The drummer sends a corrected drum stem. The new folder also contains bass, guitars and vocals with yesterday’s names. Your job is to deliver the corrected performance while keeping the mix everybody approved. Start with the two exports, the approved stereo bounce and a session in which you can audition files at their original speed.
Establish a common starting point
Create folders A and B and keep the supplied filenames inside each. Record the sample rate, channel count, duration and stated timeline start. A stem is a printed group, such as all drums together. Ask whether its effects are included, printed separately or shared across the stereo mix. Some DAWs can include returns and main processing in track exports, so two groups called “Drums” can contain different things. Ableton’s render options
Import each set onto separate tracks with time stretching off. Route the sets to separate stereo groups. Start at their documented origin and compare three landmarks, such as the first kick, the chorus entrance and the last snare. For example, an extra 0.5 seconds at the top of every B file explains a constant half-second offset. A gap that grows toward the end needs a different diagnosis. Check the source clock, file sample-rate interpretation and tempo-follow settings before moving individual hits.
Compare one group at a time
- Switch between A and B using exclusive solo. Begin with the corrected drum fill. Write down precisely where it changes and whether the following downbeat still lands with the bass.
- For two files expected to be identical, align them to the sample, set equal gains and pan positions, and invert the polarity of both channels of one file. Sum the pair through a clean bus. Identical aligned samples cancel. Disable the inversion before returning to ordinary playback.
- Listen to what remains. A nearly complete song suggests a gross level or alignment difference. A decaying snare in the residual suggests the effect tail warrants inspection. Free-running modulation, dither and nonlinear processing can also leave a difference. A residual tells you where to listen. Its presence alone cannot decide which version belongs.
Build a candidate from the approved A set plus the confirmed B drum stem. Compare its full chorus and ending with the reference. If a compressor on the original master reacted to all instruments together, the sum of independently processed stems may sound different even when the routing is correct. Record which version of every group you used and any deliberate difference from the approved bounce. Somebody reopening the project should be able to reproduce your decision.
Companion chapter · Make and check a release master
Bedroom to Stage contents and glossary