Re: [Phys-L] decibels

[John Denker <jsd@av8n.com>]

On 4/29/19 4:20 PM, Anthony Lapinski wrote:
> A student sent me this short article about the world's quietest room:
>
> www.smithsonianmag.com/smart-news/earths-quietest-place-will-drive-you-crazy-in-45-minutes-180948160/
>
> Anyone ever been in a room like this?

Yes.

Bell Labs has been doing acoustics experiments since
the days of Alec Bell.
 
> And what do negative decibels mean? The "threshhold of hearing" (10-12 W/m2)
> corresponds to 0 dB. So a sound quieter than this (say, - 20 dB) has an
> even lower intensity (10-14 W/m2)? Is there a lower limit?
>
> How does this work with volume control knobs on stereo systems?

The question incorporates a misnomer.  Strictly speaking,
a dB is a /relative/ unit.  It can be combined in a
variety of ways with some *reference level* to create
a variety of derived units:

 -- dBm		relative to 1 milliwatt of power
 -- dBA		relative to the A-weighted acoustical scale
		which is scaled to the threshold of hearing
		as a function of frequency
 -- dBB, dBC	other weighting schemes
	https://www.ocn.org.au/guide/244-weighting-dba-and-c-weighting-dbc

..... and so on.

I heartily recommend reserving the plain old dB unit for
relative measurements, such as 20dB of gain or 3dB of
insertion loss.

It is sloppy shorthand to use dB for an absolute measurement.
Sometimes you can get away with it if the reference level is
obvious from context, but I don't recommend it.  If you mean
dBA, say dBA.

Audio equipment is a mess.  There are multiple different standards
for what 0 Vu is supposed to be.  Consumer equipment is different
from professional studio equipment.  There's not even consensus
on how much headroom there should be in a digital PCM file.

Authoritative reference:
  https://www.youtube.com/watch?v=4xgx4k83zzc