naturerecordists
[Top] [All Lists]

Re: Soundscape Recordings

Subject: Re: Soundscape Recordings
From: Lang Elliott <>
Date: Thu, 07 Feb 2002 08:20:09 -0500
Vicki:

I have a lot of great recordings made with two cardioids using the near
coincident technique. The nice thing about "near-coincident" is that the
spacing of the mikes (around 7") causes the setup to capture pretty decent
interaural time and level differences that add somewhat to spaciousness.
Using two cardioids in a coincident arrangement has only one advatange that
I can see, and that is mono-compatibility, allowing you to mix both channels
to monaural without any phase cancellation.

Using near-coincident or coincident cardioid setups does result in
coloration of sounds from the sides, you but you shouldn't worry too much
about this because the resulting recordings will still sound fine for nearly
all applications.

If you decide to try a SASS setup, which will be less sensitive to wind in
comparison with your cardioids, please keep abreast of conclusions of our
group concerning microphones. To date, nobody has tried the ME62s in a SASS,
but I see no reason why they shouldn't work. What we need is some
experimentation to come up with the best midrange-priced mike that is
suitable for the SASS, before too many folks waste money on the wrong mikes.
Another possibility, of course, is Klas's EM-23.

Lang

> All,
> 
> I have learned a great deal this week about stereo recording, thanks to all
> your postings on this topic.  Your generosity in sharing such information is
> very much appreciated.  I now understand for the first time how my
> coincident pair of cardoids will have an uneven frequency response to the
> sides.  And that a SASS set up would not.  I do love the effect of binaural
> and as all my listening is through headphones I will certainly experiment
> with binaural, eventually.
> 
> I realize now I must have omni-directional mics for binaural to work.  So
> I'll start saving up for two Sennheiser ME62 mic capsules which in Australia
> are currently $ Aust 320 each (I already have the power modules).  Lang's
> favourite Sennheiser MKH 20 omni mics would cost me $ Aust 2,540 each, (plus
> power unit) so they are totally out of the question!  Meanwhile my cardiod
> pair of ME64s, home-mounted on a monopod as a coincident pair with
> adjustable angle, are easily carried and quick to set up in the field.
> Knowing of that dip in various frequencies to the side will help me to
> better consider the angle I set them at, in any given environment.
> 
> Thanks again,
> 
> Vicki Powys
> Australia
> 
> 
>> on 7/2/02 1:39 AM, Lang Elliott at  wrote:
> 
>> Walt:
>> 
>> "Uncolored" simply means that the mike itself is not coloring the sound and
>> that it's frequency response to sounds within a particular spread is pretty
>> flat. In other words, it faithfully renders the sound that arrives at the
>> mike. The SASS omni-mike setup does very well all the way out to 180
>> degrees. Beyond that sounds become muffled because of frequency variable
>> attenuation toward the rear. But at least it provides rather uncolored
>> response to signals originating out to the sides.
>> 
>> Compare this to a coincident or near-coincident stereo setup using cardioid
>> mikes, where attenuation varies considerably with frequency as sounds move
>> more and more off-axis. This is due to the uneven falloff of different
>> frequency bands typical of directional mikes. As a result, a high pitched
>> sound off to the side will be considerably attenuated, while a low pitched
>> sound from exactly the same location will have little attenuation. Thus, a
>> broadband sound coming from the side will be extremely "colored" by the
>> microphone setup itself, due to drastic attenuation of the highs.
>> 
>> Lang
> 
> 
> 
>> Re: Recording Natural Soundscapes and the 360-degree issue
>> 
>> Good comments Bernie and you're absolutely right about "360". And Walt is
>> correct in saying that the ultimate 360 is actually 3D, with sounds coming
>> from all directions, including overhead.
>> 
>> That said, it is good to keep the following in mind:
>> 
>> 1. If, as a listener, you desire more than the 60 degree spread of
>> conventional stereo sound, then you'll have to adopt some other playback
>> techique, be it headphones or multiple speakers, or other positions for your
>> two main speakers.
>> 
>> 2. All stereo microphone setups hear in a full 360 degrees, it's just that
>> highly directional setups greatly attenuate sounds to the rear. But even
>> with very directional setups, loud sounds from behind are recorded and will
>> be heard upon playback.
>> 
>> 3. Whatever soundstage the microphone setup records will be collapsed into
>> the playback soundstage. THIS IS A VERY IMPORTANT POINT, and here's a
>> example of what I'm talking about:
>> 
>> Let's say you record a soundscape using a mike setup such as the Schoeps
>> KFM6 Sphere, which is 2-channel binaural and which records a full 360
>> degrees, without any attenuation of the rear. Now, let's say that you play
>> such a recording back using a conventional 60-degree arc stereo speaker
>> setup. This results in the 360 degree recording being "scrunched" into a 60
>> degree frontal soundscape. Another way of saying this is that the 360 degree
>> natural soundstage is "compressed 6:1" into the smaller soundfield. You
>> should also realize that all the signals coming from the rear of the Schoeps
>> Sphere are simply being "superimposed" on the frontal signals. In other
>> words, sounds that happened to the rear will sound as if they're coming from
>> the front when played back using the conventional speaker setup.
>> 
>> For instance, when you're recording in the field you might notice that a
>> bird is singing loudly from behind you and to the left. Using almost any
>> stereo mike faced forward and then playing your recording back over a
>> conventional 60-degree speaker setup, the bird will sound as if it's coming
>> from just inside the left speaker, which would be perhaps 25 degrees left of
>> center, not at all cloe to where the bird was located in the wild!!
>> 
>> It is always good to think this way, even when using a stereo setup that is
>> directional. In such cases, loud muffled sounds from the rear are recorded
>> and they will appear in the frontal soundstage upon playback. Furthermore,
>> they may sound "weird" because of their muffled quality.
>> 
>> 4. When listening to true binaural recordings over headphones, one can
>> sometimes get fair imaging of sounds to the rear, but more often than not
>> the rear sounds appear to come from the front, similar to what happens with
>> speakers but within a broad 180 degree soundstage.
>> 
>> 5. Currently, it is not feasible to get true 360 degree 3D sound using only
>> two speakers. There are some very advanced recording and playback systems
>> that approach this, even using two channels, but most require multichannel
>> recorders and multi-speaker playback systems (eg. the Soundfield mike and
>> Ambisonics playback system). Conventional 5.1 Surround Sound DOES NOT
>> produce 360 degree 3D sound and it is not a good choice for nature
>> recordists for the reasons mentioned in other e-mails.
>> 
>> 6. Lang's current microphone choice for soundscape recording is his modified
>> SASS, which receives an "uncolored" 180+ degree frontal soundstage, with
>> moderate rejection of sounds to the rear.
>> 
>> 7. Lang's current speaker playback system of choice (for his SASS
>> recordings) is to place two speakers almost straight out to the sides to
>> produce a 180+ degree playback soundstage. Remember that the SASS also
>> records sounds coming from the rear, even though they are attenuated. When
>> using the 180 degree speaker setup just described, the rear sounds are
>> projected or superimposed upon the frontal sounds, creating a 2:1
>> compression, which is much less compression than the 6:1 produced using a
>> conventional stereo speaker playback setup.
>> 
>> Does all this make sense? Really, it's an attempt to clarify how one's
>> microphone hears the complete 360 degree natural soundscape (including what
>> is being emphasized and what is being attenuated), and then how one's chosen
>> playback system renders what the mike hears into an indoor virtual
>> soundstage. 
>> 
>> As nature soundscape recordists, it seems that we should gravitate toward
>> recording and playback techniques that give us the most natural and
>> realistic illusion when we listen to our recordings indoors, with a full 360
>> degree 3D effect being our ultimate goal.
>> 
>> 
>> 
>> lng
>> 
>> 
>>> Wild Sanctuary wrote:
>>> 
>>>> It should be pointed out that none of the stereo systems offers a true 360
>>>> degree listening experience by itself. To me, a realistic 360 degree
>>>> experience means that you can place the listener within a non-headphone
>>>> tethered sound field and anywhere he/she turns his/her head, they will hear
>>>> the sound accurately portrayed from that perspective. 360 degree sound is
>>>> listener-oriented and determined just as sound arrives at our ears in the
>>>> real world.  That can be currently accomplished by discrete 4-channel
>>>> recordings or (as we have recently tested) two M-S systems placed
>>>> butt-to-butt, and perhaps some other method(s) not yet well-known or
>>>> explored like, perhaps, a tetrahedral array (giving the added advantage of
>>>> height as well as depth of field on the horizontal plane).
>>> 
>>> I'm even more picky as I visualize a 360 setup as being actually a 3d
>>> volume. And I'd also prefer that more than one can enjoy it at the same
>>> time. That's going to be harder to do. For now I'm keeping my goals much
>>> simpler, good stereo. That may wander into binaural setups or M-S, but I
>>> am working on the stereo field, not surround.
>>> 
>>>> It's all a matter of taste. Test a whole bunch out and go with what pleases
>>>> YOUR ear. I recommend using systems that are durable, easy-to-use, provide
>>>> a good to excellent signal-to-noise ratio, and a reasonable to great
>>>> illusion of space. Except for the specially configured SASS system
>>>> described by Lang (and which we WILL explore when we get the bucks to do
>>>> it)
>>> 
>>> That's the big problem with Lang's system. I figure I'm going to go two
>>> steps on it, start with cheaper capsules to see if I like it at all,
>>> then try for the expensive stuff. That way I can also explore some
>>> technical issues that I'd not try using MKH20's. Or for which the
>>> MKH20's are unsuited. Working with a built up mic is a disadvantage in
>>> some ways.
>>> 
>>> The issue of practical to use is a important one for me. Does not matter
>>> how wonderful the mic, if it's hard to transport or use, or can't stand
>>> the outdoor environment, it simply won't be used. Whatever I have must
>>> fit easily in a car with lots of other stuff, preferably in a ready to
>>> go state. Due to the sort of locations I often have it must also be
>>> something that can be hand held. Those criteria weed out some types.
>>> 
>>> , we use all types and all configurations depending on the application
>>>> and what we wish the end result to be. Call me an end-resulter, if you
>>>> wish. (I won't be insulted 'cause I'm just wired that way.) I go for what
>>>> the illusion will sound like on a CD or in a public space installation so I
>>>> don't necessarily monitor for the pleasure of the moment in the field. I'm
>>>> happy enough to be there in the first place. Rather I monitor just to be
>>>> sure that the M-S recording on tape or disc is as fine as it can be knowing
>>>> that in the studio it will be close to what I imagined it should be, then
>>>> detach the headphones. The M-S system we use is more difficult than other
>>>> systems because it is not intuitive. What you hear using a simple pre-amp
>>>> in the loop is clearly not a representation of what can be got in the
>>>> studio or lab when doing the final mix. So I usually recommend simpler
>>>> stereo systems to begin with.
>>> 
>>> As I often use the stereo field in the headphones to align the mics,
>>> especially when hand holding, I've been looking at the business of how
>>> to handle the M-S system for that. Aiming by sight in the dark is not
>>> that precise. So, I'm wondering about a split system that feeds stereo
>>> to the headphones, and the regular signals to the recorder for later
>>> processing. I don't care if the stereo field I get is the best listening
>>> one, just want to aim with it. Do you have any suggestions in this
>>> regard? Hopefully something reasonably priced.
>>> 
>>> 
>>>> Lesson the long way 'round: use different mics and find one or several you
>>>> like.
>>> 
>>> I get the feeling I'm going to find I like several. But that each one
>>> will do best at different things.  So, just like I have a camera with
>>> several choices as to lenses, I am working in the direction of a
>>> recorder with several choices of microphones. And knowing those mics as
>>> well as I know camera lenses.
>>> 
>>> Each microphone (or microphone combination) is going to take some time
>>> to really get used to what it does. So, while I have initial
>>> impressions, those are subject to change.
>>> 
>>> Worst part about experimenting is how much the mics cost. I just don't
>>> think it would be effective to rent them for a short period to evaluate
>>> them. Hopefully, by using things like Ebay to keep the costs down I
>>> won't go broke doing it.
>>> 
>>> Walt
>>> 
>>> 
> 
> 
> 
> To unsubscribe from this group, send an email to:
> 
> 
> 
> 
> Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/
> 
> 
> 



________________________________________________________________________
________________________________________________________________________

<Prev in Thread] Current Thread [Next in Thread>
Admin

The University of NSW School of Computer and Engineering takes no responsibility for the contents of this archive. It is purely a compilation of material sent by many people to the naturerecordists mailing list. It has not been checked for accuracy nor its content verified in any way. If you wish to get material removed from the archive or have other queries about the archive e-mail Andrew Taylor at this address: andrewt@cse.unsw.EDU.AU