The Addicted To Audio Dictionary

Aren’t quite sure what “VTA” means? What is a “2.5-way” loudspeaker anyway? Well, you’ve come to the right place to understand a wide range of terms used in the high fidelity community. Just click on the letters below to jump to the right section of the dictionary. Or press Ctrl-F/Cmd-F to use your browser’s search feature.

This dictionary has more than 470 entries.

1 A B C D E F G H I J K L M N O P Q R S T U V W X Y Z

1
1/4” jack: See 6.35mm jack.
1/8” plug: See 3.5mm plug.
2.5mm jack: The 2.5mm TRRS jack has been a long-time standard for the balanced headphone connection on portable digital audio players. This connection is gradually being displaced by the 4.4mm jack. The main disadvantages of the 2.5mm plug are that it has no provision for shielding/ground independent of any of the signal wires and has a fairly small contact area for each of the four contacts.
2.5-way loudspeakers: A loudspeaker which divides the incoming signal into three different frequency bands for distribution to drivers, but in a different way to three-way speakers. It sends high frequencies to the tweeter in the usual way, and low frequencies to one or more woofers. But it sends the midrange frequencies and the low frequencies to an additional bass/midrange driver in the same way as is done in a two-way speaker. In other words, all but one of the large drivers handle bass only, while the last handles both bass and midrange.
33 1/3 RPM: The rotational speed of the great majority of vinyl LPs.
3.5mm plug: Or 1/8” plug. Often used on portable headphones and playback equipment and were ubiquitous on smart phones until they started abandoning analogue connections entirely. 6.35mm connections are generally to be preferred since they offer a larger electrical contact area, and are thus more reliable.
3-pin XLR: This is the standard form of XLR connection, frequently used for balanced connections.
4.4mm jack: Sometimes known as 4.4mm Pentaconn, the 4.4mm jack is a connection standard primarily for balanced headphones. It has only been around for a relatively short time, but looks set to replace the bulky 4-pin XLR, the 2.5mm TRRS jack with its small contact area, and various cumbersome 2-jack connections. As the name implies, the plug diameter is 4.4mm, and it is in the TRRRS configuration. That is, there are a total of five conductors supported, with the sleeve acting as the ground for any cable shielding, and the other four for the plus and minus for the left and right channels.
45 RPM: The rotational speed of the great majority of 7-inch vinyl “singles” and EPs.
45 rpm spindle adapter: A disc about 38mm in diameter with a hole in its centre. It slips over a turntable spindle, allowing 45 rpm records with large centre holes to be played. The reason some 45 rpm records had a larger centre hole (1.5 inches) was due to a short-lived format war when vinyl recordings were launched in 1948. Columbia's records ran at 33 rpm, while RCA Victor's were 45 rpm. 45 rpm records released in Australia rarely had the larger hole (or a punch-out section to turn it into a large-holed disk.)
4-Pin XLR: The 4-Pin XLR plug is a fairly frequently used connector for balanced headphones. Contrary to the usual XLR conventions, the signal does not flow from source to receiving device in the direction of the pins, but the other way. In other words, the male plug is on the headphone cable while the female socket is on the panel of the headphone amplifier. This connection is gradually being displaced by the 4.4mm jack.
5.1: The most common form of multichannel sound. The "5" refers to full range channels: moving clockwise, left front, centre front, right front, right surround and left surround. In Dolby Digital, DTS and MPEG 5.1 formats, each of these five channels is capable of a frequency response covering the full audible range (up to 20,000 hertz). The “0.1” refers to the LFE channel, which is not full range but designed to cover up to 120 hertz, thus the decimal point.
6.1: An enhancement of the 5.1 channel surround system, versions of this appear in both Dolby Digital and DTS. The Dolby Digital version is called Dolby Digital EX 6.1 while the DTS version appears as either DTS ES 6.1 Discrete or DTS ES 6.1 Matrix. The additional channel is intended to sit at the rear of the room (although two speakers are recommended, even though the same signal is provided to both). This provides a greater localisation of sounds from the centre rear. This is a very useful enhancement in cinemas where much of the audience are sitting off-centre, but in normal rooms with a small number of viewers, it is much less important.
6.35mm jack: Or 1/4” jack. Most familiar from the type of headphone output traditionally used on amplifiers, these are widely used in professional audio as well. They come in both mono (two conductors) and stereo (three conductors) versions. The latter type are often termed in pro-audio as TRS jacks. In that application they are often wired for balanced mono operation.
7.1: Well, after 5.1 and 6.1, 7.1 became kind of inevitable. Instead of having just one surround rear channel, 7.1 channel sound has two discrete surround back channels. It’s available in advanced surround formats provided on Blu-ray: Dolby TrueHD, Dolby Digital Plus, DTS-HD Master Audio and DTS-HD High Resolution.
78 RPM: The rotational speed of the long-standing 10-inch (sometimes 12-inch) shellac record which was the norm for music distribution through the first half of the 20th Century. This was the first standardised recording speed.

A
AAC: Advanced Audio Coding. This is a lossy codec for stereo audio, standardised under the MPEG-2 and MPEG-4 specifications. It is widely used, most famously for Apple products, including its streaming service, Apple Music, and its music store. Apple devices also use AAC for sending stereo music to Bluetooth devices, where the device supports it. Most portable digital audio players support AAC. Originally Apple used digital rights management (DRM) controls to stop the copying of files or using them on other devices, but it stopped doing that around 2009.
AC-3: AC-3 is the encoding scheme used in Dolby Digital, the name by which it is now more commonly known.
AC: Alternating Current. An electric current that reverses direction regularly. More generally, AC is also used to describe voltage sources in which the polarity of the signal reverses regularly. Power distribution networks use AC because it is relatively easy and cheap to alter the voltage (using transformers) yielding considerable economies for long distance power transmission. All the counties in the world use, it seems, either a 50 or 60 hertz frequency for their power systems. Europe, Australia and India use 50 hertz. The Americas and Japan use 60 hertz. Compare with DC.
Academy sound: Fairly soon after the introduction of talkies in the late 1920s, the movie industry settled on using an analogue optical soundtrack on the edge of the film to carry the sound. This was read by the projector and converted to an electrical audio signal. This was a very noisy medium, so in the absence of advanced noise reduction technology such as Dolby A (the cinema equivalent of, and precursor to, the Dolby B and C noise reduction systems on compact cassettes), the situation was eased by sharply cutting the treble response (at around 5kHz). The bass was also cut, which accounts for the characteristic pinched sound of old movies. The recommended system of frequency response tailoring was called “Academy Sound”.
Accurate: (Sound quality description.) This term is almost redundant, since the whole point of high fidelity sound reproduction is to deliver the sound captured on the recording truthfully or accurately. So in a review it can be high praise.
Acoustic suspension: A design for the enclosure of a loudspeaker. With acoustic suspension speakers the enclosure is sealed so that it is air tight, which causes the air within to become a very active part of the woofer’s suspension. This raises the resonant frequency of the driver and lowers its compliance. Acoustic suspension speakers tend to be less efficient than bass reflex designs, and begin fading away their bass at a higher frequency. But the rate at which the bass output reduces tends to be less than an equivalent bass reflex speaker, so they frequently produce greater bass extension. Also called infinite baffle.
Active loudspeaker: A loudspeaker with built-in amplification for all the drivers. Some speakers have an amplifier built-in for bass only, but these are not regarded as active. Most loudspeakers are passive, not active.
Active noise cancellation: Active noise cancellation is a technique used in some headphones and earphones to capture environmental noise with microphones, process it and then feed it into the music signal in such a way as to cancel out much of the noise. See here for more.
Active subwoofer: A subwoofer with built-in amplification. Most subwoofers are active, not passive.
ADC: Analogue to Digital Converter. A component, circuit or device that converts an analogue signal to a digital one, usually to some form of PCM. Compare DAC which does the reverse.
AES/EBU: Audio Engineering Society/European Broadcasting Union. A digital audio communication standard most commonly seen in professional audio applications. Electrically it is only subtly different to the consumer-oriented S/PDIF standard, the main variation being a different method of handling the clocking signal, and it generally being carried by balanced connections. Nevertheless implementations are seen that seem to be fully compatible with unbalanced S/PDIF.
Airy: (Sound quality description.) When the sound reproduction is light and clean, with a sense of space around the instruments in the stereo image. It is often associated with loudspeakers with a wide dispersion of their upper frequencies.
ALAC: Apple Lossless Audio Codec. ALAC is a lossless audio compression system developed by Apple Inc. It supports up to a bit depth of 32 bits and a sample rate of up to 384kHz, although many consumer devices which were capable of 192kHz support for FLAC were limited to 96kHz for ALAC. That is increasingly less of an issue. It also supports tagging, that is, incorporating metadata holding things like genre, album, artist, etc. ALAC is increasingly widely supported, although not quite so widely as FLAC. It is the format used for losslessly compressed tracks available in Apple Music, Apple’s streaming service.
Album: An LP or CD or digital equivalent with a collection of songs. Typically they’re from the same artist and are assembled to work together as a collection. The name derives from the days of 78 rpm recordings. They would generally only hold about three minutes of music on each side. So a collection of songs – the full songlist from a Broadway musical, for example – would be released with six or more records in a photo album style arrangement. The name stuck.
AM: Amplitude Modulation. A method of impressing a signal onto a sine wave for its transmission or storage. A constant frequency sine wave has its amplitude increased or decreased from moment to moment to correspond with the signal. The sine wave, called a carrier, must be of considerably high frequency than any component of the signal. AM radio is an application of this technique. Very simple AM receivers (consisting of little other than a crystal and a coil) can be implemented very easily to receive these signals, although of course more sophisticated circuits can produce higher quality results.
Ambience: (Sound quality description.) The degree to which the sound seems to bring with it the character of the environment in which it was recorded. For example, whether you can sense the size of the recording venue.
Amp: Short for ampere, or for amplifier. In the former sense, the amp is a unit of electrical current. Amps equal volts divided by resistance (or impedance) in ohms.
Amplifier: A component or module of a component that increases the amplitude of an electrical signal. Voltage amplifiers and current amplifiers are optimised to provide amplification for specific purposes. See also power amplifiers and preamplifiers.
Amplitude: The level of an electrical signal, usually measured in volts.
Analytical: (Sound quality description.) This is frequently deployed in reviews as a negative description, suggesting a delivery of music that while accurate and full of detail has somehow lost some of its sense of musicality.
Analogue: As in not digital. An analogue signal in an electrical facsimile of the item being represented. For example, when playing a stereo DVD, the analogue audio outputs of a DVD player trace an electrical curve that is identical, other than in sheer power, to the electrical curve that the amplifier will deliver to the speakers and this, in turn, is identical to the pattern of sound waves the speaker will create, except for distortions that creep into the system. An analogue signal can have any value within set limit, while a digital signal is constrained to a set of discrete values.
Anti-skating: A device on a turntable’s tonearm to counteract skating. This sometimes consists of an adjustable spring-loaded device near the tonearm’s pivot, but is better implemented as a weight on a string (since the torque applied by this arrangement remains constant throughout the range of travel). Why this is needed is explained here.
aptX: aptX is a lossy codec that is primarily used in Bluetooth audio devices as an alternative to the default SBC codec. It offers higher quality audio than SBC, even though it employs a similar bitrate of around 350kb/s (compared to the 1411kb/s of CD-quality uncompressed audio). It supports 16-bit 44.1kHz and 48kHz audio. The codec is quite old, first being developed in the 1980s, and does not use perceptual encoding.
aptX Adaptive: aptX Adaptive is a version of aptX with an improved compression engine, allowing support (although it’s still lossy) of sampling frequencies up to 96kHz. It uses a bitrate of either 279kb/s or 420kb/s, adjusting according the needs of the signal.
aptX HD: aptX HD is a higher bitrate, higher resolution version of regular aptX, although it is still lossy. It runs at 576kb/s (up from 352kb/s for regular aptX) and supports both 16-bit and 24-bit audio at 44.1kHz and 48kHz.
aptX LL: aptX LL is a version of aptX with reduced latency, so as to make it more usable for video and gaming applications.
aptX Lossless: aptX Lossless is an adaptive higher bitrate digital audio codec for use with Bluetooth devices. It will adjust automatically from 140kb/s up to 1Mb/s as required to ensure a reliable connection. When at 1Mb/s, CD-quality audio is transmitted losslessly (if from a lossless source). Users can choose between CD-lossless quality or lossy 24-bit, 96kHz transmission.
Articulate: (Sound quality description.) A sound quality with plenty of detail that allows every element of the music to be clearly followed.
ATRAC: A system for compressing digital audio using perceptual encoding techniques. This was developed by Sony to allow the full contents of a CD to fit onto a Minidisc, which offers considerably less storage space. In recent years a new version, called ATRAC3, has been introduced which permits greater levels of compression than the original version. This permits ATRAC-based solid state players. The bit rates used by ATRAC3 are 132 and 66 kb/s. The ATRAC compression system also forms the basis of Sony’s cinema sound system SDDS.
Attack: (Sound quality description.) An attack is when a transient occurs in the music, such as a clash of instruments or the strike of percussion. A quality system will have all the elements of the sound strike simultaneously, adding to the dynamic presentation. Although not strictly part of the attack, a well controlled system will immediately damp the sound after the attack.
Audiophile: A person who places, or would like to place if circumstances permitted, a high priority on having an audio system that performs very highly. They have generally trained themselves to be very discerning about the sound of audio systems. Some audiophiles stray into a purely subjective realm.
Auditory masking: Auditory masking is a psychoacoustic phenomenon wherein some elements of the sound cannot be heard because they are “masked” by other sounds. The most obvious case of this is when a sound is so loud, other sounds can’t be heard. But auditory masking can take place due to two sounds being close together in frequency or in time. Perceptual encoding in large part depends upon this phenomenon.
A-weighting: A system of adjusting signal to noise ratio measurements to take into account the differing sensitivity of the human ear to different frequencies. Thus an A-weighted signal to noise ratio more accurately reflects how a system’s noise performance will be perceived than an unweighted measure.

B
Baffle: A plate surrounding a driver in a loudspeaker. The lower the frequency of the sound produced by the vibrating cone of a driver, the more apt it is to simply cause air to rush from one side of the cone to the other, rather than produce the compression waves that constitute sound. By adding a baffle around the driver, this increases the length of the path that air must travel, lowering the frequency at which this destructive interference takes place. In most loudspeakers, the enclosure forms a baffle. In the case of infinite baffle enclosures, the enclosure is sealed (thus “infinite”) while with bass reflex speakers a port is carefully tuned to allow energy from the back of the cone to supplement that from the front at selected bass frequencies. In regular talk, the baffle is the front panel of the loudspeaker.
Balance: (Sound quality description.) Generally, tonal balance. That is, the bass, midrange and treble are all at appropriate levels with respect to each other, as are the narrower frequency bands of which each are constituted.
Balanced: An electrical circuit in which both the signal leads (positive and negative, active and neutral or whatever) carry equal but inverse signals produced by the source. These require three conductors: two for the signal plus a separate one for the shielding. In consumer electronics, balanced connections tend to a appear only in high-end gear. Well-designed balanced circuits provide excellent rejection of electrical interference generated in connecting wires. Balanced connections frequently use XLR plugs and sockets.
Balanced armature: Balanced armature drivers are extremely compact sound transducers used in earphones. They use a static coil to induce a magnetic field in a metal "reed" which is otherwise in balance. The reed moves on a pivot, driving a diaphragm. See more here.
Balanced headphones: Balanced headphones are, despite the name, different to other balanced connections. They do not balance the “positive” and “negative” signals for each earpiece. They instead ensure that each earpiece receives a signal entirely separate from that of the other earpiece. To that end they use a connection with at least four conductors, typically 4-pin XLR or 4.4mm Pentaconn, although sometimes they may use two separate 6.35mm plugs. More on balanced headphones here.
Band pass filter: An electrical circuit that only permits signals between two particular frequencies to pass through. An example is section of a crossover network that allows only the middle frequencies to be delivered to the midrange driver. Compare low pass filter and high pass filter.
Bandwidth: Either the range of frequencies which a component can deal with competently (often specified as the range across which the attenuation is no more than 3dB), or the frequency range required to carry a signal.
Bass: Low frequency sounds, typically below around 150 hertz, although the dividing line between bass and midrange is one of opinion. The human ear is less sensitive to bass than to midrange.
Bass extension: An imprecise term concerning how low in frequency a loudspeaker or subwoofer can still operate to produce usable output. A typical bookshelf-sized speaker may manage a bass extension of 80 hertz (say, at -10dB), a good floorstanding speaker may manage 30 or 40 hertz, an inexpensive subwoofer 40 hertz, a middling one 25 to 30 hertz, an expensive one 16 hertz.
Bass management: A facility in home theatre receivers that permits some of the speakers in a 5.1 channel system to be specified as “small” rather than “large”. “Large” speakers receive the entire signal for their respective channel, but “small” speakers have the bass stripped off and sent elsewhere. If a subwoofer is attached, this bass goes to it, otherwise it goes to the front main speakers (you will notice that most systems will not permit you to select “small” for the front stereo pair if you have the subwoofer set to “off”). Some home theatre receivers permit you to choose the crossover frequency for bass management, but many use a standard value of 80 hertz.
Bass reflex: A design for the enclosure of a loudspeaker. With bass reflex speakers the enclosure has a port that permits air to flow between the interior and exterior of the cabinet. The port is a hole, usually backed by a tube. The dimensions of the port are carefully calculated so that it permits bass at a selected frequency to be produced from the interior of the enclosure (driven by the back of the woofer’s cone). This arrangement permits a bass reflex speaker to generally achieve greater efficiency than an acoustic suspension speaker, and it extends the depth at which bass may be produced without significant attenuation. However for frequencies below the band produced by the port, the output drops off quite rapidly.
Belt: Sometimes the loop of material that is used to transmit rotational energy from motor to record platter on a turntable. Here, though, I am referring to a set of extreme audio system tweaks popularised in the 1980s and beyond by Peter Belt in the United Kingdom. These tweaks, for the most part, have no measurable effect on the sound produced by the system and there is no rational reason why they would affect, let alone improve, the sound of the system. They include such ideas as placing an extra sheet of paper in all books within the listening room, to ensure that there are an odd number of leaves in each book, placing a sheet of paper “twixt the floor and just one leg of your listening chair and, more recently, freezing and then defrosting your CDs. Try not to laugh. Some subjective reviewers have, over the years, sworn by these measures.
Binaural: This is a style of recording and playback which attempts to eliminate remove all of the usual environmental influences on the sound. Typically two microphones are used, usually compact microphone capsules placed where the ears would be in or on a head-like object. Each microphone signal forms the entirety of each of the two stereo channels. Binaural recordings achieve greatest effect when played back on headphones. That way each ear is being fed relatively unaltered the signal captured by each ear in the recording venue. Well done, this can be eerily involving. But such recordings are not common since most music tends to be better captured for normal loudspeaker playback with conventional recording techniques.
Bipole: A loudspeaker designed to offer well-dispersed sound by firing its high frequencies, and in some models its full frequency range, in two opposing directions. Unlike dipole speakers, the sound is in phase from all the drivers.
Bit: The smallest unit of digital information. A single bit can carry just one of two values: 0 or 1. There are eight bits in a byte, 1,024 bytes in a kilobyte, 1,024 kilobytes in a megabyte and 1,024 megabytes in a gigabyte. Sometimes, though, the traditional 1,000 is used rather than 1,024, leading to confusion. A bit should generally be abbreviated as lower-case “b” (compared to “B” for byte). Thus 128kb/s means 128 kilobits per second, whereas 128kB means 128 kilobytes.
Bit depth: The size of the number that records each digital sample. Since the system is digital, the number relates to powers of two. The compact disc uses a bit depth of 16, which allows 65,536 different levels to be used to track the analogue source signal. DVDs usually also use 16, but may also use 20 bits (which gives over a million levels) or 24-bits (which gives more than 16.7 million levels). The greater the bit depth, the lower the harmonic distortion and quantization noise, and the more storage space required for the signal.
Bitperfect: Bitperfect describes a state where a source of digital audio delivers the audio to a DAC without any alteration at all. In other words, each digital bit is perfectly delivered. This description became necessary with computer-based audio because, left to their own way of doing things, computers will typically resample digital audio to some other sampling frequency, perhaps adjust the volume level, and almost always mix in other computer sounds, such as notification bells and so on. It is likely that any such alteration will also degrade the quality of the audio. To enable bitperfect audio in a computer, software is required which allows the user to select the audio output device and enable an “exclusive mode” which stops other audio being introduced into the stream. See here for how to ensure bitperfect performance with Windows computers.
Bit rate: The number of digital bits a system transfers per second. In general, the higher the bit rate, the higher the quality of the signal. In every case, the higher the bit rate, the more data space required. With audio, bit rates are measured in the hundreds of kilobits per second (kb/s).
Bitstream: The digital audio output of a DVD player, when switched to outputting the DVD’s native digital audio format. Most DVD players can be switched to output a Dolby Digital bitstream, or convert the digital output to PCM.
Bluetooth: Bluetooth is the ubiquitous wireless system used for transporting audio over short distances (ten metres, but sometimes more), typically from an audio device to headphones or a portable speaker. It also allows control of the source device by the headphones or speaker. The standard codec used for stereo music is a basic one known as SBC (Sideband coding), but many devices support higher quality codecs such as AAC, aptX (and its variants) and LDAC. Both source and receiving device must support one of those higher-level codecs for it to be used.
Blu-ray: A high capacity development of the DVD which uses higher frequency (blue) rather than red light frequencies for reading the disc. The combination of shorter wavelengths and other enhancements bumps up the maximum capacity from 8.5GB for a dual layer DVD to around 27GB, allowing the storage of high definition video.
Boomy: (Sound quality description.) This is when speakers produce an excessive level of mid-bass. Often this is also quite uncontrolled, not well damped by the system, resulting in an overhang in the sound.
Boxy: (Sound quality description.) Usually to do with cheaper loudspeakers where the enclosure itself contributes to the sound in an uncontrolled manner. It adds a mid-and-upper-bass hollowness to the sound.
Breakup: (Sound quality description.) Where the physical limits of a piece of analogue equipment are exceeded you will experience breakup, often in the form of a distinctive scratchiness or buzz or worse. This is usually the case with cone breakup in loudspeakers and mistracking with turntables.
Bright: (Sound quality description.) Where the treble is emphasised by a system above other tonal bands. Sometimes if otherwise well handled this can be an attractive phenomenon.
Buzz: (Sound quality description.) This is a buzzy sound. It may be because of breakup in the equipment or some aspect of the system is emphasising narrow frequency bands in the treble range.
Byte: Eight bits. A byte can represent numbers between 0 and 255, or when interpreted as signed integers, between -128 and +127.
C
Camera Kit: An adaptor for Apple devices such as iPhones and iPads which plugs into their Lightning charge port. The adaptor provides a USB Type-A port, which can then be used for various purposes, such as plugging in USB storage. One use is to connect an external DAC for higher quality audio performance.
Cantilever: The thin rod within a turntable’s cartridge that transmits the movement of the stylus in response to a record’s groove to the interior components of the cartridge that generate the electrical signal.
Carrier: A sine wave which may be modulated by a signal to form an AM signal. The frequency of the carrier must be significantly higher than that of the modulating signal. In practice, the carrier is usually a radio frequency sine wave, and so is two orders of magnitude higher in frequency than the signal.
Cartridge: The device that converts the movements of a stylus in the grooves of an LP to electrical signals. The cartridge is a small, light-weight device, secured to the end of a turntable’s tonearm by means of two screws mounted 12.5mm apart. The movements of the stylus are transmitted through a cantilever to some form of electrical generating device. The two main types of cartridge are ceramic and magnetic. The latter is further subdivided into moving magnet and moving coil types.
CBR: Constant bit rate – as opposed to variable bit rate. The signal (video or audio) is digitally encoded so that a fixed amount of data flows each second. This has the advantage of making the space requirements for the signal easy to calculate. DTS and Dolby Digital are both CBR systems, as are the earlier versions of MPEG audio and video compression. Most MP3 files are CBR encoded, although the format does support VBR as well.
CD: Compact Disc – This is the familiar 120mm optical disc. It carries a digital PCM representation of a two channel analogue signal, along with error correction information. The analogue signal is sampled at 44,100 hertz and uses a bit depth of 16.
CD emphasis: When the compact disc was first developed, the designers implemented a rather surprising element in its specification. This was a pre-emphasis, de-emphasis cycle. In brief, this permits the treble in the source signal to be boosted before the CD is mastered (pre-emphasis), recording this fact by a special bit in the package around each segment of audio data, and cut again by the CD player (de-emphasis). When cut, it also had the advantage of reducing any noise due to the recording medium. But, in practice, there is no such noise, so initially it can be hard to see the purpose of this. Then when you consider that for most music the amplitude of the signal reduces as the frequency increases at around 6dB per octave, you can see that the amplitude of the higher frequency components would be very low and, consequently, subject to increased quantization noise. Boosting the higher frequencies significantly reduces this noise accordingly. The frequency response of the signal is pre-emphasised by boosting the signal from 50µs (microseconds – which is what engineers used to specify frequency in some contexts) or 3,183 hertz, and levels out at 15µs, or 10,610 hertz, with a maximum boost of 10.45dB. CDs that actually use this are rather rare.
Ceramic cartridge: A cartridge that produces an electrical signal through a piezoelectric effect. Such cartridges are rarely used in high fidelity applications because they require a relatively high tracking weight (usually upwards of 10 grams), have a low compliance and produce an uneven frequency response. However they do have the advantage of producing a rather higher output voltage than magnetic cartridges, and their frequency response characteristics approximate the RIAA equalization curve, allowing simpler circuitry to be used with the signal.
Chesty: (Sound quality description.)A somewhat emphasised mid-bass. The name comes from the effect this has on male vocals.
Cinch plug: Another name, used by companies that presumably don’t like to use the opposition’s brand name in their own literature, for RCA plug.
Class A: A power amplifier in which a sufficient DC bias voltage is applied to the power transistors so that the output signal always operates entirely in the positive or negative part of the cycle, entirely avoiding crossover distortion. This makes them quite wasteful of power since even at idle a considerable voltage is being generated. The DC bias is filtered out before being fed to the speakers.
Class A/B: A power amplifier in which a certain amount of DC bias voltage is applied to the power transistors so that, at low power outputs, the output signal operates entirely in the positive or negative part of the cycle, avoiding crossover distortion. Thus, at low outputs, a Class A/B amplifier operates in Class A mode. At higher outputs the signal does cross over the zero point, effectively entering Class B territory. This design is a compromise between the efficiency of Class B amplifiers (in which there is no DC bias) and the elimination of crossover distortion in Class A designs. The DC bias is filtered out before being fed to the speakers.
Class D: A Class D amplifier is a power amplifier technology that is highly efficient, in the sense of turning most of the power it draws from the supply into power output. It is also known as a switch-mode amplifier. The incoming analogue signal is converted to very high frequency pulses. The proportion of the time the pulses are on represents the amplitude of the signal. The output is low-pass filtered to remove the high frequency energy, leaving the original signal, albeit at a higher voltage and with support for enormously higher current supplies to drive loudspeakers. It is sometimes known as a digital amplifier, but that was not its intention, even though in some regards its operation is similar to Direct Stream Digital. Because of the high efficiency, such amplifiers can be lighter and with reduced cooling requirements compared to traditional designs.
Clean: (Sound quality description.) Characterised by a lack of distortion or noise.
Click: (Sound quality description.) A short, sharp unwanted noise, generally during playback of a vinyl recording. It's usually caused by a small scratch on the record surface or some dirt or dust.
Cliff effect: Where the degradation of a signal’s reception does not gradually increase with a reduction in signal quality or strength, but maintains full quality until some threshold, at which point the signal collapses into incoherence. Analogue transmissions tend to degrade gradually. Digital transmissions in modern systems (with error correction built in) tend to maintain full quality, but then cut out completely at the threshold.
Clinical: (Sound quality description.) As with analytical, this is used to describe sound felt as clean and articulate, but lacking some sense of musicality.
Clipping: When the amplitude of a signal reaches some limit determined by the equipment in use, it hits a ceiling (and floor) beyond which it cannot proceed. So the top and bottom of the wave is simply lopped off. The more it attempts to exceed the limit, the more that’s chopped off, and the closer to a square wave the formerly rounded wave begins to look. This causes it to generate lots of harmonics, and so it sounds very distorted. The graphic to the right shows a sine wave at the left, and then the same sine wave amplified by just three decibels, to the right. This relatively minor clipping generates a third harmonic of 14%, a fifth harmonic of 3%, a seventh of 1.8% and so on. Truly awful sounding. Clipping is often caused by turning up an amplifier too loud so that its power limits are exceeded.
Clocking signal: A signal used to synchronise items of equipment which are communicating digital audio or video signals to each other. The lack of a suitable clocking signal would allow their timing to drift apart from each other since their internal clocking signals would generally not be identical, so digital samples would be lost.
Closed-in: (Sound quality description.) The opposite of airy, where the sound seems to lack any atmosphere or space around it. Often associated with speakers that tend to "beam" their sound directly towards the listener.
Cloth ears: A purported characteristic of a person, the possession of which is alleged to account for his or her inability to notice differences in the sound quality produced by different pieces of equipment. Those so accused, assuming they have taken care to listen carefully to the demonstration, should forthrightly reject the allegation.
Coarse: (Sound quality description.) A system that lacks detail, but sometimes seems to attempt to make up for it by adding a grainy texture to the sound.
Coaxial Digital: The digital audio output signal of a DVD player in an electrical format, rather than optical. The data format accords with the S/PDIF specification.
Codec: Compression/Decompression. A system which compresses a signal in some way for storage or transportation and then decompresses it at the point of delivery. Examples are MPEG, Dolby Digital and DTS. These systems use a codec to reduce the amount of data in the signal. Other forms of codec, particularly in the days of analogue audio systems, compressed and then decompressed the dynamic range of the signal, not to reduce the size of the signal but to reduce noise levels. One consumer system was called “dbx”.
Coherent: (Sound quality description.) When all the elements of the sound are brought together by the system into a natural-sounding and detailed "whole".
Colouration: An unwanted alteration in the character of audio. Significant colouration of sound can make instruments and voices sound unrealistic. It may be caused by harmonic distortion, vibrations of component parts (for example, the panels of an inadequately braced loudspeaker enclosure) or, most commonly, through an uneven frequency response.
Comb filtering: When two drivers in a loudspeaker are producing the same tone they will interfere with each other in interesting ways. If the distances between each of them and your ears are identical, their sound will simply sum, making it louder and of a similar balance to that produced by the drivers. But if your ears are at an angle such that the distance between them and one of the drivers is different to the distance between them and the other driver, the sound from each driver may end up somewhat or entirely out of phase with the other. In phase or out of phase is a function of the difference between those distances along with the frequency of the sound. Sound frequency is inversely related to the wavelength of the sound. Since it is frequency-dependent, some frequencies will be enhanced (constructive interference between the two drivers) and some will be reduced (destructive interference). The net result, plotted on a frequency response graph, shows a series of spikes and dips, thus the name "comb".
Compact cassette: A neatly packaged magnetic tape recording and playback system for audio introduced in the 1960s. This uses a narrow plastic tape with four tracks running at a speed of 47.6mm/s (1 7/8 inches per second), contained in a sturdy plastic enclosure. In consumer equipment, only two of the tracks are accessible at any one time, providing stereo recording and playback. The tape is turned over (or an auto-reverse transport moves the recording and playback heads sideways) to access the other two tracks. Some semi-professional four track recording systems allow all four tracks to be recorded and played back at once, or even separately to permit multi-track recording. All cassette decks capable of recording (some are playback-only) have at least two heads: a record/playback head and an erase head. Some better decks include three heads, with separate record and playback heads to allow close-to real-time monitoring of the recorded material directly from the tape, and allowing the heads to be optimised for their different functions. While initially very low in fidelity, in the early 1970s tremendous advances were made in tape formulations, first with Chromium Dioxide magnetic materials and then later with “Metal” tapes, which extended the high frequency response and improved their saturation characteristics. The addition of Dolby B noise reduction (and later Dolby C) reduced the inherent problem of high noise levels due to the low tape speed.
Compander: Compressor/Expander. A noise-reduction system that works by compressing the dynamic range of the audio before recording, and expanding it again by an equivalent amount during playback. An example of this was the dbx system.
Compliance: The degree of “springiness” in a mechanical system. For example, the cone of a loudspeaker driver with a soft suspension which can move to and fro relatively easily is more compliant than one with a stiff suspension. The stylus of turntable cartridge with a high compliance tends to follow the excursions of the groove more easily than a that of a cartridge with low compliance.
Compression: Reduction. Air is compressed by squeezing it into a smaller space. Digital signals are compressed by reducing the amount of data space required to hold them. Some digital data compresses readily, due to easily identifiable redundancy within the data. So, for example, a text document typically has a lot of space characters and “e” characters, so other ways of expressing these can be found. Digital audio and video signal tend not to carry a lot of redundancy, so systems that compress these highly rely on eliminating some of the data and are consequently called lossy compression systems. See also dynamic range compression.
Compression driver: A kind of loudspeaker driver. Rather than using a speaker cone to directly vibrate the air in a room, a compression driver has a throat in front of the moving part that causes a relatively small amount of air to compress and rarify rapidly in response to the movement of that part. This, in turn, drives the vibration of the air in front of it into the room, through some form of horn. Most compression drivers use piezo effects for the initial vibration, although some use more or less conventional cones. Compression drivers offer significantly higher efficiency than conventional drivers, allowing high output levels for a given input power, but can tend to reduce the dynamic range of the input signal and colour the sound.
Compression ratio: The extent to which a signal (particularly digital signals) is compressed, expressed as a ratio of the uncompressed size to the compressed size of the data. MP3, for example, with a bit rate of 128kb/s has a compression ratio of about 11:1.
Constructive interference: Where two signals, added together, act in sympathy with each other to boost the signal level. For example, bass from a subwoofer may, at certain frequencies, bounce from a nearby wall and that reflected signal may interfere with bass still coming directly from the subwoofer to effectively increase the output at that frequency. But it is also likely, at other frequencies, to result in destructive interference.
Control: (Sound quality description.) This concerns the ability of a loudspeaker to exactly deliver the signal. And that depends on the amplifier and loudspeaker working together to ensure that the speaker's cone movements exactly track the signal. So transients should hit precisely with power, and the speakers should not continue to resonate after the signal has gone (overhang). Much of the work here requires an amplifier with a low damping factor to keep on top of the speaker's behaviour.
Counterweight: The counterweight of tonearm is a cylinder that is fitted its rear to balance the weight of the cartridge at the other end. It is usually adjustable, generally by rotation, so that the tracking force can be set. See here for more details.
Crossover distortion: A small discontinuity (or, at least, nonlinearity) in a signal when a transistor-based amplifier circuit switches from positive to negative operation. This is addressed by Class A and Class A/B designs. Sometimes called “zero cross distortion”.
Crossover frequency: The frequency at which a signal is handed from one component to another. This applies in crossover networks and bass management systems.
Crossover network: The set of components in a loudspeaker that divides up the incoming signal, sending the bass to the woofer, the treble to the tweeter and, sometimes, the middle frequencies to a midrange driver. The crossover network uses resisters, capacitors and inductors (coils) to divide up the signal. Some high end systems use active crossover networks. In such cases, the division is made before the signal is amplified, so a separate amplifier is required for each driver.
Crosstalk: Where a signal (particularly audio) leaks from one channel to another. Thus a voice may be intended to be entirely in the left channel of a stereo recording, but some portion of it appears in the right, dragging the apparent position of the voice somewhat towards the centre of the sound stage. Normally specified by the inverse of crosstalk: separation.
Crunchy: (Sound quality description.) This is not a widely-used term, but one that I employ. It means sound which is laden with higher frequency noise, often not harmonically-related to the real signal. If sound is crunchy, something wrong is happening.
Current: The quantity of electrical charge moving through a circuit over a given time. The unit for current is amps or amperes.
D
DAB+: Digital Audio Broadcasting +. This is an upgraded version of the original DAB system and was the first introduced to Australia. It uses the HE-AAC v2 lossy audio codec, an improved version of AAC. “HE” stands for High Efficiency, which means higher quality sound from lower bit rates. For example, ABC talk radio channels on DAB+ are typically around 64kbps, while ABC Classic is 120kbps.
DAC: Digital to Analogue Converter. A component, circuit or device that converts a digital signal to an analogue one. Compare ADC which does the reverse.
Damping factor: A specification for power amplifiers which suggests the degree of control that the amplifier exercises over a connected loudspeaker. It is the ratio of the nominal impedance of the speaker (and is typically quoted for eight ohms) to the internal impedance of the output stage of the amplifier. A high internal impedance for the amplifier means that its frequency response will vary with real-world speakers since their impedance varies across their frequency range. It also means that the driver, which wants to do its own thing under the influence of air, its suspension and so forth, rather than what the signal is telling it to do, will face a relatively high impedance to the voltage it is generating back into the amplifier. Consequently it will be freer to do its own thing, rather than what the amplifier is telling it to do. However the damping factor quoted for amplifiers does not take into account the impedance of the wiring between amplifier and loudspeakers, nor the impedance of the speakers’ own voice coils. Consequently there is only a modest performance gain between a damping factor of, say, 60 and one of 600.
DAT: Digital Audio Tape. A compact tape developed in the late 1980s by Sony which stores audio in PCM format. The tapes look very similar to the MiniDV tapes used in digital video cameras, but are somewhat larger, measuring 71mm wide by 53mm deep and 10mm thick. The reading part of the tape is protected by the casing. In order to pack sufficient capacity on, the tape is read using a helical scan head in the same way as a VCR. The default format for DAT is a 48 kHz sampling frequency and 16 bits of resolution, but they can also be switched to 44.1kHz.
dB: Decibel. (See decibel.)
dBFS: Decibel - Full Scale. The level of a signal, measured in decibels, with reference to the maximum possible level of the signal. With digital audio the maximum recording level is 0dBFS, so all measurements of the signal are negative values.
dBSPL: Decibel - Sound Pressure Level. A measure of sound intensity. This is a logarithmic measure. To increase the sound level by three decibels, it is necessary to double the power output. A 10dB increase in sound level roughly equates with a perceived doubling of volume level. A figure of around 120-130dB is normally considered to be the threshold of pain. dB is sometimes loosely used as a synonym.
dbu: dBu is kind of a unit of voltage more common in Pro Audio, but also sometimes used in consumer audio. It’s not your typical unit because it expresses a logarithmic relationship to a fixed value of approximately 0.7746 volts (a historically typical number, since it delivers 1mW into a 600-ohm load). So 0dBu is 0.77 volts, 1dBu is 0.87 volts and you have to go all the way to 6dBu to double the voltage to 1.54 volts. dBu measurements are often negative: -6dBu is 0.39 volts. A useful online converter is here.
dbv: Like dBu, dBV expresses the ratio of the voltage to a fixed figure, in this case to 1 volt. So 0dBV is 1 volt, 6dBV is 2 volts and -6dBV is 0.5 volts. A useful online converter is here.
DC: Direct Current. An electric current that retains its level and direction, at least in the short term. More generally, DC is also used to describe voltage sources that deliver this kind of current. DC is generally used at low voltages within electronic equipment. Batteries deliver DC. Compare with AC.
DCC: Digital compact cassette. An attempted replacement for the compact cassette introduced by Philips in the late 1980s, early 1990s. This recorded audio in digital format, but used a lossy compression system. It failed to make headway since Sony’s Minidisc had the advantage of direct access and, more recently, the recordable CD provides convenient and cheap recording with higher quality.
Decay: (Sound quality description.) After a transient note, the sound fades away naturally. This is called decay.
Decibel: Abbreviation dB. A logarithmic measure of ratio. To determine the decibel relationship of, for example, two voltages you use the formula dB=20*log(V1/V2). If the result is negative, V1 is less than V2. If positive, then V1 is greater than V2. For power the formula is dB=10*log(P1/P2). dB is often used loosely as shorthand for dBSPL.
Depth: (Sound quality description.) One of the things surprising to people new to high quality audio is that a fine stereo system can convey not just an aural sense of the left and right positioning of instruments, but also their layering, front to back, in the sound stage, and also their vertical height. That layering is often called stage depth.
Destructive interference: Where two signals, added together, act in opposition to each other and reduce the signal level. For example, bass from a subwoofer may, at certain frequencies, bounce from a nearby wall and that reflected signal may, if arriving back out of phase, interfere with bass still coming directly from the subwoofer to effectively reduce the output at that frequency. But it is also likely, at other frequencies, to result in constructive interference.
Detail: (Sound quality description.) The finer, more subtle elements of the sound. Poor systems tend to hide detail due to their lack of resolution and general vagueness. Some also talk about "micro detail". I can't say that I understand why there should be any division here.
Dialog normalization: In Australian, “dialog normalisation”. Frequently called “dialnorm”. A metadata indication in a Dolby Digital bitstream of the volume level of the dialog in an audio signal. Dolby Digital decoders can use this “flag” to adjust the volume level of the whole audio stream, so that the dialog levels of different program sources remain the same. Most DVD movies have this flag set to “27”, which means -27dBFS. If the decoder is set to act upon this, it will reduce the level of the signal by 4dB because it means that the dialog is set 4dB higher than the calibration level of -31dBFS. It is important to note that this volume adjustment is made not just to the dialog, but to the entire sound track. It is also important to note that it does nothing to the sound other than adjust the overall volume level. Whether or not the dialog normalization feature actually works as hoped – in the sense of bringing the voice levels of all programs into line with each other – depends entirely on the engineers who mix the sound setting the dialog to the appropriate level in the first place. Dolby Digital encoders typically default to the value of 27, and must be explicitly changed for a different value. Many home theatre receivers report dialog normalization with reference not to 0dBFS, but to -31dBFS (the calibration level), and so a level of -27dBFS is reported as -4dB.
Digital: As opposed to analogue. It is a method of representing real-life signals (which are generally effectively infinitely variable) by using discrete numbers, usually binary numbers (a pattern of 1s and 0s). Holding discrete values, rather than the infinite number of intermediate levels used by analogue, makes digital signals relatively resistant to distortion and noise. The reason is that if any inaccuracy creeps in, unless very severe it will not affect the signal enough to throw it off. Consider a binary system. If all data is represented as either 0 volts or 1 volt, then it doesn’t matter if some interference causes the 1 volt level to be sometimes 1.1 volts, sometimes 0.9 volts. The receiving module will regard any voltage as greater than 0.5 volts as 1 volt and treat it accordingly.
Digital audio: Any one of a number of systems for recording sound using a digital representation of the sound. Some digital audio systems are straightforward representations of the analogue signal. Examples of these are PCM, DSD and MLP. Other systems take a simple digital signal (usually PCM) and process it heavily to reduce its size. Examples are Dolby Digital, MPEG audio and DTS.
Digital Audio Player: A Digital Audio Player is a portable music player, typically with capacious storage for thousands of music tracks. Originally that space was small, with early models being limited to a few tens of megabytes, and even those employing hard disc drives offering only 5 to 10GB of storage. So music was typically stored in MP3 or some other lossy format. These days they are likely to allow a terabyte or more of storage to be used, permitting many thousands of high resolution, losslessly compressed tracks to be enjoyed anywhere.
DIN: Deutsche Institut fuer Normung. A German standards body. DIN frequently appears in specifications to give an indication of how measurements were conducted, and also applies to a number of connectors approved by the body.
Dipole: A loudspeaker designed to offer a diffuse, non-directional sound by firing its high frequencies in two directions, out of phase with each other, so that a listener receives few aural clues as to their exact location. To make such speakers work optimally, they should be placed so that the axis running through the front and rear tweeters is at 90 degrees to a line drawn from the speaker to listener. This will maximise the cancellation (see destructive interference) of direct radiation from the speaker to the listener’s position.
Direct sound field: A speaker system in which the great majority of the sound that you hear is coming directly from the loudspeakers, and very little from reflections from surfaces within the listening space. Direct sound field speakers tend to deliver a more accurate reflection of the source, and sharper stereo imaging, than reverberant sound field speakers. Direct sound field sound can be achieved by choosing speakers with restricted dispersion and placing them close to you.
Dispersion: The degree to which loudspeakers spread their sound production in all directions, rather than directly to their front. All loudspeakers widely disperse their bass. But as the wavelength of a sound nears the size of a driver’s cone diameter, the sound tends to become more directional.
Distortion: An inaccuracy in the reproduction of a signal. In the case of audio, it is normally regarded as being composed of harmonic distortion and intermodulation distortion. But used more broadly, it can also encompass frequency response variations and noise. In the case of lossy compression technologies, some distortion consists of spurious noise (not harmonically related) surrounding the signal. When “distortion” is quoted as a specification without qualification, it normally refers only to harmonic distortion.
Dither: Very low level noise, usually “white” in character, added to a digital audio signal to reduce harmonic distortion. It typically is just a random variation in the least significant bit of the digital signal. In some systems, such as Sony’s SBM, the noise is shaped to yield a lower noise floor in the more easily audible midrange and low treble, by pushing much of the noise into the near-ultrasonic. Read more.
Dolby Atmos: Dolby Atmos means several things. It started principally as a multichannel audio encoding scheme but has since taken on many of the attributes of a branding exercise. When first launched its main benefits were touted as it supporting sound “objects” and providing considerable channel independence, with support for up to 32 output channels, including height channels. Sound “objects” could be controlled by the mixing engineer and they’d be held within the audio stream as objects. The Dolby Atmos decoder would “dynamically render” the acoustic image of the objects in the appropriate place, according to the loudspeaker layout in the listening room. In most implementations, the maximum number of channels supported was 7.4: with a conventional 7.1 loudspeaker layout, plus four ceiling or Dolby Enabled speakers. Or some variant of this with a reduced number of channels. These days even some TVs are labelled as supporting Dolby Atmos.
Dolby Digital: Dolby Digital is an encoding scheme invented by Dolby Laboratories as a way of compressing digital audio so that it uses a lot less data space. It is also known as AC-3. Originally developed as an audio compression system for US digital television, it achieved prominence by allowing multiple channel sound tracks to fit onto standard 35mm cinema film prints (in between the sprocket holes on the film!), it has become the de facto standard for DVD. The compression system uses perceptual encoding, similar to DTS, MPEG audio (including MP3) and Sony’s SDDS and ATRAC. It can carry up to 5.1 channels of sound, but does not necessarily have that many. Dolby Digital 2.0 (that is, stereo) can be encoded with Dolby Pro Logic surround sound. The Dolby Digital bitstream can also carry codes (metadata) to control playback parameters in the Dolby Digital decoder. Dolby Digital apparently supports bit rates of up to 640kb/s, but on 5.1 (or higher) channel DVDs the bit rates actually used at 384kb/s and 448kb/s. (Note, the “k” here stands for 1,000, not 1,024). Dolby Digital bitstreams also include metadata for controlling the operation of the decoder.
Dolby Digital EX 6.1: A new surround sound standard which provides the usual 5.1 channels plus an additional channel: the centre rear channel. Unlike DTS ES 6.1 Discrete the additional channel is not carried discretely but is encoded into the two rear channels in a similar way to the front centre channel is encoded into a Dolby Pro Logic sound track. Movies prepared for Dolby EX 6.1 presentation in cinemas should have the same encoding on DVD. However EX 6.1 is compatible with 5.1 channel systems in the same way that Dolby Pro Logic is compatible with stereo systems. More correctly, this should be termed LucasFilm THX 6.1 since it was developed by them, but the name above seems to have come into common currency.
Dolby Digital Plus: Dolby TrueHD is an enhanced version of Dolby Digital, introduced with Blu-ray and HD DVD. It supports up to 15 discrete channels and, while still a lossy format, it supports a bitrate of up to 6.44Mbps (Dolby Digital is limited to 640kbps on Blu-ray). And it also supports Dolby Atmos encoding.
Dolby Enabled speaker: Recognising the difficult some homes would have installing height loudspeakers, when Dolby introduced Atmos, it also introduced support for Dolby Enabled speakers. These are largely upwards-firing, although slightly tilted, loudspeakers that sit atop the regular front left and right speakers, plus also on the surround left and right speakers for the full four height channels if desired. While relying to some extent on “bouncing” sound off the ceiling, the Dolby Atmos surround processor adjusts the signal significantly for these speakers to increase the sense that the sounds are coming from above.
Dolby HX: Dolby Headroom eXtensioin. Not to be confused with the various sorts of Dolby noise reduction, Dolby HX is a process that improves the high frequency, high level recording performance of magnetic tape, particularly compact cassettes. In order to provide reasonable performance linearity, magnetic tape needs to have a “bias” signal applied to it during recording. This is an ultrasonic sine wave. Magnetic tape also has limits to the level at which it will record a signal. The problem is tape “saturation”. You can only make the particles so magnetic, then they won’t magnetise any further. Tape saturation tends to afflict higher frequencies more than mid and low frequencies. That is why many good quality cassette decks will maintain a flat frequency response all the way to 20,000 hertz at an indicated -20dB recording level, but roll off badly before that frequency at 0dB. Some proportion of a tape’s signal-holding capacity is used by the bias signal. The Dolby HX system monitors the frequency and level content of the signal during the recording and when these reach certain levels, the system reduces the level of the bias signal, consequently allowing higher recording levels, particularly at high frequencies, to be reached before tape saturation. It relies on the signal to be self-biasing.
Dolby noise reduction: It was the invention of an effective noise reduction circuit by Ray Dolby in the 1960s that got Dolby Laboratories onto the road to where it is today. This circuit was developed into Dolby Type A noise reduction, which became very widely used for professional analogue recording onto tape, prior to the advent of digital recording. Most analogue recording media produce background noise, typically like white noise. The simple solution would be to boost the middle and upper frequencies during recording, then cut them on playback, thus also cutting the hiss. But this causes overloading problems. Dolby noise reduction systems rely on the masking effect of sounds. If the signal was loud, the hiss would be largely inaudible anyway. So Dolby’s system tracked the level of the signal, leaving it unaltered when it was strong, but introducing the boost/cut system when the signal was low. Dolby noise reduction later made it onto consumer equipment in the form of Dolby B, C and S noise reduction systems which all work along similar lines. It was Dolby B, in particular, that allowed the compact cassette to become an established media. Dolby HX, which also appears on recent cassette decks, is not to be confused with the noise reduction systems.
Dolby Pro Logic: An improved version of Dolby Surround which decodes two channel Dolby Stereo sound tracks to four channels. In addition to the front left, front right and mono surround channels, it also extracts a front centre channel signal. The use of a centre channel improves the localisation of sound, particularly dialogue, for those viewers not seated directly in front of the screen.
Dolby Pro Logic II: A further enhancement of the two-channel based Dolby Pro Logic, this cleverly decodes separate left and right surround channels from the original signal and eliminates the 100 to 7,000 hertz bandwidth limitations of the older system.
Dolby Stereo: The original name for Dolby Surround, as it was used in cinemas. In cinema usage “stereo” tended to mean some form of surround sound, although it was usurped by the home entertainment industry to mean two channels at the front of the room.
Dolby Surround: Originally the home equivalent of Dolby Stereo, it was the original system for decoding sound tracks for surround sound in the home, usually from video cassette. That original form of Dolby Surround carried three channels of sound, matrixed into two channels. The decoder extracted the additional channel, known as the surround channel, and sent it to a pair of rear or surround speakers. This surround channel was limited in frequency range to 100-7,000 hertz. This was replaced by Dolby Pro Logic. With the introduction of Dolby Atmos, the name Dolby Surround was reintroduced as a general surround decoder to replace the previous Pro Logic forms. Dolby Surround now attempted to generate a fully immersive hemispherical sound field from whatever number of channels of input were available, using whatever loudspeakers were available.
Dolby TrueHD: Dolby Digital Plus is losslessly compressed multichannel digital audio format for movies, introduced with Blu-ray and HD DVD. The stream contains an embedded Dolby Digital stream for backwards compatibily. It supports up to 16 discrete channels. Its core lossless technology is based Meridian Lossless Packing, which first appeared on DVD Audio. It supports Dolby Atmos encoding.
Driver: The moving part, or parts, of a loudspeaker. These are usually woofers (bass drivers), tweeters and midrange drivers. There are a number of different driver designs. Virtually all woofers use the traditional speaker cone (some light material, often paper pulp or polypropylene) held in place by a suspension and backed by a coil inserted into the magnetic field of a strong permanent magnet. The amplifier’s signal is fed into the coil, generating its own magnetic field, causing the coil and the attached cone to move. Midrange drivers, which are relatively rare these days, usually use either cones or domes (often polypropylene or a light metal such as magnesium), although there are some ribbon midrange drivers. Tweeters are most commonly domes (often polypropylene, silk or some other textile, or a light metal such as aluminium or titanium), but cheaper ones use cones. Some use inverted domes (that is, they are concave rather than convex), while some expensive speakers use ribbon tweeters. There was even, for a while, a “plasma” tweeter where the high frequencies were generated by a pulsating ball of superheated air. Some speakers do not use what could be conventionally called drivers, for example electrostatic speakers.
DRC: Dynamic Range Control. (See Dynamic Range Control.)
DRM: Digital Rights Management. A catch-all name for various systems that control the distribution of digital audio and video content. Usually based on secure(-ish) keys and encryption.
DSD: Direct Stream Digital. The digital audio format used in the SACD. Unlike the PCM system normally used, DSD uses a stream of single bits of information. The momentary level of the analogue wave form being represented by the bitstream is determined by the density with which the bits are “on” rather than “off”. It is modified by using noise shaping to increase the effective dynamic range in the main audible band. DSD uses for each channel a bit rate of 2,822,400 bits per second. For a deeper dive into DSD, see here.
DSP: Digital Signal Processor. A computer-type processing unit optimised to perform 24 or 32 bit floating-point operations on digital audio signals. This allows it to perform Fast Fourier Transforms and other complicated operations in real-time to apply frequency response adjustments, generate reverberation and even split-out certain frequency bands into separate channels. Many home theatre receivers incorporate DSP programs to generate ambient multichannel sound from stereo sources.
DTS: Digital Theatre System. A high quality digital surround sound compression format capable of carrying multiple channels of audio. While using perceptual encoding like many other systems, it uses much lower levels of compression. DTS claims that it first uses non-lossy compression techniques to reduce or eliminate the need for perceptual encoding. Many users consider that it produces higher quality sound than Dolby Digital (see here for discussion of the validity of that view). DTS sound tracks come on DVD encoded at either 768kb/s or 1,536kb/s. (Note, the “k” here stands for 1,000, not 1,024). The principle advantage of DTS over Dolby Digital is the implementation of DTS in the cinema. Rather than the digital data being optically recorded onto the film itself, DTS audio is recorded on a CD. Special CD players attached to the cinema’s film projector keep the audio and film synchronised by means of time sync signals on the film. The advantage lays in the fact that in distributing foreign language sound tracks, only the CD needs to be different for each language, not the film.
DTS 96/24: A variation on DTS. However rather than using DTS’s normal 20 bits of resolution and 48kHz sampling frequency (of the PCM data before it is encoded), it uses a 96kHz, 24-bit source signal, offering a theoretical frequency response to in excess of 40,000 hertz. DTS suggests that some of the higher resolution offered by this PCM standard carries through into improved audio quality. I have grave doubts about this. 96kHz, 24-bit PCM demands a 2,304kb/s bit rate for each channel. Since DTS 96/24 is offered as 5.1 channels, the equivalent PCM would need 11,520,000kb/s plus data for the subwoofer, yet DTS 96/24 is delivered with a compressed bit rate of 1,536kb/s, which means that a compression ratio of 7.5:1 has been employed. I would far rather all the available bits of data be employed in the most important (that is, audible) band of up to 20,000 hertz. (Note, the “k” here stands for 1,000, not 1,024). DTS 96/24 DVDs are backwards compatible with standard DTS decoders.
DTS ES 6.1 Discrete: A 6.1 channel version of DTS where a centre-rear channel is held in a discrete audio channel. DTS ES 6.1 DVDs are backwards compatible with standard DTS decoders.
DTS ES 6.1 Matrix: A 6.1 channel version of DTS where a centre-rear channel is encoded into the two normal surround channels in a similar way to that employed by Dolby Pro Logic to hold a centre channel encoded into the two front channels. DTS ES 6.1 DVDs are backwards compatible with standard DTS decoders.
DTS-HD High Resolution: DTS-HD High Resolution is an enhanced version of the DTS multichannel digital audio format, introduced with Blu-ray and HD DVD. It supports up to 7.1 discrete channels and, while still a lossy format, it supports a bitrate of up to 6Mbps and carries audio sampled at 96kHz.
DTS-HD Master Audio: DTS-HD Master Audio is a losslessly compressed multichannel digital audio format for movies, introduced with Blu-ray and HD DVD. The stream contains a regular DTS core, which can be used instead for backwards compatibily. This core forms the basis of the sound. The rest of the stream is used to adjust it to lossless exactness. It supports up to 8 discrete channels on Blu-ray and supports up to 192kHz sampling.
DTS Neo:6: A processing system that endeavours to extract an engaging 6.1 channel surround experience from two channels of audio. Unlike Dolby Pro Logic, this is not an encode/decode system, but more like a DSP system to generate something new.
DTS:X: DTS:X, like Dolby Atmos, supports sound objects, and dynamically adjusts the sound so that the sounds appear in the correct places, according to the actual loudspeaker system being used. There are several other similarly-named processses, such as DTS Neo:X which, in various ways analyses and derives multichannel and height information from the stream, regardless of the source number of channels, and attempts to create a full three-dimensional surround field.
Dull: (Sound quality description.) Usually suggesting that the sound has a recessed level of treble. The opposite of bright.
DVD: Digital Versatile Disc, the 12cm optical disc capable of holding masses of data for computer use, video, still pictures or audio.
DVD Audio: Digital Versatile Disc - Audio. This is the form of DVD which primarily holds DVD Audio material, all of which resides in a folder on the disc called “AUDIO_TS”. The material may be either PCM or, more commonly, MLP encoded. The DVD Audio material can only be played on a DVD Audio player. DVD Audio discs almost always also hold a repeat of the material in DVD Video format so that they can be used in DVD Video players as well. Typically, though, the audio in this section is recorded in a lower standard of PCM, Dolby Digital or DTS.
Dynamic: (Sound quality description.) Dynamic sound pertains to the difference between the quiet and loud elements of the music. Weaker systems tend to restrict the level of the louder parts, dynamically compressing the sound. It's not just the difference between loud and quiet sections of the music that matter, it's also the almost instantaneous level variations, for example percussion strikes in the midst of the music. Preserving this dynamic character of the music adds enormously to the sense of reality, or real presence.
Dynamic driver: A dynamic driver is the most common kind of loudspeaker or headphone driver. It uses a coil wound into a cylinder which is nestled within a strong magnetic field. The signal is fed to the coil, so it moves. A diaphragm of some kind is attached to one end of the coil, so it moves as well, and thus also moves the air.
Dynamic power: In some respects dynamic power is a similar measure to PMPO, but remains far more realistic. It is quoted in conjunction with continuous power outputs and shows how much power the amplifier or receiver can deliver for a very brief instant, such as when a crescendo is played in music. An amplifier with a reasonably low continuous power output can often sound quite good at reasonably high levels if it has a high dynamic power rating.
Dynamic range: The differences in the volume level between the loud bits and the quiet bits of a movie or some music. On DVD these differences can be quite marked. This can lead to problems when trying to listen via your TV’s speakers. The term is also used as a specification for DVD and CD players. In this sense it means the range between the loudest and the quietest sounds that the player is capable of producing, and is determined by the noise floor of the player and the medium.
Dynamic range compression: A technique used by music producers to reduce the range between the loud and quiet sections of music. The reasons include allowing a higher average level of music playback, and in analogue systems helping deal with background noise. Some home theatre systems also provide for dynamic range compression in playback.
Dynamic range control: A facility in Dolby Digital to the reduce the dynamic range of the audio content in order to allow the entire program to be heard in adverse conditions. It does this by reducing the loudness of the parts of the program which are louder than the level set by the dialog normalization setting, and boosting those parts quieter than that setting. The parameters for reductions are carried in Dolby Digital metadata. This feature is useful for appreciating movies without disturbing neighbours, and when the audio from a DVD is being heard through a limited sound system. In particular, if DRC is available on a DVD player, it should always be switched on if a TV’s speakers are being used to listen to the program.
E
Earphones: Earphones are compact devices for personal audio listening. The ear pieces usually weigh only a few grams and have tips, often made of silicone that conforms to the listener’s ears, that are inserted partially into the outer ear canal. Most stay in place through friction, but many use ear hooks or small “wings” which lock them in place. Earphones may be wired or wireless (ie. connect with Bluetooth).
Efficiency: Used in connection with transformation of energy from one form to another, efficiency is the ratio of energy output to energy input in the transformation process. In regard to speakers, it is normally expressed as speaker sensitivity.
Effortless: (Sound quality description.) An effortless performance of a system is one in which there is no apparent distortion or other weakness. Normally used when a system is pushed to a level which would distress some other systems, while remaining clean and controlled.
Electrostatic loudspeakers: A form of loudspeaker in which the driver is a panel that responds to a varying electrostatic charge. Such speakers incorporate electronics which convert the moderate voltage, moderate current output from an amplifier into the high voltage, low current signal required to drive the panels. Electrostatic speakers are noted for producing little harmonic distortion and an excellent amount of detail in the reproduction of music. However they often present a difficult load upon amplifiers, can be subject to arcing (producing electrical sparks) in high humidity situations and tend to be somewhat inefficient. Because they use a large panel, these speakers do not employ an enclosure and so are bass limited. For this reason many models are hybrid, with standard woofers in an enclosure for bass. Electrostatic speakers are by their nature dipolar in operation since each side of the panel is generating an out-of-phase sound wave to the other.
Enclosure: The box of a loudspeaker. In any half decent speaker, this is not merely a device to keep the drivers off the floor, but an integral part of the design of the whole loudspeaker, contributing greatly to – or, if done badly, detracting from – its performance. The most common enclosure designs are bass reflex and acoustic suspension. Good quality enclosures include strong bracing to resist sympathetic vibrations in their panels, which can lead to sound colouration.
EP: Extended Play. A 7-inch vinyl 45 rpm recording with, typically, two songs on each side.
Error correction: Redundant data included with a signal in transmission or storage that allows the signal to be reconstructed even if some data has been damaged or lost. Error correction is used in most digital systems, from CDs through to digital television.
EQ: Equalization (or, in Australian, Equalisation). An adjustment to the frequency response of some piece of equipment or process in order to achieve a desired outcome. It might be part of a standard process (eg. RIAA Equalization) designed to overcome physical limitations, or it might be used to correct unwanted frequency response inaccuracies imposed by a room.
Equal-loudness contour or curves: Experimentally derived results which plot the perceived loudness (by humans) of sounds across the audio spectrum. These suggest that our ears are very much more sensitive to midrange and low treble sound than they are to the bass and treble frequency extremes. The peak sensitivity occurs at around 3,000 to 4,000 hertz. At the extreme, we can hear much quieter sound at those frequencies – as low as -6 decibels SPL! By contrast, a tone at 50 hertz has to be at a level of more than 40dB SPL to be heard. These curves were formerly known as the Fletcher-Munson curves, after the researchers who conducted the first research on the subject in the 1930s, but their curves have been superseded by later work.
Etched: (Sound quality description.) Etched sound is precisely and sharply defined by the system.
Excursion: The degree of fore and aft movement of a mechanical component in an audio system. Loudspeakers, for example, produce sound by vibrating their drivers‘ cones or domes. The distance between the furthest the cone, say, protrudes forwards and the furthest it moves back into the housing is the excursion of the driver. Woofers tend to have a greater excursion than drivers for higher frequencies. The drivers in some compact subwoofers deliver an excursion of up to 50mm. This wide excursion is necessary because output level is related to the area of the cone and the excursion. A smaller driver in a subwoofer can only provide equivalent output levels by allowing a greater excursion of the cone. Another example of excursion is the stylus of a cartridge. When playing a stereo LP it moves from side to side and up and down. Each of these is referred to as the excursion (horizontal and vertical) of the stylus.
F
Far field: Rarely used term to essentially mean "not near field". Basically, most loudspeaker systems are far field. Compare near field.
FFT: Fast Fourier Transform. A set of mathematical techniques for deriving an close approximation of the Fourier series for a real-world signal so that processing can be performed upon it. FFT is, with suitable hardware, sufficiently fast to allow real-time processing of signals. DSPs are based upon its mathematics (although, of course, they also do much processing work on the result of the FFT as well).
Filter: A circuit that reduces the levels of some frequencies. They are ubiquitous in audio gear. For example, the crossover network in a loudspeaker is a combination of filters. They are also required in digital audio to restrict the bandwidth of the sound to that able to be handled by the system's digital resolution, both before analogue to digital conversion and after digital to analogue conversion. Some DACs allow this operation of this final filter to be altered. See here for more.
FLAC: Free Lossless Audio Codec. FLAC is a common lossless compression system for digital audio. It supports multichannel audio, but is mostly used for stereo. It also supports up to 32 bits and 655,350 hertz sample rates. Of the industry standard rates, that includes 384kHz, but not 768kHz. It also supports tagging, that is, incorporating metadata holding information like genre, album, artist, etc. FLAC is widely supported by just about all quality audio playback devices ... even most smart TVs. As the name suggests, FLAC is free to use and implement. See here for how lossless audio codecs work.
Flat: In regard to the frequency response, the graph depicting the response is flat, meaning no frequencies are favoured over other frequencies. For most equipment that's a virtue to be sought.
Fletcher-Munson curves: Experimentally derived results which plot the perceived loudness (by humans) of sounds across the audio spectrum developed in the 1930s. Later research has largely replaced these. See Equal-Loudness Contours.
Flutter: A defect affecting analogue audio signal sources that rely on rotating the medium, particularly LPs and compact cassettes. Flutter is a rapid, repetitive speed variation, typically repeating at least ten times per second. If an LP or audio cassette undergoes this, it produces rapid variations in the playback frequency. At the slower end of the scale, it can sound like an unwanted tremolo or vibrato. At higher rates it can add an unpleasant harshness to the sound. Flutter is specified in per cent and specifications of more than around 0.1% are unacceptable. Digital sources such as CDs are immune to flutter because they lock their playback speed to a solid-state timing device.
FM: Frequency Modulation. A system of storing or transporting a signal by using it to modulate a carrier in the frequency, rather than amplitude, domain. FM radio, high-fidelity audio on VHS, and numerous other systems are based on FM. FM tends to be more resistant to interference than AM.
Focus: (Sound quality description.) Another of the somewhat vague terms used in reviews, this is mostly about how the virtual stereo images appear. Focused imaging suggests very cleanly defined and stable placement of the instruments on the sound stage. It can also be used to refer to a high degree of coherence in the sound.
Fourier: Actually, Fourier analysis. Fourier found that every repeating signal could be expressed as the sum of some particular sine wave, called the fundamental frequency, plus its harmonics at various levels. See, for example, square wave. This understanding permitted the development of all forms of signal processing, whether digital or analogue.
Four-way loudspeakers: A loudspeaker which divides the incoming signal into four different frequency bands for distribution to drivers. It sends high frequencies to the tweeter, upper middle frequencies to a small midrange driver, lower middle frequencies to a larger midrange driver, and the lowest frequencies to one or more woofers.
Frequency: The rate at which a repetitive signal repeats, measured in cycles per second or hertz (or its convenient multiples). Audio frequencies are generally in the range of 20 to 20,000 hertz because those are the frequencies to which the human ear sensibly responds. Equipment measurements often extend this range to 0 to 100,000 hertz. The frequency of a signal can be calculated by dividing the speed with which the signal propagates through a medium by its wavelength.
Frequency range: A specification somewhat similar to frequency response, but typically more vague. A good frequency response measurement specifies a decibel range containing the specified frequencies. If it doesn't, then feel free to consider it a mere frequency range. A frequency range simply states the range of frequencies the device is capable of reproducing at some level. A compact speaker is usually capable of producing 20 hertz at some level. But that level is way too low to be audible.
Frequency response: A measure of how accurately a system reproduces different frequencies. In the case of audio in a home theatre system, it is desirable for the frequency response of a whole system, including speakers and subwoofer, to be from 10 hertz to 20,000 hertz ±3dB. This performance requires a very expensive system indeed and, in practice, very few systems will produce bass down to anything like that bottom limit. Manufacturers who claim a frequency response for speakers of, say, 20 to 20,000 hertz without specifying decibel boundaries are telling you nothing. A tinny two inch transistor radio speaker can reproduce that range, although you won’t actually hear it at either extreme because its output will be so low. Even subtle variations of less than half a decibel across the audio band can be quite audible, especially if they’re spread over a fairly wide band of frequencies, and can thus change the character of the sound. Indeed, with speakers the single measure most closely related to their sound is the frequency response.
Fundamental frequency: Any musical tone primarily consists of a particular simple tone (a sine wave) and a series of higher frequency tones, where the frequencies of those higher tones are whole multiples of the first-mentioned tone. The frequency of that simple tone is called the fundamental frequency. When we say that the Middle C of a piano has a frequency of 261 hertz, we are actually talking about its fundamental frequency, not that of the harmonics without which a piano note would have no distinguishing character.
G
GB: Gigabyte. (See gigabyte).
Gigahertz: A measure of frequency: 1,000,000,000 hertz.
GHz: Gigahertz. (See gigahertz).
Gigabyte: A measure of memory storage, a gigabyte equals (ie. 1,024 x 1,024 x 1,024 or 2³⁰) bytes, 1,048,576 (ie. 1,024 x 1,024 or 2²⁰) kilobytes or 1,024 megabytes. However, if used as a measure of hard disk storage, the “giga” prefix normally means a round billion.
Glare: (Sound quality description.) The aural equivalent of the visual phenomenon described by the word. It means a marked emphasis on the mid to upper treble, beyond just being bright, and is often due at least in part to actual distortion in the signal (all harmonic distortion products are higher in frequency than the sound being distorted.)
Glassy: (Sound quality description.) Also an emphasis in the treble, somewhere between being bright and exhibiting glare. The word also conveys a slight sense of fragility.
Grainy: (Sound quality description.) Where sound should be smooth and even but has a detectable texture or overlay. The cause of this is far from certain.
Gritty: (Sound quality description.) Like grainy but more so.
H
Haas Effect: Also known as the Precedence Effect. The human ears do not rely solely on different loudnesses to determine the direction from which a sound is coming; they also use timing. It is often the case that timing trumps loudness. If the same sound arrives from two different directions, the ear will tend to identify the direction of the source as that of the first sound to arrive, even if the other sound, arriving a few milliseconds later, is significantly louder. This is why it is important to properly adjust the time alignment of surround speaker systems. This effect is also made use of in high quality sound reinforcement systems in concert halls, since by delaying the sound coming from speakers near the back of the hall, it can be made to sound as though the high volume coming from those reinforcing speakers is actually coming from the front of the hall.
Harmonic: A tone the frequency of which is a whole multiple of another tone with which it is associated. Virtually all physical sound producers (including, unfortunately, loudspeakers) produce complex sounds which consist of a fundamental tone and many harmonics. Harmonics are also known (more in musical, than home entertainment, contexts) as overtones.
Harmonic distortion: When a signal (usually audio) is not reproduced perfectly, it is said to be distorted. Harmonic distortion is a specific, and common, type of distortion in which a given frequency that is supposed to be in the signal has added to it overtones, or additional unwanted signals which are whole multiples of its original frequency (harmonics). Equipment suppliers often quote a measurement called THD or Total Harmonic Distortion. The lower the figure the better … in general. But relatively high levels of evenly numbered harmonics are easier to tolerate – indeed, can often be attractive, making the sound “warmer” – than quite low levels of odd order harmonics.
Harsh: (Sound quality description.) Unpleasant sound quality, generally involving considerable distortion and some treble emphasis, that makes the music tiring to listen to. It also can make it seem louder than the actual listening volume.
HDCD: High Definition Compatible Digital. This is an encoding/decoding system intended to improve the resolution of CDs (and other digital audio sources) while retaining full CD compatibility. It is said to provide performance equivalent to a PCM system with a resolution of 20 bits, while retaining CD’s bit depth of 16. The trick is in re-introducing the old-fashioned compander, albeit carefully implemented in a sophisticated digital fashion. This applies a “limiter” to the peaks of the signal so that they can be reduced by 6dB, providing an effective two more bits of resolution. It also increases the level of low-level signals in the -45dBFS to -65dBFS range by up to 4dB, giving greater clearance from the noise floor. When played back with a HDCD decoder, the peak limiting and low level boost are reversed (a HDCD switching signal is embedded in the dither noise). HDCD also incorporates a dynamic low pass filter optimised for different types of signals (its operation is also signaled by the dither noise signal) and the dither itself is noise-shaped to provide a very low floor of noise and harmonic distortion below 16,000 hertz. I haven’t analysed any HDCD CDs, but it seems likely that such severe noise shaping would produce a rather high noise floor between 16,000 hertz and the 22,050 hertz cutoff. Whatever benefits HDCD CDs may deliver when decoded with a HDCD decoder, it seems that the processing of the signal would adversely impact on playback in regular CD players.
HDCP: High-bandwidth Digital Content Protection. An encryption protocol developed by Intel and accepted for DVD use by the DVD Forum. This encrypts digital video output in real-time solely for the purpose of decrypting at the other end of a HDMI cable. It provides for two devices to conduct a handshake, establish an encryption key, and then feed the audio and video signal in digital format. This system is supposed to protect against high quality digital video copying.
HDMI: High-Definition Multimedia Interface. A connection standard for feeding signals from sources to output devices in digital format. This carries both uncompressed digital video and uncompressed digital audio in PCM format. With a suitable receiver, HDMI can also carry Direct Stream Digital audio.
Headphones: Headphones are devices for personal audio listening. They sit on or over the listener’s ears. Headphones may be wired or wireless (ie. connect with Bluetooth).
Heavy: (Sound quality description.) Sound which has too much bass.
Height: (Sound quality description.) One element of a stereo sound stage. In addition to presenting width, a quality stereo system can add depth and height to the stereo image. With height, certain sounds or instruments appear to be floating at a higher level than others.
Helical scan: A system of recording tracks of data or signal onto magnetic tape in which the tape is wrapped partway (up to about three quarters) around the circumference of a spinning cylinder. The cylinder houses one or more reading and writing heads. The tape is also drawn linearly across this contraption. This results in the tracks being written diagonally across the face of the tape. It has the advantage of allowing an effectively higher relative speed of tape to head, allowing a much greater density of data to be held on the tape than a linear system. Used by VCRs and DAT.
Hertz: Cycles per second – a measure of frequency. Young healthy humans can hear from around 20 hertz to around 20,000. Frequencies lower than that can often be detected by other parts of the body.
Hi Fi: High Fidelity. This is a tricky set of four letters. “Hi fi” is frequently used a generic noun for any stereo sound system. But it should more properly be regarded as an adjective, taking due note of the Latin roots of the word fidelity, which means “truth”. Originally high fidelity described a sound reproduction system that gave a more accurate rendition of the recording than was commonly available. The greater the accuracy, the higher the fidelity. Some audiophiles mistake pleasing sound for accuracy. I use it in the original meaning. Note that in absolute terms, it changes over time. As audio systems improve, so do those properly described as “high fidelity”. A very fine high fidelity system from 1970 would not qualify for the term today.
High pass filter: An electrical circuit that impedes signals below a particular frequency. In other words, it lets signals above that frequency pass through. An example is the circuit in a home theatre receiver that stops deep bass from going to a centre channel speaker. Compare low pass filter and bandpass filter.
High resolution audio: Any digital audio that exceeds a bit depth of 16 bits and a sampling frequency of 48kHz. Some quite common high resolution audio formats are 24-bit, 96kHz and 24-bit, 192kHz, as well as DSD in all forms. Some brands – Sonos in particular – regard 24-bit, 48kHz audio as high resolution, although many would dispute that.
Home theatre receiver: A multichannel amplifier with a digital decoder and tuner built in. The decoder always provides at least PCM and Dolby Digital decoding, and almost always DTS as well. A few also incorporate MPEG audio decoding.
Horn loaded driver: A loudspeaker driver in which a horn is placed over the front of the driver. The primary advantage of this is to increase the efficiency of the transfer of mechanical energy from the driver to the air. However it usually results in reduced dispersion of sound, particularly for higher frequencies which tend to be “beamed” from the horn, and can colour the sound. Nevertheless, some speaker makers have specialised in using horns, especially Klipsch, and they are frequently used in professional sound-reinforcement installations. The increase in efficiency can be quite marked. Klipsch loudspeakers tend to offer around nine decibels higher output than equivalent non-horn loaded loudspeakers, which means they can produce the same output from just one eighth of the power required for conventional speakers. This can, in turn, mean lower harmonic distortion because driver excursion is significantly reduced.
Hum: A low frequency continuous noise with a fundamental frequency of 50 hertz along with various harmonics (100 hertz, 150 hertz etc). This may be due to either electrical interference from mains power, or perhaps a very low level leakage. The interference could be within equipment or cabling. Turntable cables are most susceptible to this because they carry very low level signals, and the phono preamp boosts bass a great deal in accordance with RIAA equalization.
Hump: (Sound quality description.) This describes the shape of a frequency response graph where a broad range of frequencies are at a higher level than the norm. This hump could be in any place, bass, midrange or treble.
Hybrid loudspeakers: Loudspeakers in which two significantly different driver technologies are used. For example (and most commonly), loudspeakers which use electrostatic panels for midrange and treble combined with conventional bass drivers.
Hz: Hertz. (See hertz).
I
IMD: Intermodulation distortion. (See Intermodulation distortion.)
IEMs: In-ear monitors See In-ear monitors.
Imaging: The sensation produced in a stereo or surround system of sounds coming from between the loudspeakers. The imaging is described in various subjective ways relating to how tightly focused those sounds appear, whether they seem to offer a fore-aft depth, whether they give an impression of height as well as width and depth.
Impedance: A measure of electrical resistance and reactance. These are the properties of a component that limit the amount of current that can flow through a circuit. Resistance affects the DC part of the electrical current, while reactance affects the AC part. Measured in ohms.
Impulse: An element of sound involving an extremely sharp change in level, such as the strike of a percussion instrument. A "perfect" impulse is a single sharp spike in sound from a floor of nothing. Such a "perfect" impulse is constituted by all frequencies simultaneously. For strong high fidelity performance, all the frequencies within the available bandwidth of the system which constitute the impulse should be in perfect timing with each other, otherwise the sound may be smeared.
In-ear monitors: In-ear monitors – commonly known as IEMs – are wired earphones engineered for high quality sound, high sensitivity and good noise isolation. IEMs were originally used by performers on stage for “foldback”, but have made their way into the consumer space at the quality end of the market. Some custom IEMs are available, in which the casing is moulded specifically for the purchaser. That ensures and ideal fit and acoustic seal.
Infinite baffle: Another term for acoustic suspension. The term is descriptive, in that if the baffle of a loudspeaker were to be infinitely extended in all directions, there could be no movement of air between the front and back of the driver. Of course, with a real infinite baffle speaker the baffle is wrapped around into a convenient package.
Infrasonic: Audio tones of frequencies lower than capable of being detected by the human ear, generally below 20 hertz.
Interconnect: The cables used to transfer an analogue line level signal from a source component in a home entertainment system to an amplifier or recording device. Thus a set of interconnects is commonly used to plug a CD player into the analogue inputs of a home theatre receiver. They are normally two sets of electrically shielded cables with RCA plugs on each end, however some systems have adopted the XLR plugs and sockets used in much professional equipment.
Integrated amplifier: An amplifier consisting of both a preamplifier and a power amplifier.
Intermodulation distortion: If two tones are produced at the same time as each other, they can interact in a piece of equipment to produce other tones. Those tones are the sum and difference between the two original tones and, consequently, are generally not harmonics of either tone. As such, intermodulation distortion is generally more audible, and objectionable, than harmonic distortion. If the standard SMPTE test tones are applied (60 hertz and 7,000 hertz sine waves), then you can expect to see intermodulation distortion peaks at 6,940 and 7,060 hertz. In practice, small levels of harmonic distortion in the original signals will lead to other IM peaks, so you might see one at 6,880 hertz (ie. 7,000 - 2 x 60) and so on.
J
Jitter: Jitter is a type of distortion that can occur in digital audio. It is produced by slight divergences from the perfect timing of intervals between digital samples. With CD audio, the time between each pair of samples should be exactly 1/44,100th of a second, or approximately 22.67574 microseconds. With jitter, samples may be placed slightly earlier or later on the timeline. Unless extreme, jitter has no effect on audio quality while the signal remains in the digital domain, but manifests audibly when converted to analogue sound. If specified, it is usually stated as some average (often RMS) of the time variation in nanoseconds or picoseconds. The lower the number the better. (Jitter is also an issue in networking contexts, but there the units are more typically milliseconds.)
K
kB: Kilobyte. A measure of memory storage, a kilobyte equals 1,024 (ie. 2¹⁰) bytes.
kb/s: or kbps – Kilobits per second. A measure of the data flow rate for digital audio from a DVD. Stands for thousands of bits per second. Generally, the higher the number the better the quality.
kHz: Kilohertz. A measure of frequency: 1,000 hertz.
L
Latency: Latency is delay. When an audio signal is delayed, it may fall behind the video signal to such an extent as to introduce lip sync problems. Normally it’s the other way around, with TVs performing so much video processing that they delay the display of the video, potentially leading to similar lip sync issues. Most audio equipment designed for use in home theatre systems – especially home theatre receivers and surround processors – includes an audio delay feature. Some TVs can notify the audio device via HDMI of their inherent delay.
Layering: (Sound quality description.) Another term for talking about stage depth in a stereo image, used because it conveys how the sound stage can present as layers of instruments at different distances from the listener.
LDAC: LDAC is a digital audio codec developed by Sony for higher quality audio transmission via Bluetooth. It automatically adjusts its bitrate between 330, 660 and 990 kb/s (for 48kHz or 96kHz content) and between 303, 606 and 909 kb/s (for 44.1kHz or 88.2kHz content) to ensure a reliable connection. It is claimed to support up to 32 bits, although it's hard to see why. It is usually lossy, but if a 909kb/s connection is in operation, it is lossless with CD-quality 16-bit, 44.1kHz audio.
Lean: (Sound quality description.) A slightly lower level of bass than would be fully accurate. This can give a sense of tighter, more controlled sound.
LFE: Low Frequency Effects channel, sometimes imprecisely known as the subwoofer channel. This carries audio information covering frequencies up to 120 hertz and is designed to provide a substantial foundation to the bomb blasts, rumbling trucks and the like in movies.
LHDC: Low Latency High-Definition Audio Codec. LHDC is an audio codec developed for use with Bluetooth audio devices by Savitech – which seems to be associated with Huawei. It automatically adjusts its bitrate between 400, 560 and 900 kb/s and supports a bit depth of 16 or 24 bits, and sampling frequencies up to 96kHz.
Linearity: In any home entertainment system, at each stage the input and output signals should be precisely proportional to each other (except where specifically provided such as RIAA equalization or Dolby noise reduction). Thus if a momentary 0.5 volts is fed to the input of an amplifier and it produced a three volt output at its speaker terminals, then a momentary 1.0 volt input should produce a six volt output. Any other output means that there is a non-linearity which will manifest itself as harmonic distortion. Another example relates to digital to analogue converters. Each particular quantization level in a PCM digital signal has a precise analogue voltage level associated with it. Any variation means distortion.
Linear turntable: A turntable with a tonearm designed to eliminate tracking angle error. Rather than being pivoted at one end, with a linear turntable the tonearm is attached to a rail of some kind that allows the whole arm to move across the LP while remaining exactly parallel to the grooves. In some cases the turntable is a mere stub to hold the cartridge, which again tracks in a parallel manner across the record.
Line level: An electrical signal at a voltage level suitable for transferring a signal between components of a home entertainment system (for example, CD player to amplifier. Most modern equipment operates with a maximum of level of not much more than two volts RMS. Line level signals demand high impedance inputs. At least 10,000 ohms is suitable, although the de facto standard is 47,000 ohms.
Lip Sync: Short for “Lip Synchronisation”. With A/V material, the action on screen should be timed to match the audio. That could be voices matching lip movements, as well as other noises. Humans tend to not notice timing discrepancies for as much as 30 to 80 milliseconds – there is a wide variation between individuals – but once the individual threshold is reached, the effect of a timing mismatch is distracting.
Lossless compression: Lossless compression is to do with reducing the amount of digital data required to carry a digital audio stream, not anything to do with its dynamic range. Audio data does not typically compress very well using conventional information technology techniques. That’s why lossy techniques have been so popular. They can reduce the amount of data by more than 90% with only a modest effect on sound quality. However we now have much larger and faster systems for coping with digital data, so the typical 40% to 60% reduction in file size is tolerable. As the name suggests, once the data is decompressed, it is identical to the data as it existed before compression. More on lossless compression here.
Lossy compression: A common term for a system of compressing data using perceptual encoding.
Loudness control: A sad, sad circuit on a preamplifier that boosts the bass and, to a lesser extent, the treble of an audio signal. The idea was to overcome the reduced sensitivity of the ear to bass and treble at low volumes, but in fact this could only be achieved with any degree of realism by providing a complex means of calibration, which was never actually provided.
Loudspeaker: A device for transforming electrical energy into acoustic energy (ie. vibrations in the air). A loudspeaker normally consists of one or more drivers, a crossover network and an enclosure.
Low pass filter: An electrical circuit that impedes signals above a particular frequency. In other words, it lets signals below that frequency pass through. An example is the circuit in a home theatre receiver that allows bass frequencies to go to a subwoofer. Compare high pass filter and bandpass filter.
LP: Long Play record. The vinyl record that was introduced in the early 1950s to replace the shorter-play records previously used. Rather than their 78 rpm rotational velocity, the LP rotated at 33 1/3 rpm and used narrower grooves. Accordingly, it gave a playing time of between 15 and 30 minutes per side, depending on the how heavily the signal was modulated, the amount and character of its bass content and the amount of material available.
LPCM: Linear Pulse Code Modulation. For all practical purposes, the same as PCM. This tends to be called LPCM when dealing with DVDs.
LSB: Least Significant Bit. The bit of the binary number that varies the value of the binary number by no more than one (which is why it is the least significant). This bit is often randomised in PCM digital audio in order to add dither to the signal.
M
Magnetic cartridge: A turntable cartridge that produces the signal by moving a magnet and a coil with respect to each other. The movement is caused by the stylus in the grooves of an LP. The main types are moving magnet cartridge), moving coil cartridge and moving iron cartridge.
Mass loading: This is where mass is added to loudspeakers or stands to improve performance. A small number of loudspeakers are designed with a space inside the enclosure for the addition of loading material to increase their weight. And many loudspeaker stands feature hollow upright sections, designed to be filled. Commonly sand is used. This can increase the stability of the loudspeaker and stand and reduce possible resonances, particularly in hollow stands.
MB: Megabyte. A measure of memory storage, a megabyte equals 1,024 kilobytes or 1,048,576 bytes. However, if used as a measure of hard disk storage, the “mega” prefix normally means a round million.
Mb/s: or Mbps – Megabits per second. A measure of the data flow rate for digital video from a DVD. Stands for millions of bits per second. Generally, the higher the number the better the quality.
MC cartridge: Moving Coil cartridge. (See Moving coil cartridge.)
MDF: Medium Density Fibreboard. A material frequently used in the construction of high fidelity components, particularly loudspeakers and to a lesser extent turntables. It is made from breaking wood products down into their fibres and then combining them with a binder under high pressure and temperature into a workable wood-like product. They tend to be stable in shape, relatively devoid of resonances and easily workable. Many loudspeakers appearing to made of wood are actually veneered MDF.
Mellow: (Sound quality description.) Sound delivery where the midrange and lower treble is slightly recessed compared to the other tonal bands. It was long associated with the sound of English loudspeakers, although this kind of uniformity of sound is less common these days.
Memory stick: A small, removable flash memory (ie. non-volatile) cartridge developed by Sony, used in a host of digital storage situations, notably in digital cameras and digital audio players.
Metadata: Formally, Metadata is data about the data. So, it includes things like the “flags”, or predefined digital bits, carried in a Dolby Digital bitstream that provide instructions to the decoder. They do not carry actual audio data, but guide the decoder in the interpretation of the audio data. Examples include a flag for dialog normalization and some for dynamic range control. It also includes things like the tags in digital audio files identifying the artist, track title and so on.
MHz: Megahertz. A measure of frequency: 1,000,000 hertz.
Midrange: The audible frequencies typically constituted by frequencies of between 150 and 5,000 hertz, although the dividing lines between midrange and bass at the bottom end, and midrange and treble at the top end, are ones of opinion. The human ear is most sensitive to midrange frequencies.
Midrange driver: A middle-sized driver designed to reproduce the important midrange frequencies between the bass and treble notes. If this is omitted, the loudspeaker is called “two-way”. If present, along with a woofer and a tweeter, it’s called “three-way”. Some loudspeakers use a midrange driver that looks identical to the two or more woofers in the speaker box. However these drivers are usually tuned differently.
Minidisc: A digital audio music delivery system using a disc mounted inside a small, robust plastic cartridge. The audio is encoded using ATRAC. Interestingly, the audio may be carried on the disc in two different ways. Sony designed it both for acting as a recorder, and playing back commercially-produced Minidiscs. So in the former role it acts as a magneto-optical recorder. For the latter, it uses discs which have been pressed in a similar manner to commercial CDs.
Mistracking: This is where a turntable cannot handle the groove modulations on a vinyl recording. This can be the case with some LPs which have relatively extreme modulations, or when the stylus, cartridge or tonearm have not been set up correctly. One common cause is too low a tracking force. In extreme cases -- allegedly, and it's something I've never experienced -- the stylus can be flung from the groove. See tracking.
MLP: Meridian Lossless Packing. A digital audio standard used for DVD Audio. Also known as Packed PCM or PPCM. This provides superb sound, from mono through to multichannel, with a number of technical advantages over Dolby Digital and DTS. The middle word, “Lossless”, is important. It is not a perceptual encoding system that abandons some of the original source material to achieve space savings. It preserves the original signal perfectly, even through multiple encoding/decoding cycles.
MMC: Multimedia Card. A small, removable flash memory (ie. non-volatile) cartridge, used in a host of digital storage situations, notably in MP3 players. This seems to be going out of favour.
MM cartridge: Moving Magnet cartridge. (See Moving magnet cartridge.)
Mono: Or Monophonic. Where just a single channel carries all the sound. Even if played back on a system with two or more speakers, the sound remains mono because the left and right front speakers are delivering identical signals. Contrast with stereo and surround sound.
Moving coil cartridge: A magnetic cartridge in which the stylus moves a coil via the cantilever, while the magnet is fixed in position. Moving coil cartridges tend to have lower moving mass than moving magnet cartridges, but also tend to be lower in output by an order of magnitude. There are, however, high output MC cartridges available that produce comparable levels to MM cartridges. Because the coil must be wired to the outputs, they also tend to be somewhat lower in compliance than MM cartridges, so are not normally amenable to very low tracking weights (they typically operate best at around two grams).
Moving Iron cartridge: The most common turntable cartridges are moving magnet and moving coil models. But a third kind – the moving iron cartridge – has certain benefits of both. Rather than a heavy magnet fitted to the cantilever (as in MM cartridges) or the coil attached to the cantilever (MC cartridges), a light-weight armature is attached to the cantilever. A powerful nearby permanent magnet induces a magnetic field in the armature. Its movement in turn induces an electrical signal in the fixed coil. So its moving parts are lighter than for MM, but its output is much higher than that of MC.
Moving magnet cartridge: A magnetic cartridge in which the stylus moves a magnet via the cantilever, while the coil is fixed in position. Moving magnet cartridges tend to have a higher output than moving coil cartridges, but also tend to have a higher moving mass (possibly reducing their ability to deliver fine detail from record). Because the magnet is able to freely move, MM cartridges generally offer a higher compliance than MC cartridges, so the cartridges with the very lowest tracking weights (0.75 to 1 gram) come from the MM camp.
MP3: MPEG 1 Layer 3. This is a lossy digital audio compression standard that achieves high levels of compression of mono or stereo sound through perceptual encoding techniques. Close to CD quality sound can be achieved for most kinds of music with only 9% of the data carried by a CD, making this format good for Internet music transportation and small portable solid-state players. While MP3 supports constant bit rates of 8 to 320 kb/s, as well as variable bit rates, the most common encoding rate is 128 kb/s CBR.
MPEG: Motion Picture Experts Group. In practice, this stands for a number of audio and video compression standards. The video on DVDs is compressed according to the MPEG-2 standard, which permits not only compression of each single frame of the picture, but higher levels of compression by comparing frames with each other and only encoding their differences. MPEG audio also appears on some DVDs. This permits up to 7.1 channels of audio to be encoded, although only 5.1 channels is ever seen in practice and its use is now obsolete. MP3 is one particular form of MPEG stereo.
MQA: Master Quality Authenticated. MQA is a lossless digital audio compression system as well as a system to authenticate the quality of the digital audio. The main compression feature is the ability to “fold” ultrasonic content down into the lower frequency bands at low levels, where they can be detected by the process and unfolded again to yield content with up to 384kHz sampling. There has been some controversy about the true losslessness of the system. MQA-content is most readily available from TIDAL, and many DACs support it.
Multichannel sound: Often called surround sound. Any system designed to deliver more than two channels (stereo) of sound to the consumer. The most common forms these days deliver 5.1 channels of audio. Some systems can process two channels, or even one channel, of sound to produce the effect of multichannel sound.
N
Near field: This is where the listener sits very close to the loudspeakers, typically within a metre or so. Because the sound heard directly from the speakers is so much louder than the sound reflected from room boundaries such as walls and the ceiling, this form of listening can deliver a more accurate and precise rendition of the sound. Audio engineers often use near field monitor speakers in their work for this reason.
Network Audio Controller: A network audio controller is typically an app that runs on a smartphone or tablet. It interfaces with a network audio server and a network audio renderer, telling the server which audio to send to the renderer. It controls both devices, but the audio itself does not run through the app. It goes direct from server to renderer. The data the app does receive from the server is information about its contents, such as lists of artists, cover art and so on. For more detail, see here.
Network Audio Player: A network audio player is a device capable of reading information from a network server, presenting it to the user through some interface so they can select the content they wish to play, and then rendering the digital audio. These also typically have DACs and amplifiers built in, operating more or less as all-in-one devices. For more detail, see here.
Network Audio Renderer: A network audio renderer is a device capable of accepting a digital audio stream provided by a server and converting it into regular digital audio, capable of being decoded to analogue by normal audio components. Often a renderer will have that capability built in. Often, a renderer is also a network audio player. For more detail, see here.
Network Audio Server: A network audio server is a combination of hardware storing digital audio, and software which provides that digital audio to other devices. It may be a dedicated device from a high-end audio company, software which runs on regular network attached storage devices on a person’s home network, or even digital audio stored on a computer and served up by software running on that computer. There are various protocols employed for this functionality, but the most common is DLNA/UPnP. For more detail, see here.
Network audio streaming: Network audio streaming is the delivery of digital audio from an Internet service (for example, Spotify, TIDAL, Apple Music and many others, including tens of thousands of Internet radio stations). The delivery may be managed by an application on a computer, by an app on a smart device, or by functionality build into an audio device. There are many dedicated network streamers on the market, including high-end audiophile models. In most cases, they also support streaming digital audio from local network servers.
Neutral: (Sound quality description.) A style of sound that exhibits no significant tonal bias.
Noise: Noise is a thing you don't want in your high fidelity sound system. Noise is any extraneous sound added to the real signal, apart from such distortions as THD and IMD which are accounted separately. Often it will appear as white noise, but it may have other characteristics, for example bass components related to mains frequency. But since noise is inevitable and unavoidable, the real-world solution is to reduce it to the lowest possible levels, preferably below the ability of the human ear to perceive it.
Noise shaping: A system of adding dither noise to a digital audio signal so that it is biased towards high frequency noise. This permits the noise floor to be lower in the range of frequencies to which the ear is most sensitive, at the cost of poorer noise performance in the higher frequencies where the ear is less sensitive. Read more.
Nominal impedance: A specification of the load a loudspeaker places on an amplifier, measured in ohms. It tends to oversimplify the actual situation because the impedance of a loudspeaker varies according to frequency. So a speaker with a nominal impedance of eight ohms may actually present a load of six ohms at some frequencies and thirty ohms at others. Common loudspeaker impedances are four, six and eight ohms. Back in the 1960s, higher impedance speakers – 16 and 32 ohms – were fairly common.
NOS: Non oversampling. Some DACs feature a NOS option or in some cases are fixed in that mode. The term describes a legitimate and long-standing method of digital to analogue conversion, but these days is associated with the lack of the usual output filter. That leads to excessive ultrasonic noise in the output. Nonetheless, it has become fashionable amongst some audiophiles to suggest that these filterless DACs sound better. See here for more.
O
Octave: A range of frequencies of some form of repetitive wave, where the highest frequency is precisely twice the lowest frequency. In music, for example, the fundamental frequency of A below Middle C is 220 hertz. The A above Middle C is 440 hertz.
Ohm: A unit of electrical resistance or impedance. In a DC circuit, the number of ohms of resistance offered by a component can be calculated by dividing the voltage across the component by the current (in amps) flowing through it.
Optical digital: A method of communicating digital audio between components using light carried on optical fibre. See TOSLink. The data format accords with the S/PDIF specification.
Optical soundtrack: The traditional medium for carrying a film’s soundtrack. The sound was converted to markings which were developed onto the edge of the film. The majority of optical soundtracks were analogue (in the form of squiggly lines, like the track on an LP). However Dolby Digital is also carried on the film in optical form. Rather than replacing the analogue soundtrack (which is left on for broad compatibility), the Dolby Digital data is optically recorded between the sprocket holes on the edge of the film.
Opus: Opus is a relatively recent (2012) lossy audio format. Like FLAC (it comes from the same organisation, Xiph.org), Opus is an open-source, free codec. It was designed to provide high quality for low bitrates, and very low latency. Few people will be aware of it, but come across it in everyday life since most YouTube videos use Opus, as do such as the voice-calling feature in WhatsApp.
Oscillation: A movement to and fro at some particular frequency. It could be physical or electrical. Musical tones are sometimes generated by means of electronic oscillators. A guitar string oscillates.
Out of phase: Typically this refers to a set of stereo loudspeakers which have been wired incorrectly. If one channel has the + and - connections wired correctly, and the other has them wired in reverse, the sound will be quite strange. It will tend to float around in a diffuse way, and bass will be quite weak. That's because any shared content between the channels -- which with just about all recordings is most of it -- there will be a lot of destructive interference between the channels. The solution is simple: switch the + and - connections on one of the channels.
OTG: On the Go. A capability of many Android phones and tablets to support external devices via their USB (usually Micro-B or USB Type-C) charge ports. This is often such things as USB storage, but one use is to connect an external DAC for higher quality audio performance.
Overblown: (Sound quality description.) Excessive bass, particularly mid-bass.
Overhang: (Sound quality description.) When a note, particularly an impulse or transient continues to sound even though it should have stopped. This is more common with bass with poorly damped systems.
Oversampling: In the early days of the CD, many criticisms were levelled at the format for perceived deficiencies in sound quality. One claim was that the sharp analogue output filter required between 20,000 and 22,500 hertz caused phase shift and other problems. A range of improved filtering techniques were developed, including oversampling. That generates a higher sampling rate to define the signal, and then employs a less sharp filter to avoid the original problems.
Overtone: See harmonic.
P
Parametric EQ: A sound equalisation system primarily distinguished by the ability to set one or more frequency bands to be increased or reduced. Typically there are three parameters for each band: the centre frequency of operation, the level of increase or decrease, and the "Q", which is the width of band. This allows a tremendous amount of flexibility in adjusting EQ.
Passive loudspeaker: A loudspeaker without built-in amplification for all the drivers. Most loudspeakers are passive, not active.
Passive subwoofer: A subwoofer without built-in amplification. Most subwoofers are active, not passive.
PCM: Pulse Code Modulation. This is standard, uncompressed digital audio of the kind that is on compact discs. However while normal CDs are always encoded with 16 bits and 44,100 hertz sampling rate, on DVD bit depths of up to 24 and sampling rates of up to 96,000 are permitted.
Perceptual encoding: A compression technique for digitally recorded sound. Digital audio does not compress at all well using conventional “non-lossy” schemes because there is very little redundancy in the data stream. Perceptual encoding techniques rely on extensive psycho-acoustic analyses of how the human ear and brain detect and interpret sound. In essence they manage to achieve high levels of compression by discarding elements of the sound that such analyses suggest cannot be heard. In this way they can reduce the data volume of the sound by a factor of 10 or more without obviously degrading the sound quality. Examples of perceptual encoding compression systems are Dolby Digital, DTS (although DTS claims that it first uses non-lossy compression techniques to reduce or eliminate the need for perceptual encoding), MP3, SDDS, Windows Media, RealAudio and ATRAC.
Period: In relation to a repeating wave form, the reciprocal of the frequency (ie. 1/f).
Phantom image: A phantom image is another name for the sound that appears to come from between your stereo loudspeaker. When it's dead-centre, this sound is a result of both loudspeakers producing an identical signal. This wasn't really a major term in high fidelity audio until home theatre introduced the centre channel loudspeaker, which obviously tied the formerly "phantom" image to a specific speaker.
Phase: The alignment or otherwise of two signals in time. “In phase” means that they are aligned. “Out of phase” means that one or the other is shifted in time by one half the period (inverse of frequency) of the tone in question. If an out of phase pair are played together, they will cancel each other out. Phase is usually measured in degrees. A zero degree shift is in phase, while a 180 degree shift is out of phase. Causes of phase shift include filters (especially those in loudspeaker crossover networks) and the (lack of) time alignment of drivers in a loudspeaker.
Phon: A unit of sound which measures perceived loudness. Unlike the more commonly used decibels of SPL, the phon is tied to human perception. Since we all differ somewhat, it based on the standard equal loudness contours which assess human sensitivity to sounds of different frequencies. By definition, a 1kHz tone is the same level whether represented as phons or as decibels SPL. For most other frequencies, the phon level will be lower than the decibel level.
Phono: From phonogram. An adjective relating to systems for playing vinyl recordings, such as LPs. Thus the phono input of an amplifier is the input into which a turntable is plugged. This is usually equipped with an RIAA equalization circuit.
Phonogram: An old name for a record player, that is, a combined system with turntable, amplifier and speakers, usually all contained in one cabinet.
Piezo tweeters: High frequency loudspeaker drivers that, rather than using conventional voice coils, rely on the ability of some crystalline materials to physically respond to the application of a signal. They are usually horn loaded. Such tweeters have the advantage of having a very high power handling capability, so they are often seen in professional sound reinforcement installations.
Pink noise: Random audio noise where the average amount of power is the same for each octave. The power level of pink noise falls away by three decibels for each increasing octave. Frequently used as a test signal, it more accurately reflects the energy content of real-world audio than white noise.
PMPO: Peak Music Power Output. This is a fairyland measure of power output, sometimes quoted for home stereo systems. Sometimes figures of well over 2,000 watts PMPO are claimed. This is in the same class of reality as saying my Nissan Nomad van will travel at up to 240 kilometres per hour. It will, but only after it has fallen from a very tall cliff.
Port: The hole in the enclosure of a bass reflex loudspeaker.
Power: The amount of energy expended per unit time. The common unit is the watt. Power (in watts) in a DC circuit equals voltage times current (in amps). Things are a bit more complicated with AC. See here.
Power amplifier: A device that increases a line level signal (typically around two volts into a high impedance) and boosts it in voltage, while allowing sufficient current to be supplied to drive low impedance loudspeakers (that is, impedances nominally in the four to eight ohm range). Most power amplifiers are analogue in design and most use transistors, although some analogue amplifiers use valves. New digital amplifiers are appearing which effectively use a form of pulse width modulation, low pass filtered to reduce the ultrasonic noise, to drive the loudspeakers.
Power supply: Within a home entertainment device, the section that provides a suitable voltage and current for the operation of the device. This typically consists of a transformer, rectifier (to turn AC into DC) and regulator (to smooth out fluctuations in the voltage). Power amplifiers in particular are heavily dependent upon the quality of their power supplies, which contribute a significant proportion of their cost and weight.
PPCM: Packed Pulse Code Modulation. Another term for MLP.
Preamplifier: Sometimes called “Control Amplifier”. Nominally this boosts a low level signal to a level suitable for delivering to a power amplifier, but in these days of fairly high output source components (such as CD players), their main function is to provide a volume control and switch between different sources. Some preamplifiers may incorporate such things as loudness and tone controls and mute buttons. Many preamplifiers also incorporate a phono preamplifier which boosts the low level signal (say 5 to 100 millivolts) of a phono cartridge’s output and applies RIAA equalization.
Precedence effect: See Haas Effect.
Presence: (Sound quality description.) A nearly indefinable sense of reality in the music, almost as though you're in the presence of the performers.
Psychoacoustics: The study of the human perception of sound. This has informed a wide range of high-fidelity technologies, including the development of the perceptual encoding systems as the core of lossy digital audio compression formats such as MP3, and the development of the equal loudness contours.
Q
Quantization: Or “quantisation” in Australian. The numerical value assigned to a particular analogue voltage input at an instant of time in an analogue to digital converter, or the process of making such an assignment. The range of numerical values available, and consequently the accuracy of the assignment, depends on the bit depth of the PCM system employed.
Quantization noise: Or “quantisation noise” in Australian. A form of harmonic distortion produced in all digital systems. Low level signals do not use all the bits available in the digital system (that’s why they are low level signals!) Consequently they operate as though the digital system has a low bit depth (say, an eight bit system or even four). This increases their harmonic distortion. By its nature, harmonic distortion generated by digital systems tends to give relatively high levels of odd-order harmonics, which are far more irritating than the even order harmonics to which analogue systems tend. This is routinely addressed by adding dither to digital signals. See more here.
R
RCA plug/socket: The most common audio and video connection standard. In audio gear they are most commonly used carrying analogue stereo or multichannel sound from component to component. For stereo they are typically colour-coded white and red for left and right. Also frequently used for S/PDIF digital audio. Sometimes called a “cinch” plug/socket.
Reactance: The aspect of impedance which varies according to the frequency of the signal it is affecting. Measured in ohms.
Receiver: In the stereo world, an amplifier with a tuner built in. Often used as short hand for home theatre receiver.
Recessed: (Sound quality description.) When some element of the tonal band is lower in level than the rest, that area is said to be recessed.
Resistance: The aspect of impedance which remains constant regardless of the frequency of the signal it is affecting. In the case of DC voltages, the resistance is the same as the impedance. Measured in ohms.
Resolution: The amount of detail a system is capable of recording or producing. For digital audio, it is largely determined by the bit depth.
Resonance: Certain frequencies of sound oscillation are favoured by particular systems. This is most obvious in mechanical systems. Where a weight is on a spring, the weight will oscillate up and down at particular frequency – number of times per second – which is determined by such factors as the stiffness of the spring and the amount of weight. Loudspeakers are mechanical systems, so they resonate, sometimes in quite complicated ways. Well-designed loudspeakers take account of that. Some electrical circuits also have resonances for signals of particular frequencies.
Resonant frequency: The frequency at which a system offers the lowest impedance, allowing the highest power transfer. In the case of physical systems, it is the frequency at which the vibration is greatest for a given input. For loudspeaker drivers, the resonant frequencies of the drivers are very important characteristics to take into account in the design process, avoiding placing them in the range of operation for higher frequency drivers, and carefully designing the enclosure to make best use of the resonant frequency of the woofer.
Reverberant sound field: A speaker system in which a significant amount, perhaps most, of the sound that you hear is coming from reflections from surfaces within the listening space rather that directly from the loudspeakers. Reverberant sound field speakers tend to have a larger sweet spot, and produce a more rounded stereo image with greater depth than direct sound field speakers. Reverberant sound field sound can be achieved by choosing speakers with a very wide dispersion, such as bipole speakers, or by purpose designed speakers from companies such as Bose.

RIAA Equalization: Record Industry Association of America Equalization. When converting an electrical signal to mechanical motion, there is an inconvenient fact: the lower the frequency, the higher the excursion of the moving part for a given signal level. This is important in several fields, particularly with LPs. The grooves of an LP would wobble very widely if the bass frequencies were cut in direct proportion to the signal. The playing time would, consequently, be very short and turntables would have a great deal of difficulty in holding their styli in place within the groove. So before LPs are cut, the signal is subjected to RIAA equalization. This reduces the bass by an enormous amount (-17dB at 50 hertz), and increases the treble equally hugely (+13dB at 10,000 hertz). A circuit built into the phono preamplifier of the playback equipment reverses this EQ, boosting the bass back to its proper level and handily cutting the treble. This last has the welcome effect of substantially reducing the hiss produced by the scraping of the stylus within the groove. Note: old-style ceramic cartridges did not require this second step of reversing the EQ because their frequency response characteristics roughly approximated this anyway.
RMS: Root Mean Square. This is a measure of voltage or (incorrectly) power output. The latter is often quoted in the specifications for an amplifier or home theatre receiver. Measuring voltage is simple with direct current, even if it varies in level. You can simply average the values over time. But alternating current, when arithmetically averaged, gives you a value of zero or close to zero. If you square each value (multiply it by itself) before taking the average, and then take the square root of that average, you not only get a reasonable representation of the voltage, but a value that provides effectively the same amount of power as a similar DC voltage into a resistive load. What is called “RMS power output” for amplifiers is calculated using the RMS voltage delivered by the amplifier to test loads, so in that sense the term is not entirely misleading (however something like “average sine wave power output” is far less irritating). Such figures should include a measure for the amount of distortion generated at that power. This should be well under 0.5%. If the figure is 10% (often used with home entertainment systems) the figure should be discounted substantially.
Roll-off: The behaviour of the output at either frequency extreme in virtually all systems. When the bass descends below a certain frequency or the treble extends beyond one, the output diminishes, often quite rapidly depending on the design. With some DACs the steepness and style of high frequency roll-off is user adjustable. See here for the effect of these adjustments.
Rumble: Low frequency noise, usually from a turntable. Given that RIAA equalization boosts the bass signal from a cartridge by an enormous amount (17dB at 50 hertz), the highest quality bearings and excellent isolation from the turntable’s motor are required to control rumble.
Run in groove: The silent groove near the outer circumference of an LP or other record. This section of groove precedes the start of the music and is provided for the stylus to be lowered to the surface of the record and find the groove before the music commences.
Running in: A frequently recommended process for the installation of new equipment. Essentially, running in is operating the new equipment for some hours, or tens of hours, to bring it up to peak performance. As with many such areas of performance, there is both validity and silliness in this. Loudspeakers are physical systems with suspensions and surrounds that do benefit from being run in. But running in speaker cables achieves precisely nothing. Those who claim that their speaker cables sounded better after being run in are imagining the improvements. Of course, I could never prove this because A-B testing is impossible.
Run out groove: The silent groove near the label at the centre of an LP or other record. This section of groove follows the end of the music (although some records did include material in it for reasons of novelty or art). Initially a spiral for a few revolutions, it ends up turning into a completed circular groove around the label. With a manual turntable, it is up to the user to lift the tonearm to preserve the stylus. Some turntables offer various degrees of automation to lift the arm when it reaches a certain point, or turn off the turntable motor.
S
SACD: Super Audio CD. A new audio format developed by Sony and Philips. This takes on DVD Audio by offering very high resolution sound, using DSD digital encoding. SACDs can deliver up to 5.1 channels of sound, but even if they do they are required by licence to also carry a two channel sound track. SACDs are most commonly made in a two layer hybrid format. One layer provides SACD music (stereo and, optionally, multichannel) while the other is fully compatible with CD players. The SACD specifications also provide for some video-carrying capability, but neither disc nor player offering this has yet been seen.
Sample: One from a series of digital measurements taken of an analogue signal at regular intervals. The timing of the intervals is determined by the sampling frequency. The accuracy of the sample is determined by the sample’s bit depth and the quality of the ADC.
Sampling frequency: The number of digital samples taken each second of an analogue audio signal. For the compact disc, this is always 44,100 samples per second (usually expressed as hertz). For DVD it is typically 48,000 hertz, but may be 96,000. The higher the figure, the more accurately the analogue source is recorded, giving an extended high frequency response, and the more space the signal requires.
Satellite speakers: Small loudspeakers designed to deliver only midrange and high frequency audio. They are intended to operate in conjunction with a subwoofer which delivers the bass.
SBC: Subband Coding. Or, more fully “Low-complexity subband coding”. This is the standard lossy codec used by Bluetooth for carrying stereo music. It supports 48kHz audio at 16 bits, and employs a bitrate of around 345kb/s. Other codecs can be used instead if supported by both the transmitting and receiving device.
SBM: Super Bit Mapped. An analogue to digital conversion system developed by Sony that incorporates a form of noise shaping, designed to deliver very good performance results in the middle frequencies, at the cost of a poor signal to noise ratio in the high frequencies.
SD: Secure Digital. A small, removable flash memory (ie. non-volatile) cartridge, used in a host of digital storage situations, notably in digital cameras and digital audio players. Physically it is very similar to the MMC, but incorporates support for digital rights management. Major non-Sony Japanese consumer electronics companies tend to support this format.
SDDS: Sony Dynamic Digital Sound. A discrete channel surround sound system developed by Sony, based on its ATRAC sound compression technology for the Minidisc. It provides eight channels of sound and is optically printed on the film, using the space outside the sprocket holes. The eight channels are allocated as LFE, surround left, surround right, and five front channels. This adds a centre left and centre right channel to the three front channels provided by other systems. SDDS is used only in cinemas, not home theatre. It uses perceptual encoding compression techniques to reduce the amount of data space required to carry the signal.
Sensitivity: A measure of the efficiency with which loudspeakers turn the electrical energy provided by a power amplifier into acoustic energy. The more sensitive, the greater the volume for a given amount of power. This is normally measured as the sound pressure level in decibels (dBSPL) achieved by the loudspeaker in an anechoic chamber at a distance of one metre with a 2.83 volt 1kHz signal applied. (2.83 volts is the voltage required to deliver one watt to an eight ohm load.) Using my own methods, I have measured sensitivities ranging from not much more than 80dB to nearly 100dB. Each 3dB increase in sensitivity is equivalent to doubling the amount of power, so for a loud system it is far better to choose sensitive loudspeakers rather than pay for a higher-powered amplifier.
Separation: A measure of the degree to which leakage from one channel of sound to another channel (crosstalk) is limited. This is typically measured in decibels (eg. -90dB at 1kHz). While great emphasis is placed on this figure, the reality is that very modest figures like -20 or -30dB (typical of LP records) provide excellent stereo separation and imaging. More important is that the separation should not vary widely between different frequencies, since this could lead to a positioning mismatch between the fundamental and harmonic frequencies for particular instruments.
Shibata stylus: This is a turntable stylus with a very narrow profile, but a relatively long contact length with the sides of the groove of an LP. The Shibata stylus was developed in the early 1970s for playback of LPs produced for the CD-4 quadaphonic four-channel sound system. This system relied on a frequency response out to 50kHz. The narrow profile of the Shibata allows such high frequency modulations of the groove to be negotiated, while the reach increases surface area, reducing wear.
Shielding: A finely woven mesh of thin wires, or a conductive foil wrapping, around a signal cable under the outer layer of insulation. The shielding is earthed and acts to protect the signal-carrying wires within from electrical fields which could introduce noise into the signal.
Sibilance: An undue emphasis on "S" sounds, in extreme cases making them almost hiss. This is due to an output peak in the lower to mid treble. Note, often the sibilance is actually in the recording which a fine system will faithfully reproduce.
Signal to noise ratio: A specification for the level of noise produced by a system. This is normally expressed in the decibel difference between the measured noise and some reference signal.
Sine wave: An electrical signal or tone that follows a sinusoidal shape. The shape is ubiquitous in nature. All repeating waves can be generated by a combination of a sine wave with various harmonics of that wave (see Fourier).
Skate: The rather unintuitive tendency of the stylus of a turntable cartridge to seek to slide towards the centre. This is due to the geometry of the tonearm and results in a greater force being applied to the inner side of a groove than the outer side. Decent tonearms have an anti-skating device. This subject is explored in greater detail here.
Slam: (Sound quality description.) This refers to a bass delivery powerful and fast enough to deliver a visceral impact to the listener's body.
Slew rate: This is a measure of the rate at which a voltage can change. It is sometimes specified for power amplifiers and in that context is usually expressed as volts per microsecond. A high slew rate can make for impressive looking square wave performance. Typically the slew rate is largely determined by the bandwidth of the device – a wider bandwidth permits a higher slew rate. The important measure here is bandwidth. However with power amplifiers, it can also be limited by an inability to instantly respond with the appropriate amount of current. So, in short, a high slew rate is a good thing.
Smart Media: A small, removable flash memory (ie. non-volatile) cartridge, used in a host of digital storage situations, notably in digital cameras and digital audio players. It should be noted that many MP3 players reformat Smart Media cards in a way that will not permit them to be compatible with digital cameras.
Smearing: (Sound quality description.) Where a system lacks coherence, the different elements of sound can be slightly out of time with each other, smearing the sound. The result is a loss of both detail and dynamism.
SNR: Signal to Noise Ratio. (See Signal to noise ratio.)
Sound Stage: (1) A movie set where audio is recorded along with video. But in the home entertainment context, the sound stage is area between a pair of stereo speakers from which they appear to make the various sounds appear. Some speakers manage the feat with some recordings of making the sound stage actually wider than this space, and providing sensations of both vertical sound placement and depth in the stage.
Sound Stage: (2 - Sound quality description.) This the space between and around your stereo loudspeakers in which the stereo image is presented.
Space: (Sound quality description.) This is a quality in the sound in which each instrument can be heard separately from the whole, including a sense of air around the instrument. Excellent control and resolution is required of a system to achieve this with more complex music and large ensembles.
S/PDIF: Sony/Philips Digital Interface Format. A widely used digital audio protocol. It is used as the protocol for all consumer home entertainment equipment. It is distinguished from the professional AES/EBU protocol by incorporating the clock timing information in the main signal. Originally designed for 44.1 and 48 kHz and 16 bits, it now carries up to 96kHz and 24-bits, plus the bitstreams for the various compressed digital audio standards. The connections used are generally coaxial or optical.
Speaker: Short for loudspeaker.
Speed of propagation: Physical waves, whether sound or electromagnetic, have a typical speed of propagation through various media. This varies depending on the medium. Light travels through a vacuum at 3x10⁸ metres per second. Sound travels through room-temperature air (20C) at 343.5 metres per second. The speed varies slightly according to temperature, increasing to 349.3m/s at 30C, falling to 337.6m/s at 10C.
Spindle: The short vertical rod at the centre of the platter of a turntable. The spindle is usually a close match for the hole at the centre of an LP or other recording.
SPL: Sound Pressure Level. A measure of volume or loudness in decibels. 0dBSPL is the quietest that can be heard. A rock concert may produce up to 120dBSPL. The threshold of pain is a little above this. Hearing damage is a function of loudness and the frequency and period of exposure.
Square wave: A signal, the shape of which when displayed is, well, square. In other words, it holds to a negative value for a time equal to half the wave’s period, then switches abruptly to a positive value which it holds for the same length of time, then switches back negative again, and so on. How quickly it switches between these states is determined by how far into the higher frequencies the equipment in use can extend. Square waves rarely form part of music, and are primarily used as diagnostic tools, since inspection of how well equipment handles a square wave can reveal a lot about its high frequency handling and whether it shifts phase across its frequency handling range. For those who are mathematically inclined, a square wave can be defined by the formula, where f is the fundamental frequency and n is an odd natural number: sin(f) + 1/3 sin(3f) + 1/5 sin(5f) + ... 1/n sin(nf).
Stage Depth: (Sound quality description.) This is a sense of depth that can be apparent in the sound stage of a high quality stereo image. Also known as layering.
Stereo: Or Stereophonic. In the home, an audio system which delivers two channels of music, left and right, to create the illusion of a plane of sound facing the listener. Some carefully produced recordings can use these two channels to also give a sense of fore-aft depth in that space between the speakers, occasionally generating an illusion of sound even behind the listener. Confusingly, in the professional cinema context, stereo means surround sound. Contrast with mono and surround sound.
Stereo imaging: See imaging.
Stylus: A small diamond on the end of a cartridge‘s cantilever. This sits within the groove of an LP and picks up the vibrations recorded therein. The stylus is generally spherical or elliptical in shape, although some variations have been developed, all with a view to more accurately tracking the groove while reducing damage to it. Elliptical styli are only suitable for tracking weights of less than around two grams because their low contact area with the groove can cause damage.
Stylus Rake Angle: Closely related to VTA. It’s to do with setting a tonearm’s height so that the angle of the stylus with respect to the surface of the record matches that of the cutting device which first produced the record. This can be difficult to do since different cutting machines used different angles – although 92 degrees is considered to be the most common – and the angle changes depending on the thickness of the LP being played, any warp in the record and even the tracking weight. For those not seeking to follow this particular rabbit hole to the bottom of the burrow, it’s generally best just to follow cartridge manufacturer’s instructions, which is usually to set the height of the arm (if adjustable) that ensures the cartridge is parallel to the surface of a typical LP. The extent to which this adjustment affects sound quality is a subject of some dispute within the high-fidelity community.
Subjective reviewing: Subjective assessment is a vital component in judging any piece of audio equipment. We have not established a set of tests which fully and accurately describe the perceived performance of audio or video components. Particularly with things like loudspeakers, an experienced listener can come to a valuable view of the performance just from a brief listen. That said, subjectivism does not negate the equal value of objective tests and has far too often been taken to ridiculous extremes. One giveaway is when the reviewer starts talking about his or her feelings. If a system is claimed to be more able generate an emotion in the listener than another system (except, perhaps, for anger and irritation when it fails to work properly), then it is time to stop reading. Emotions are primarily generated by the interactions between the music, the listener’s tastes and the listener’s mood. These things will swamp any due to the subtle differences between two CD players or amplifiers.
Subliminal: (Sound quality description.) Sounds which aren't consciously audible, but may have an impact on the listening experience at some deeper level.
Subsonic: Less than the speed of sound. Often incorrectly used when infrasonic is meant.
Subwoofer: A speaker designed to produce only deep bass frequencies. Most subwoofers are “active” models, which means that they have an amplifier built in. “Passive” subwoofers require an external amplifier. Active subwoofers also have a level control and, usually, an upper frequency control, although an increasing number are appearing without features not required when connected to a home theatre receiver’s Dolby Digital (or whatever) LFE output, which covers a frequency band from 3 to 120 hertz. But to achieve useful output from a subwoofer below around 20 hertz, expect to pay big dollars.
Surround Sound: An audio system which delivers sound from behind, as well as in front of, the listener. This can be artificially generated by “virtual surround” systems which process a stereo signal to produce the effect of sounds from the rear, even with just two speakers. More commonly, though, it refers to systems such as Dolby Pro Logic, Dolby Digital and DTS which were designed to deliver specific channels of surround sound from speakers placed slightly behind and to the sides of the listener. These typically also use a centre front channel and, in some newer versions, an additional centre rear channel. In the professional cinema context, surround sound is often called stereo. Contrast with mono and stereo.
Sweet: (Sound quality description.) A smooth treble with no sense of harshness, allowing things like violins to appear with their full tonal complexity intact, while avoiding any scratchiness.
Sweet spot: The seating position or positions at which a stereo or surround sound system produces its best effect, particularly with regard to imaging.
Switch Mode power supply: The switch mode power supply is now quite common in audio systems. These use transistors flicking between full on and full off at extremely high frequencies. The output voltage corresponds to the proportion of the time that the transistors are on. The output is regulated to remove (or at least soften) the output bumps. Because the full on and full off states of transistors use very little power, switch mode power supplies are very efficient. However, poorly designed ones can introduce noise into a system and they may prove inadequate in high-current-draw applications. In general, high end audio gear tends to use more conventional power supplies, based on heavy duty transformers.
T
Theile and Small parameters: Back in the 1960s the Australian engineer Neville Theile and American engineer Richard Small developed a method of modelling loudspeakers. This was groundbreaking work, allowing the bass performance of loudspeakers to be explicitly designed, rather than developed through trial and error. Many decades later their work remains the basis of loudspeaker design. To use their design techniques a number of characteristics of the woofer to be used must be known. These include such things as the driver‘s resonant frequency, its compliance, its “equivalent volume” and so forth. These are collectively known as the Theile and Small parameters and are stated by all reputable driver makers for those drivers they sell (although not necessarily for drivers that are only sold installed in loudspeakers).
Three-way loudspeakers: A loudspeaker which divides the incoming signal into three different frequency bands for distribution to drivers. It sends high frequencies to the tweeter, the middle frequencies to the midrange driver, and the low frequencies to one or more woofers.
THD: Total Harmonic Distortion. (See total harmonic distortion).
THX: A certification standard controlled and operated by LucasFilm. It covers certification of both cinemas and home entertainment equipment to meet given minimum standards. Originally, on the home entertainment front, there was just one standard: “THX Certified”. But in recent years this has been replaced by two standards: “THX Ultra” certification, which is the highest level and is similar to the old standard, and “THX Select” which is a lesser standard. Note that many brands and products which could easily achieve THX certification decline to seek it for reasons of brand self-respect, or to avoid the cost involved, so THX certification does not necessarily mean that a piece of equipment is superior.
Tight bass: (Sound quality description.) Bass that is well controlled by the system and is well balanced. Control is the main ingredient though, particularly in damping bass tones so that there's no overhang.
Timbre: The particular sound of an instrument, largely defined by the different levels of the harmonics within the tone. That's why a violin sounds different to a clarinet, even though they're both playing the same sustained note. Any imperfections in the tonal balance of a sound system can alter these proportions. In extreme cases, it can be hard to identify which instrument is being used.
Time alignment: Systems for ensuring that audio signals from various loudspeakers or drivers arrive at the listener at the correct time. For example, with surround sound it is important that the sound from the surround speakers not arrive early, even though these speakers are often situated closer to the listener than the main speakers. So home theatre receivers incorporate a system to allow the sound to these speakers to be (adjustably) delayed by some milliseconds. Likewise, some high end loudspeaker makers will recess the tweeter further into the enclosure to ensure that the high frequencies arrive at the listener at the same time as lower frequencies, with a view to delivering a more coherent sound.
Tone controls: Labelled “bass” and “treble”, fitted to a preamplifier these provided a means of boosting or cutting the bass or treble of the signal. They typically provide up to ten decibels of boost or cut at 50 hertz for bass and 10,000 hertz for treble. Best avoided.
Tonearm: The long thing on a turntable which is pivoted at the back and at its end has provision for a cartridge to be attached. These normally provide for an adjustable tracking weight by the use of a spring-loaded or weight-loaded mechanism at the back, and usually have a damped cueing lever to allow the stylus to be gently lowered to the surface of the record. Over the years some makers have provided parallel tonearms which do not use a pivot, but a motor-driven rack at the back which gradually moves the rear of the tonearm as the width of the LP is traversed.
Toe-in: When loudspeakers are toed in, they are turned so that rather than firing directly into the room perpendicularly to the rear wall, they point towards the listener. There can be different degrees of toe-in. Some speakers sound better somewhat but not fully toed in.
Tonal balance: This refers to the levels at which different bands of frequencies are reproduced in comparison to other bands. A fine system will ensure that all are equal without favouring any in particular.
Toroidal transformer: See transformer.
TOSLink: DVD players can be connected to digital audio decoders by means of a digital signal cable. One kind uses wires and sends an electrical signal. TOSLink cables use optical fibre and send an optical (or light) signal. More correctly, TOSLink refers to the kind of plug on the end of such cables. This is roughly square shaped with a small ridge to allow it to click into, and be firmly held by, a socket. Some portable Minidisc recorders and CD players can receive or generate an optical digital signal but these generally require a cable with a different plug, shaped somewhat like a standard 3.5mm stereo headphone plug.
Total harmonic distortion: All the harmonic distortion components added together to give a useful, if imperfect, summary measure. The imperfection resides in the fact that, audibly, some of the harmonic components are worse than others. In particular, odd-ordered harmonics from the fifth and up are particularly poor. Solid state amplifiers and the like typically offer THD ratings of less than 0.1% at rated power output, although many all-in-one systems cheat by quoting their power outputs at a ludicrous 10% THD. This much distortion suggests that the system has been pushed well into clipping.
Track: The divisions within the program material of a CD or an LP.
Tracking: This is the ability of a turntable, tonearm and cartridge combination to handle high groove modulations on a vinyl recording. Some grooves can be so difficult to track that audible distortion can occur if the equipment cannot cope. This is called mistracking.
Tracking angle error: With any turntable with a conventional pivoted tonearm, the cartridge and stylus combination will only be at an exact tangent to the groove at two points. At other points during playback there will be a tracking angle error of up to a couple of degrees with a properly setup system. Linear turntables are designed to eliminate this.
Tracking Force: That’s the amount of downwards force applied by a tonearm to the stylus while it is playing an LP. On a modern, high quality turntable this is typically between 1 and 2.5 grams. In general, the greater the force, the greater the wear on the LP. Old-style ceramic cartridges frequently required a tracking force in excess of 10 grams. However, setting a force lower than recommended by the cartridge maker risks other damage to the LP and the stylus, since the stylus may not maintain its position in the track securely. See here for more details on setting tracking force.
Transformer: A device used to alter the voltage of AC electricity. This typically consists of an iron ring of some kind with two coils of wire wound around it. The input current is fed to one of the coils, which generates a magnetic field in the iron ring and which, in turn, generates a voltage in the other coil. The proportion of input voltage to output voltage is the same as the proportion of the number of coil windings on the input (called the “primary”) and output (“secondary”). Most transformers use a square-shaped ring with the primary and secondary windings on opposing sides. Some transformers, which tend to appear in high quality audio equipment, are toroidal, which means that the ring is shaped like a donut and the primary and secondary windings cover the entire surface, overlapping each other. Transformers do not work with DC electricity because while DC can generate magnetism in the iron core, a magnetic field cannot in turn induce electricity in a wire unless it is changing (or the wire is moving with respect to it).
Transient: A transient is a rapid -- almost instantaneous -- change in level in the sound. It is somewhat like an impulse. Poor handling of transients softens the dynamic responsiveness of the system.
Transmission line: A system of tuning the bass response of a loudspeaker enclosure that involves a labyrinthine internal structure, with a long internal passage between the rear of the woofer and the outside air. This can enhance bass, although it tends to result in significant phase delays in the deep bass.
Transparent: (Sound quality description.) The ideal in sound reproduction, where the system is so accurate your system effectively becomes a window through to the recording.
Treble: The audible frequencies typically constituted by frequencies above about 5,000 hertz, although the dividing line between midrange and treble is one of opinion. The human ear is less sensitive to treble than to midrange frequencies.
TRRRS: Tip Ring Ring Ring Sleeve. This describes the connection layout of the 4.4mm balanced headphone connection. The tip and the closest ring are for the left channel (+ then -), the next two rings for the right channel (+ then -), and the sleeve is for the ground.
TRRS: Tip Ring Ring Sleeve. This describes the connection layout of the 2.5mm four-pole plug. When used for balanced headphone connections, the tip and the closest ring are for the right channel (- then +) and the next two rings are for the left channel (- then +). There is no provision for grounding.
TRS: Tip Ring Sleeve. Another name for a 6.35mm (or 1/4”) stereo jack. This terminology tends to be used in professional audio and these plugs tend to be wired for balanced mono operation.
True Wireless earphones: Bluetooth earphones in which the left and right buds are completely separate, with no wires running between the two. They communicate with the source device and each other wirelessly. They come with a charge case which holds them when not in use and can charge them up again several times. The charge case itself must be charged using a standard phone charger.
Tuner: A component (or module within a component) that can receive an AM or FM radio signal, demodulate it and deliver an analogue audio signal to an amplifier.
Turntable: A device used in a home entertainment system to rotate at the correct speed a vinyl recording, such as an LP. More generally, the word can refer to the turntable itself along with an installed tonearm and cartridge. The platter on the turntable (the rotating part) is powered by a small electric motor. Different types of turntables are defined by the drive mechanism used to connect motor to platter. The three most common types are idler-wheel, belt drive and direct drive. An idler wheel is a rubber-like wheel, perhaps two or three centimetres in diameter, that presses against the shaft of the motor and a rim on the underside of the platter. These are never used in high quality equipment because of speed variations and noise problems. Belt drive turntables use a rubber-like belt or band running around a pulley on the motor shaft and a rim on the underside of the platter. These appear in a wide range of turntables, from inexpensive ones through to some of the most prestigious models available. In direct drive turntables the motor runs slowly and the spindle at the centre of the platter is connected to the shaft. These appear in some very high quality turntables and offer particular advantages of high acceleration to speed, plus electronic speed control, prized by DJs.
Tweak: A subtle change to a home entertainment system intended to improve the sound. This could range from merely adjusting the system’s controls, through replacing cables and experimenting with speaker positions, to all kinds of weird mystical stuff. See Belt for examples of the latter.
Tweeter: A small speaker driver designed to produce high frequency (or treble) sounds. This typically operates from 2,000 to 6,000 hertz, depending on the other drivers, up to and sometimes beyond the limits of human hearing at 20,000 hertz. The deeper notes are routed by a crossover network to the midrange driver (if any) and woofer.
Two-way loudspeakers: A loudspeaker which divides the incoming signal into two different frequency bands for distribution to drivers. It sends high frequencies to the tweeter and low frequencies to one or more woofers.
U
Ultrasonic: Audio tones of frequencies higher than capable of being detected by the human ear, generally above 20,000 hertz.
Ultrasonic record cleaning: A record cleaning system in which the grooved section of an LP or other recording is rotated through a water bath, usually containing various additives, to which a fairly powerful ultrasonic agitation is applied. This causes “cavitation” or tiny bubbles to form at the surfaces within the bath, which then implode and dislodge materials such as dirt.
UPnP: Universal Plug and Play. Used these days interchangeably with DLNA. See DLNA for more.

V
VBR: Variable bit rate. As opposed to CBR or constant bit rate. The flow of digital data increases or slows over time, according to the complexity of the encoded signal. Has the advantage of allocating more of the scarce data space to those moments of video or audio that most need it, while economising on sections that can get by with less data.
Visceral: (Sound quality description.) When you can feel the impact of the music in your body, typically with deep and powerful bass.
Voice coil: The coil of wire that is attached to the back of the cone or dome of a loudspeaker driver. This is surrounded by a strong, close magnet so that when electricity is fed into the coil, it moves to and fro, moving the attached cone or dome to and fro, generating sound.
Volt: The standard unit for electrical potential.
VTA: Vertical Tracking Angle. This is the angle in degrees between the cantilever of a cartridge and the surface of an LP. The cantilever is the thin rod which protrudes from the body of the cartridge and has the stylus on the end. The back of some tonearms can be raised or lowered to adjust the VTA. VTA is closely relative to the Stylus Rake Angle. The extent to which this affects sound quality is a subject of some dispute within the high-fidelity community.
W
Watt: A unit of power. For DC, it is equal to the current multiplied by the voltage (one watt equals one volt times one amp).
Wavelength: The end-to-end physical measurement of a cycle in a repetitive signal, measured in metres (or convenient multiples). Audio wavelengths are generally in the range of 17mm to 17 metres (at air temperature 20C). The human eye responds to light in the wavelength range of 380 to 780 nanometres. The wavelength of a signal can be calculated by dividing the speed with which the signal propagates through a medium by its frequency.
White noise: Random audio noise where the average amount of power is the same across all audio frequencies. Sometimes used as a test signal, but it is weak because real audio has power characteristics more like pink noise.
WMA: Windows Media Audio. A group of several audio compression formats developed by Microsoft, originally a lossy audio format operating in competition with MP3. It included digital rights management back when that was all the rage. There were several upgraded versions and ultimately a lossless version called, unsurprisingly, WMA Lossless. The original WMA was widely supported in non-Apple portable devices, but in recent years support has become desultory, with MP3 maintaining its dominance, followed by AAC.
Woofer: The largest driver in a loudspeaker box, sometimes called the bass driver. This provides the bass sounds while the higher sounds get routed by the crossover network to the midrange driver (if any) and tweeter. Many modern loudspeaker designs use two or more smallish woofers rather than one large one.
Wow: A defect affecting analogue audio signal sources that rely on rotating the medium, particularly LPs and compact cassettes. Wow is a slow, repetitive speed variation, typically repeating at less than once per second. If an LP or audio cassette undergoes this, it produces slow variations in the playback frequency. Wow is specified in per cent and specifications of more than around 0.1% are unacceptable. Digital sources such as CDs are immune to wow because they lock their playback speed to a solid-state timing device.
X
XLR: A professional audio connection standard in which the signal pins are not connected to the shielding earth. This allows them to carry balanced audio signals. Most commonly seen in three pin versions (for mono) although there are also five pin (stereo) versions. Sometimes called Canon plugs.
Z
Zero cross distortion: See Crossover distortion.

Tags: Definitions, Dictionary

Previous article Next article