View Single Post
Old 06-10-2019, 05:47 AM   #7
Human being with feelings
Join Date: Jul 2010
Location: Silicon Valley, CA
Posts: 2,086

The "pro standard" is 24/96 and virtually all DAWs work in floating-point internally. I "doesn't hurt" to make & keep a 24-bit or floating-point master.

Put a half volume signal into a 16 bit container and you use 8 bits. That's an 8 bit recording.
No. Half the volume (-6dB) "looses" 1 bit. With 8-bits you can count from 0-255 and with 8 bits audio you can hear quantization noise. With 16-bits you can count to 65,535. 16-bit audio holds values from −32,768 to 32,767 and you can't hear the quantization noise with 16 (or 15) bits.

The guys who do scientific, blind, level matched, ABX Tests have pretty-well demonstrated that nobody can hear the difference between a high-resolution original and a copy downsampled to "CD quality". Audiophiles (and pros) who claim to hear the difference usually haven't done scientific blind listening tests... And, it can be humbling to find out your ears aren't as good as you think!

Mixing is done by summation. (Analog mixers are built-around summing amplifiers.) If you mix two 16-bit files without changing the levels you need 17 bits (that's assuming the original files are at a level that requires all 16 bits.) Or, you could say you gain 1-bit of resolution. With floating-point you actually are gaining resolution. If you mix 4 16-bit files, you need 18-bits, etc.

Consumer format is 24 bit fixed... so get your levels in order at the end of your mixing chain!
With fixed (integer) formats you are hard-limited to 0dB so you have to "watch your levels". Floating-point can go way-way over 0dB. But, DACs are always limited to 0dB so no matter what your consumer/release format is you shouldn't exceed 0dB, otherwise the listener can clip (distort) their DAC.

The "consumer format" could be almost anything from MP3 or AAC to 24/192 on a Blu-Ray.

Last edited by DVDdoug; 06-10-2019 at 05:57 AM.
DVDdoug is offline   Reply With Quote