Pro Tools gives me the option of 16 bit or 24 bit. Is there a reason I shouldn't just always use 24-bit sessions to get the highest quality sound?
migrated from avp.stackexchange.com Jan 27 '14 at 15:02
This question came from our site for engineers, producers, editors, and enthusiasts spanning the fields of video, and media creation.
Let me generally explain the use of bit resolutions.
24 bits is giving you a higher dynamic range: you can store audio information until a level of -144dB FS instead of -96dB FS in 16 bits. That is the quality improvement you get, so the precision in dynamic range improves. This is useful if your recorded material is too soft, and you want to increase the volume while keeping the information in the less significant bits (i.e. the soft passages of the music) intact.
Recording in 24 bits does have two major drawbacks: 1. The audio files consume 50% more disk space 2. If your master needs to be 16 bits (i.e. if you are mastering to an audio CD) you need to add dithering noise to your recording in order to mask away rounding errors when truncating the least significant 8 bits.
In general, if you have the disk space, it is reasonable to record in 24 bits, since the dithering options available today are very good and almost not audible to people.
There is one situation I can think of where 16 bits recording would be better: if you are going to record audio and you have no planning on increasing the volume at any point, so no faders > +0dB, no expanders, EQs, or anything else that can gain the signal. (Reverb is fine since it never increases the original signal strength). In this situation, using 24 bits has no additional value and you might as well save yourself the disk space and the dithering noise by recording directly on 16 bits.
the short answer:
no. record and edit at 24 bit when 16 or 24 are your options.
the longer answer:
there are a few exceptional cases where it may be worth considering
simply put, 16 bit is a holdback from last millenium if you are working on anything serious, and today's hardware is cheap.
the human ear is responsive to about 120dB of dynamic range at the right volume; that is equivalent to 20 bits.
your ADC and DAC convertors are likely to have a signal to noise ratio from 90 to 120 dB.
you will want to record live sources with ample headroom, to avoid clipping.
you will want to perform destructive editing. destructive editing at 16 bit can quickly introduce audible artifacts.
altogether, that means you'll want to record at 24 bit to keep errors imperceptible (for practical purposes).
during mixing, playback, and processing, ProTools increases your audio file's samples to its native stream format (which varies by the PT software/hardware you use), this internal stream is better than 24 bits. this stream is truncated when passed to the D/A - either by PT, or by the DAC.
finally: i don't agree with Pelle - do not worry about dither. dither is a very good thing when used correctly.
Check your hardware... if you're not converting at 24-bit, then there's no point to a 24-bit session.
(Right? Or would it still be worthwhile for internal processing?)