Is stereo necessary? Rode NT4 for documentary, interviews...

[Dave Largent]

Originally Posted by Jon Fairhurst

This technique should give a clean dialog track, a natural, wide stereo ambience without the hollowness of a shotgun
http://PoorlyProjectedPictures.com

Not sure I understand. You plan on using two separate mic's?
Three separate mic's ("shotgun plus M-S pair")?
It seems to me you'd still have that
hollowness if you are using a shotgun, either if it's an
M-S mic or not.

[Jon Fairhurst]

The M-S would be a standard condenser mic. Not a shotgun.

The reason a shotgun sounds hollow in a small room is because the ambience loses its highs. Only the direct signal has a flat(ish) frequency curve. By using noise reduction, you get rid of a lot of the dull ambient noise from the shotgun. By mixing in some flat ambience from the M-S mic you get back to a more normal sounding environment.

* Clean dialog in the center
* Flat, controllable ambience at the sides
* The ability to mix down to mono without phase problems

Does that make sense?

Something like this would work...
http://www.studioprojectsusa.com/lsd2.html

Though, I'll probably go for two cheaper large condensers with switchable patterns and a couple of stands. The large condensers can do double duty for studio voiceovers.

You certainly won't get a hollow ambience from large condenser mics. My only real concern relates to phase issues when mixing this in with the processed shotgun.

Maybe the best solution - especially if you only have a stereo recorder - is to record "silence" with the m-s pair before/after the shoot and use it as a pad underneath the noise reduced mono shotgun track.

-Jon Fairhurst
http://PoorlyProjected Pictures.com

[Dave Largent]

[QUOTE=Jon Fairhurst]The reason a shotgun sounds hollow in a small room is because the ambience loses its highs.http://PoorlyProjected Pictures.com[/QUOTE

You sure about that? That's not how I'd describe
the hallowness, such as you say "the ambiance
loses its highs". I could be wrong but it sounds
to me more like a reverb effect such as "inside
a tank".

[Douglas Spotted Eagle]

Since shotguns become omni's in the low frequencies, the low frequencies are enhanced, and maybe this is being perceived as a lack in high freq's vs the actual bump in the low mid and low?

[Ray Ambrosi]

Originally Posted by Gian Pablo Villamil

Whatever you get, you should check out this free software: http://www.voxengo.com/product/msed/

It is an M-S to L-R encoder/decoder in a VST plugin. So you can run it in Vegas Video, for example, and any other editing software that accepts VST.

It decodes M-S to L-R, allowing you to control spatialization, and encodes L-R to M-S (the other way round). It also has an "inline" mode, which simulates putting two instances of the plugin in sequence, hence allowing you to adjust mid-side values even if your source is L-R stereo. It is very cool!

Thanks Gian! I'm pretty new to all this so I have had to do a lot of reading where I can and will need a crash course in audio recording before I head off to do my project.

That software sounds interesting, but I'm not sure exactly what it does, or what advantages it offers. Could I use that software with the Sony ECM-MS957? Can anyone tell me if the specs on this mic are good or not? Any practical experience? Should I make a new posting? I really do want to find out if the Sony is worth buying and whether it can provide the type of M-S recording that can be manipulated to bring out the sounds and degree of stereo separation that I want to emphasize (this is the advantage of M-S as I understand it). Please advise!!

Originally Posted by Steve House

Just a thought - the NT-4 is essentially two NT-5 cardioid heads mounted together. If you were to get a stereo matched pair of NT-5's you'd get two mics for a total cost that is actually a few dollars less than the NT-4. A stereo bar would let you put them on a stand for X-Y recording just like the NT-4 but with the additional flexibility of other mic placements like ORTF, A-B, etc, or to use them singly for interviews, instrument mics, etc. If one of them craps out on you on location, you still have the other so you can at least record in mono while if all you have is the NT-4 and it fails, you're toast. When traveling light the more different things a single piece of gear is capable of doing the better off you are. Two mics that can be used either together or separately in a variety of jobs is a better bang for the buck spent than 1 mic that is fairly specialized. Of course, the NT-5s don't have a battery and required phantom so that might be an issue but if you have the Beachtek along it's not a big deal.

Thanks for that suggestion Steve. It sounds like a good idea. I think I'd use two Rode NT3 mics because they can run on an internal 9v. (yes I'm fairly obsessed with the mikes having an internal battery). The only problem with putting them on a stereo bar is that I'd have to set it up each time I want to use it. Is it hard to get the mic placement correct? Would the NT5 mics sound much better than 2 NT3 mics? I had planned to use either my Rode NTG-2 shotgun on my video camera to focus on the person speaking and the stereo mic on the MD recorder to capture the ambient sound. Or is should I use another arrangement? So I was thinking I could get away with only carrying 2 mics to save weight and bulk.

[Dave Largent]

Originally Posted by Douglas Spotted Eagle

Since shotguns become omni's in the low frequencies

Is this true of the cards and hypers as well?

[Steve House]

Originally Posted by Dave Largent

Is this true of the cards and hypers as well?

No. While pressure gradient microphones like cardioid and hypers do exhibit a "proximity effect" where bass is boosted at close working distances, their directivity is not frequency dependent. They develop their directivity on the pressure differences caused by sound waves arriving at both the front and back of the diaphram. Shotguns use a line-gradient principal with a pressure transducer (an omni capsule) located at the bottom of a tuned interferenece tube. Sound waves enter the tube through the front and at the various ports on the tube. Those coming from the front are in phase with the portion of the same sound wave entering through a side port. But sounds arriving from the sides and rear enter the tubes from the ports and from the front at slightly different times so there's a phase difference between the portion of the wave from the side port and the portion from the front port. As these two portions of the sound wave travel through the tube to the diaphram they interact with each other inside the tuned tube and cancel each other out. Since it is the wave interactions inside the tuned tube that does the work they are dependent on the frequency of the sound and the dimensions of tube, just like the tone of an organ pipe is dependent on its dimensions. Off-axis high and mid frequencies are affected much more than are low frequencies from the same directions because low frequency wavelengths are too long to be affected very much by the tube. As a result off-axis mids and highs are supressed while lows from all directions reach the diaphram undiminished.

[Gian Pablo Villamil]

Originally Posted by Ray Ambrosi

...That software sounds interesting, but I'm not sure exactly what it does, or what advantages it offers. Could I use that software with the Sony ECM-MS957? ...

It is a plugin that works within most audio applications, and some video applications. Basically it takes two channels of audio, recorded as mid and side, and converts it to left and right stereo. In the process, it lets you vary the mix between mid and side, to achieve more of a "wide" stereo ambiance, or a more "narrow" ambiance.

It can also do the reverse transformation - extract mid and side information from a left/right recording. Finally, it does the equivalent of sticking two instances of the plugin in a row - encoding L/R stereo as mid/side, adjusting the mid/side balance, then re-encoding it back to L/R. So starting with a conventional L/R stereo recording you can modify the apparent spatial distribution.

It will work with any mid/side mike or any L/R stereo mike. You might need to invert the phase of one channel, but that is easy to do.

[Steve House]

Originally Posted by Ray Ambrosi

...
That software sounds interesting, but I'm not sure exactly what it does, or what advantages it offers. Could I use that software with the Sony ECM-MS957? Can anyone tell me if the specs on this mic are good or not? I really do want to find out if the Sony is worth buying and whether it can provide the type of M-S recording that can be manipulated to bring out the sounds and degree of stereo separation that I want to emphasize (this is the advantage of M-S as I understand it). Please advise!!
...

While the Sony mic uses the M/S principle to achieve its stereo output, it decodes the signals internally and so its output is conventional L/R stereo. AFAIK you don't have access to the original mid and side signals coming off the individual mic capsules. That means the post recording manipulation that's possible when recording the "raw" mid and side signals is no longer available.

[Dave Largent]

Originally Posted by Gian Pablo Villamil

It can also do the reverse transformation - extract mid and side information from a left/right recording. Finally, it does the equivalent of sticking two instances of the plugin in a row - encoding L/R stereo as mid/side, adjusting the mid/side balance, then re-encoding it back to L/R. So starting with a conventional L/R stereo recording you can modify the apparent spatial distribution.

It can take an X/Y recording and convert it to M-S,
so that you can work with it as you normally would
work with an M-S recording? I'd be curious to
hear what some others here have to say about
this. I guess we don't need M-S mic's anymore.

[Dave Largent]

Originally Posted by Steve House

While the Sony mic uses the M/S principle to achieve its stereo output, it decodes the signals internally and so its output is conventional L/R stereo. AFAIK you don't have access to the original mid and side signals coming off the individual mic capsules. That means the post recording manipulation that's possible when recording the "raw" mid and side signals is no longer available.

Except if you use that program that Gian
recommended ... maybe. For some reason
I feel a bit skeptical about it but I don't
really know.

[Steve House]

Originally Posted by Dave Largent

Except if you use that program that Gian
recommended ... maybe. For some reason
I feel a bit skeptical about it but I don't
really know.

Because the stereo L/R pair is derived from a mid/side signal matrix it can also go the other way.

Mid = (L+R)/2
Side = (L-R)/2

[Gian Pablo Villamil]

Originally Posted by Steve House

Because the stereo L/R pair is derived from a mid/side signal matrix it can also go the other way.

Mid = (L+R)/2
Side = (L-R)/2

Exactly.

If you've ever encoded an MP3 using the joint stereo mode, you've used a type of M/S encoding. See here: http://harmsy.freeuk.com/mostync/

The concern with, say the Sony mike or the AT 835ST in L/R mode, is that they don't do a straight transform between M-S and L-R: they apply multipliers to give the narrow or wide field effects. (It would be nice to know what these multipliers are...) So you'd have to do some tweaking to reconstruct the exact, original M-S inputs. However, since the MSED plugin is doing two symmetrical transforms, you can still use it to tweak "spatialization" of an incoming L-R signal, even if you don't know how that signal was constructed from the original M-S. You're basically applying differences.

[David Ennis]

Originally Posted by Dave Largent

...[The plugin Gian mentioned]...can take an X/Y recording and convert it to M-S,
so that you can work with it as you normally would
work with an M-S recording?...I guess we don't need M-S mic's anymore.

Well, a single point M/S mic would still be a more convenient way to get M/S signals then fiddling to set up an X/Y pair and then deconsrtucting the recording. Plus,the mid and side signals you might extract from any given X/Y pair recording will be different sounding than the corresponding mid and side signals you'd get from any given M/S mic, because different mics are still different mics. However, that doesn't mean that I might not prefer the ultimate result from M/S reprocessing the recordings from my X/Y'd AT3031s over the result from an AT835ST. I don't have an AT835ST to play with, but I just tried the MS encorder plugin on one of my AT3031 pair recordings and I like it a lot. I had been contemplating getting the 835ST but now I don't know....

[Steve House]

Originally Posted by Fred Retread

Well, a single point M/S mic would still be a more convenient way to get M/S signals then fiddling to set up an X/Y pair and then deconsrtucting the recording. Plus,the mid and side signals you might extract from any given X/Y pair recording will be different sounding than the corresponding mid and side signals you'd get from any given M/S mic, because different mics are still different mics. However, that doesn't mean that I might not prefer the ultimate result from M/S reprocessing the recordings from my X/Y'd AT3031s over the result from an AT835ST. I don't have an AT835ST to play with, but I just tried the MS encorder plugin on one of my AT3031 pair recordings and I like it a lot. I had been contemplating getting the 835ST but now I don't know....

The Waves S1 plugin gives you the same image manipulation tools normally associated with M/S using a conventional stereo input. It'll also decode M/S inputs to stereo but doesn't work the other way around.