Production guides · Meme Plugins · 2 November 2026
Serial processing passes one processor’s output into the next. Parallel processing splits the source into separate paths and combines their outputs. Choose the route according to which sound each processor should receive, then judge the combined result at a comparable level.
1. Draw what feeds each effect
Write a simple path before moving devices: source → delay → reverb is different from source → delay and source → reverb. In the chain, the reverb receives the delay’s output. On separate paths, each receives its own source feed.
Ableton’s Rack reference describes devices in series within a chain, with parallel effect chains receiving the same input and their outputs being mixed. A return-based setup can implement the same routing idea, with different controls for the feeds and returns.
2. Control dry signal deliberately
A processor’s dry/wet control affects what leaves that path. When the original stays audible on its own route, an effect-only return often needs its fully wet setting. An unnoticed dry copy can make a parallel setup louder before it adds any useful effect.
Read insert effects versus send effects for that distinction. A parallel path can itself contain several devices in series; these terms describe signal relationships, not exclusive project types.
Example: one tap, delay and reverb
Imagine a short original percussive tap. This is a hypothetical routing exercise, not auditioned audio. Keep the dry tap on the same separate route in every trial.
- Serial effect route: feed the tap into a fully wet delay, then a fully wet reverb. The reverb receives the delay’s repeats. Start with short, manageable feedback and decay.
- Parallel effect routes: feed the tap separately into the same fully wet delay and reverb, then sum those two returns. The reverb now receives the original tap directly, alongside an independent echo return.
- Compare: adjust final listening level rather than assuming equal fader positions create equal loudness. Hear the first echo, its surrounding space and the ending.
- Reverse the chain: as a separate test, place delay after reverb. Listen to how the already reverberant sound is repeated. Do not change several settings while deciding what order does.
The purpose is to identify the input to each processor. It does not establish one universally better sound, and matching output level does not make the routes mathematically equivalent.
3. Check interactions after the sum
EQ, saturation or dynamics placed after the sum acts on the combination. Moving it into one branch changes the task. If a blend becomes hollow, inspect timing and polarity and the host’s delay compensation before adding more processing.
For a dynamics-specific two-path example, follow parallel compression. The dry path can remain present while the final envelope still changes.
4. Verify routing in playback and export
Solo each branch briefly, restore the complete arrangement and check a sparse passage. Reopen the export and listen through the final tail. Keep the routing whose behavior you can explain and whose result serves the song.
Does parallel mean no interaction?
No. Paths interact when summed and when later devices process the combined signal. Their levels and timing still matter.