gives a rather poor
frequency response of around 7 kHz and a dynamic range of 70 dB.
AV Macs and the 7500/8500 PowerMacs provide high-quality inputs and outputs with 16-bit sampling at 44.1 kHz — this is optional on 6100, 7100 and 8100 PowerMacs.
The hardware in some Mac’s may produce nasty clicks, although this can be minimised by using the latest system software. Other clicks may originate from inside a particular sound file. To determine if a click’s in a file just play it repeatedly — if it clicks every time, the problem’s in the file and can be removed by means of a sound utility. You can either edit out the click or introduce a short fade at the beginning or end of the sound.
§ Sounds may not record properly with older versions of HyperCard on a PCI PowerMac.
— just record your sounds with SimpleSound and then import them into HyperCard.
§ Only Macs with an Apple Sound Chip can play sounds directly from a disk drive.
§ With some software combinations on a PowerMac you may have to turn off speech
recognition before recording a sound.
§ Not all PowerMacs can record compressed samples, even though the system indicates
that it’s possible — and some models may not support Automatic Gain Control (AGC).
Sound in Detail
qqqqqqqqqqq
Sample Rates
wwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwww
  The higher the sample rate the greater the frequency response — meaning better
reproduction of high-pitched sounds. Most Macs can work with sample rates of
up to 65,535 Hz. Apple’s preferred rates are:-
Sampling Rate (Hz) Abbreviated to (kHz) or (kHz)
5563.6363 5.5 5
7418.1818 7.4 7
11,127.27272 11.1 11
22,254.54546 22.2 22
44,100 44.1 44.1
Of these, 11 and 22 kHz are the most common rates but for high-quality sound you must use 44.1 kHz, the rate used by audio CDs. Other rates that you may encounter include:-
Sampling Rate (Hz) Abbreviation (kHz) Purpose
4000.0000 4
8000.0000 8
11025.0000 11 Multimedia CD-ROMs
22050.0000 22 Multimedia CD-ROMs
24000.0000 24 Voice Recognition
32000.0000 32 Broadcasting Links
48000.0000 48 Digital Audio Tape (DAT)
64000.0000 64
å Multimedia CD-ROMs often provide 16-bit quality, sometimes 8-bit.
å If you sample a sound whose frequency is higher than the sample rate you’ll hear
unpleasant sounds with a frequency equal to the difference between the sound’s
frequency and the sample rate. This effect is known as aliasing.
($2B77.45D1). Some applications ignore this and produce sounds of the wrong pitch!
In some applications you can adjust the sampling rate during playback, usually by reducing playback speed in the ratio of 1:2, 1:3 or 1:5. A lower rate makes the sound deeper and increases its playing time. Raising the speed increases pitch, makes voices sound like chipmunks and decreases the playing time.
The Sampling Process
wwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwww
  The number of bits used for a sample sets the resolution or quality of the sample.
16-bit sampling gives very high-quality when used at a fast sample rate. 8-bit
samples often sound awfully granular but the files only use half the disk space!
Each sample is represented by a sound frame, usually in one of the following types:-
Type Size (bytes) Format
8-bit mono 1 Single byte for sample
8-bit stereo 2 Byte 1 = left channel Byte 2 = right channel
16-bit mono 2 Two bytes for sample
16-bit stereo 4 Two bytes for each channel
Some utilities can increase resolution from 8 to 16-bits — the original 8-bit data is used for the most significant 8 bits of the 16-bit sound and the remaining bits are set to zero. This doesn’t improve sound quality but makes it easier to mix with other 16-bit material.
File Compression
wwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwww
Data compression can be used to save disk space. Some applications don’t accept all kinds of compressed files or 16-bit stereo sounds. Apple uses lossy algorithms that work by discarding some sound data — so you’ll hear the effects! 3 :1 and 6 :1 compression is used for 8-bit material and 4 :1 for 16-bit sounds.
A 16-bit sample at a rate of 7 kHz with 4:1 compression occupies a similar amount of disk space as an 8-bit, 22 kHz, 6:1 sample — but has a much greater dynamic range.
Sound Documents
qqqqqqqqqqqqq
A sound can be stored either in a file’s data fork or as a resource in the resource fork. Apart from system sounds most files only use the data fork.
  You can use a utility such as SoundApp, SndSampler or Sound Mover to convert
between various types of sound file or snd_ resources.
The following documents may be encountered — their Type codes are in brackets:-
System Sound (sfil)
  A system sound file includes any sound in the System file, or any alert you’ve
recorded using the Sound control panel. They have a Creator code of movr and
contain a single Type 1 snd_ resource.
§ A system sound file always plays when you double-click on it in the Finder — other
files don’t! They may produce a continuous loop of sound or a musical sequence when
played from the Finder, but not when used as a system alert!
§ System sound files can be placed in the Startup Items or Shutdown Items folders.
Other sound files can’t be used this way.
Audio Interchange File Format (AIFF)
Apple’s format for any number of sound channels at any sampling rate, using signed 16-bit samples in the data fork. Sound duration is only restricted by the maximum permitted file size accommodated by the system — the current 2 G limit permits over three hours of high-quality stereo sound. AIFF files are similar in size to system sounds.
Audio Interchange File Format – Compressed (AIFC)
Similar to AIFF with compressed samples using MACE 3:1 or 6:1 algorithms. This is a lossy process that removes some detail from the sound — the results are debatable but it saves lots of disk space. ACE 2:1 and 8:3 compression is only used on the Apple II GS computer.
CD Audio Track
As found on any audio CD. They can be extracted using a suitable application in conjunction with QuickTime. Each track can be converted into a movie and then into another format — but each track often uses a huge amount of disk space!
DVI ADPCM
These files use an unusually fast form of Adaptive Differential Pulse Code Modulation (ADPCM) employed by Intel. Lossy 4:1 compression is used in a 16-bit format and the sounds are often sampled at 8 kHz.
FSSD (FSSD)
Used on the Mac prior to the introduction of system sound files. They’re also called SoundCap, MacNifty, SoundMaster, SoundEdit or SoundWave files by virtue of associated hardware and software.
z SoundCap, MacNifty or SoundMaster
A mono format first used with the SoundCap audio digitiser. An uncompressed file
consists of 8-bit unsigned bytes in the data fork. The compressed version uses the
Huffman algorithm and includes a checksum together with data about the sample
rate — they’re known as HCOM files since they start with these four characters.
Sounds are sampled at 5.6, 7.4, 11.1 or 22.2 kHz.
These files have a Creator code of FSSC.
z Studio Session Instrument
Similar to the SoundCap files but with an extra eight-byte header. They’re mainly
used with Super Studio Session for storing sampled instruments.
z SoundEdit
Originally the same as an uncompressed SoundCap file, but with extra information
in the data fork giving details about format, sample rates, looping segments,
colours and labels.
In a stereo file the left and right-hand samples are kept adjacent in the data fork.
SoundEdit Pro and SoundEdit 16 can sample at up to 48 kHz with 16-bit resolution.
IFF
Used in the Commodore Amiga computer for mono 8-bit sound at any sampling rate. Samples are encoded as signed values with optional lossy 2:1 compression using the Fibonacci delta compression algorithm.
Àù These files usually have a filename extension of .8SVX.
MIDI (Midi)
Contains a sequence of musical notes without any digitised sounds for the instruments. To hear a General MIDI file on the Mac you must first convert it into a QuickTime movie. A suitable QuickTime application, such as SimpleText, can then play the sequence from the movie. Both the QuickTime‚Ñ¢ and QuickTime‚Ñ¢ Musical Instruments files must be in the Extensions folder (in the System Folder) for music playback via a QuickTime application.
√π See the Music chapter for more on MIDI and musical sequences
MOD (STrk)
  Used by the SoundTracker application on the Commodore Amiga computer for
storing a musical sequence, including the associated digitised instruments. Long
compositions only need a comparatively small file.
Each file can accommodate up to 32 separate sound tracks using 32 instruments with up to 128 patterns and positions. It can use musical notes in the range C2 to B5 and stores the song name as text within the file.
MOD files are used by Mac applications such as ModPlayer, Player PRO, MacModPro and Sound-Trecker. They’re often downloaded in a LHA compressed format that has to be decompressed using MacLHA. Although the resultant files don’t have any File or Creator codes you can easily add these using FileTyper, Snitch or a similar utility.
If you can find an application that opens these files you may be able to convert them into Standard MIDI Files!
The Sound-Trecker plug-in for the Mac supports the following formats:-
Amiga MED/OctaMED: 4 to 16 track MMD0/1/2 formats
Amiga SoundTracker
Amiga StarTracker: 4 and 8 track
IBM FastTracker: 4, 6 and 8 track
IBM TakeTracker: 1 to 32 track
StarTracker
NoiseTracker
Oktalyzer: 4 to 8 track
ProTracker: 4 track
There are also MTM and S3M formats. Slower Macs may have problems playing MOD files with more than four tracks.
√π See the Music chapter for more on MIDI and musical sequences
PSION sound file
Used in the Psion Series 3 personal organiser. The file starts with a short header followed by string of a-law encoded samples at 8 kHz.
Àù These files usually have a filename extension of .WVE.
QuickTime Movie (MooV)
Apple’s format for fast-moving multimedia material, including any combination of movies, sounds or musical sequences. To play a sequence in a QuickTime MIDI file you must first place QuickTime™ and QuickTime™ Musical Instruments files in the Extensions folder (inside the System Folder).
MIDI data can’t be converted directly into a sound file using QuickTime.
Àù Sound2Movie can convert a sound resource into a movie or vice versa.
√π See the Music chapter for more about MIDI and musical sequences
√π See the Movies chapter for more about QuickTime
Sound Blaster
Format for hardware of the same name as used in a PC. The sample rate is a multiple of the clock rate in the hardware and samples are encoded as signed values. Sounds can be segmented and looped, or silent portions can be added.
Àù PCs use a filename extension of .VOC.
Sound Designer II
As used in the professional Mac sound editor of the same name. These can use any number of channels, rates or bits. Samples are encoded as signed values with extra details about the sound stored in three STR_ resources.
Sun Audio and NeXT
For any number of channels, rates or bits using linear or logarithmic (log) encoding. An 8-bit sample with log encoding gives the same dynamic range as a 12-bit linear sample — but may suffer from noise problems and the files can be slow to decompress.
When a log file is converted into another format its dynamic range exceeds that of an 8-bit system. Even with a conversion factor you’ll get quiet portions or clipping distortion in loud passages. You can avoid both problems by converting the file into a 16-bit format.
US telephone systems use µ-law encoding whilst Europe employs a-law. A filename extension of .al indicates a-law encoding unless shown otherwise by the header. Some older Sun Audio files may lack a header (or it may begin with .snd), in which case most applications assume the file’s sampled at 8 kHz, is mono and is µ-law encoded.
CCITT lossy compression standards such as G.721, G.723-24 or G.723-40 for Adaptive Differential Pulse Code Modulation (ADPCM) may be used for 8 kHz 14-bit samples sent over a 32, 24 or 40 kbit/s data stream.
Àù These files often have a filename extension of .AU.
Text (TEXT)
An ASCII file in which the first line contains the number of samples in the file — the remaining lines each contain an 8-bit sample. The sample rate defaults to 22.255 kHz. This file type is useful for transferring samples between different types of computer.
WAVE
Standard format for sounds on a PC with Windows, accommodating any number of channels, rates or bits. Samples are encoded as signed values and are stored with low-bytes coming first — the latter as required by Intel processors.
Various compression algorithms are used — Microsoft’s version of Adaptive Differential Pulse Code Modulation (ADPCM), with lossy 4:1 compression is most common.
Àù These files usually have a filename extension of .WAVE.
Other Formats
A range of sound files, all with sounds in the data fork, have been used. These include files of Type ASND, DEWF (Jam Session/Studio Session instrument files), GSND, SMSD and WAVE (MacKeyboard).
Sound Resources
qqqqqqqqqqqqq
  Sound resources are used to store sounds inside applications and other software.
The snd_ resource was introduced at the time of the Mac II computer and is most
commonly found in system sound files.
å The _ at the end of snd_ represents a space character.
å The maximum size for a snd resource is 16 M.
Sound Resource Types
wwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwww
A snd_ resource can come in two types:-
z FORMAT 1
Also known as a normal resource. It contains instructions for the system’s Sound
Manager. These can be any combination of sound samples or synthesised sounds
made using frequency-modulation (FM) or wave tables.
z FORMAT 2
For use with HyperCard, this type can only contain a sound sample. It’s really
obsolete and may not sound correctly, especially with 16-bit stereo sounds.
å The Sound control panel uses either type, but HyperCard only plays format 2 sounds.
Sound Resources in Detail
wwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwww
  The way data is stored in a snd_ resource depends on the sample type. 8-bit samples
are kept as unsigned bytes in the same way as FSSD files. 16-bit samples are
signed in the same way as AIFF files — MACE 3:1 or 6:1 compression is optional.
The clipboard always converts and stores every sound as uncompressed format 1— even when you copy a 3:1 compressed format 2 sound.
In some applications you can edit the snd_ resource header. This can set the playback sample rate without touching the actual sound data. The header depends on the sample:-
Header Bits Format Compression
Standard 8-bit Mono No
Extended 8-bit Stereo No
Extended 16-bit Mono/stereo No
Compressed 8/16-bit Mono/stereo Yes
Sounds that are cut or copied end up with the minimum header necessary for the sound. For example, if you paste an 8-bit mono sound with an extended header into a document the final result will have a standard header.
If a 16-bit sound is reduced or downsized to 8-bit, the extended header remains. If an 8-bit sound with a standard header is upsized to 16-bit an extended header is introduced.
Application Sound Resources
wwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwww
Many application files use snd_s or other resources. Here are some examples:-
