what is the DAW?

A Digital Audio Workstation is a pice of computer software that can take the job of almost any type of analog equipment you might find in a recording studio. A DAW can function as a mixing console, tape recorder, MIDI sequencer, synthesizer, drum machine, guitar amplifier, reverb chamber plus an entire rack full of effects units like compressors, EQ’s, chorus, delay and pretty much any audio equipment that you can imagine except for microphones and speakers. Some of the practical differences between a DAW and a studio of analog equipment are:

  • We are doing almost everything on a computer. Instead of routing audio through a mixing console and moving physical faders we are using images of faders in the software to mix tracks together.
  • Instead of using patch cables to route audio to effects we are choosing where to send it within the software.
  • Instead of tweaking real buttons and knobs that adjust transistors and tubes in hardware effects, we are mousing over virtual buttons and knobs in the DAW to tweak software code that in many instances emulates analog gear.

One of the advantages of using a DAW is simply because they utilise digital audio instead of analog. Namely, they enable Non-Linear recording and Non-destructive editing.

When you record on an analog tape, the tape runs linearly one second after another and each sound recorded on that tape has to be in sequence physically in order on the tape. This is not the case when recording digitally. A digital file is stored on a hard drive. Your DAW can call up any part of any audio file on the HD at any time and play it back along with hundreds of other audio files, separately or all at once and in any order. The main window of the DAW functions as a map. Each waveform we see on the screen refers to section of a file stored on the HD, ready for the DAW to play it back at any given moment. This is non-linear recording.

In non-destructive editing when you cut and splice sections of digital audio you are not actually slicing up the audio files. Instead you are only redrawing the map that tells the DAW when to play which section. You always have to original file to go back to.

Editing

With analog tape to cut and paste means that we literally have to cut the tape with a razor blade and splice the pieces together. The only way to copy and paste in analog is to set up two tape machines, connect the output of one to the input of another and press play on one machine and record on the other. Every time we copy analog audio the equipment retraces and reinterprets the waveform. Each pass through the copying process changes the frequency response, creates speed and pitch fluctuations and increases distortion and background noise. On the computer, all that we do is to copy a list of numbers from one place to the other. It is incredibly quick and easy to do and each copy sounds exactly the same as the others because it is the exact same list of numbers.

To do processing in the digital domain we do not need to purchase, store and care for dozens of different analog audio processors each of which is wired with electrical components to do one specific thing. Instead, the programmers who make the audio software define mathematics that act on the numbers making up the sound. All of those complex mathematical operations can be carried out in a tiny fraction of a second on the computer.

Different Categories of DAWs

The abundance of available DAW have different histories, strengths and weaknesses. We can sort the software in 5 categories based on their particular strength.

  1. DAWs with roots in MIDI – most of these DAWs started in the 1980s as MIDI sequencers and over time added audio recording. Historically, these DAWs have offered the most powerful composition and MIDI manipulation tools. Logic, Digital Performer, Cubase and more fall into this category.
  2. DAWs with roots in audio – Most of these started in the 1990s as audio only and overtime added MIDI sequencing and other capabilities. These programs have efficient workflow for recording, editing and mixing multitrack performances because they were originally designed to take the place of a mixer and multitrack tape. Pro Tools, SAWStudio, REAPER and many others fall into their category.
  3. The third category of DAWs is of more recent development, starting around the early 2000s. These are DAWs build not as analog substitutes but as original digital tools. They specialise in live performance, looping and manipulating audio in ways that are difficult or impossible in the analog domain. Ableton Live, Propellerhead Reason and Bitwig Studio are three examples. Emerging new DAWs often fall into this category but not always.

Generally, the workstations in these three categories have evolved into each other. Some specialised tasks are only possible or easy in certain DAWs but for the most part workstations have grown enough alike that for day to day tasks it mostly comes down to personal preference.

4.   The fourth category is the single track audio editor, also known as a wave editor, two track editor or a destructive audio editor. The first examples of theses started in the late 1980s and were among the first generally available audio editing applications. They are usually used for post-production, editing and mastering. The most common examples are -Audacity, iZotope RX, Adobe Audition and Sound Forge.

5.   The fifth category is music notation software. These programs don’t exactly record audio or sequence MIDI, instead their focus is on printing sheet music. This type of software has been around since the 1980s and it is used primarily by arrangers and composers. Usually these programs include a synthesizer to playback the music you have written but the primary focus is on the look and readability of the sheet music as opposed to the sound from the computer. Some of the programs out there in this category are – Finale, Sibelius, Noteflight, MuseScore and others.

If we need to decide on what software to use, it helps to consider what type of work we are doing.  Research is suggested and networking with people who have used the various software to get a sense of which program will suit our needs the best. At the end, it does not matter which software we use because all of them are usable tools. There is one circumstance where choosing a software is important. It can be problematic to transfer projects between DAWs. Each one has its own proprietary file format and important details can be lost in translation. When we are working with other engineers, producers and songwriters, it matters less which DAW we are using but more that everyone is using the same one.

VSTs and Effects

In a traditional analog studio we can find a lot of music instruments to produce sound and a lot of equipment like mixers and effects units to process sounds. In the digital studio, software versions of those instruments and processors are available inside of the DAW. The software versions are called plugins. The quality of the work done in the studio will depend much more on our decisions and skills rather than the type (analog or digital) of equipment that we use. The very best plugins may be even better that their analog counterparts because they have no unintended noise or distortion or because they do things that are impossible in the analog world. There are also some distinct practical advantages of using digital plugins.

If we want to use the same effect on 20 tracks, we do not need 20 hardware units of the effect in a rack of gear or build into a mixing console. We simply need to create duplicates of the software plugin, called instances, on the tracks. Another advantage is that all of the settings that you choose on a plugin are saved in the DAW file and we can automate all of the parameters in the plugin which is impossible in the real world unless you have people moving the controls in real time. There are two main categories of plugins:

  1. Effects plugins – adds an effect to the sound we send through it(reverb, echo, EQ, compression, pitch control, amp simulation and man more).
  2. Virtual software instruments – these either work as a Synthesizer: creating sounds from scratch using maths; or like a Sampler: playing back pre-recorded sounds from a sampled sound library. At its input the virtual instrument accepts MIDI data. The output of the instrument is audio, the sound of the notes being played.

Besides these two most common categories there are a few special kind of plugins that we may encounter. There are plugin that send digital audio to a DAC in your audio interface, or receive it from an ADC. Those are often used together to process specific sounds with an analog processor. There are also plugins that interface with hardware synthesizers, plugins that send and receive digital audio over a computer network and plugins that interface with lighting systems. Plugins can be programmed for different computer systems (OS) and in different formats.

Common systems are Apple Mac OS X, Microsoft Windows and Linux. Common plugin formats include: Audio unit, VST, AAX, RTAS, MAS, Direct X, LADSPA and LV2 among others. Some of these formats are only available on one OS.

controlling computer resourses

Digital audio relies on computers to do an incredible amount of processing. Sometimes the computer can run out of one or more resources while doing all of that processing resulting in slow performance or even crashes. The computer uses three main resources ,if we can generalise them , in processing digital audio. They are:

  1. Processing Power (CPU)

The processing power refers to how much math can the computer do in a certain amount of time. Everything a computer does is math and the CPU manages all of the numbers that the PC uses. A CPU has one or more cores were each core can work on a different math problem all at the same time. The processor also runs at a certain frequency usually measured in gigahertz, meaning billions of cycles per second. Each cycle computes part of a mathematical operation. The CPU power is partly determined by the speed and number of cores but some models of CPUs are more efficient (specially the new ones). This means that the same math takes fear cycles. Because of that different models of CPU can have different amounts of processing power even if they have the same number of cores and the same speed in gigahertz. In the digital audio realm, more processing power means the computer can handle a greater amount of plugins. It also allows for effects that are more complicated in computer terms. For example, a simple effect might be a simple echo. In that case the computer only needs to copy samples to create the echo, divide the copied samples to make the echo quieter and then add the echo shortly after the original sound. On the other hand, to do automatic pitch correction, it has to do calculus, trigonometry and other advanced math that uses much more processing power.

Memory and storage are commonly confused with each other because both of them store information but they behave differently and are used for different purposes.

  1. Memory (fast and short term; RAM)

Memory refers to short term working memory. It is very fast to access but temporary. If the PC is turned off, the contents of memory are lost. RAM stands for Random Access Memory. Adding RAM can sometimes make the computer faster because if you do not have enough RAM to to fit everything we are working on, the CPU has to work on one smaller piece at a time which slows it down. If we have enough RAM to fit everything, adding more will not make it faster. Each DAW’s documentation will tell us its minimum RAM requirements.

  1. Storage (slower and long term; HD or SSD)

Unlike memory, storage is archival. It is slower to access but it is permanent. Information stays there when the power is off. A Hard Disk records information magnetically on spinning circular platters. An SSD uses a different technology called flash. USB drives usually use a cheaper version of flash. Other storage mediums can be optical media like CD, DVD and BluRay disk and older technology like floppy disks and magnetic tape. The storage capacity of the medium (not the physical size) tells us how much information it can hold. It is measured in Bytes. Each byte is made up of 8 bits. Digital audio takes different amount of space depending on the format.

At 48 kHz and 24 – bits (48 000 samples per second x 3 bytes per sample x 60 seconds) audio takes up 8 640 000 (8 MB) bytes per minute per mono audio track. Information on a PC is constantly moving back and forth between memory and storage. When we save a document this means copying the information from temporary memory to permanent storage. Likewise, loading or opening a document means copying it from the slower permanent storage to the quick temporary memory so the CPU can work with it. This works simple enough when loading small files like word documents. Audio projects are quite a bit larger and more complex. Most of the time when recording and playing back audio tracks the audio files are streamed more or less directly to or from the HD. That means that the speed of the HD or SSD is a big factor in determining how many audio tracks can be recorded or played back simultaneously. Storage and memory come into play for virtual instruments too but in a different way. Some virtual instruments use samples. Long term storage are usually too slow to (especially HD) find and play back the samples in time to respond to the key being pressed, especially if we need to hit multiple keys at the same time. In this case, it makes more sense to load all of the samples into RAM so that the computer has instant access to them at any time. Generally speaking, the amount of RAM that a computer uses determines how many different sampled instruments can be actively playing at once. Not all instruments use samples. Some instruments instead of playing pre-recorded sounds synthesise their own sounds, using math. This means that instead of using HD or RAM they use the CPU’s processing power.

It is possible for a computer to have more than one memory storage unit. In that case, it makes a difference how we spread the work across drives. If for example we have two drives, it will be most efficient to put our audio recordings on a different drive from our libraries of sampled virtual instruments. Using two partitions on a single drive does not count for this. Only separate physical drives. To summarise:

  1. CPU – How many effects plugins or non sampled instruments we can use at once.
  2. RAM – How many sample-based synthesizers we can use at once.
  3. Storage – speed = how many audio tracks we can use at once.

                    – size (capacity) = how much we can store.

With an understanding of the three main resources we can make better informed choices about computer hardware based on our specific needs.