From: Syd Curtis <>
> I have not followed this technical discussion, but wonder if I'd be
in order suggesting a possible qualification to one minor non-technical
bit of Walter's advice. (If this has already been covered, I apologise.)
Walter wrote:
>
>>> From: Walter Knapp <>
>>> To:
>
>
>>> I really think that copying all that large amount of recording into the
>>> computer is not the way to go. It's going to eat hard disks like mad,
>>> and will use time that it's not necessary to use. You are much better
>>> standardizing on a system where the initial sort is done by listening t=
o
>>> the recording on the original medium and only selectively transferring.
>>> Best if you have people who can do the selective transfers on the fly a=
s
>>> they listen. That is unless you can come up with computer software
>>> that's reliable enough at doing the original analysis. Even there it
>>> might do that without transferring.
>
>
> If one does have a generous HD capacity, it may be worth while copying al=
l
> of a large recording into a computer, providing what one looks for is eas=
ily
> recognisable on the computer screen.
>
> Two species of cicadas, and two species of birds presently sing
> intermittently in my garden. I was able to add an 80 GB HD to my Mac G3 =
for
> a surprisingly small cost. With PEAK LE software I can easily see on the
> spectrograph trace on the computer screen, when a cicada or bird was sing=
ing
> nearby. So ...
>
> I can set my Tascam to record a 125 minute DAT while I do other things;
> then set the G3 to copy the Tascam's digital signal, again doing other
> things while that is done. Peak allows the Spectrograph trace to be
> expanded or contracted at the click of a mouse, and to jump forward a scr=
een
> at a time again with the click of a mouse. I can quickly scan the whole
> recording in that way.
>
> I note the time at the beginning and end of each period where nothing is
> happening, and after I've thus gone through the whole recording, I can th=
en
> delete and replace with a second of silence, each period that I don't wan=
t.
>
> This leaves me with the useful parts of the original 125 minutes, and the
> facility, if I wish, to relate any part of it to the actual time when it
> happened.
>
> At that stage, assuming it's now down to less than 80 minutes, I can set =
the
> G3 to copy it to a CD if I don't want to work on it immediately, and when
> that is done, delete it from the HD and reuse the 125 min. DAT.
>
> I hope this is more help than nuisance.
This works well for recordings of the length you specify. But Graham is
working on how to deal with hundreds of hours of recordings. Any way
currently available to transfer the recording is going to take large
blocks of time. Each recording is 10 hours long. Peak can still handle
this, but it does slow it down a lot. I'm estimating that the best
current compromise is not to try to bring it in, but bring in clips.
Even if you were to bring it in unattended. If Graham can come up with a
workable fast transfer, then it might be different, hard to tell. It's
one of those things you have to just try several ways and see what you
stumble over. Graham is stumbling over the transfer times right now. As
well as some other issues.
Your advice is good, if, as you say you can see what you want in the
waveform and never listen to most of the recording. As long as they
still have to listen to it all, I believe my way would be faster.
I do a variation of your method. I use the transfer time, MD is just
like DAT in the time it takes, to listen critically to the recording,
usually for the first time since I recorded it in the field. My
recordings typically don't have long usesless sections. By the time it's
transferred I have come to some decisions about parts to cut out and
such like. And find those by scrolling the waveform or the time
counters. Then I do a additional listen through to see if I missed
anything and it's ready to archive. I don't do filtering and such like
before archiving, but on copies.
In my recording practice I don't normally make very long recordings, so
I bring in each recording and process it before doing the next. As these
accumulate they get copied to backup optical disks in the original aiff
format. And organized into audio CD's with a printfile for a booklet for
the case. That involves things like selecting a photo for the front,
copying info from the field cards. I often will have my database open at
the same time and enter the data from the field cards into that at the
same time.
I've currently got 7 - 72 gig Ultra160 SCSI hard disks and one 200 gig
ATA hard disk connected into my mac. So I have no room problems, and the
disks are fast enough for large files. I have 6 of the SCSI 72's in two
external cases, so I can power them down when not needed. Which is most
of the time. It's really nice to have lots of fast disk space.
Walt
________________________________________________________________________
________________________________________________________________________
|