At 1:51 PM -0400 9/2/06, Walter Knapp wrote:
>Posted by: "Rob Danielson"
>
>> The cricket frog calls are similar to "jingling keys"-- a sound
>> often used for testing sibilance and high frequency transient
>> response. As one of the more telling tests of high end performance,
>> its impressive to see the 3032's hold their own.
>
>That may be the impression from listening, but the calls are quite
>different from the precussive harmonics from bits of metal. They are
>made with vocal cords vibrating in air. I've not made a study of cricket
>frog calls, but roughly the basic components are small bursts of a set
>frequency about 1-2 thousandths of a second in duration. No after
>vibration as you would have with the metal.
>
>As a biologist and having the advantage of hearing the actual calls with
>my own ears, the MKH-20 rendition is a little more true to life than the
>3032's. The 3032's render the calls a little too grainy. Still not bad
>and easily recognizable.
>
>We have two species of cricket frogs here, and it can be tricky to tell
>them apart at times. So I'm used to dissecting the calls to determine
>the species.
>
>> Interesting theory. Perhaps the reflected and direct waves being in
>> phase provides a similar advantage to both mics and lessens
>> differences.
>
>I actually think the 3032 is probably at a slight disadvantage on the
>reflected and direct wave part of a boundary mic due to it's hard,
>slotted diaphragm cover vs the screen cover of the MKH-20. I don't think
>the acoustic coupling of the 3032 to the surrounding boundary is quite
>as good. The diaphragm of the 3032 is aligned to the ends of the slots
>but slightly lower so the coupling has to be over this slight lip
>through slots. The MKH-20 diaphragm is more exposed and even with the
>top of the housing so aligns better through it's fairly open screen.
>
>Note that the boundary of the SASS considerably modifies the polar
>patterns of the mics. From the all around omni to more or less a
>hemisphere. This is going to modify the frequency response curves of
>each mic some.
>
>> Your comparisons are likely to encourage more experimentation with
>> 3032's, but probably not with the advantages of the mkh-20 reference.
>> Eric measured 8dB(A) self noise on the 3032's so there's a decent
>> chance the similarities could persist in quiet locations.
>
>As I noted I did not have the opportunity to try the mics in a really
>quiet site so cannot really do more than speculate. I would have loved
>to have them set up for comparison on the Florida site at 3 AM. That was
>a quiet site.
>
>I'm willing to believe the mics are lower self noise than the 16dBA spec
>from the two examples I tried. But for every example to be 8dBA I think
>will turn out to be a big stretch. The two I had did not sound that
>quiet by ear. Maybe as quiet as a 30 or 40, but I won't swear to that even=
.
Eric has measured it 8dBA. I've used mine beside my mkh 40 and it had lowe=
r
>
>I do think they are worth more experiment for nature recording. Even if
>the self noise was 16dBA that's a lot lower than mics folks keep talking
>about. And they do have pretty good sound.
>
>> Your rig's ability to get completely above the trees in some
>> instances is ingenious. Thanks also for the breakdown on the SASS
>> construction on your website! Rob D.
>
>I would not say it get's above the trees too often. But the brush is
>generally lower and so are lots of ground reflecting/absorbing surfaces.
> From measurements I made many years ago ground clutter modifies sound
>levels considerably up to 6 or more feet in natural (forest)
>surroundings. We were doing a highway traffic noise study along the
>North Cascades Highway in Washington State, a preliminary to a new
>campground which was in forest off the highway. We normally measured
>with the meter at 5' and needed to know the falloff to calibrate our
>noise model. As part of that we did measurements in the forest from
>actual ground level up to 10'. The height data was never written up and
>published and I don't have it anymore.
>
>Note like so much of what I do this is not a original idea with me. Klas
>was the first I remember in the group who talked about using a high mic.
>Though he was talking about using a high parabolic pointed downward to
>minimize unwanted noise behind a wanted caller. The greater clarity
>became obvious when I started trying it.
>
>Walt
>
Hi Walt--
When Eric measures and prints "8dB(A)" self noise, I tend believe
him. He has an pseudo-anechoic chamber and quality test gear. What
character of noise The AT3032 produces at that level, we'll know more
and more about over time.
I do recall Klas first mentioning the height advantages back in 2001.
Not long afterwards, though, you built your first 16' rig.
I re-read Crown's SASS material and looked at Bruce Bartlett's
comparisons again
http://www.tape.com/Bartlett_Articles/stereo_microphone_techniques.html
in which he attributes the SASS ("Baffled OmniPairs") these features:
>*It uses two omnidirectional mics a few inches apart, separated by a baffl=
e.
>
>*Level, time, and spectral differences between channels produce the
>stereo effect.
>
>*Images are sharp.
>
>*Stereo spread tends to be accurate.
>
>*Low-frequency response is excellent.
"Level, time, and spectral differences" are especially impressive.
The SASS discriminates timing difference below 700Hz (much like our
ears) and level differences in the higher frequencies. Interestingly,
the SASS, the Jeckiln variations people are experimenting with and
Curt Olson's "head-spaced parallel barrier array"
http://www.trackseventeen.com/images/mic_arrays/3032_parallel.html
all seem to fall within Bartlett's "Baffled Omni Pairs" category
with these benefits. Rob D.
|