[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [ARSCLIST] 2GB limit for audio file formats
At 08:11 PM 6/24/2005, Eric Jacobs wrote:
As a rule of thumb, I do find that higher sampling rates of 88.2 kHz or
more make a considerable difference when it comes to removing impulse
noise on mechanical carriers (records with pops, clicks, and crackle).
That makes absolutely perfect sense as there is information from
these artifacts out there. They weren't put on the disc by the
recording process, they were "recorded" later and may have very sharp
rise times. Here is a case where more bandwidth is better. The
original discussion was tape.
Unless you're planning on digitizing the bias, I don't see the need
for recording at a bandwidth that the reproducer isn't reproducing.
As I said, I am seeing stuff in the 30kHz region on 15 in/s tapes
(bells and cymbals on band organs was one test, also cymbals in a
symphony orchestra). I got into a long discussion on the Ampex list
about this and ended up doing measurements. It seems that most mics
fall off rapidly (12dB per octave might be a good approximation)
above 20kHz, with the exception of some measurement mics.
However, do higher sampling rates help remove broadband noise on
magnetic media?
IMHO, the broader bandwidth merely adds more noise to remove. I'm
generally running the Algorithmix NoiseFree Pro plugin ahead of other
equalizers because then I don't have to guess what needs to be
removed pre-NoiseFree (i.e. what NoiseFree won't "get", but with less
powerful noise reducers, I have limited and filtered (using both DC6
and Sound Laundry).
Doing the EQ after Noise Free lets me do it once - I am working on a
cassette made in the mid 1970s that was recorded with the mic built
into the cassette machine. Oh joy!
Once any processing has been done, I will continue to work in 32-bit
(or 64-bit in Cube-Tec) and save intermediate files with the
longer wordlengths, and as the very last step dither down to 24-bit.
I try and do all the processing in one pass in the virtual working
mode in Samplitude. The program works in 32 bit, but we have the
original input files and the output files. No intermediate files if I
can avoid it (I hate re-doing intermediate files when the first step
doesn't quite sound as good as I thought it did.
Richard L. Hess email: richard@xxxxxxxxxxxxxxx
Vignettes
Media web: http://www.richardhess.com/tape/
Aurora, Ontario, Canada (905) 713 6733 1-877-TAPE-FIX
Detailed contact information: http://www.richardhess.com/tape/contact.htm