16 June 2013

Bits and Bobs and Bytes

Today's tidbit is only tangentially related to sound-for-picture. Usually, production sound mixers do not find themselves caught up in esoteric debates about sampling rates and bit depths, or steadfast proclamations of the superiority of analog formats over digital.

Aside from certain specialized SFX gathering, we live and breathe at one set of data specs: 48 kHz sampling rate and a 24 bit depth. (Yes, once upon a time there were off-speed sampling rates to deal with pull-up/down for 35mm film, but these have mostly been superseded by modern DAWs' capability to re-sample cleanly on-the-fly).

Me mixing on set. (Ha ha, no. I don't have that haircut anymore).

In the music world, it's a different story. Numerous articles and reviews extolling the "air" of higher sampling rates, or the flat-out accusation of inferior playback have been levied at the venerable redbook CD format, and its cousin visiting from out of town, the MP3.

Don't even mention vinyl. We'll be here for hours.

Do I have a personal attachment to either? Nope. But I do have a certain bemused schadenfreude when an engineer can systematically demonstrate what others vehemently rail against, as in the case with the following video from the wonderfully nerdy folks over at Xiph.org (and this is good viewing for anyone who wants a clear, concise explanation of how digital audio works).

To reiterate: there are specific reasons why we on the pro side of things need 48 kHz and 24 bit: 48 mathematically agrees with digital video formats better, and we need 24 bits of headroom for the processing that will inevitably occur with production audio tracks (which are handled in most DAWs with an internal precision of 32 bit floating point). But for the folks at home? Rewind your CDs and give them another spin. Assuming better mastering of latter-day discs, they'll sound just fine.

Link to video.

No comments: