DAW Basics
This covers the basic functionality of most professional DAW (Digital Audio Workstation) software. It's probably nothing new to anyone already producing music, although some sections might be interesting. This is a quick overview for the noobs.
Contents
Timeline View/Transport
The default window/view your DAW will display should be the timeline view, also called the transport. This consists of a big empty box (called the timeline) with numbers running along the x axis, either above or below. These numbers indicate the time in terms of measure numbers and/or minutes and seconds - it is literally the timeline for the project. The y axis is divided into rows called tracks, covered below. Each track usually consists of a small track control panel to the left of the timeline. Each track can have multiple blocks or events of audio or MIDI data, as well as automation envelopes. The events (audio, MIDI) look like rectangles inside the track's row in the timeline, spanning whatever time they consume. Envelopes operate in parallel to the audio/MIDI data, and usually appear as sub-rows under each track row but can be collapsed/hidden inside the track row.
The timeline is usually adjustible, so that you can zoom in/out for either the x or y axis. Zooming out allows you to see a greater portion of your project. For instance, zooming out on the x axis means the amount of time visible inside the timeline gets larger, while the events appear smaller horizontally. Zooming out on the y axis means more tracks are visible, but the control panels and events shrink vertically. In contrast, zooming in allows you to see greater detail per track, making it easier to view/edit audio or MIDI events and their content as well as edit track settings.
Transport Controller
The timeline may have built in controls, or they may appear in a separate window that can be moved around the screen or hidden. These controls are collectively referred to as the transport controls or simply transport (although sometimes the timeline window is also called the transport). This has standard controls like jump back, rewind, play/pause, record, stop, fast forward, and jump forward. It may have additional features or displays, such as the HH:MM:SS marker for the current point in time selected in the timeline, as well as the tempo.
Timeline Cursor
The timeline has a cursor that typically appears as a small arrow/triangle above and/or a vertical line overlaid on top of the timeline, indicating the current place in time. When playing back the project, you'll see it move from left to right. The cursor can be moved using the timeline controller mentioned above. Typically it can also be moved by keyboard buttons and shortcuts, as well as by clicking in the numbered timeline above the timeline window.
Timeline Selection
Click-dragging in the timeline above the track window selects a span of time, often called a region. This selection can be looped, bounced, or processed independently of the rest of the project. Often the region is bound by separate start and end markers in the numbered timeline.
Tracks
The term tracks is a holdover from the days of multi-track tape, where each instrument could be recorded (or tracked) indpendently. This allowed flexibility to process each track separately before mixing them all down into the final stereo or mono mix. Tracks in a modern DAW are much more flexible and aren't devoted to recording audio. For example, a track can contain MIDI data, which is sent to virtual instrument to generate audio. The typical types of tracks allowed by a DAW include:
- Audio (Mono)
- Audio (Stereo)
- MIDI (Includes Virtual Instruments but they may be listed as separate types)
- Group Tracks
- Adjustment Tracks
Tracks typically have a control panel with the features below, closely resembling a mixing board. Some DAW's display limited control panels for every track to the left of the workspace. Others require highlighting or double-clicking the desired track to make the control panel appear. Depending on the level of zoom and the screen space given to the control panel portions of the tracks, certain features may be hidden.
Record Arm
The most basic feature of a track is to record arm the track. This tells the DAW to record audio or MIDI input into the track when you press the record button on the timeline controller. Tracks that are not record armed will simply playback whatever audio or MIDI data they already contain instead of recording whatever data is flowing in through their specified input. You can record arm multiple tracks to simultaneously record them, even if they use different input sources.
Solo/Mute
The solo button essentially mutes all other tracks. This allows you to focus on that single track, so you can notice imperfections or edit it without the distractions of all the other tracks playing back. You can solo multiple tracks simultaneously. You can still record on any track while one or more tracks are being solo'ed.
The mute button, as expected, simply mutes playback of the track. A track can still be recorded while muted.
Also, see Mixing Boards/Consoles#Solo and #Mute.
Input Monitoring
There is often an input monitor button per track. This sends the input stream from the audio track to the software output when the track is record armed. Essentially, what you hear before you press record is exactly what will be recorded.
Hardware interfaces generally provide a software mixer to control the hardware. It also usually includes an option to enable direct monitoring (or hardware monitoring) of any of its inputs. This routes the audio going into that input directly to the output without having to be routed through the computer's operating system and other software. The advantage of direct monitoring is that there is minimal latency. For non-acoustic instruments that use heavy audio processing like electric guitar, this is a very good thing. Trying to play guitar but not hearing your playing until a 10th of a second or more throws off the feel and timing of the player.
When trying to set up your DAW and track inputs, direct monitoring can be misleading. The sound from the direct monitor might be fine, but the track you want to record that audio might be set incorrectly, so that it only records noise, the wrong audio source, or silence. Rather than trying to record a test clip, then play it back; it's often simpler to disable direct monitoring on the interface and enable software monitoring. This may mean you are hearing more latency, and it is difficult to play in-time while recording; but it will quickly demonstrate if your track is recording correctly. What you hear from the software monitor is exactly what your recording will sound like. You will also hear any processing the DAW performs on the signal, such as level changes, panning, or insert plugins, which are bypassed in direct monitoring.
When actually recording, I recommend turning off software monitoring and using your device's direct monitoring to remove the disorienting nature of latency. If your interface has very low latency, however, the difference may be negligible. Then, software monitoring is preferable, as it makes sure you don't accidentally record a difficult part perfectly only to later find out you weren't properly set up to record it. If both direct and software monitoring are enabled, you may get a phasy or comb-filtered type sound because both sources will be sent to output. They will be slightly out-of-sync, causing the unwanted effect. Make sure to use only one or the other.
NOTE: Modern DAW software is equipped to know what the audio latency from your input source is and time-shift the resulting recording so that you never know it existed. When recording, you should play in-time with the click track and/or other tracks.
For more, see Interface Setup.
Input/Output or Send/Receive
Each track allows input/output selection, also called send (output) and receive (input) to normalize the terminology with mixing boards. Possible inputs/outputs are set up in your DAW settings and should correspond to your physical input/output connections on your hardware audio interface. This includes audio and MIDI I/O's. Mono tracks will allow stereo inputs (such as S/PDIF) to specify either the left or right field.
Most tracks are set to output to the master bus by default. However, tracks do not necessarily have to output to output buses - they can also output to input buses, and multiple input buses can be set up to capture the same physical inputs. It's a good idea to set up several additional input buses with no physical input attached, designed for routing multiple tracks into. For example, you may have several tracks with percussion instruments and want to set all their outputs to a percussion input bus. Another track is set up to use the percussion input bus as its input and receives the output of all those tracks. This single track can apply a uniform reverb or compression preset as inserts and send the output to the master bus.
Output buses on the other hand are routed out of the DAW and to the interface (as software return inputs), where the interface will handle which physical outputs the signal is actually sent.
Obviously, this can all get very, very confusing for the novice user. It's best to work up to complexity rather than trying to dive right in. Start out by only using inputs linked to your interface's physical inputs and outputting to the master bus. Once you are comfortable with that, then try more advanced routing.
While DAW tracks appear similar to mixer channels, they tend to be much more powerful (and thus possibly more complex). This is because they are not limited by hardware. You could create several hundreds of tracks and dozens of input buses for a single project, routing signals through multiple tracks before reaching the master bus. In practice, your computer is unlikely to have the ability to process all that in real-time, and you would have to resort to bouncing to get the same result.
Inserts
Inserts are basically in-line routing of the track to algorithms (virtually always called plugins) that process a track's audio or MIDI events. For instance, you can apply a reverb plugin as an insert on audio track to add reverb to it. The inserts function independently of the track events, and the input for the track is recorded before the signal reaches the inserts. In other words, it does not matter if you have inserts on the track at record time - the track is only recording the raw audio or MIDI data. Inserts only affect the playback/monitoring of the track and can be changed or removed later on. Each track is usually capable of having numerous inserts, which process in a serial order. Some inserts process before certain standard track controls, such as the volume/level or EQ settings, while others process after.
Audio
Audio clips in the workspace are typically called audio events. Each audio track may contain a number of audio events, even overlapping ones (although only one of them will playback in this case). Each event appears as a colored (or themed) rectangular box with a black or white visual representation of the audio waveform inside it.
Fade/Trim
Most DAW software provides quick ways to fade or trim audio events. A fade is a transition from no volume to full volume (or vice versa). It typically involves click dragging from the top corner at the beginning or end of the audio event towards the other end. This creates a linear fade over that time, which is visually represented by a diagonal line on top the audio. There is often a click point in the middle of the line allowing it to be warped into a curve, changing how the fade processes. More advanced fading is typically available through a specialized fading tool.
Fading is useful to achieve a seamless transition into and out of the event. Audio often carries background noise, and if the full volume of that noise pops in and out of the mix instantaneously, it is often quite noticeable.
A trim is simply removing some of the beginning or ending of the audio event. Just click drag from the edge of the event, not on the corner. This will cause part of the audio to seemingly disappear. The DAW is not actually shortening the underlying audio file; only the playback inside the DAW is affected. If you click-drag the opposite direction, you'll see the trimmed audio reappear. This is useful to remove leading/trailing noise from an event.
Level Adjustments
Similar to quickly editing fades and trims, an audio event can be click-dragged from a click point at the top of the event directly at its center. This adjusts its volume, capable of quickly normalizing your events visually, regardless of their source. Each event can be edited independently, even if they reside on the same track. Further track-wide volume adjustments can then be fine-tuned in the mixer or through automation. NOTE: if your peaks are visually exceeding the event box, you likely have digital clipping in the recording. Lowering the event's level will not remove the clipping but make it less noticeable. Clipped recordings should be deleted and re-recorded with proper levels.
Slicing
Slicing is a DAW feature that cuts an audio event into two or more independent events. For example, if you have an audio event of knocking on a door then someone saying, "Who's there?"; you may want to use the knocking and "Who's there?" in different locations in your song. Or you may want to widen the delay between them. Or you may want to move one to another track to apply a plugin to it but not the other. Or you may want to level them differently than how it was recorded.
Plugins
Plugins are software algorithms designed to process audio or MIDI, similar to stomp boxes or fx units. Plugins are more flexible, however, since they can be altered after the original audio recording and automated by the DAW. They are most often found as inserts for tracks. Plugins cover the same ground as most guitar effects:
- Compression/Limiting
- Noise Gating
- Distortion
- Modulation (Phaser, Flanger, Chorus, Tremelo)
- Pitch Shifting (including the infamous auto-tune)
- Delay
- Reverb
- EQ
Yet others cover grounds related more strictly to audio production:
- Dithering
- Noise Reduction
- Mastering
- "Glitch" Effects (bit crushing, reverse, slice/stutter, cutoff filters)
Many plugins come with their own graphical user interface (GUI), which often resembles their real life counter-part. For example a plugin simulating a specific distortion pedal may use a visual mock-up of the pedal with virtually knobs and switches that rotate and switch via click dragging.
The term VST is often synonymous with plugin. VST stands for Virtual Studio Technology by Steinberg, which is one of the most popular API's for programming plugins. However, there are other types of plugins, such as Audio Units (AU) (Apple), Real-time Audio Suite (RTAS) (Digidesign), and Rack Extension (RE) (Propellerhead). For a complete list of plugin format compability with DAWs, see here. Also, a piece of VST software may be a synthesizer/sampler rather than a plugin, although these are technically called VSTi's, a type of virtual instrument.
Virtual Instruments
Virtual Instruments (or VSTi's - VST Instruments) are software algorithms that act as synthesizers or samplers. They usually take MIDI information as input and require an independent track in the workspace, generating audio output.
The most clear-cut application is to act as standalone synthesizers, with selectable/editable waveforms, cutoff filters, resonance, and envelopes. However, their use in studios also created some strong incentives to develop complex yet simple-to-use samplers. Multi-sampling instruments at a number of different pitches and volume points and interpolating allows samplers to successfully emulate real instruments. Piano, orchestra, drum, and bass virtual instruments are exceedingly common. Orchestra VSTi's are particularly welcome, as the cost/space necessary to record an entire orchestra is far beyond the budget of most productions.
MIDI
Like audio, DAW's feature MIDI tracks and events. MIDI events are blocks of classical musical notation, with each note being described by pitch, duration, and velocity. Additional data such as pitch bend or modulation can also be recorded.
MIDI input is often assisted by a piece of hardware called a MIDI controller. This typically has at least a small keyboard and pitch and mod wheels. It may also include vertical sliders, drum pads, knobs, a ribbon controller, a touchpad (also called xy pad), or other specialized controls.
MIDI events can be recorded from a midi controller (or the computer keyboard acting like a keyboard (although this approach will not capture note velocity)). MIDI can also be edited directly inside the DAW's (or an external) MIDI editor software. Double-clicking a MIDI event will open the DAW's MIDI editor. This typically appears as classical musical notation.
It can also appear as a piano roll grid where the y axis appears like a piano keyboard turned sideways. Each row corresponds to a specific note and horizontal space represents time. Notes appear as rectangular blocks (similar to events in the DAW workspace), whose length indicates the note duration. There may also be a specialized piano roll for drum editing, where notes appear as small diamonds. For drums, duration is meaningless, but aligning hits exactly at the right moment in time is crucial. Below the grid is often a small view for velocity, representing how strongly each note is struck. Click-dragging is used to select multiple notes, which can then have their pitch, timeline location, duration, and velocity altered at the same time.
Quantization
Most DAWs feature quanization of time, in terms of tempo, attempting to keep everything lined up with the beat. This essentially means that timeline selections and MIDI data are "snapped" to the nearest beat (or sub-beat division) instead of being placed exactly as specified. For example, if I set quantize to the nearest quarter note and try to set the timeline cursor a bit later than the 1st beat of a measure, the DAW will actually set the cursor to directly on the 1st beat. If I click a bit after halfway in the 1st beat, it will set it to exactly the 2nd beat, which is the nearest quarter note. This affects moving and slicing audio events as well.
MIDI recordings can be quantized as well. So if I'm playing in a slight swing feel but have quantize turned on to even beat divisions, the recorded MIDI will not reflect the swing feel - those slightly delayed notes will be snapped back to their non-swing locations. The same will happen with editing MIDI using the mouse or keyboard. You would have to turn quantize off or set quantize to 8th note triplets to get the notes to align how you want.
Quantization is a very useful feature to make sure everything lines up to the beat, but it can occasionally be too rigid. In this case you can decrease the size of each quantized block, for example switching from quarter notes to eighth notes. Or if you need a tuplet (such as a triplet), there are options to quantize in terms of fractions of the beat, like 3:2 (common triplet) or 5:4 (complex tuplet).
When you are dealing with audio, quantize may actually prevent you from getting an event to line up directly on the beat. For instance, consider a sample of a jet sound then an explosion. You want the explosion to line up on the beat, but the jet sound doesn't fit perfectly into your project tempo. So quantize is lining up the beginning of the jet sound instead of the explosion, and the explosion never lines up with the beat. Here's a case where it's best to turn off quantize and do what you need to do.
Quantize only affects current actions - it will not automatically align all the events/notes that have already been placed.
DAWs often have features to automatically create timing offsets on MIDI data. So for instance, you can take a straight on-beat MIDI event and choose a swing-feel feature, then adjust the lag of every note that occurs on the 2 and 4 beats. Even though the MIDI data is not changing, its playback gets the intended effect. Like inserts, this is handy to be able to adjust things after-the-fact, without having to re-record or edit lots of individual notes.
There are also randomize features to "humanize" virtual drums or other MIDI data, moving notes away from their quantized position in time.
Automation
Automation is a way to control some editable parameter on a track, such as track controls as well as insert or virtual instrument settings. This is usually done using an automation track, which is often tied directly to an audio or MIDI track.
The automation track uses an envelope to control the parameter, which is essentially a series of points with varying magnitude connected by interpolating lines. For example, I can create an automation track as a child of my audio track with an envelope that has a starting point of 0 and an end point of 100% that controls the track volume. When I playback my track, the audio will start inaudibly low and gradually get louder and louder until it hits max volume when the audio ends.
Of course, automation can do far more than create simple fades. It can do panning sweeps, swift audio drop-outs (stutter-type effect), and EQ changes. And this is just when controlling the track's basic parameters. It can also control parameters of plugins or virtual instruments. For instance, it can control the amount of reverb, create complex pitch shifting patterns, create complex cutoff filter movements, etc.
Automation can also be controlled by a configurable LFO (low frequency oscillator). This can be used to simulate most mod effects, as well as do achieve more original sounds when attached to different plugin/virtual instrument parameters.
Automation events can be manually edited or recorded like MIDI. Single automation events can control multiple parameters, or multiple envelopes can control the same parameter. The sky is the limit. Just be sure to stop tinkering at some point and finalize that track before you die!
For more, see Automation.
Mixing Board
Some DAWs feature a mixing board window that can be viewed at the same time as the timeline; others have a dedicated view. This looks like a typical tangible mixing board, with all the tracks laid out side by side with vertical sliders to view and edit their levels. This is most useful for balancing the volumes of the multiple tracks, but can also quickly get to the tracks' inserts, panning, or other features accessible from their control panels.
Overdubs
Looping
The timeline controller features a button that looks like a U laid sideways, or an arrow that turns around and points backwards, or a circle with an arrow in it. This button enables looping. With looping enabled, whatever section of the timeline is selected will loop indefinitely. When playback reaches the end of the selection, it automatically jumps back to the beginning of the selection. Playback can start before the selection, but will start looping once it reaches the timeline selection.
Looping is a good way to mix or edit an individual track, so you can listen to playback as you edit without having to keep jumping back to the timeline controller to play the appropriate part of the project.
Bounce/Render
When you playback your project, the DAW software must process each track, with all its required audio synthesis, plugin processing, panning, volume, fades, trims, and automation, as well as mixing all the resulting audio down to stereo in real time. As you start adding more and more tracks and effects, this begins to take its toll on system resources. If the load exceeds the CPU's ability to process in real time, playback will stutter, lag, or otherwise experience undesirable output.
The solution to this problem is to buy a new computer. ...Just kidding, you need to bounce or render smaller parts of your project. Bouncing basically records the output of a track or an output bus to an output file or a new track. For example, let's say you want to add one more guitar track, but this causes your DAW to reach the CPU limit. You have a virtual instrument track for drums with a lot of reverb and some other effects that takes up a lot of CPU, but you know it's finalized. So you can bounce that track down to an audio in a new track, then mute the drum track. The DAW will use far fewer resources playing back an audio event than it would computing the reverb and other effects on-the-fly.
This is also how you will get the final .wav or .mp3 file from your DAW to distribute to your millions of adoring fans. You bounce the master bus of the entire project to a file.
Tempo
Tempo is typically handled one of two ways, selectable by the user. Either the tempo can be set to fixed at a user-specified bpm or the tempo is set by the tempo track. Fixed is simple - just set it and go. The problem is that your song cannot change tempo at any point, which is common for intro's and outro's to emphasize the build-up or climax of the song.
The tempo track looks identical to any other automation track, but it is tied firmly to tempo. The same way you edit any other automation track you would edit the tempo track. So for instance you can set the initial tempo to 60 bpm, which ramps up to 120 bpm.
Be wary of changing the tempo once you have laid down audio tracks. MIDI data doesn't need to change, as it is only tied to tempo and not real-time. Audio, however, is fixed to a length of real-time. To time-shift it to fit different lengths of time must increase or decrease the pitch of the audio. Then to revert it to its original pitch requires using pitch-shifters, which typically leave noticeable artifacts.