From:
>
> In a message dated 2/13/04 6:44:03 PM Pacific Standard Time,
> writes:
>
>
>>> There are those that worry that sound levels will not have enough detail
>>> in the lower levels, but that's quite a way down. I've cranked in
>>> 30-40dB of gain into a quiet recording in post and everything I hear is
>>> still plenty good
>
>
> Dear Walt,
>
> The above reflects my concern. On the Portadisc, if one records at -50dB will
> the signal be sufficient to capture all the sound? I guess there is some
> point where the input is just not enough to allow the machine to digitize
> sufficient "bits" to make a good recording. Any idea where this might
> be..-60..-50..-40?
I can tell you it's not a specific point, but varies depending on what
you are recording. And the best way to work it out is to get out and
record. No amount of rules will help much, they are just a start to the
learning, it's a feel you get after using the equipment and hearing the
results.
Please note I don't find places where I'd be recording with a top
reading of -50dB (some 40+dB of headroom). I record a great deal in what
most would consider very quiet sites in the quietest time of the day,
the middle of the night. Between the frogcalls it may be 10-20dB at
times. Though if my sound meter is to be trusted, usually it's actually
louder. And I've almost never found a place where I could not set to my
usual average sound level of -15dB. I do find the occasional place where
the mics are too noisy once that level is set, and if I record the mic
noise will intrude, and that with MKH mics.
Pohlmann, in his book does not seem to find this the kind of issue that
it's made out to be. Maybe he's like me, and sets record levels
correctly. He does make it clear there is no such thing as using only
some of the bits. You are using a bit even if it's 0. And such
discussion he has that I've found on this subject revolves around
designing and fine tuning A/D converters which he goes into in
considerable depth. The only graph he shows of the linearity of a A/D is
of one that's linear down to -100dB. From what he's saying, other things
than where you are on the bit scale are in play. He has long discussions
of all the effort that's going into accurately recording the sound right
down to the lowest increment, and a lot of that is getting the lowest
increment right.
Then you get into what's a good recording. For some of us that's what
will play nicely on our usual listening equipment. For others it's clear
it has to meet some test numbers, they hardly listen seems to me. And
all kinds of variations in between. I kind of pity studio types trying
to get nature to behave like their studio.
There is a big difference between a very weak recording being amplified,
and a recording that contains a range of sound levels and was recorded
with normal headroom. You won't be amplifying a recording made right
much, or you will clip. I won't get into the idea of compressing the
dynamic range as I think that's inappropriate for nature recording. If
your recording is weak you can amplify it some because you have headroom
to do so. It's well to think of the recording as a dynamic range. If you
are recording a typical site of, say, 50dB dynamic range, then that
get's placed somewhere in the 94dB available range. If you record it at
a top of -50dB, then it should be obvious you have lost the low end of
the dynamic range, and nothing will get that back. If, on the other hand
your highs are at -10dB then the lowest sounds are at -60dB, and
probably partially masked by the louder sounds, and you have all the
dynamic range of the site in the recording. You do want to always set
gain for as strong a recording as possible without clipping. Since
clipping is so awful, you err on the side of caution, as we have
discussed many times. So, your recording will have a unused upper margin.
I would say that if your top level on the meter at the peaks is
consistently below about -30 to -40, then you will probably notice some
compromise in the sound. And the weaker it gets the worse it will sound.
How much is degradation of intermediate levels and how much is simply
loss of dynamic range off the low end I don't worry too much about. I
maintain my record levels so that's not something I have to deal with
much. In actual practice I'm nearly always setting the gain on the basis
of not clipping off the top, clipping off the bottom is not my problem much.
Focus on getting good levels on the meter, and you will have done as
good as you can. Remember, no matter how bad it gets it will probably
beat the best we could do 30 years ago.
On the question of using bass rolloff vs headroom, throwing out part of
the sound to gain headroom is throwing out the baby with the bathwater.
Use bass rolloff with a great deal of caution that you are not tossing
the natural sound you want.
Walt
________________________________________________________________________
________________________________________________________________________
|