> I think it is not enough to just look at the self-noise specs - the
> sensitivity matters as well. More sensitive the microphone, the less gain=
> you need less noise.
Jonas,
How come cheaper mics often give better specs than professional ones? The
only specs worth looking at compare the mic noise to the thermal noise of
the air itself, but there is no agreed way of doing this consistently.
The best professsional mics quote weighted noise above thermal noise at bes=
t
Message: 10dBs.
Subject: The record for a B&K laboratory mic is 5dBs. Believe specs over 15dB=
s
and doubt the rest. The big snag is that for wildlife we need the low level=
high frequencies which are removed by the weighting curves.
The real test of mic noise is to listen to the recordings, comparing the
high frequencies with a "pile of bedclothes" test. There is another clue. I=
f
you make an audiogram of a noise only section, you will sometimes see a
6dB/octave tip up above 15KHz or so with a high quality mic. This may be th=
e
thermal noise of the air which is what the theoretical mic noise specs are=
supposed to be based on - but which they filter out.
Allowable noise depends also on the sounds being recorded, and the only tes=
t
for this is your ears, not the specs. The "noise weighting" curves do not
follow the response of most human ears in practice. They well may mean
something in quiet studios, but in the open environment, I find the most
objectionable noise is above the main frequencies of the noise graphs - the=
hiss.
One reason for this is that outdoor naturel noise is usually bass heavy
because the higher frequencies are absorbed with distance.
It is possible to remove hiss from wildlife recordings using a sound editor=
(I use Audacity). Take a sample from the quietist second or two and put the=
sample through a high pass filter. The noise reduction will only apply to
the higher frequencies and will be inaudible in practice if done carefully.=
There is one sure way to beat mic noise and that is to use an amplifying
device such as a parabola, or even a rectilinear horn such as my cardboard=
prototype "Brinibox" which boosts the incoming sound by about 12 dBs. The
noise of =C2=A310 mics is comparable to Sennheisers.
Listen to:
http://www.stowford.org/recordings.htm#stereotest
David Brinicombe
|