Choose sample rate, bit depth, and dither for the handoff in front of you

Ask what the file is for

Choose handoff settings from the next job, not from a slogan about the highest number. A mix engineer may need a project-matched multitrack set. A picture editor may specify a video workflow. A mastering engineer may request a particular delivery format. A listener preview has different needs from an editable archive. Get the destination specification in writing before converting anything.

Within a session, avoid unnecessary sample-rate changes. Apple notes that mismatches can cause incorrect playback and that higher rates cost more processing and storage. That is a workflow reason to match a project’s established rate where possible, not proof that one rate is universally better sounding.

Keep the working master distinct from copies

Export a clearly named, lossless master at the project’s intended handoff settings. Make encoding copies from that master rather than repeatedly converting a compressed or already converted file. Record the file type, sample rate, bit depth, channel layout, version, and whether master-bus processing is included. WAV is commonly convenient across systems; an agreed alternative is also fine when the recipient can open it.

Bit depth belongs to the same decision. Logic’s current recording documentation distinguishes 16-, 24-, and 32-bit float recording choices; export options depend on the version and operation. For a transfer, the meaningful question is what the receiving stage supports and needs. Do not reduce a high-resolution editable source merely because a consumer service will later create its own delivery encodes.

Use dither at the reduction, not as decoration

Dither is relevant when reducing bit depth. Check the effective word length at the final output, not only the input file’s label: processing can occur in floating point even when the source and destination both say 24-bit. A floating-point handoff for further processing and a reduced fixed-point final master are different decisions. If a requested final file requires a lower fixed-point depth, make one intentional final conversion, document the tool and setting, and retain the pre-conversion master. Do not stack multiple dithered conversions while sending versions around.

Worked example: two destinations, two files

A producer has a 48 kHz, 24-bit project for a short film and also needs a client preview. They export a 48 kHz, 24-bit WAV print mix and aligned stems for postproduction, with no unrequested rate conversion. Separately, they make a clearly labeled review file for email. The review copy is not substituted for the post file, and the archive retains the exact source master. If the mastering or picture team asks for a different specification, they generate a new derivative and log it.

Limitations

Services, distributors, editors, and clients can change their requirements; no single sample rate, bit depth, or dither setting guarantees platform acceptance or sound quality. Recheck the actual destination. For delivery structure, pair this with stems that line up; for critical listening after mastering, use a level-aware master comparison.