Composite Recording: How to Build the Perfect Part

Build the perfect part by recording multiple takes, then editing them into a single, composite take

By Craig Anderton

Composite recording lets you record multiple takes of a part, usually in quick succession so you can get into a “groove,” then edit the best sections together into a single, cohesive composite part (e.g., take one’s beginning, take three’s end, and take two’s middle). Different DAWs handle the mechanics differently, but there are two crucial artistic factors: 

• Don’t obsess.I’ve seen engineers piece together parts on an almost note-by-note basis, but a great part evolves time. Generally, the longer the chunks you piece together, the more coherent the part. Pasting together too many little bits can create a Frankenpart.

• You can’t create a great take; you record a great take. Editing exists to give something that’s already magical the extra 10\% to make it transcendent—it’s not about turning garbage into better garbage.

THE COMPOSITE RECORDING PROCESS

Define the region you want to loop (i.e., where playback or recording jumps back to the beginning after reaching the end, then continues). Allow a few measures before and after the part so you can “regroup” after each take. With most DAWs, turning on looping prior to hitting record is sufficient to enable loop recording. Some DAWs place each take in its own track, while others create separate layers within one track. 

Here’s why it’s good to record no more than a half dozen takes at a time: 

• You won’t have to wade through too many takes when trying to locate the best sections.

• If you can’t get a good performance in six or seven takes, you may need to practice the part more, or re-think it.

• It’s good to hear what you’ve done before committing to too many tracks, so you don’t waste your time if they’re going in the wrong direction.

EDITING

After recording, compare takes to choose the best sections. Various programs employ different workflows; I use Sonar's "speed comping" mode to assemble a collection of “good” clips (see Fig. 1). Selecting a clip automatically mutes other clips in the same column, and you can use the arrow keys to navigate among rows and columns.

Fig. 1: With Cakewalk Sonar’s composite recording, the highlighted clips have been selected, while the gray clips are muted. The waveform at the top shows the overall composite waveforms from the various selected clips.

Once you’ve isolated the clips, delete anything you don’t want to keep. For smooth transitions between sections, you may need to add fades at the beginning or end of the “winning” clips. Sometimes crossfading two clips yields the best transition. Finally, bounce all the best bits into a single track.

DOUBLING, TOO

Doubling your part can create a more animated sound. Although you can do this electronically, “real” playing usually sounds better. 

Check if any other takes are equal to, or almost as good as, the chosen takes. If so, drag the “secondary” takes to another track to create the doubled part. Another option is to learn the composite part, then play it again to create the doubled track.

In any event, always remember is to choose individual phrases based on musical continuity, not just musical perfection. The technology should serve you—not the other way around.

Craig Anderton is Editor Emeritus of Harmony Central. He has played on, mixed, or produced over 20 major label releases (as well as mastered over a hundred tracks for various musicians), and written over a thousand articles for magazines like Guitar Player, Keyboard, Sound on Sound (UK), and Sound + Recording (Germany). He has also lectured on technology and the arts in 38 states, 10 countries, and three languages.