C

Cache RAM

Most of the RAM (Random Access Memory) that computers use is inexpensive dynamic RAM. In modern computers, dynamic RAM is actually too slow to keep up with the bus speeds. To compensate for this, computers incorporate a small amount of expensive Cache (or Static) RAM, which is fast enough to keep up with system speeds (this is also called a Level 2 or L2 cache). All modern CPUs have a small cache built into the chip itself (8-16k). External cache memory can also be added to improve performance. PowerPC's require at least 256k per CPU (or more). Pentium Pro chips have a built-in L2 (256 or 512k) cache that runs at a full 200 MHz, greatly speeding performance. How much cache RAM you should have depends on the amount of total RAM your computer has. Unless you are running serious graphics systems, 256 or 512k should be fine. Increasing to 1 Mb will only provide a minor performance improvement.

Capacitor

An electronic component, also sometimes called a condenser. Capacitors, which come in many shapes and sizes, basically do nothing more than store electrical voltage, somewhat like a battery. They are comprised of two (or more) electrical conductors (plates) separated by an insulator. Current cannot flow "through" a capacitor without breaking it (by breaking down the insulator). Thus they do not pass any DC, but do pass AC in varying degrees by the charging and discharging of the two plates. In electrical circuits their behavior in terms of the time it takes them to store and release their voltage charge can be used in more ways than your credit card at the mall. Their use can be as simple as removing a DC component from an AC signal or short-term power for memory in electronic instruments, but they are also used in very sophisticated audio and digital circuits (as control elements). Capacitors are important ingredients in most EQ circuits, power supplies, oscillators, clocks, filters, and the list goes on and on. They are everywhere. A typical mixing board could have hundreds or thousands of capacitors.

There has been a great deal of eyebrow raising (an exercise much practiced in the audio industry) about the relative merits of different types of capacitors used in audio circuits. Many audiophiles hold that the quality of these components can have a significant impact on the audio quality (especially long term) of a device. This type of thing is one of many that may separate more expensive equipment from the cheap stuff.

Capstan

In a tape recorder the capstan is the rotating shaft that drives the tape past the heads. Usually the tape is squeezed between the capstan and a rubber wheel known as the pinch roller. The capstan is what controls the speed of the tape and is usually attached to a "flywheel" so that a smooth and consistent tape speed is maintained.

Interestingly (or maybe not), the tape speed is not always equal to the surface speed of the capstan as one would expect. A phenomenon known as "belting" can cause tape to be moving faster than the capstan turns, especially with small diameter capstans. Also, thick tape will move slightly faster than thin tape. This means that a very slight, but audible, rise in pitch occurs whenever a spliced (therefore thicker) piece of tape moves over the capstan.

Cardioid

A microphone polar (pickup) pattern. Characterized by strong sensitivity to audio from the front of the mic, good sensitivity on the sides (at 90 degrees, 6 dB less than the front), and good rejection of sound from the rear, the cardioid pattern can almost be visualized as a "heart-shaped" pattern (hence its name).

The ability to reject sound from the rear makes cardioid patterns very useful in multi-miking situations, and where it is not desirable to capture a large amount of room ambience. Popular in both studio and live use (where rear rejection cuts down on feedback and ambient noise), cardioid mics are used for a very high percentage of microphone applications.

Keep in mind that like all non-omnidirectional mics, cardioid mics will exhibit pronounced proximity effect (see WFTD archives, "Proximity Effect").

Cart Machine

Broadcasters' slang for "Tape Cartridge Machine," which is a playback machine that uses endless loops of tape in plastic cartridges. Commercials and announcements have historically been recorded on "carts," and some of the playback machines can handle a dozen or so of them, something like a record changer. Nowadays there are various forms of digital technology taking the place of these old machines, but many are still in service.

CD-RW

Stands for CD rewritable, the latest incarnation of CD writing technology. CDRW is pretty much like regular WORM (see WFTD archive WORM) CD writers except the media can be overwritten with new data.

CD-RW technology supports a variety of recording modes (disc at once, track at once, multisession, linked multisession, and packet writing). It also supports disc formats such as Red Book (CD audio), Yellow Book (CD-ROM, CD-XA), Green Book (CD-i), Blue Book (enhanced CD, CD Extra), White Book (Video CD), Orange Book Parts II and III (CD-RW), CD-ROM XA (Photo CD), mixed mode, hybrid, and bootable discs.

Cent

In musical terms a cent is 1/100 of a semitone. Cents are a common and convenient way of describing very small increments of pitch in musical terms. Keep in mind that the relationship between frequency and pitch is not linear so describing small pitch differences or changes in terms of frequency can be very confusing.

Channel Separation

The crosstalk, or bleed of audio signals from one channel to another. The amount of channel separation is inversely related to the item's crosstalk spec; i.e. a low crosstalk spec indicates high channel separation.

Chorus

In musical terms a chorus is: A) a composition in four or more parts written for a large number of singers. B) A refrain in which an audience joins a soloist in a song. C) A repeat of the opening statement of a popular song played by the whole group. D) A solo section based on the main melody of a popular song and played by a member of the group. E) A body of singers who perform choral compositions. F) A body of vocalists and dancers who support the soloists and leading performers in operas, musical comedies, and revues.

In the technology discourse a chorus is also an effect that modifies the sound of a single instrument to simulate a group of the same instruments. The subjective effect of a real chorus is caused by the fact that each of the sounds being mixed together have different frequencies and/or timbres. The mixture becomes extremely complex as the relative phases of the signals cause partial cancellation and reinforcement over a broad frequency spectrum.

The synthetic chorus effect was first obtained by subjecting the sound to a series of very short time delays and mixing them. The delays were then modulated to increase the uncertainty of the combined pitch. This "time based" method takes quite a few delays to achieve a good, rich effect. More recently pitched based chorusing has been introduced, which is sometimes more convincing as a true chorus. These units work by splitting the signal into many frequency bands and then randomly varying the phase and amplitude of each before recombining them. Both methods are valid and can produce good results, but there is often a fundamental difference in the sound between the two.

Circumaural

Used in reference to headphones. "Around the ear". Circumaural headphones encircle the ear, and provide a good seal. Typically, circumaural phones use a "closed" design, and provide good audio isolation.

Class A

A type of amplifier design. When an amplifier's stage devices are passing current at all times, including when the amplifier is at idle (no music playing), whether the amplifier is single ended or push-pull, the amplifier is said to be biased in Class A. Because the current is flowing at all times, an input signal causes the current to be immediately diverted to the speakers, and therefore, the sound is very "fast". In the case of a push-pull amplifier, there is also less crossover distortion when the signal passes from the positive to the negative or negative to positive, since each side of the push-pull section is already "on". If all stages of the amplifier are biased in Class A, and the amplifier operates in Class A to full output (enough current flowing at idle that could be required for full output), it is said to be a "Pure Class A" amplifier. Pure Class A designs are understandably expensive to build and are usually only found in high-end audiophile equipment.

Class AB

A class of amplifier output design. As its name implies, it is sort of a combination of Class A and Class B operation. If an amplifier operates in Class A for only a portion of its output, and has to turn on additional current in the devices for the remainder of its output, it is said to operate in Class AB. Most amplifiers are in this category, and are said to be Class A/B amplifiers, since they operate in two classes. In class AB and B, the amplifier is slower than in Class A because there is a finite time between the application of the input signal and when the devices are turned on to produce a flow of current to the speakers. However, Class AB and Class B are more efficient than Class A and do not require such large power supplies.

Class B

A type of amplifier design. Class B differs from Class A in that there is no current flowing when the output devices are at idle, and as a result, they have to turn on from a zero current state when signal is present. In a push pull Class B design the output devices would each produce half of the audio waveform (one set for the positive half, and another for the negative half) and would not have any current flow when the other half is operating. Class B designs tend to have a slower slew rate and more crossover distortion but are less expensive and require less robust power supplies.

Clipping

A specific type of distortion. If a signal is passed through an electronic device which cannot accommodate its maximum voltage or current requirements, the waveform of the signal is sometimes said to be clipped, because it looks on a scope like its peaks have been clipped off by a pair of scissors. A clipped waveform contains a great deal of harmonic distortion (see WFTD archive harmonic distortion) and often sounds very rough and harsh. Clipping is what typically happens when an audio amplifier output is overloaded or its input over driven.

Interestingly, light to moderate clipping does not usually reduce the intelligibility of some signals, especially speech. In fact, it has been shown that clipped speech is easier to understand than normal speech in noisy environments. A probable reason for this is the increased high frequency content that accompanies this type of distortion, which can make a signal stand out more among other sounds and noises. Aphex and some other companies have been using this principle for years in their "exciter" type products. By adding the right amount of distortion at the right frequencies a signal will sound almost clearer and more distinct amidst other sounds, thus standing out more in a mix.

Clock

A timing source. In digital audio or video a clock serves as a timing reference for a system. Everything in digital audio and video is related to a clock. There are specified numbers of operations per period of time and they all must be carried out at a consistent speed in order for the device to work properly. In digital audio the relevant clock is the sample rate, or word clock of the audio (44.1 kHz, 48kHz, etc). In video it may be the frame rate of the video, or the field rate, or some other video specific rate. The clock or timing of information is an ongoing issue when synchronizing systems with each other.

CMRR

Abbreviation for Common Mode Rejection Ratio. In lines there is always a positive signal on one cable and a negative or opposite polarity signal on the other. The idea is that anything common to both wires will eventually get canceled at the receiving end (as noted in WFTD Balanced). The degree of cancellation at the receiving end, however, is dependent upon a number of factors relating to the integrity of the circuit involved. CMRR is a spec that defines to what degree common signals are canceled at the input of a balanced system.

Coincident

In audio terms, coincident is normally used in the context of stereo microphone pairs. The idea is to get the capsules of the two mics as close together as possible to minimize phase problems in the final recording. Often the mics are directional (i.e. cardioid) and are "stacked" one atop the other, commonly at an angle of 90 degrees. Another coincident miking approach is called "MS" or "Mid-Side". Here a bi-directional (figure 8) and cardioid mic are placed close together. By combining the outputs of the two mics in varying amounts, the apparent width of the stereo field can be changed.

Comb Filter

A comb filter is a filter that has a series of very deep notches in its frequency response with the spacing of all of the notches at multiples of the frequency of the lowest notch (they are all harmonically related). It got its name from looking like a comb when plotted on a frequency response graph. Comb filters are produced when a signal is time delayed and added back to itself. Some frequencies will cancel and others will be reinforced, which can dramatically change the tonal color of the sound. In practice this is common problem that occurs when a stereo mix is collapsed to mono because many stereo effects, such as chorus and flanging, achieve their stereo imaging by using some form of the Haas effect. A static comb filter will make its audio sound kind of hollow or "phasey" depending upon how severe it is. Add some modulation and you have a flanger. Comb filtering is one of the main ingredients in the distinctive sound of a jet aircraft passing overhead. The difference in the time arrival to your ears of the direct sound versus the sound reflected off of the ground causes various frequencies to be cancelled or reinforced. As the plane moves these distances are all changing, thus causing the coloration of the sound to change. Again, it's the same principle used in a flanger.

Companding

A type of noise reduction used in audio equipment, a compander circuit is a combination of a COMPressor and an exPANDer. The signal is compressed before recording it to tape (which maximizes the signal to noise ratio), then expanded as the tape is played back. As the signal is expanded, tape noise tends to be "pushed down," resulting in a quieter signal.

Compressor

A compressor is a device that reduces the dynamic range of an audio signal. First a threshold is established. When the audio signal is louder than this threshold, its gain is reduced. The amount of gain reduction applied depends on the compression ratio setting. For example, with a 2:1 ratio, for every 2 decibels the input signal increases, the output is allowed to increase only 1 decibel. A variety of other parameters in the compressor will also affect its performance processing specific signals; attack time, release time and others are very important.

There are a variety of uses and applications for compressors, the most obvious one being to control the dynamic range of a live performance so that it will fit into the fairly narrow dynamic range of recorders, etc. Other applications include making a signal's average level higher, increasing the apparent sustain on a guitar, evening out a vocal or bass guitar performance, fattening up sounds, and on and on. The list of possibilities is extensive!

Condenser Microphone

The condenser microphone is a very simple mechanical system, with almost no moving parts compared to other microphone designs. It is also one of the oldest microphone types, dating back to the early 1900's. It is simply a thin stretched conductive diaphragm held close to a metal disk called a backplate. This arrangement basically produces a capacitor, and is given its electric charge by an external voltage source. This source is often phantom power, but in many cases condenser mics have dedicated power supply units. When sound pressure acts on the diaphragm it vibrates slightly in response to the waveform. This causes the capacitance to vary in a like manner, which causes a variance in its output voltage. This voltage variation is the signal output of the microphone. There are many different types of condenser microphones, but they are all based on these basic principles.

Constant Q

On most graphic equalizers, changing the gain of a frequency band also changes the Q or bandwidth of that band. (As the slider is pushed up, the width of the band affected becomes wider. This also increases the overlap between adjacent frequency bands). On an EQ with Constant Q, the bandwidth remains constant no matter how far the gain is boosted or cut. This allows the EQ's effect to remain more predictable and controllable as there is less interaction between adjacent bands.

Continuous Controller

In MIDI terms, a continuous controller (CC) is a MIDI message capable of transmitting a range of values, usually 0-127. The MIDI Spec makes 128 different continuous controllers available for each MIDI channel, although some of these have been pre-assigned to other functions. CC's are commonly used for things like MIDI controlling volume (#7), pan (#10), data slider position (#6), mod wheel (#1) and other variable parameters.

Use of continuous controllers in performance and sequencing can be a major factor in adding life to MIDI music - but beware, over-use of CC messages can result in MIDI log-jam, where the amount of data being sent is more than the bandwidth of MIDI can support. (Most sequencers support commands for "thinning" CC data if this becomes an issue)

Interestingly, pitchbend is technically NOT a continuous controller. Because of the greater resolution wide bends require (to prevent "stair-stepping"), pitchbend has been assigned its own dedicated MIDI message type...

Coupling

In electronics, coupling refers to ways of connecting circuits or subsystems to one another. For example, gain stages of an amplifier may be directly coupled or may have capacitors or transformers in line. The capacitors and transformers eliminate a direct connection, but still provide coupling that allows the signal to be transmitted from one stage to the next. There are many different types of electronic coupling.

CPU

Abbreviation for Central Processing Unit. The chip on a computer's motherboard which ultimately controls all the activity of the computer. Standard Macs have a 680x0 chip (x = 0, 2, 3, or 4) manufactured by Motorola. PowerPC Macintoshes use a new RISC (Reduced Instruction Set Computing) chip designed by a conglomerate of computer hardware manufactures, including Apple, IBM, and Motorola. Most IBM compatible computes use a chip based on Intel's X86 architecture. These days most electronic instruments (keyboards, drum machines, etc.) and digital tape machines have a CPU which controls all of the functions of the machine.

Critical Distance

When dealing with acoustics, critical distance is the point at which the volume of a sound source is equal to the volume of reflections from that source off of other surfaces. Control of the volume and timing of these reflections is an important part of creating an accurate listening environment.

Crossfade

A technique commonly used in editing audio. One sound is faded out as another fades in, allowing for a smooth transition between the two. Crossfading is also common in samplers, where it is used to smooth loop transitions (crossfade looping), and sound design to create hybrid sounds (one sound morphing or turning into another). While we often think of this as a digital process, audio engineers have been using two channel faders on a mixing console to crossfade between two signals or tracks for many years.

Crossover Distortion

A type of distortion pertaining to anomalies that occur when a sound "crosses over" from the positive portion of its waveform to the negative portion in audio circuits. It is most widely known as a phenomenon of push-pull amplifier designs where separate banks of output devices handle each half of the waveform. There is a region near zero where the signal is transferred from one to the other that, if not done smoothly, can produce a discontinuity in the audio that sounds a bit like harmonic distortion. Since this distortion is at a constant level it is much more obvious at low levels. As such crossover distortion is most thought of as a problem only with low-level signals. Digital converters are also sometimes plagued with this problem, though for different reasons. In digital it is more a manifestation of quantization error at low bit depths (volumes).

Crosstalk

In multi-channel audio systems, crosstalk is signal bleeding or leaking from one channel to another. Mixers, tape recorders, and many other pieces of gear are all susceptible to this problem. In most modern gear, crosstalk is not a major concern, but be aware that older gear can have significant amounts of bleed between channels!

Current

A term used in speaking about electricity or electrical concerns. In electrical and electronic circuits, the signal consists of electric currents. These are the flow of electricity produced by an electrical charge (voltage). Technically, it is actually the flow of the charge, but we'll deal with that later. Current flow is measured in amperage.

Cutoff Frequency

In a filter, the cutoff frequency is the point where the response is 3 dB down in amplitude from the level of the passband. Beyond the cutoff frequency, the filter will attenuate all other frequencies, depending on the design of the filter. On a sweepable shelving EQ or filter, what you are "sweeping" (or changing) is the cutoff frequency. To our ears, this changes the point at which the filter is operating.