From A to F
You're here : Learning >Dictionary
The ECVST Dictionnary is a digest of the knowledge dispensed on the Espace Cubase Mailing Lists. These definitions are given by the subscribers wheither to answer a question or spontaneously. But you can bring your contribution too, by posting a definition or giving a different version of an existing definition. To do this, click here...
The other pages
LEXICON
ADAT : Alesis Digital Audio Tape. Norm created by the firm Alesiss. It is a digital interface format which can manage 8 audio channels (16 ADAT recorder can be chained). The recording is done on S-VHS video tape at the resolution of 20bits and 48kHz. This recording format has become a standard in many professional structure and even in some home-studios.
?subject=Commentaire Dico ECVST : ADAT">CeedJay
AFL : After Fade Listen. On a mixing board, it is a solo listening or a auxiliary send after the volume fader. It is synonymous to "effect send". It is used to put the sound which is passing through the mixing board channel in an effect unit, allowing to set the effect in level. It can also be used as a monitoring out, but if the level of your tracks is modified when a singer is recording his part, he may turn crazy because he will hear all you are doing. In that case, it is better to use a Pre Fader Listen.
?subject=Commentaire Dico ECVST : AFL">CeedJay
ALGORITHM : In maths (so in computing too), it is the process which enbles to solve a problem by organizing the succession of basic operations. That is what is done by a computer processor. It only knows how to reproduce algorithms (succession of additions and subtractions...). But it does this so quickly he can solve extremely complex problems. As far as music is concerned, when effects are used (guitare multi-effects for example), the algorithm is the organization and the order of the different treatments in a program. For instance, a multi-effect won't calculate a reverb in the same way depending it is before or after any other effect...
?subject=Commentaire Dico ECVST : ALGORITHM">CeedJay
A/N : abbreviation of Analog / Numerical. Concerns any conversion from an analog sound to a numerical one.
?subject=Commentaire Dico ECVST : A/N">CeedJay
ANALOG : said of a sound as it is in the real, characterized by an infinity of vales representing it, contrary to a numerical sound which takes a single value off and on.
?subject=Commentaire Dico ECVST : ANALOG">CeedJay
ASIO : that means Audio Stream In/Out. It is a protocol invented by Steinberg when Cubase 3.0 was released, to enable this new program to communicate with soundcards which had the necessary drivers (ASIO driver). The principle is the same for Mac and PC. Without ASIO, Cubase must use the system ressources (for instance on Mac, with the standard soundcard, the MacOS Sound Manager) to be able to send something to or receive something from the card. With ASIO, the dialog is directly done, without using the system.
As anyone knows, shortcuts are done to gain time... In the "real life" it is not necessary the case, but with ASIO... It works! As Cubase (in fact the computer processor) gained time, well, it has more time for your effects, eq, etc... Life is sweet, isn't it ? ?subject=Commentaire Dico ECVST : ASIO">Arnaud LE BOULANGER
As anyone knows, shortcuts are done to gain time... In the "real life" it is not necessary the case, but with ASIO... It works! As Cubase (in fact the computer processor) gained time, well, it has more time for your effects, eq, etc... Life is sweet, isn't it ? ?subject=Commentaire Dico ECVST : ASIO">Arnaud LE BOULANGER
AUTOMATION : on a mixing board (analog or numerical, hardware or software), this process allows to memorize all what is done on the machine: faders movements, pans, effects levels... Then, when the song is played, the recorded operations are automatically done in real time. Automation is the ideal tool to perfect a mix.
?subject=Commentaire Dico ECVST : AUTOMATION">Ceedjay
AUXILIARY : additional circuit which is present on mixing boards, it can help in many ways. Most of the time it is used as an effect loop. Each channel can be treated by the same effect processor, setting the send and the return. It can also be used to plug a source which doesn't need particular settings (using only the Aux return). When the signal is taken before the level fader, this circuit is called "Monitor". When it is taken after, it is called "Effects" (FX). It has a "Send" and a "Return".
?subject=Commentaire Dico ECVST : AUXILIARY">Ceedjay
BANDWIDTH : it is all the frequencies (bass, medium, treeble) included between two extreme frequencies which can be reproduced or generated by a machine. For instance, if a speaker has a bandwidth from 15Hz to 22kHz, it would be able to reproduce all the frequencies between these two limits (if the amplifier can produce the same bandwidth).
"Bandwidth" also desifgns the capacity for a telecommunication network to transfer data. It is given in bits/second. For instance, the ADSL network has a theoric bandwidth of 1024kbits/second. In a general way, the bandwidth represents the capacity for an electronic component to transfer a maximum volume of data per second (for example, the memory bandwidth). One synonymous in this context is the "debit". If we are trying to reduce the debit of an audio file by compressing it, we also reduce the bandwidth, so the sound is less good. ?subject=Commentaire Dico ECVST : BANDWIDTH">Ceedjay
"Bandwidth" also desifgns the capacity for a telecommunication network to transfer data. It is given in bits/second. For instance, the ADSL network has a theoric bandwidth of 1024kbits/second. In a general way, the bandwidth represents the capacity for an electronic component to transfer a maximum volume of data per second (for example, the memory bandwidth). One synonymous in this context is the "debit". If we are trying to reduce the debit of an audio file by compressing it, we also reduce the bandwidth, so the sound is less good. ?subject=Commentaire Dico ECVST : BANDWIDTH">Ceedjay
BOUNCING : pre-mixing several voices already recorded to reduce them to one or two tracks. On the "old" 4-tracks tape-recorder, this operation (also called "ping-pong") enables to free up some tracks on order to record new ones (we "bounce" the tracks 1, 2 and 3 on the track 4 to record new ones). The problem on these analog tools is the loss of sound (especially in treeble) that follows the operation. Thanks to present audionumerical tools, we can do as many bouncing as we want without losing quality. The operation is useful on little powerful computers, because it enables to play less audio tracks in the same time (by pre-mixing the drums elements for example).
?subject=Commentaire Dico ECVST : BOUNCING">Ceedjay
BUS : a bus can refer to two rather different things. On a mixing board, a bus gathers all the signals routed to a same destination: stereo out bus, effects bus, auxiliary bus... On an electronic card, it is a link constituted by a series of lines (physical thread) which links the elements of a computer (memory, processor, co-processor, chipset, extension cards...). There are three types of computer bus: data bus, address bus, command bus.
?subject=Commentaire Dico ECVST : BUS">Ceedjay
BYTE : Measure unt for the memory capacity. One byte equals eight bits of data, that is to say it can get 256 different values (2^8). You have to be careful when you are talking of Kilo-bytes. One kB does not equals 1000 bytes, but 1024 (2^10). In the same way, one MB does not equal 1 million bytes but 1048576 (2^20), and one GB equals 1073741824 bytes (2^30). Generally, these values are rounded, but it is good to know ;-) ?subject=Commentaire Dico ECVST : OCTET"> Ceedjay
CHIPSET : on a computer mother board, it is a circuit which seconds the porcessor by undertaking to do tasks which are not calculation, as the management of the different in/out bus (memory, PCI, AGP...), the management of hard drives, the power supply... It is actually almost as important as the CPU, because the quality of a configuration very depends on this component. If it is badly built, your machine won't function well, even with a very big processor.
?subject=Commentaire Dico ECVST : CHIPSET">Ceedjay
CLIPPING : it is the distortion of a signal, represented by a signal level above 0dB. In an analog machine, the sound is distorted, but it remains listenable to a certain extent, whereas in a numerical machine, everything above 0dB is purely cut. So in this context, clipping has to be avoided.
?subject=Commentaire Dico ECVST : CLIPPING">Ceedjay
CPU : central Processing Unit. It is your computer processor.
?subject=Commentaire Dico ECVST : CPU">Ceedjay
CLUSTER : a cluster is a stack of several notes (an aggregate), away from one or a half key. It appeared in the beginning of the 20th century, I think in the free atonality period, even before the serialism (although it can certainly be found in a "sturm und drang" piece). So the cluster has no determined tonal function, but we can found it: C9#11 played Bb-C-D-E-F# for example... I think these are ot really clusters in Monk's, but simply dispositions which mix several tensions in a restricted interval.
?subject=Commentaire Dico ECVST : CLUSTER">Régis LAMORA
COMPUTER : Machine which comes in by the door and comes out by the window. ?subject=Commentaire Dico ECVST : COMPUTER">Kempfred
DAC : Digital to Analog Converter. Electronic component which will decode numerical information (0 and 1) contained in a hard drive or in a memory, to transform them in electrical impulse which will be decoded by an amplifier (for example). It is the opposite of the ADC.
?subject=Commentaire Dico ECVST : DAC">Ceedjay
DC OFFSET : I think it is the shifting between the curve of the sound and its zero... The DC Offset has to be as next to zero as possible... (which is generally the case). Mathematically, we shift the curve thanks to a simple vertical translation by adding to each sample a positive or negative value, which actually corresponds to the adding of a continuous signal C(t) on the signal S(t). If the signal is expressed as C(t)+S(t) where "t" is the time, then C(t) would represent the DC Offset... to be confirmed, as I am not sure.
?subject=Commentaire Dico ECVST : DC OFFSET">Vincent BUREL
DI BOX : small electronic box (Direct In Box) which enables by adapting the in/out levels and impedance to plug a bass or an electric guitare directly in a mixing board, with no loss of signal nor degradation of the mixing board channel components. This configuration is interesting to have clean sounds, because then we can add the possible effects in the auxiliary circuits of the mixing board (compressor for the bass, reverbs and delay for the guitare)
?subject=Commentaire Dico ECVST : DI BOX">Ceedjay
DITHER : when, for example, a signal is recorded on 24bits, and it has to be reduced to 16bits (to burn it on a CD for example), we can do it in different ways, the simplest is to suppress the last 8 bits (24-8=16). But, by doing this, we lose all the signal part which was inferior to -96dB in relation to the maximum possible amplitude (-96dBFS - indeed 16bits allow a maximum dynamic of 96dB: 6dB x 16 = 96dB).
Pragmatic people will say that it is not important, that because of the background noise we cannot hear anything below -96dB... But it is less and less certain, thanks to the increasing quality of audio equipments. Moreover, opinions diverge about the ability of the human ear to hear signals drowned in background noise (the restaurant syndrome: if we prick up our ears, we can understand the next table conversation, even if the background noise is louder and covers it). In short, purists don't like to lose signal, even at -96dB, and there is the use of the "dither".
To keep a part of the signal below -96dB when it is cut, the "dither" technics consists (paradoxically) in adding to the 24bits signal a very low amplitude random noise, before cutting this signal to 16bits. This adds a little noise, but improves the final 16bits signal imression of clearness. Explaining in details how it works would be very long, I would say it has to do with the modulation of a signal by another signal. We can mention there are several types of noise which can be added (white noise, pink noise or brown noise), so there are several different types of "dither".
I gave the example of a shift from 24 to 16bits, but it is true for the shift from a bigger to a smaller number of bits. The only thing which changes is the dynamic cutting threshold, linked to the smallest number of bits chosen (-96dB for 16bits, -48dB for 8bits, -144dB for 24bits, etc...).
Pragmatic people will say that it is not important, that because of the background noise we cannot hear anything below -96dB... But it is less and less certain, thanks to the increasing quality of audio equipments. Moreover, opinions diverge about the ability of the human ear to hear signals drowned in background noise (the restaurant syndrome: if we prick up our ears, we can understand the next table conversation, even if the background noise is louder and covers it). In short, purists don't like to lose signal, even at -96dB, and there is the use of the "dither".
To keep a part of the signal below -96dB when it is cut, the "dither" technics consists (paradoxically) in adding to the 24bits signal a very low amplitude random noise, before cutting this signal to 16bits. This adds a little noise, but improves the final 16bits signal imression of clearness. Explaining in details how it works would be very long, I would say it has to do with the modulation of a signal by another signal. We can mention there are several types of noise which can be added (white noise, pink noise or brown noise), so there are several different types of "dither".
I gave the example of a shift from 24 to 16bits, but it is true for the shift from a bigger to a smaller number of bits. The only thing which changes is the dynamic cutting threshold, linked to the smallest number of bits chosen (-96dB for 16bits, -48dB for 8bits, -144dB for 24bits, etc...).
To conclude, please note that :
- Cubase VST doesn't manage the "dither" (whatever version, PC or Mac, VST24 or VST)
- On PC, Wavelab manages the dither
- On Mac, Master pulg-ins enable to manage dither, such as Dynasone for example. We can also use SonicWorx Studio to do it
- A useful link concerning these notions of signal treatment : http://www.digido.com. ?subject=Commentaire Dico ECVST : DITHER">Arnaud LE BOULANGERDTD Direct To Disk : numerical recording process, during which data are directly transfered on a mass support (quick) such as the hard drive of a computer. Nevertheless, there are DTD systems without a computer (numerical portable studio for example). ?subject=Commentaire Dico ECVST : DTD Direct To Disk">Ceedjay
FADER : linear pots which are on a mixing board. They amplify or soften the sound which pass through the circuit. They are prefered to rotary pots because they are easier to manipulate and more precise. We can find virtual faders on the virtual mixing board of software such as Cubase or Logic. ?subject=Commentaire Dico ECVST : FADER">CeedjayFEEDBACK : consists in sending back in (differently measuring out depending on the effect we want) the out signal. It is used in the effects, especially for chorus and flangers. Feedback can also design the larsen effect. In that case, we will try to avoid it. ?subject=Commentaire Dico ECVST : FEEDBACK">CeedjayNote from the person who puts these pages into form : L'Espace-Cubase regular customers will find another explaination for Feedback, far more interesting and witty ! See ya later, here or elsewhere !
FREQUENCY : electricity (or electronic) term which designs the number of times where an event repeats in a given period (generally in one second). The sampling frequency is the number of samples taken during a analog/digital conversion or reproduced during a digital/analog conversion. Frequency is expressed in Hz. Audio CDs sampling rate is of 44,100 Hz (44.1kHz), that is to say that the sound is cut 44,100 times in a second. The more high is the sample rate, the more the numerical sound (discontinuous by definition) will be close to the anaolg original sound (continuous by definition). We can also talk of frequency to define the height of a sound. The more a sound is high pitched, the more its frequency will be high. For example, the A in the phone has a frequency of 440 Hz. ?subject=Commentaire Dico ECVST : FREQUENCY">CeedjayFULL DUPLEX : said of a machine which can do an operation and its reverse at the same time. Indeed, a soundcard will be Full Duplex if it can play and record at the same time. Most of the actual soundcards are. This enables to hear a playback (instruments already recorded) during the recording of a voice, for example. ?subject=Commentaire Dico ECVST : FULL DUPLEX">CeedjayPage viewed 4552 times