Sound in Computing -- A Technology Overview

Logitech Inc.
November 1993

-----------------------------------------------------------------

Overview

Music, sound, and speech are no longer locked up in arcades, and they are
no longer strictly the domain of gamers. PC sound in computing has enjoyed
tremendous market success and technical advancement over the past few
years. More PC users are upgrading their systems with audio abilities, and
more PC software programs are adding stereo soundtracks to presentations
or spoken comments to spreadsheets and documents. At the same time, there
has been increasing demand from the gaming community and the multimedia
community for quality and diversity in both hardware and software.
Manufacturers tapping into this market have caused a sound explosion of
affordable product choices with increasing levels of quality,
compatibility, and sophistication. Whether it is used in business
multimedia or educational programs, games or music software, audio is
bringing applications to life in a way that redefines the computing
experience.

Even for basic audio applications, sound is still very much an add-on
feature in the PC market. Most IBM-compatible PCs come with an extremely
poor-quality internal speaker that's positioned ineffectively and has no
volume control. Originally, the only sound produced on an IBM computer was
"beep," used for boot-up and error messages. Later "chirps and squawks"
were added to support computer games, then rudimentary sound cards were
included in many systems.

There are several different approaches to adding sound to the PC. Sound
boards play a critical role by providing PCs with the capability of
recording and playing back sound. These boards are actually cards that fit
into an empty expansion slot. Other solutions come in the form of
easy-to-install external devices that plug into parallel ports instead of
requiring you to install the card in the PC case. Adding sound to personal
computer applications has never been easier thanks to the introduction of
Windows 3.1, which provides an application programming interface (API) for
producing digitized sound and music output at the system level.

The Basics of Sound

Generally speaking, there are two basic ways to make sound or music with a
computer: digitizing and synthesizing.

The digital audio section of a board accepts input from an analog source
(e.g., a microphone or a tape player) and turns it into a digital file
stored on the hard drive. The file can then be played back, with the board
turning the digital data back into analog signals that can go to
headphones, an amplifier, speakers, or any other device that accepts audio
input. Two factors determine how accurately sound is reproduced; sample
rate and resolution. The sample rate, measured in KHz (i.e., thousands of
cycles per second,) determines how many times per second the sound board
samples the incoming audio signal and digitizes it. Resolution refers to
the number of bits used to store each sample.

In more technical terms, special purpose integrated circuits, called
analog-to-digital converters (ADCs,) convert analog sound waves into
digital bits that the computer can understand; digital-to-analog
converters (DACs) change them back again for human ears. The quality of
the sound is controlled by the bit resolution of the samples (16 bit for
compact disc) and the recording frequency (44 KHz for compact disc.) Sound
recorded at 11 KHz (capturing 11,000 samples per second) is by definition
fuzzier than audio sampled at 22 KHz or, better yet, 44 KHz. Most audio
boards capable of dealing with digitized sound use 8-bit resolution, and
audio is generally recorded anywhere between 5 and 22 KHz. This does not
mean that the board fits into an 8-bit expansion slot instead of a 16-bit
slot, but that it uses 8-bits to digitize a sound sample. That translates
to 256 possible digital values, or substantially lower fidelity than the
over 65,000 values allowed by a 16-bit audio board. (8-bit audio is
adequate for speech; 16-bit sound is superior for music.) Whether or not a
user should choose an 8-bit or 16-bit sound solution depends on the
application's requirements for sound quality and the user's budget. The
advantage of digitized sound is that you can simply record whatever you
want. The disadvantage is the space requirements.

Synthesized sounds, as their name implies, are created, not merely played,
by the computer. The musical results range from simple to sophisticated.
The term synthesis refers to the technology of simulating the sounds of
various musical instruments. There are different types of synthesis, but
the input is always the same. Peripherals with a Frequency Modulation (FM)
chip are a popular solution for producing synthesized sound. FM synthesis
was built on mathematical algorithms that try to artificially create what
a sound wave looks like. Yamaha's FM chips are popular choices to generate
mono or stereo sounds.

A newer form of synthesis, called wave table synthesis, is leading edge
technology that uses true instrument recordings and offers superior
quality over FM synthesis. Instead of starting with an equation, or
computer generated sound, it starts with a very small actual recording of
an instrument. For example, with a piano it could start with a short
recording of middle 'C', and a 'C' octave above, and then it builds from
that, using algorithms to interpolate the rest of the notes in between.
Wave table synthesis is now the highest quality available in a sound
card.

Most wave table synthesizers are general MIDI (Musical Instrument Digital
Interface) compatible, meaning that they follow an established MIDI
standard. MIDI is a protocol for the interchange of musical information
among musical instruments, synthesizers and sound boards. Many cards, like
Logitech's SoundMan Wave comes bundled with software titles that allow you
to record, edit, and play with MIDI music.

Logitech is one of the first developers to write directly to the general
MIDI standard through the MPU-401 interface. This is the new DOS standard
emerging for synthesized sound. This allows game players to immediately
access the latest technology and hear wave table quality in some of the
hottest games today, such as X-Wing, 7th Guest and Space Quest 5. The
advantage of a MIDI file is the tremendous amount of space that is saves.
Game developers use synthesized music files because it allows them to
include a lot of music in their games without using up a lot of space on
the diskette.

Sound Choices

The idea of using frequency modulation to synthesize instrument sounds was
developed by computer music researchers at Stanford University in the
mid-1970's. Although they knew that FM synthesis could produce a wide
range of complex musical tones, their computers couldn't perform the
necessary mathematical operations fast enough to create music. The process
soon captured the interest of engineers at Yamaha, the electronics,
industrial, and musical instrument giant. Yamaha had the know-how to
create a single electronic chip that could perform all of the computations
needed for FM synthesis, and to make it inexpensive enough to use in
moderately-priced musical instruments. The first generation of sound cards
carried the Yamaha OPL-2. The OPL-2 offered 11 voices but only two
operators. Later the OPL-3 had the technology for 20 voices and stereo
outputs, and can combine up to four operators per voice, while still being
able to emulate the two operator chip. The latest generation of Yamaha
chips is the OPL-4 which features FM synthesis, wave table synthesis, and
general MIDI compatibility all on the chip. If you have a sound card in
your PC, it most likely has a Yamaha chip. These chips give you AdLib
compatibility allowing you to hear encoded music in your PC games.

The first major player in the sound market was AdLib. AdLib was the first
with a sound board that had a synthesizer on it. It allowed the gaming
market to add music, but the card didn't have any digital audio
capabilities, so voices could not be recorded. The next generation brought
Creative Labs' SoundBlaster. They took the same card and added digital
audio capabilities. This allowed game manufactures to include both music
and pre-recorded sound effects in their games This set the standard for
all future sound cards.

The increasing demand for sound quality from both the gaming community and
the new multimedia market have made the PC sound industry explode. AdLib's
standard Yamaha FM Synthesis and SoundBlaster compatibility are still the
most important features of any sound card, mostly because game
manufacturers still write to that code.

The Complete Sound Solution

If you hope to compose a multimedia masterpiece, or use your card for
business applications, you will want a sound device that pays as much
attention to input (recording) as to output (playback). At minimum, a
credible board should include a microphone input jack, output jacks for
headphones, software for recording and manipulating sound files, and
Windows 3.1 support. MIDI ports and sequencing software are added bonuses.
Another consideration that users look for in sound cards is an interface
that allows you to hook up a CD-ROM drive.

As the industry moves to the next generation of sound in the DOS world, you
will find Logitech on the leading edge. By choosing the Yamaha OPL-4 chip
for their newest 16-bit stereo card, SoundMan Wave, they are able to offer
the highest quality and most realistic sound available today with 44-wave
table and FM voices; full SoundBlaster and AdLib Compatibility; 16-bit or
8-bit stereo, or mono digital audio; 44 KHz sampling; a MIDI interpreter
for general MIDI/MPU-401 compatibility in DOS; a five-channel stereo
mixer; and a SCSI CD-ROM interface. They have eased the installation
process by adding sound and graphics to the installation process, and
support both Windows 3.1 and DOS, with separate controls and drivers for
each. Both drivers talk to each other for equal settings in both
environments.

Logitech first entered the sound market in 1992 with the introduction of
AudioMan. AudioMan is a business audio product that has been winning rave
reviews for its intelligent design and portability. Designed for the
corporate market, AudioMan is compact and portable, and makes it easy to
add sound effects, music or spoken messages to presentations,
spreadsheets, and E-mail. Recently upgraded, AudioMan now incorporates
true text-to-speech software for PC users.

SoundMan 16 was first introduced in February 1993. Targeted for the retail
and home recreation market, this product rounded out Logitech's Senseware
family of products with a high-quality, 16-bit stereo sound board. An
upgrade to SoundMan 16 was announced in September, with added features
such as improved user installation with sound and graphics, making it even
easier for users to add sound cards to their systems, and new software
bundles including; MCS MusicRack, Icon Hear-It, ScreenCraze II and
AddImpact. Also in September 1993, Logitech increased its sound product
offerings with two new additions to the SoundMan product family.

SoundMan Games, is an 8-bit stereo sound card specifically developed for
newcomers to the PC game market. This product offers high-quality sound
and full SoundBlaster Pro and AdLib compatibility, and a Sony CD-ROM
interface.

SoundMan SuperPack, designed to offer a complete sound solution for games
and multimedia users, ships with the high-quality SoundMan 16 sound board,
plus Labtec speakers and microphone, and several software packages.

SoundMan Wave, a 16-bit stereo card based on wave table synthesis is the
most recent addition to the SoundMan family.

Logitech, the leader in "Senseware," designs products to improve the
interface between the user and the computer. Logitech offers the complete
sound solution for entertainment and multimedia. The SoundMan and AudioMan
product lines are high-quality sound products that are easy to install and
come packaged with extra software titles. Logitech sound products support
the most common sound standards AdLib and Sound Blaster. These products
are not the last in the sound category. You will be hearing more from
Logitech.

-----------------------------------------------------------------

GLOSSARY OF TERMS

Analog: An audio signal is an electrical replica, or analog, of the
waveform of the sound it represents. The voltage of the signal varies up
and down (negatively and positively, in electrical terminology) the same
way as the sound pressure varies up and down at the microphone. As long as
the signal is in this form, i.e., is a voltage that varies directly with
the sound pressure, it is an analog, and audio devices which use such
signals are analog devices.

Analog to Digital Converter (ADC): A converter that turns sound frequencies
into digital information.

Digital to Analog Converter (DAC): A device which converts digital
samplings into an analog signal.

Digital Signal Processor (DSP): A device which can be used to manipulate
digital signals, providing special effects, real-time echo, harmonics,
etc.

Digitize: The process of converting an analog value into digital (numeric)
representation of that value. Example: If an analog input ranged from 0 to
a peak of 6.2 volts, an input value of 4.5 volts would be digitized to a
numeric 186, under an 8-bit sample resolution, (256 x 4.5 / 6.2.)

Direct Memory Access (DMA): A feature which allows a device to access RAM
independently of the microprocessor, thereby achieving better, more stable
performance, and reducing the load on the microprocessor.

Dynamic Filtering: Helps eliminate electronic emissions from the PC that
can show up as noise in the sound board output.

Frequency Modulation Synthesis (FM Synthesis): A technique for producing
sounds by modulating sine waves to produce harmonics.

Interrupt ReQuest (IRQ): A mechanism whereby a peripheral can notify the
microprocessor of some event. The processor would be interrupted to
perform some service for the peripheral.

Musical Instrument Digital Interface (MIDI): A protocol for the interchange
of musical information among musical instruments, synthesizers and sound
boards.

MIDI Synthesizer: A device producing music according to a MIDI stream.

MPU-401: Popular MIDI interface standard for musical instruments.

OPL-2: OPerator type L2; a Yamaha FM Synthesis chip.

OPL-3: Operator type L3; a Yamaha FM Synthesis chip, backwards compatible
with the OPL2, but with twice the capabilities.

OPL-4: OPerator type L4; a Yamaha FM Synthesis chip backwards compatible
with the OPL3 with on-chip Wave Table synthesis. FM synthesis and Wave
Table synthesis one chip.

Pulse Code Modulation (PCM): Refers to any type of digital encoding and
decoding of a signal.

Sampling Rate: The frequency at which analog samples are converted to
digital values during digitization. The higher the rate the more accurate
the sample. Speech is typically sampled at 7000-8000 samples per second,
CD quality audio is sampled at about 44,000 samples per second, (44 KHz.)

Small Computer Subsystem Interface (SCSI): A bus-type connection that can
be used to connect computers with peripheral devices; hard disks, CD-ROM
drives, scanners, etc.

Synthesis: A term that refers to the technology of simulating the sounds of
various musical instruments.

Terminate and Stay Resident (TSR): Programs which remain resident in
memory, such as hardware "driver" software. Excessive TSRs are likely to
interfere with games.

Wave Table Synthesis: Incorporates digitized samples of instruments stored
in ROM to simulate the sounds of various musical instruments.
-----------------------------------------------------------------
Logitech Inc, 6505 Kaiser Dr, Fremont, CA 94555
510-795-8500;  fax: 510-792-8901

 ============================================================
 From the  'New Product Information'  Electronic News Service
 ============================================================
 This information was processed from data provided by the
 above mentioned company. For additional details, contact 
 the company at the address or telephone number indicated.
 OmniPage Pro is now used for converting all printed input! 
 ============================================================
 All submissions for this service should be addressed to:
 BAKER ENTERPRISES,  20 Ferro Dr,  Sewell, NJ  08080  U.S.A.
 Email: RBakerPC (AOL/Delphi), rbakerpc@delphi.com (Internet)
 ============================================================
