At 7:49 PM -0400 7/20/05, Walter Knapp wrote:
>From: Rob Danielson <>
>
>> Sorry, I'm having diificulty picking out the sound quality to fit
>> your description in the first 7 minutes. Maybe if we both refer to
>> one short, uncompressed clip? Here's a 16/48K aif with two excerpts.
>>
>> (1) section with lowest presence I could find (from section
>> represented by the 7 minute, mp3 clip you have).
>>
>> (2) section from 20 minutes in where the right mic is breaking up.
>> The sound I'm referring to is pretty apparent, but there's prominence
>> ~ 400Hz.
>>
>> http://www.uwm.edu/~type/Temporary/7723EQ2_Pres3m_BreakUp20m.aif
>> (1.4mb)
>
>Here's the Spark XL sonograms of the two mics:
>http://gruagach.home.mindspring.com/Left.pdf
>http://gruagach.home.mindspring.com/Right.pdf
>
>The dividing line between the two clips is visible so you can compare.
>
>I believe the sharp spikes may be what you are calling the mic breaking
>up. And they could be. But were not what I'm talking about.
>
>I'm looking at all those bands of sound that make up the background of
>this recording. Note there are bands from 20hz up to 12khz or so. The
>bands are solid and unvarying, or not varying much. Those below 300hz
>are the background rumble that I assumed was a freeway, some distant
>large engine or such like. But the rest don't sound like any sound the
>site may have had. They are not insect calls, for instance. My
>description of crackly applies to them. I'm curious as to the source of
>this.
>
>You have a recorder that's not supposed to do things like this. And
>describe the Rode's as being so clear. So, where's the hash coming from?
>When I first played the recording, I, of course, listened for frogs.
>Having the distant frogs masked so badly became evident quickly.
>
>Note after the 3.8 sec mark, the sonogram of the "pickerel" call in the
>Left mic.
>
>Note also, when dissecting a recording, I use my Sony MDR-V900
>headphones. They are closed phones so keep down the distractions of the
>room ambiance. And I find their sound reproduction highly accurate. I
>can hear the sound in question in speakers as well, but details like
>this are much easier to hear and evaluate with headphones.
>
>Walt
>
>
>
Thanks for taking time to make the pictures and pose good questions.
Its fascinating to consider sources for the banding and other things
going on. Neither of us has time to really get to the bottom of it,
but here are some of my thoughts you can read over when you get rhe
chance.
(1) The approximately 9 bands 5K and band ~12k. Seems like these
would have to stem from a combination of mic self noise, high
frequencies from the environment and possibly dithering noise. To my
ear, the "zizz" I'm hearing above 6K is roughly a 50/50 blend of mic
noise blending with very subtle sounds. Horizontal sonogram banding
in this range around this much water under a canopy of leaves is
expected. Are the mics representing these high frequencies as well
as we can imagine? Clearly not, but we're asking a lot out of the
recording system (mic, pre, medium etc). Most of the of the total
amount of sound energy is concentrated below 160Hz, so most of the
resulting ~1% file saturation of the original 24/48K recording
(background ambience only) results from energy from 160 hz and
below. Its hard to interpret with one color acrossa good part of
sonogram range, but it looks like the bands above 5 K could be around
-30dB to -40dB relative to this those at 50Hz. If the ambient
background sound level is around 25 dB (a workable estimate based on
a test with a good meter once and the resulting saturations) and the
difference between the low energy and that above 3k is only 20dB,..
the environmental sounds or noise, take your pick, are very close to
the self noise rating of the mic.
(2) The ~7 horizontal bands between 83 and 500Hz. A harmonic tone
can heard it whenever its really quiet in both mics. In a short
snippet, it does sounds a lot like a very distant highway-- like
truck tires. I checked places when a truck comes through valley and
there's no point that really sounds the same and when the sound
character is most like this drone, the file saturation is much, much
greater. Take a look at the "checkered" pockets or amplitude clusters
in the bands between 80-500.They're thicker (longer) at the lowest
frequencies and they get continuously shorter each band up where they
blend into the picture resolution on my screen around 600Hz. This
pace of the pulsing is quite structured and if its noise from mic
moisture or even distortion noise, its very unlikely to be so similar
in both mics. The orange spikes at the bottom have much less of a
pattern- more like one would associate with noise and the lower
spikes correspond with higher spikes too. Accounting for these clues
and the way it sounds, I take these bands between 83 and 500hz as
mostly a blend of distant low hz energy being reinforced by local
acoustics. The other part might be overtones produced by way these
particular mics "bounce" in response to very low level, low frequency
sounds. All mics have a "bounce" like this in my experience.
What low Hz energies? There's a visitors center with HVAC ~ 1/5 mile
away, a truck garage/farm behind the dam about 1/5 of mile, a new
dairy/cheese factory on the hill a half mile away and close edge of
LaFarge (pop 700) is 1.5 miles away. I'm not discerning any one
source, but compressors at the dairy/cheese factory is my best guess
for the harmonic structure. Can you see a 60HzX component? Your
sonogram app spreads out the range as well as mine can. I know its
not a truck idling anywhere near because I have heard hours of that
on other recordings.
The local acoustics, primarily, would be created by the 12-15' high
limestone river bank on the close side and 4-9' mostly bare clay bank
on the far side, a stand of small trees along the far bank, a stand
of pine trees behind the mics and a dense deciduous canopy over the
river.
No mics are without significant artifacts that can be heard in
headphones with high amounts of amplification, but I don't think the
source of the banding stems only from self noise or moisture induced
problems with the mics. They're certainly not without wrinkles, but
the "reach" of the Nt1A's, especially after I've done some eq to tone
down the local resonance and the sound source is on axis with higher
Hz content, can be shocking. I picked this night to work on because
the action was thinner and it was very quiet. The mics are high
ground too. I would not be surprised if the pickerel (we hope) was
floating in the open water in the wetland 80 to 100 yards away. This
qualifies as "distant" to me. Rob D.
________________________________________________________________________
________________________________________________________________________
|