Skip to content

Sample rate: which one to export at

Send feedback

Last updated 2 min readTags:Getting startedExport

Choose a sample rate for a Sonalix upload and understand how it affects spectrum and true-peak readings.

Session or source file Export for Sonalix
44.1 kHz session 44.1 kHz
48 kHz session 48 kHz
Session at 88.2 kHz or higher A 48 kHz copy, with the original kept for delivery
Existing 32 or 22.05 kHz file Upload it as it is, or make a fresh 44.1 or 48 kHz export from the original session if available

For delivery, use the rate requested by the mastering engineer, distributor or pressing plant. For the remaining upload settings, see how to export a mix for analysis.

What the rate sets

Sample rate is the number of samples per second. Half the rate is the Nyquist frequency, the upper limit for frequencies the file can represent. At 44.1 kHz it is 22.05 kHz. At 48 kHz it is 24 kHz. Both sit above the commonly quoted human hearing limit of about 20 kHz.

Upsampling an existing 44.1 kHz mix to 96 kHz leaves its captured bandwidth unchanged. Downsampling removes content above the new limit. If the destination needs a different rate, convert from your best available source after mixing and processing. If you also need a lower bit depth, do the rate conversion first and apply dither at the final reduction.

24-bit stereo WAV Per minute Four minutes
48 kHz 17.3 MB 69 MB
96 kHz 34.6 MB 138 MB
192 kHz 69.1 MB 276 MB

Doubling the sample rate doubles the PCM data at the same bit depth and channel count. Converting an MP3 to a higher rate or to WAV leaves the information lost to lossy encoding missing.

What changes in a Sonalix report

Spectrum. Sonalix measures the decoded file at its native rate. A 32 kHz file runs out of bandwidth at 16 kHz. A 22.05 kHz file runs out at about 11 kHz. Bear that in mind if the top of the curve drops away. The file format guide covers other export effects.

True peak. The level engine resamples to 192 kHz for true peak. Sonalix validates readings at 44.1 and 48 kHz against the EBU Tech 3341 test cases. It reports 32, 88.2, 96 and 176.4 kHz as provisional even though they passed the same synthesised cases. At other rates the reading is unavailable. See measurement methodology.

Results from a single export test. Sonalix's 48 and 96 kHz versions gave the same loudness, true peak and correlation in that test. In its five-band results, the 96 kHz bands generally read about 3 dB higher, with Sub 5.8 dB higher. Sub stereo width read 0.084 at 96 kHz and 0.012 at 48 kHz. The test does not establish the cause or predict the difference for another mix. Use the same sample rate when comparing successive exports.

Was this page helpful?

    Type to search titles, headings and summaries.