Two3 many bits used to encode sound? I say yes!

A while back I analysed the lossless audio encoding of a few hundred Blu-ray discs. The aim? To find out how efficient lossless compression of 24 bit audio was compared to 16 bit.

The results were not entirely unexpected. The first 16 bits score high compression ratios. The next eight bits score low ones.

The reason? I posit that most of the least significant eight bits are simply encoding noise. Noise is random, and so unpredictable. The efficiency of the lossless algorithm depends heavily on predictability. Therefore, much or most of those eight extra bits do not compress well.

Here’s the full article (a version was previously published in Sound and Image magazine).

(I’ve also changed the menu bar above so that this kind of thing will now appear, along with the ‘Sharpness‘ piece, under ‘Articles‘ above.)

This entry was posted in Admin, Audio, How Things Work, Imperfect perception. Bookmark the permalink.

Leave a Reply

Your email address will not be published. Required fields are marked *