Posted by: "Rob Danielson"
> "Controlled" monitoring is probably the weakest link in the recording
> chain after mic placement and mic/pre. The end-user controls
> monitoring and this shapes everything.
As far as the nature recordist is concerned this is out of their hands,
emphasis needs to be on what we can control. The weakest link is
undoubtedly the field environment itself. There is a lot to learn about
mic placement for recording this environment. One should spend lots of
time critically evaluating what's picked up by mics in the field,
learning how to use them.
> I used headphones, Sony MDR-V600's mostly (for web discussions like
> this). I have two other pair, sony 7502, and Senn 280's that I pop
> on at the end to see if there are some discrepancies. They're usually
> pretty much in agreement.
Serious processing work I do with Sony MDR-V900's. I then check the
result with everything from boom boxes to my magnaplan speaker system. I
don't have a properly acoustically designed studio where I could at all
trust speaker systems.
> I used to fear that the qualities I was discriminating were
> "monitor-specific." They are, of course, "monitor-centric," but with
> more and more fringe applicability as I learn use the tools. Its all
> we have to work with. The technique helps we understand what the
> mics' prejudices are at the very least.
I find that a lot of what you report is not reproducible on my
monitoring nor shows up in sonograms, where it should be easy to spot.
It's one thing to report what you find in your system, quite a different
thing to report it as if it's a universal characteristic of a particular
brand and model of mic. I've concluded that at least some of what you
report is a characteristic of your monitoring, your own hearing's
variables, or the processing you have already done to the file. As a
scientist It's a big strike against something if it's not reproducible.
It is not insignificant that your own attitudes and moods change what
you hear. Our hearing is strongly colored in this way. Just think what
happens to a new nature recordist when they start recording and suddenly
discover that the world is full of man made noise. Until the recording
brought it out their hearing was blanking all that out. Once you start
to think something occurs your brain will happily oblige and you will
"hear" what may or may not be there. This is one of the big reasons why
I always use sonograms and other such tools to show the sound in a way
that is independent of my hearing. I also verify what I hear by having
others listen without telling them what I'm trying to verify. I also
test my own hearing so I have some idea how it will respond at different
frequencies.
> Higher >1600Hz as a rule and the character varied with amplitude. I'd
> bring up the mid and there would be spots where everything would
> cancel-out and then kick back in. Very odd. I messed with all of the
> obvious variables, simplifying the chain and some peculiarity was
> always there. As all three plugs did this. I tend to think that
> encoding MS in the field and decoding in post is probably not the
> best of options. Rob D
One has to realize that any processing done on a soundfile has some
error. The ideal is to have absolute minimal processing on the raw
files. For M/S this means recording the separate mid and side channels
in the field directly from the mics. Your archival copy will then be
those channels. Always consider any processed file off that as a dead
end and return to the archived raw if you change what you are doing to
the file. Don't process what you put in the archive with any filters
etc. That way you will do the minimum to the sound.
As a parallel, do the minimum processing to any soundfile. Filters, EQ
and so on are mathematical constructs that do damage to the sound. Done
well the damage is minimal, but still there. This can be particularly
evident in things like the delicate sounds of quiet ambiance, or quiet
calls in such ambiance.
The ideal is that you record your final product directly in the field
with quality mics and don't post filter it. Obviously with M/S you do
have to decode it, but processing beyond that is questionable if
accurate reproduction is the goal.
As far as M/S you have the additional problem decoding in the field that
the decoders in recorders or pre's are either preset at one ratio
setting, or will cause you to not match the gain settings to the ideal
for optimum recording from each mic. Record each mic with optimal
settings in your recorder, even though that often will not be the ratio
you will use when you decode. In other words make the best independent
recordings of the mid mic and the side mic. Set your mic headroom
appropriately.
For this reason I've gone to not routinely monitoring decoded stereo
from my M/S in the field. I monitor the mid by feeding it to both sides
of the headphones in mono. The side is set by the metering and not
monitored. That way I can still check my aim by listening but am not
tempted to fiddle with the gain and throw the recording off. My M/S
recording has improved and become more consistent since I started doing
this.
Walt
|