naturerecordists
[Top] [All Lists]

Re: Digital clipping; transfer to CD; increased gain

Subject: Re: Digital clipping; transfer to CD; increased gain
From: Syd Curtis <>
Date: Wed, 26 Jan 2005 22:22:56 +1000
Thank you, Dan Dugan and Walter Knapp for your expert advice on my
questions.  (Both responses on the 25th if anyone reading this wishes to
read them in full.)

Walter has pointed out that my digital data will be changed if there is any
resampling.  Fortunately, my Tascam DAT recorder offers 44.1 kHz as well as
48, and I stick to 44.1 and also work with 44.1 in Peak.  So I'm happy ther=
e
is no resampling.

Walter wrote of my sometimes very soft (below -24dB) field recordings:

> Is this a average, or the peak value? What you need to look at is the
> peak values. A site with a few loud peaks, the average may be quite low,
> just as it was at the site. A site with fairly even sound the difference
> between peak and average may be very small.

I was thinking of peak values: from a recording situation where I don't kno=
w
what is going to happen.  A specific example: I go to a favourite area of
subtropical rainforest, arriving just before daylight, to record the dawn
chorus.  I know what species to expect; I don't know where the individuals
will be with respect to my mic.  I may get a very loud note unexpectedly
close.  And if I don't, everything may be below -24dB.  So I'm pleased to
have your assurances, Dan and Walt, that increasing the gain is not changin=
g
the data except in that one regard:

>
> The data will be changed, but by a even amount. It will be changed in
> terms of it's mathematic value, but not in how it sounds (except for a
> volume shift).
>
Being technically inept, I tend to use only that which I know I have to use=
.
So I never even noticed 'clipgard'.  Certainly will from now on.  Thanks
Walter:

>
> In peak's gain adjust dialog there is a button for clipgard. If you
> click that it will calculate the margin between the loudest sample and
> 0dB. You can then set for whatever margin you want, but never more gain
> than is shown by clipguard or you will get clipping. I often choose -2dB
> for my adjusted file. By simply subtracting 2 from the value shown by
> clipguard.
>

Sure beats my (previous) primitive technique of picking a few loud passages
and trying various gains before applying to the whole.

Walter, you pointed out that "in general, digital recording has a lot of
dynamic range to play with.   Far more than the analog tape had. With analo=
g
tape you had to keep the signal at the top to keep from losing some of the
dynamic range off the bottom".

That is something that has given me great satisfaction.  None of my analogu=
e
recorders - Uher, National, Marantz, Sony WM -  could cope with the dynamic
range of an Albert's Lyrebird singing at a mic.  A level that would pick up
his softest notes overloaded on the loudest.  But no problem with the Tasca=
m
DAT.  Set it to -4dB for his territorial song and I can be sure of getting
everything with no clipping, though in fact, I try to make it -6 for a bit
of extra safety.

Glad to have your reassurances that relying on the DAT recorder meters is
sensible.  Loved your comment Dan.  Had some incidental relevances for me:

> Digital is good that way. It would be helpful to have a sorcerer's
> apprentice around to listen. I've been told Isaac Newton was color
> blind and used an assistant to describe the colors of the spectrum.
> And then there was Beethoven.
>
Lacking a sorcerer's apprentice, I tend to trade on the good nature (and
good hearing!) of our (Australian) AudioWings tape editor, Vicki Powys.
Anything real

I've often thought how sad it is, that music was so far ahead of science:
maybe today's science could have saved Beethoven's hearing, and who knows
what he might then have achieved, considering the glorious music he wrote
when totally deaf.

Thanks again, Walter and Dan.

Syd

(BTW, why was 44.1K chosen as a sampling rate?  My guess had been that it
approximated a power of 2.  But if my elementary arithmetic is correct, the
15th power is 32,768 and the 16th 63,536.)



________________________________________________________________________
________________________________________________________________________

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

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