Klas, you wrote,
>A simple way to learn the difference is to put an attenuator on the mic an=
d
>lower the mic self noise until you start to hear the MD / DAT /Edirol
>/whatever mic amp noise. Then you can hear the difference.
Excellent plan.
>2/ When reading noise figures, consider that "A-weighted" noise or
>"CCIR-weighted" noise means that one has compensated for the frequencies
>which are most audible to the human ear.
At low sound levels!
>For naturesound recordists there should be another way to "weight" a noise=
,
>compensating for the noise which is most masked by nature atmosphere. A
>high frequency noise can be audible "on top" of the atmosphere, while a lo=
w
>frequency noise is masked by it and not possible to hear at all.
A-weighting is suitable for that, it mostly rolls off the bass.
>3/ When I send my postings on this list, I don't feel that I am understood=
.
>I do not succeed in finding the right words for my logic. Let me try this =
one:
>
>Suppose a microphone has a self noise of 16 db(A). Fine. That cannot be
>changed by any pre-amp.
>Then the question must be: "How sensitive (output =3D mv/Pa) must that
>microphone be to "noise-dominate" over the MD/DAT/ Etc mic input noise?
>That is the question which needs to be answered!
>
>I mean - the noise you hear should always come from the mic!!! The mic mus=
t
>always be the weakest link in the chain!
Yes.
>4/ Sensitivity must be presented as mV/Pa, NOT -db, whatever - as there is
>too much cheating on the market, using different reference levels for the
>db measuring.
I'd prefer a dB rating to a common standard, but I agree entirely
that mic sensitivity specifications are a morass of confusion. I use
a Shure cardboard slide rule to convert from one reference to another.
>I spent two hours on the web a few days ago (with broadband- finally!! her=
e
>in the woods!!) - to straighten out my questions about "sensitivity"
>expressed as "db". I found myself in a total mess of different standards.
>If you like to try - please visit
>http://www.acoustics-noise.com/dBA-B-C-D-U-G-U-AU-tables.shtml for example=
.
That's about -weighting-, not sensitivity standards. The weightings
other than A and C are rarely encountered and we needen't worry about
them.
>When the data sheet says: "10 mv/Pa at 1000 Hz" - fine, that is a fact! On=
e
>can understand it. But when it says: "Sensitivity -64" db or something, I
>don't care about it. I don't see any good reason why somebody should leave
>an established and accurate expression (mV/Pa) in favor of a strange figur=
e
>(db), unless there is fraud going on.
dB figures are easier to understand and translate to effects, -when-
they are using the same reference.
>We must all understand that: When one manufacturer starts manipulating wit=
h
>data, even lying, all the others have to follow. Or die.
Example: the use of dB scales for mic directivity polar plots. Makes
no sense. A linear scale (0 to 1) shows the real shape of the pattern.
-Dan Dugan
________________________________________________________________________
________________________________________________________________________
|