Visualization of various stereo mic configurations

[DigiGal]

I heard from my friend (Rob) today.  He said making a version for iOS 4.2.1 is doable.  He is cranking on some other projects right now but thought that after New Years he would likely have some time.  Also, we will hopefully have some other feedback we could incorporate at that time.  Sorry for the wait.  I really wish you could use it NOW.

I've been researching how to derive shotgun polar patterns.  I've got a good start on it.  I'll see if we can get them into the app at the same time.  I'll post a pic of the shotgun pattern I'm thinking of using in a few.  I need to make sure it looks like the ones that are published for common shotguns first.


-MIQ

Great news thank you, looking forward to the update.  I personally will use either a AKG CK98 or Sennheiser ME66 for a shotgun mid on occasion. 

[Gutbucket]

I've been researching how to derive shotgun polar patterns.  I've got a good start on it.  I'll see if we can get them into the app at the same time.  I'll post a pic of the shotgun pattern I'm thinking of using in a few.  I need to make sure it looks like the ones that are published for common shotguns first.

Patterns between various interference tube (shotgun) mic vary greatly between models and manufacturers.  They also vary dramatically with frequency - far more than 'classic' first order microphone patterns do, even the ones that don't hold their pattern well.  I'm pretty sure all of these configuration visualizers assume a textbook pattern that is invarient over the full frequency range, which is rarely the reality of the situation, but close enough for jazz, rock'n'roll, bluegrass, classical. 

[MIQ]

Thanks Gutbucket

I noticed the freq dependency in the equation I'm using.  Also the length of the interference tube plays a big role in the response.  The thing that is most surprising to me is that the rear attenuation gets worse between 1k to 4kHz and then starts to get better above 4k on the plots from AKG and Senn.

The first order patterns we used in the Mid-Side Visualizer are assumed to be ideal and I mention that that is not reality in the info page.  I reference the paper by Trevor de Clercq here http://www.midside.com/pdf/nyu/masters_thesis.pdf  It might have been you who turned me on to that paper. 

-MIQ

[MIQ]

Attached is a first stab at a shotgun polar response.  I tried to hit the following 1kHz response points from the CK98

-5dB @ 60 deg
-20dB @ 90 deg
-15dB @ 180 deg

The ME66 is not quite as directional.  It's front lobe is almost exactly the same as a supercard but the rear lobe has more attenuation.

-MIQ

[John Willett]

Attached is a first stab at a shotgun polar response.  I tried to hit the following 1kHz response points from the CK98

-5dB @ 60 deg
-20dB @ 90 deg
-15dB @ 180 deg

The ME66 is not quite as directional.  It's front lobe is almost exactly the same as a supercard but the rear lobe has more attenuation.

-MIQ

This chart is very misleading, unfortunately.

It only shows a theoretical pattern at a single frequency which is totally meaningless for an interference tube gun mic.

A gun mic. will normally have a super-cardioid capsule, so the low frequency response will be the same as a super-cardioid.

However, it has an interference tube on the front - this will make the microphone more directional as frequency rises.

This is the polar-pattern of the MKH 60 which shows this effect very clearly:-

(low frequencies up to 1kHz are on the left and high frequencies 2kHz and above are on the right)


This is the polar-pattern of the ME 66:-

(low frequencies up to 1kHz are on the left and high frequencies 2kHz and above are on the right)

I hope this helps.

[MIQ]

Thanks John,

This is exactly what I noticed.  The low freq response is basically that of a supercard.  Thanks for confirming this since I was assuming manufacturers start with something like a supercard capsule design and then stick an interference tube on the front of it.  Is that close to true?

The response of an interference tube follows the basic formula:

R = sin(pi/wavelength)*(length of tube - length of tube*cos(theta))
      ---------------------------------------------------------------------------------
       (pi/wavelength)*(length of tube - length of tube*cos(theta))

where wavelength is the wavelength of sound and theta is the angle of the sound source (incidence).  From Olson via Eargle.

The wavelength gets tiny as freq goes up and this explains the strong changes in polar pattern as freq increases.  It also explains why the response of these mics is simply that of something like a super card at 1000Hz and below.  The interference tube has little affect at these frequencies and so the capsules response dominates.  Below is a plot showing the response of an interference tube following the formula above.   Tube length = 110mm, Frequency = 8000 Hz
It looks somewhat like the ME66 response at 8kHz.  The ME66 interference tube looks about half of its 221mm total length.

John, in your opinion, is it meaningless to try to include a "shotgun" pattern (at a single frequency) for a mid side visualizer like the one we've created?  I realize there is an even stronger affect on the polar response vs freq compared to the first order mics.  Does this make it meaningless?  I appreciate your thoughts.

-MIQ

[John Willett]

Thanks John,

This is exactly what I noticed.  The low freq response is basically that of a supercard.  Thanks for confirming this since I was assuming manufacturers start with something like a supercard capsule design and then stick an interference tube on the front of it.  Is that close to true?

In essence, yes.

The longer the tube, the more directional it gets at high frequencies.

See the MKH 60 and MKH 70 side by side:-

John, in your opinion, is it meaningless to try to include a "shotgun" pattern (at a single frequency) for a mid side visualizer like the one we've created?  I realize there is an even stronger affect on the polar response vs freq compared to the first order mics.  Does this make it meaningless?  I appreciate your thoughts.

Yes, I really *do* think showing a single frequency is meaningless.  The interference tube of a normal short gun has virtually no effect at all at 1kHz and pulls in above this.

A long gun will start working at lower frequencies and be more directional at higher frequencies.

You also need to understand the what the side lobes do.

Your diagrams *do* look very good, though, but would be better to show at different frequencies.

[F.O.Bean]

Very cool, +T

[MIQ]

Thank you John,

It does look like having a single pattern displayed is not enough info for a shotgun mic.  Displaying more than one pattern, say at 2kHz and 8kHz, at the same time could be possible but might make the diagram too busy.  There is so little space on an iPhone.  The real problem IMHO is that there is not a single set of "shotgun" patterns, even if you could display the pattern at more than one freq.  As you've shown, the length of the tube changes everything.  Having an input for freq, like a selector for 1k, 2k, 4k, 8kHz etc could be possible and another for tube length in mm would be neat to play with.  But these are more complicated to implement.  To be honest we haven't seen much interest in the visualizer as it stands.  It's hard to justify sinking more time into it.

-MIQ

[Javier Cinakowski]

Is there anyway to look at thisis on a computer ir android? Id pay a lot more than $1 for an android or windows app...

[MIQ]

Hi Javier,

The visualizer was first developed as an Excel sheet to work out the algorithms and prove the concept.  The polar plots with a white background I've been posting here have been cut and pasted from that sheet.  While it's a way to see it on a computer, it is hardly a commercial product and not protected in any way. 

-MIQ

[DigiGal]

I heard from my friend (Rob) today.  He said making a version for iOS 4.2.1 is doable.  He is cranking on some other projects right now but thought that after New Years he would likely have some time.  Also, we will hopefully have some other feedback we could incorporate at that time.  Sorry for the wait.  I really wish you could use it NOW.

I've been researching how to derive shotgun polar patterns.  I've got a good start on it.  I'll see if we can get them into the app at the same time.  I'll post a pic of the shotgun pattern I'm thinking of using in a few.  I need to make sure it looks like the ones that are published for common shotguns first.


-MIQ

Do you know if he has been able to start work on making this compatible with iOS 4.2.1 yet?

[MIQ]

Hi DigiGal,

We talked about it after Xmas but I was working at CES and Rob is at NAMM so not much progress has been made.  Fortunately the update should not be too difficult.  I will keep you posted.

MIQ

[DigiGal]

Hi DigiGal,

We talked about it after Xmas but I was working at CES and Rob is at NAMM so not much progress has been made.  Fortunately the update should not be too difficult.  I will keep you posted.

MIQ

I see a new version available today is now compatible with iOS 3.2+.  For some reason it won't load on my iOS 4.2.1 device.

[MIQ]

DigiGal,

Unfortunatley the app is not supported on generation 2 devices even if they are running iOS3.2 or higher. 

If you upgrade to a gen 3 or higher device, the app will work for you.  We look forward to hearing your comments when you upgrade.   

-MIQ