Questions we often hear about music mastering
File formats, bit rates.. what the hell is going on here?
As far as I’m concerned, the only hard and fast rule is you don’t send in masters that are clipped. At either the recording or mixing stage.
All mastering engineers have a different preference as our gear and plugins is set to start working from a certain level. So unless it’s clipped, there’s no reason it can’t be accepted, we’ll just adjust the level to where we prefer out peaks.
That being said, it saves the mastering engineer a step if you send it in at their preferred peak level.
Mine is -3dBFS in case you wanted to send in a song for mastering.
Just send what your recorded in. You won’t gain anything by converting a 16 bit recorded and mixed file at 24 bit.
BUT – don’t send 16 bit, if you recorded at 24 bit.
If you are recording at 32 bit integer, you are either wasting a lot of drive space. Even at 32 bit floating point, the dynamic range is more than 5 times what is physically possible in earths atmosphere (https://www.sounddevices.com/32-bit-float-files-explained) 32 bit float is great for the internal processing of a DAW but not useful as a file format.
If you are deciding what to record in , 24 bit gives you 6dBFS of headroom above 0.. That really should be enough.
Easy – record at 48Khz, or 44.1Khz if you can’t do 48.
There’s reasons why in the past 96KHz was a good idea but it’s not so relevant now. It’s only worth sending in 96Khz files if you recorded in that sample rate OR you are planning to release in HD formats 24/96.
Higher sample rates?
Dan Lavry of Lavry engineering, arguably the best AD and DA converters you can buy, explains technically why theoretically there’s no reason to record higher than 96Khz.
Whether we can perceive a frequencies above 20Khz is arguable, however the average person can’t hear above 17Khz by the time they are 24.