John Hartog wrote:
>
> ...More a hiss from the left that seamed to encroach, I am guessing
>it was aspen leaves in the wind. The localization seamed missing,
>just left channel but no sense of distance. I am not trying to knock
>Kevin's recording here: I think the recording as nothing less than
>amazing.
Yes, it sounds as if there's a closer aspen on the left when the wind
picks-up. There HF is quite evenly represented across the center of
the field. This distinction cropping up is a good indication of it.
> > Front-rear confusion typically becomes a concern when one wishes to
>> locate sound sources accurately opposed to capturing as many events
>> as clearly as possible. For example SASS produces (a) significantly
>> lower amplitude when the sound arrives at angles greater than 90
>> degrees off axis. (b) tonality that is noticeably warmer and (c)
>> depth cues (such as reflections from surfaces in front of the array).
>
>It is also worth considering when trying to capture a perspective
>that just sounds natural. Even the SASS with MKH 20's will produce
>an unnatural bounce back effect with direct flyovers.
I'll listen for something like this. Ground reflections would be
normal. I've noticed that I can hear another order or sometimes two
of reflections with my SASS mod that I cannot hear with my Jecklin
Disk when they're in the same quiet location. I sometimes notice
reflections with the SASS that I didn't notice in the flesh.
>Think a billiard ball bouncing of the rail - not very natural for a
>mallard duck. Lang Elliot pointed that one out to me several years
>back a a NSS workshop.
>
>> M-S arrays with a cardiod mid mic produce some of the (a) and (b)
>> effects. X-Y and ORTF with cardioid mics do as well.
>> Front-back distinction with Perpendicular to Boundary arrays is a
>> challenge and therefore these arrays are good when wants to pick-up
>> sounds from the rear with almost equal clarity. Blumlein is very good
>> for this too. Jecklin Disk and Spaced Omnis produce less of the (a)
> > and (b) effects than M-S and X-Y in my opinion.
>
>Maybe off axis attenuation and/or coloration have a place in
>reducing front-back confusion.
If we're modeling human listening, there is direct analogy. Human
listening has a forward prejudice. We turn to listen to important
sounds even if we know we cannot see the source.
Trying to make convincing surround installations helped me realize
that the first industry array designs to "listen out in all
directions" create imagery that is too bright in surround and feels
quite unnatural and disorienting to the listener, especially a seated
one. Ambisonics arrays like the ST-350 system Martyn is more
challenged to make front-rear distinction. Treating sound from all
directions equally is a neat hyper-fidelity "effect," but I feel that
being able to feel that a creature is moving behind the rig is
equally as informative and more experientially compelling. It kind of
makes the rig more animal-like. I've become so interested in front
rear imaging that often I set up with my "rear" stereo array spread
to another sweet spot and directed forward-- the same direction as my
front array.
There are many, many ways to affect the overall tonal balance and
phantom imaging in post with spaced arrays, this is not unique to
Ambisonics. The same can be done to improve off-axis and center field
coloration to your tastes with a stereo pair. I'd play with Kevin's
recording but the lowest octaves were rolled off.
>I certainly have played with that with NT1-A's, though I'm not sure
>I made much progress. More practical perhaps is careful scrutiny on
>location for potential off axis problems.
The extra directionality of the larger diaphragm NT1-A's is something
you picked-up on very quickly and was instrumental in the decision to
make a more uniform HF spread in front with a narrow 60 degree angle.
That ORTF-ish array or with forward-facing NT1-A's between a barrier
has pretty good front back imagining. I used it as a "pointing out"
rear pair in my surround rig for years. Its short-coming in terms
of rear imaging is probably associated more with a significant dip
in lower midrange response with NT1-A's. Those frequencies are
important for location reflections.
Its about becoming aware than an image of the "rear" is also created
when micing. We are so front-prejudiced that we continue to mic that
way when we understand better. Rob D.
>John Hartog
--
|