evertveldhuis wrote:
>
> Hi,
>
> After reading most of the online postings from the very start to the
> latest I have now some questions.
>
> I'll put them in a example that relates to another thing I do
> which also involves the ingredients analog, digital, computer editing
> and storage: photography.
>
> When I take pictures with my digital camera I try to capture what I
> see; I rather just take the time to make a great visual composition
> (without altering the scene) and just point and shoot as opposed to
> taking a picture and later spending lots of time digitally editing it
> to `get it right'. I believe that if the composition is right
> then there is no need for large editing, maybe a little cropping
> (cutting away parts around the edges) or doing a slight brightness
> and contrast adjustment to make the colors somewhat more true to the
> real world.
> But I hate to spend lots of time behind the computer trying to get
> something out of it that isn't really in there in the first place.
I'm a old hand at photography and do much the same thing. Though I'm not
above washing the bits of stuff off the frog, etc.
Particularly since I went digital I do find I use some post processing.
Even if I take a hundred photos of a subject, the best one selected out
of that can probably be improved. A little unsharp mask, a little levels
adjust, a little contrast adjust, maybe saturation. And a much improved
photo results. I do these tweaks almost automatically now.
> The same goes for my approach (and most of you guys I think) of audio
> recording.
> I rather wait until the right circumstances are there to record a
> good nature recording and wish to edit it at my computer as little as
> possible (maybe some EQ, some volume level change but nothing all to
> drastic) I know that with today's computers and software it is
> very
> tempting to edit hours for a recording that actually last maybe 3
> minutes...
> But how good is that?
You should note that waiting for the right circumstances is much easier
for photography than for sound. For sound, in a lot of circumstances you
may be waiting forever. Or for something like Sept.11 to ground all
planes, a very rare event we hope. You can photograph a beautiful flower
right next to a rotting dead animal and no one will know. Sound is not
so isolated. And time dependent, a certain frog species may call at only
a few locations and only for a few nights a year...
> I learned that the folks on this list who do birding and frogging are
> trying to have a library of scientific storage, so post-editing is
> forbidden. But what about sound artist (like Aaron with his Memories
> from Annapurna recording)? How do you guys feel about editing? How
> much is acceptable, and what types of sound-editing is done?
As a rule, post editing other than minor gain adjusts or trimming for
length is not done in stuff destined for scientific documentation.
That's more so whoever else may want to listen to it can mangle it
themselves, and not because scientists don't use filtration. And that's
really what it is, documentation, not analysis. You would store a raw
file for this. Note that folks have over inflated imaginations about the
value to science of their recordings. Actual detailed bioacoustic
analysis generally takes custom calibrated equipment these days, and we
don't use that sort of stuff, way too expensive. The value to science of
our recordings is primarily as documentation of the existence of a
species at a location, or simply communicating to others what the call
sounds like. As such, the criteria is primarily that the calls be
identifiable. The actual more critical part is the field record,
documenting the exact location (use a GPS) and preferably describing the
habitat, abundance of the species, etc. There is a newer use of
recording habitats as a whole, generally called soundscapes, but the use
criteria is pretty much the same. Simple identification of what's there.
Now just because you keep a raw, which I do recommend, does not mean you
cannot process every recording for listening as much as you want. That's
a derived copy, not the master. You would keep it separate and
identified as processed if necessary.
How much do you do for listening? Generally either whatever it takes or
alternatively as much as you can get away with without messing the sound
up. It's much easier where you just want calls of a species as you can
then trim away everything of a different frequency, and do some light
application of noise filters to the call frequency. For the greenhouse
frogs I did recently, I applied two notch filters, boosted the frequency
of the calls, and cut the gain on other frequencies, applied a absolute
top and bottom cutfilter, passed the result twice lightly through
adaptive noise filters and twice through declick filters. It was all
done as a single pass as the FX Machine in Spark XL is designed for you
to use linked arrays of plugins and can run it all in realtime. I
adjusted it all while listening and watching the plugin I put last,
Spark Sonograph. When everything was adjusted I set it to make the
filtered copy to disk. I'm not completely happy with the result for just
listening, but it's intended audience is people wanting to ID calls.
Part of the filtered result in mp3 form is linked off this page:
http://wwknapp.home.mindspring.com/docs/greenhouse.frog.html
And there is a very short bit out of the two original tracks here:
http://frog_recordist.home.mindspring.com/naturerecordists/GFrog.mp3
I believe like you, in that you get the cleanest, best recording you can
get out in the field. The place to put your money is into mic setups and
learning how to choose and get the most out of them. Unfortunately this
very often does not mean the recording is in ready to listen condition.
Many things are simply not available in pristine habitats. Even habitats
you think are pristine will suddenly show new faults when you are into
recording sound. And we do resort to computer processing. It's not hours
and hours once you learn what to do. I could set up and filter that
greenhouse frog track from scratch in 10 - 15 minutes. Of course the
first time I tried doing this sort of thing it did take hours/days. Not
only was I dealing with individual filters, which I already knew well,
but I was dealing with filter interactions.
It's much harder if you want ambiance as you want all frequencies.
Often what's done there is mix more than one recording after
individually cleaning them as much as will work. Bernie, for instance,
talks about recording a near and far recording of surf and mixing them
to make proper sound of surf. This sort of thing is not done in
photography, or not often. For natural looking photos each one is one
take. But mixing is more common in sound. The call may be from one
recording, the ambiance from one or more others. Of course doing this
you have to understand the natural world, so you don't mix things that
should not be together, like hollywood uses pacific treefrogs for
habitats throughout the world.
Walt
________________________________________________________________________
________________________________________________________________________
Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/
|