Why you should never use multi pattern mics

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.
Coming very late to this discussion, I noticed a few posts, particularly the OP mentioning "phasing issues".
Phase is an old endemic scapegoat, often used by barely qualified people to support their criticism of a particular equipment.
I'm not saying it's the case for kingkorg or Recording Engineer.
If there was no phase, there would simply be no cardioid mics.
Single diaphragm cardioids are cardioid because of an acoustic phase device, when dual-diaphragm rely on simple spacing, but in the end it's the same principle that is at work. Of course they have a different response but they both have "phase issues".
Just to be clear about "phasing issues" I mentioned in the op. I am talking about introducing signal coming from rear diaphragm which is separated by the thickness of the backplate from the front one. So there is slight delay, although small it falls in the audible range.
 
Just to be clear about "phasing issues" I mentioned in the op. I am talking about introducing signal coming from rear diaphragm which is separated by the thickness of the backplate from the front one. So there is slight delay, although small it falls in the audible range.
For sure, but if it wasn't there, there would be no directivity. Actually, there would be no signal either.
It's because the rear pressure is phase-shifted compared to the front pressure that the diaphragm moves.
 
ears have all sorts of phase and resonance issues too. the brain just corrects for them. stand in the bathroom with the fan on and cover your ears fully for about 15 seconds, then slowly tilt your hands away (listening to the pitch of the noise go up as the resonance changes). there will be a point at which you can still hear the pitch, but your hands are fully off your ears. congratulations, you can now (temporarily) hear the acoustic system of your ear.
 
Resonance, yes, but phase???
The very notion of phase implies there is a reference; what would be this reference?
i was actually agreeing with you and disagreeing with kingkorg a bit here. it's just a bit of a joke. the reference would be a signal coming from a pressure transducer in the same location with none of the the acoustic system around it, which would be completely inapplicable to the material reality of the ear and how it works. that was my point
 
@soliloqueen @abbey road d enfer ,

I am not even sure if i agree or disagree on this, or if we are even talking about the same thing.

Here's an example of a typical k67 behaviour. This is the response of cardioid vs omni. Both are campensated so that cardioid seems flat, just for the effect to be more obvious. Both sides are just summed together at same level. Green is cardioid, blue is omni. More than 5db notch at 5k is created as a result of just summing them in a traditional way. Very uneven response above 2k.
20230315_202344.jpg

This is the same exact situation, but both diaphragms are "virtually" alligned by adjusting the time delay between the two. The omni response is much smoother. It sounds much better, and off axis sound is way more even regardless of angle of incidence.
20230315_202444.jpg
 
@soliloqueen @abbey road d enfer ,

I am not even sure if i agree or disagree on this, or if we are even talking about the same thing.

Here's an example of a typical k67 behaviour. This is the response of cardioid vs omni. Both are campensated so that cardioid seems flat, just for the effect to be more obvious. Both sides are just summed together at same level. Green is cardioid, blue is omni. More than 5db notch at 5k is created as a result of just summing them in a traditional way. Very uneven response above 2k.
View attachment 106490

This is the same exact situation, but both diaphragms are "virtually" alligned by adjusting the time delay between the two. The omni response is much smoother. It sounds much better, and off axis sound is way more even regardless of angle of incidence.
View attachment 106491
makes me want to design a capsule that has intakes for the rear at almost the same place as the front diaphragm, routes it around the back and then back to the front. i wonder what would happen.
 
Here's an example of a typical k67 behaviour. This is the response of cardioid vs omni. Both are campensated so that cardioid seems flat, just for the effect to be more obvious. Both sides are just summed together at same level. Green is cardioid, blue is omni. More than 5db notch at 5k is created as a result of just summing them in a traditional way. Very uneven response above 2k.
OK, I didn't read properly your post. I thought you were questioning the use of applying back pressure to the rear of the diaphragm. Chalk iot up to language barrier and not enough sleep.
 
This is the same exact situation, but both diaphragms are "virtually" alligned by adjusting the time delay between the two. The omni response is much smoother. It sounds much better, and off axis sound is way more even regardless of angle of incidence.
kingkorg, which K67 capsule is this?

Can you explain a bit more about your "adjusting the time delay"?
If this is an "on-axis" response and the delay is adjusted for this, surely the "back" response will now have the delay 2x the original.
So the "back" response, and other directions, will be even worse.
 
kingkorg, which K67 capsule is this?

Can you explain a bit more about your "adjusting the time delay"?
If this is an "on-axis" response and the delay is adjusted for this, surely the "back" response will now have the delay 2x the original.
So the "back" response, and other directions, will be even worse.
No, the sound from all directions is more coherent. The back response is not adjusted, it won't have the 2x of the original. I basically delay the front diaphragm in time so it gets virtually alligned with the rear one. The goal is to make one virtual diaphragm out of the two physically separated. Like when you have two mics on a guitar cabinet, but one is say 1cm further. You delay the one closer to the speaker in order to allign it with the other one.
 
No, the sound from all directions is more coherent. The back response is not adjusted, it won't have the 2x of the original. I basically delay the front diaphragm in time so it gets virtually alligned with the rear one. The goal is to make one virtual diaphragm out of the two physically separated. Like when you have two mics on a guitar cabinet, but one is say 1cm further. You delay the one closer to the speaker in order to allign it with the other one.
Then, as ricardo mentioned, the rear pick-up is altered.
 
Thank you all for this very interesting thread. Lots of diverse, inspiring approaches, explanations and insights. These are great times to learn new things!!

At this stage I want to chime in, to state that lately I am using omni mics - be it multi pattern/dual diaphragm omni or true single diaphragm omnis - more often in close pickup/soloist recording situations. I tend to preferring those omni recordings over my cardioid ones, that I did for decades, because this was "the way you had to do it". At least to my ears and in my control room, these omni recordings seem to have more space and sound fuller, more natural and more complete than all cardioid mics - both single and dual diaphragm - that I have in the locker. I have to admit, I have an amazing sounding room, though. And high end preamps. And this doesn't apply to all multi pattern mics I have access to. So this experience might not translate to everyone as a general rule of thumb; we should always consider there are way more things factoring in than we can ever think about. And I am strictly talking about solo overdubs, because that's what I am recording 95% of my time. Of course there are perfect reasons NOT to use omnis in any kind of ensemble recordings (as severe bleeding leads to phase interference leading to cancelations leading to warped amplitude attenuation and maybe even phasey sounding tracks).

When it comes to multipattern mics, I see a split picture: Some of them sound way better in cardioid and I would never use them in omni, while others really shine there. To me, my EF47 build with Thiersch M7 (PET) capsule sounds mind blowing for close solo recordings in omni. As the distance between the diaphragms of the M7 is something around 6mm, I reckon (does anyone know the precise value?), maybe this smaller distance results in a less warped FR?! At the same time I am coming to the conclusion that impulse response/time domain is often way more crucial than FR. So if there was actually some FR warping going on in dual diaphragm omnis, this might not be such a bad thing after all: F*** the phase!! Literally: I found myself using an allpass filter warping the phase of almost every track I have been mixing over the last 10 years. I started this, because it sounded better and I had no idea that I began to screw up my recordings, technically. To me those tracks simply started to sound bigger, more pronounced, more like carved into the listening space. Same goes for those EF47 omni recordings I am doing now. Is this similarity a mere coincidence? I don't know.

Judge for yourself; here you can listen to some standardized recordings* of a variety of my EF47 mic builds (file names should be self explanatory) and of a bunch of some other well known mics.
Standardized_Mic_Demo_Rec

*here's how I made them:
A poor (wo)man's microphone measurement equipment

Enjoy! I am curious to hear what you guys might think about all that. Btw. I just yesterday I made another recording (they keep coming under the same Dropbox-Link) using a self made cable ("RVX-cable" in the file name). Can you hear a difference?

Roman
 
At this stage I want to chime in, to state that lately I am using omni mics
There is no doubt that omni capsules are much easier to make right than cardioids. As to multi-pattern set in omni, the results are variable; as someone else mentioned (kingkorg?), the fact taht the two halves are not perfectly coincident results in some cancellation. Note also that when the two halves are used, the transient response is no more Minimum Phase, which may or may not be detrimental.
these omni recordings seem to have more space and sound fuller, more natural and more complete than all cardioid mics
For sure, the ratio of reflected to direct sound is higher. If these reflections are well controlled, it is generally perceived as more pleasing.
Remember that cardioid mics are generally used to mitigate undesirable acoustic reflections.
 
As my opinion multi pattern mics is never go like best sound solution ever possible. It made just for some special scenario of recording when you need record sound from different direction.
 
Just to throw a bit of a curve ball into this discussion: you can use ambisonic mics as sometimes very precise multi pattern mics, once decoded. The capsules themselves are fixed cardioids (usually) but the resultant patterns are pretty much infinitely variable once you put them through software.

As a performer who likes to make recordings of rehearsals and gigs, I find ambisonic techniques immensely useful as I’m not in a position to monitor while I’m also part of a performing group. Obviously it won’t completely save you – you have to get the mic position right as you can’t alter the actual distances of the microphone relative to the performers or the space - but after the event you can choose infinite polar patterns, even synthesise multiple microphones from one mic to give yourself stereo or multichannel outputs.

I appreciate we’re talking about individual capsules as opposed to sets of them but the synthesised cardioid / omni / fig8 patterns from ambisonic mics are pretty impressive.
 
Last edited:
I used a Soundfield Ambisonic mic in the late '80s and enjoyed it quite a bit. There are mics I like better, but I found it useful and versatile.

BTW, "never" for a multi-pattern mic is a pretty strong statement. I use mostly single pattern mics but will readily use a multi-pattern one to get "the sound."
 
Back
Top