Re: [LAD] twice as loud

On Fri, Jul 23, 2010 at 11:34:43AM +0200, Philipp Überbacher wrote:

> > > We could think about what makes judging twice the
> > > loudness more difficult and maybe find a relation to another phenomenon
> > > this way. The limits of hearing apply to everything, but what about
> > > factors like the time between two sounds or the length of the sounds?
> >
> > All of these affect both masking and loudness.
>
> Yep, but maybe some of the other possible factors match one phenomenon
> but not the other.

Indeed. As I said, this relation between 'loudness' and masking
is pure conjecture, I have no hard arguments pro.

> May I ask why you used 10*log(2/1) in your two person example?

You mean why power and not amplitude ? Two persons talking would
produce twice the power, since the signals are not correlated.
So if our idea of 'twice as loud' would be determined by such
experiences (but it clearly isn't), it would refer to power.

> Do nearfield effects matter?

Probably yes, but don't ask me how !

Joern's remark that the phrase 'twice as loud' doesn't make
sense is to the point. We only accept it because it is 'well-
formed' at the language level. But there is no a priori
numerical value for loudness (indeed we are trying to find
one !), so 'doubling' it is in fact undefined.

Ciao,

-- 
FA
There are three of them, and Alleline.
_______________________________________________
Linux-audio-dev mailing list
Linux-audio-dev@email-addr-hidden
http://lists.linuxaudio.org/listinfo/linux-audio-dev