[Top] [All Lists]

Re: Making a parabola

Subject: Re: Making a parabola
From: "Scott Fraser" scottbfraser
Date: Fri Sep 21, 2012 8:12 am ((PDT))
<< I=B9m wondering how you=B9d do this ? Why wouldn=B9t you roll off the ba=
ss post-recording instead? I really don=B9t see how you could simply put  b=
ack bass that was never recorded.>>

To further elaborate on David's & Klas' responses, wind rumble on a mic is =
a very high amplitude signal. If you adjust the record gain so that the win=
d rumble doesn't distort, your desired sounds are recorded at a needlessly =
very low level, at a point where preamp & converter noise start interfering=
 with the sound. If you don't adjust level for the rumble you get preamp ov=
erload & converter clipping which add upper harmonics throughout the spectr=
um, resulting in a ruined recording. I.e. it's not just bass which can be s=
imply high pass filtered out at this point. So we use bass cuts (high pass =
filters) in order to preserve headroom in the original recording chain, hea=
droom which can't be reconstituted in post production.
Additionally, a high pass filter does not eliminate bass frequencies. It at=
tenuates them. So bass is still recorded through a high pass filter, but at=
 a lower level, & this is what David refers to by putting back bass in post=
. It's there & can boosted by the same amount as the bass cut on the mic/pr=
eamp/mixer in order to restore natural lows, if needed. Generally, nature h=
as rather little energy in the lowest octave of human hearing most of the t=
ime, except for thunder, waterfalls, waves & a few other geologic events. S=
o rolling off lows primarily serves to attenuate human vehicular noise. Win=
d rumble goes all the way down to DC & the gentle 6dB/octave curve David me=
ntions, even if at 50Hz is enough to eliminate the most pernicious effects =
of rumble while still allowing a reasonable sense of low frequency reality =
to be recorded.
So, a long answer to your query, but the short version is that we have to s=
tart with the best possible recording of the original event, & not plan on =
fixing problems later, since many of the problems relating to wind actually=
 can't be fixed adequately in post.

Scott Fraser

<Prev in Thread] Current Thread [Next in Thread>

The University of NSW School of Computer and Engineering takes no responsibility for the contents of this archive. It is purely a compilation of material sent by many people to the naturerecordists mailing list. It has not been checked for accuracy nor its content verified in any way. If you wish to get material removed from the archive or have other queries about the archive e-mail Andrew Taylor at this address: andrewt@cse.unsw.EDU.AU