In fact, there is no such thing as digital sound. It would be more accurate to say that digital sound processing technology can produce analog output.
All sound is of an analog nature. Sound is created by continuous variations of pressure in the air around us. These changes in air pressure vibrate the ear drum and are converted into neurons by the mechanisms of the inner ear, which in turn are passed to the brain for interpretation. Sound is analog by its very nature. It cannot be digital.
An accurate definition of the word analog in describing the
processing of sound would be
The representation of sound by continuously varying amplitude and
frequency.
The key words here are "continuously varying". When a sound is
digitized, or converted into numbers, the continuous voltage and
frequency changes are recorded as a number with a timing interval. This
information is not continuously variable, but a number of 'snapshots'.
Much more detailed information on this subject is covered on
the Understanding
Digital Audio page.
Every soundcard with audio inputs and outputs has some analog electronics on board. These will usually provide the user with line-level and microphone-level inputs and outputs for connection to a mixer, Headphone amplifier, monitors and perhaps a direct output headphones.
It is just as important to understand about the analog components in the recording chain as it is to understand what is going on in the digital domain. The advantages which can be gained in this area are:
In order for us to process sound in the digital domain, it is transformed from one form to another by analog to digital (AD) and digital to analog (DA) convertors. Detailed explanations of these processes are widely available on the internet and (thankfully) there is no reason to go into detail about them here.
Our main concern in this respect, is to understand the way in which the gain structures are important in relation to the analog elements of the recording chain.
If you have not yet read the Understanding Metering page on this site, now may well be a good time to do so.
Most soundcards and professional mixing desks use the standard of +4dBu as the as the voltage which will correspond to 0VU on the meter. The illustration below is of a peak meter on a mixing desk. This should not be confused with the Digital Peak meter in your DAW sofware. The peak meter should display a reading of '0' when the output voltage is 1.23V, or +4dBu.
This in turn should display a reading of -18dBFS in your DAW software when connected to an input on your soundcard. That is to say: a reading of '0dBVU' should correspond to '-18dBFS' with the '+4dBu' standard adopted for use.
Equally, when your DAW is set with its master outputs set to +4dBFS the output from your soundcard should be 1.23V, or +4dBu.
This is what is known as a Unity gain structure. It allows the user to maintain a certain amount of gain though a number of pieces of equipment so that the actual voltage will be comparable at every stage.
Unlike digital equipment, which has no gain above its pre-defined maximum level of '0', analog equipment is always designed so that it has a degree of extra voltage capability available. This is because analog equipment has inherent distortion which increases with voltage. It is better to operate it at some distance below its maximum voltage capabilty to avoid this becoming excessive. Analog equipment also has a much higher signal to noise ratio (SN ratio) than digital equipment. That is to say, the amount of hum and hiss inherent in analog design is louder compared to the amount of electrical signal it can emmit. So it is important to make sure that the most useful area in this range is used. In so doing we attempt to use the area of dynamic range which provides the cleanest output, both in terms of background noise and distortion. It is a bit of a balancing act, but quite easy if you have your equipment (both analog and digital) set up to take the best advantage of both worlds.
Exactly the same thing applies to the analog components of a sound card. Often there will be independent input-level controls for each line input. If they are turned up very high and the equipment plugged into the input is adjusted downward, you will probably notice excessive hiss in your recording. If the instrument plugged in is turned up too high, you may well distort the analog input on your soundcard. This distortion will not be a factor of the digital recording process, but of the analog process which pre-ceeded it.
Exactly the same principle can be applied to signals leaving the outputs of your soundcard and arriving at a mixing desk, monitoring or headphone amplifier. So it is important to take advantage of the gain structure all the way through the recording process.
The final consideration to take into account is that of microphone levels. Microphone pre-amplifiers vary in quality to a high degree. One area worth considering very carefully is the amount of gain which should be used with a preaplifier. Sadly, the microphone pre-amps. on many cheaper soundcards really don't live up to the standards which would be required of a professional recording situation. This is not always the case. The pre-amps on soundcards such as the RME series, MOTU, and some others are truly world-class, but these are very good all-round designs and cost substantially more than mid-range models, with good cause.
A microphone preamplifier amplifies the impulse from the microphone to line-level, from just a few millivolts. It is not unusual to find that the maximum gain obtainable from such a preamplifier is in excess of +60dB, or a gain of 1000 times, even though it would be very unusual to use a gain setting of this order used in practice because the signal to noise ratio involved would make the background noise clearly audible. By balancing the amount of gain in the microphone pre-amp. against the amount of gain in the next stage, the line-amp, we can tune the stages to obtain a high enough gain with as little background noise as possible and also ensure that there is enough head-room to accomodate any un-expected loud peaks without distortion. This is quite a different approach to the digital part of the recording chain, where recording at a high level really is not anywhere near as important.
Remember that if you record a microphone without considering these factors, any analog noise which is on the recording will be present on the track. This will sit in the background, masking the detail of the mix and has the potential to make your mixes sound muddy and lifeless.
Good attention to the Analog stages of your recording and playback are just as important as the Digital ones.
This site is under on-going development. More pages will be added soon. Thanks for visiting inthepipeline.net