12 Jul
2013
12 Jul
'13
10:34 p.m.
On Fri, Jul 12, 2013 at 9:20 PM, Ricardus Vincente <wizardofgosz(a)gmail.com>
wrote:
Further, if the D/A has sufficient headroom (modern
good D/A should) I
would think it's not
really a problem.
Found it! I was thinking of a presentation I'd seen somewhere, where the
details of inter-sample peaks is properly explained: Slide 22 of Fons' EBU
loudness meter presentaton (given at LAC2011 Maynooth, by Jorn
Nettingsmeier):
http://lac.linuxaudio.org/2011/download/lm-pres.pdf
The wave is pretty much predictable: until a (strange) *double 1.0* value
sample appears. The D/A conversion goes crazy, a spike to value of approx
3, while the range should be -1 to 1.
The worse thing is, the longer this "predictable" part is, the *higher* the
spike gets.
So in short, I tend not worry about it. :-)
Me
neither! Posted this for the sake of completeness. -Harry