Insert Effects vs. Send/Return Effects
There are two ways to apply an effect to a signal. The distinction matters more than most beginners realize.
Insert effects sit directly in the signal path. The audio flows through them — what goes in comes out processed. An EQ on a vocal track is an insert. A compressor on a drum bus is an insert. The entire signal passes through the effect, and what you hear is the processed result. If you bypass the effect, you hear the original signal.
Inserts are for effects that need to process the entire signal: EQ, compression, gating, saturation. These are “series” effects — the signal flows through them in sequence, each one acting on the output of the one before it.
Send/return effects work differently. You send a copy of the signal to a separate track (called a return, aux, or bus), where the effect lives. The original signal stays dry on its channel. The effect output — the wet signal — lives on the return track. You blend the two with the fader and send level.
Sends are for effects where you want to share one instance across many tracks — reverb is the classic example. Instead of putting a separate reverb plugin on every track (which wastes CPU and makes each track sound like it’s in a different room), you create one reverb on a return track and send varying amounts of each channel to it. The vocal gets more reverb, the drums get less, but they all share the same space.
This is parallel processing — the dry and wet signals run in parallel, not in series. You’ve heard the term before (Chapter 7 mentioned it). Now it becomes practical.
Two signal flow diagrams: insert (signal through effect in series) vs send/return (signal splits, dry passes through, copy to effect on return track).
Pre-Fader vs. Post-Fader Sends
When you send a signal to a return track, that send can tap the signal at one of two points:
Post-fader: The send level follows the channel fader. Turn the channel down, the send level drops proportionally. This is the default for most mixing sends — if you pull a vocal down in the mix, you usually want its reverb to decrease too.
Pre-fader: The send level is independent of the channel fader. Even if you pull the channel all the way down to silence, the send still feeds the return at whatever level you set. Pre-fader sends are useful for headphone mixes, monitoring setups, and special effects where you want the wet signal without the dry.
Most of the time, post-fader is what you want. Use pre-fader deliberately, not by accident.
Two signal path diagrams showing where the send taps the signal relative to the fader — pre-fader vs post-fader.
Grouping and Buses
Vocabulary
Bus
A channel that collects the outputs of multiple tracks into one — letting you control, process, and manage a group of related signals with a single fader. Also called a group or submix.
A bus collects the outputs of multiple channels into one. Route all your drum tracks to a drum bus, all your vocals to a vocal bus, all your guitars to a guitar bus. Now you can:
- Control the level of the entire drum kit with one fader
- Apply processing (compression, EQ, saturation) to the drums as a group
- Mute or solo all drums with one click
Buses also give you a sane way to manage large sessions. A session with 40 tracks and no buses is chaos. A session with 40 tracks routed through 5-6 buses is organized — you can see the shape of the mix at a glance.
Nesting: You can route buses into other buses. Individual drum tracks → drum bus → instrument bus → master. This gives you multiple levels of control. Adjust the snare relative to the other drums on the drum bus. Adjust the drums relative to everything else on the instrument bus. The flexibility scales with the complexity of your session.
Hierarchy diagram showing individual tracks (kick, snare, hat, toms) routing to drum bus → instrument bus → master bus.
Resampling, Freezing, and Flattening
Sometimes you need to commit a processed signal to audio — to “print” the effects so you can move forward without the CPU load or the temptation to keep tweaking.
Freeze: Most DAWs let you freeze a track, which renders the audio with all its effects in place but keeps the original accessible. You can unfreeze later to make changes. It’s a temporary commit — useful when your CPU is struggling.
Flatten/Bounce in Place: Renders the track to a new audio file with effects permanently applied. The original is gone (or replaced). This is a permanent commit. Use it when you’re sure you’re done with a sound and want to move on.
Resampling: Recording the output of one track (or a bus, or even the master) into a new audio track in real time. This captures everything — effects, automation, routing — as a new audio file. Resampling is essential in electronic production for creating new material from processed sources, and it’s useful in mixing for capturing complex effect chains as a single audio file.
Audio Compression vs. Data Compression
Two completely different things that share a name. In a chapter about signal flow, this is worth clarifying:
Audio compression (dynamics processing) reduces the dynamic range of a signal — making quiet parts louder and loud parts quieter. That’s Chapters 16-20.
Data compression reduces the file size of audio. It comes in two flavors:
- Lossless (FLAC, ALAC, WAV): The file is smaller but the audio is identical to the original when decoded. No quality loss. Use lossless for production work.
- Lossy (MP3, AAC, OGG): The file is much smaller, but audio data is permanently discarded. The encoding algorithm removes information it predicts you won’t hear. At high bitrates (256-320 kbps), the loss is minimal. At low bitrates, it’s audible — high frequencies get smeared, stereo imaging narrows, transients soften.
Related: sample rate (how many snapshots per second — 44,100 Hz is CD standard) and bit depth (the resolution of each snapshot — 16-bit for distribution, 24-bit for production). Higher values mean more data and more fidelity, but with diminishing returns. Record at 24-bit, 44.1 or 48 kHz. Export finals at whatever the destination requires.
What to Practice
- Build a mix template. Create a session with buses for drums, bass, vocals, instruments, and effects returns (one reverb, one delay). Color-code everything. Route your channels to the appropriate buses. Save it as a template. Open it every time you start a new mix.
- Set up a send/return reverb. Put a reverb plugin on a return track. Send three different instruments to it at different levels. Listen to how they share the same space while maintaining different amounts of wetness. Now compare this to putting a separate reverb on each track — hear the difference in cohesion.
- Freeze and unfreeze a track. Pick a track with heavy processing, freeze it, and notice the CPU drop. Unfreeze it and verify nothing changed. This workflow saves sessions when your computer is struggling.
