Oddball microphone technique (OMT) - part 3

[heathen]

Couldn't one adjust the front/rear balance of a figure 8 center by doing the same sort of processing one does when "manually" mixing down to mid-side?  Reverse polarity of the figure 8, etc?

Only if you had an omni coincident with it to do "Polar-flex" style M/S.  Then you could choose any pattern facing directly forward and rearward.

With a single fig-8 in the middle between the wide-spaced omnis, flipping its polarity won't do much.   It may change the sound as it changes it's phase/polarity relationship with the omnis, but that phase relationship is already  highly randomized at high frequencies.  I expect you might get more mid bass cancellation in one polarity orientation versus the other.

I'm probably just revealing my own ignorance here, and it's been a bit since I manually did a mid-side mix, but can't the figure 8 channel alone be used to make discrete left and right channels?  I'm struggling to remember if that's how it worked.  If so, though, the front lobe would end up in one channel, with the rear lobe in the other, and the level of each could be adjusted then mixed back down to one channel right?

I feel like it's entirely possible that I'm so off base here I come across as not much different than a crazy person standing on the corner yelling at traffic.

[heathen]

It's only when that output signal is sum/difference mixed with another phase-correlated signal (from a second coincidently-placed microphone) with a different directional orientation, that you get an output of two signals with different directional vectors. <meaning normalish patterns facing in particular directions.

Bingo.  That's the part I was forgetting.  Sorry for wasting everyone's time, and thanks for setting me straight.

[Gutbucket]

No worries.

[edit- oops accidentally deleted this post while editing it to add the pseudo-stereo stuff (which is now in the post that follows this one), so I'm reposting it here]

Dude, watch out for that garbage truck!

No. Its only a single channel.  The mic doesn't really know which direction it is facing and is equally sensitive to both directions. The front positive-polarity lobe and the rear negative-polarity lobe are "mixed together" in the output.  If you flip polarity the microphone is still facing both ways, except the waveform is inverted.

It's only when that output signal is sum/difference mixed with another phase-correlated signal (from a second coincidently-placed microphone) with a different directional orientation, that you get an output of two signals with different directional vectors. <meaning normalish patterns facing in particular directions.


You might be thinking of how some bidirectional microphones are formed by the differential summing of two well-matched back to back cardioid elements.  But in this case it's really two microphones facing opposite directions.  Typically the summing is done in the microphone and there is only a single output (the omni component of the two cardioids cancels out, leaving just the sum of the bi-directional components of the cardioids), yet some mics have outputs for the individual capsule which allows to you do the differential summing in the mixer, or use the outputs directly as two back to back cardioids. 

[Gutbucket]

It's only when that output signal is sum/difference mixed with another phase-correlated signal (from a second coincidently-placed microphone) with a different directional orientation, that you get an output of two signals with different directional vectors. <meaning normalish patterns facing in particular directions.

^ I struggled with the wording there.  You can do mid/side sum/difference processing of any two signals and they needn't be phase-correlated with each other.  But if they aren't you won't be synthesizing well-behaved 1st order virtual polar patterns. 

I've posted in these threads about using the rear-facing mic as Side channel in combination with the forward facing Mid.  Doing that "pseudo-stereoizes" the single rear-facing channel, but that doesn't equate to virtual X/Y hypercards or whatever.  If you could plot out a polar that would probably look something like an exploding asterisk at mid and high frequencies, varying with frequency.  At low frequencies where the spacing between the two mics is acoustically insignificant, you'd get summing to omni in the left channel and differential summing toward null (though not perfectly) in the right channel, which is exactly how it is supposed to work for M/S with coincident channels (they are effectively coincident for long wavelengths), yet is a problem with that sterilization technique.  Fixes for that could be monoizing below a the frequency where the two start to come into phase correlation, or using a +90° phase-shift on the left channel copy and -90° phase-shift on the right channel copy rather than flipping polarity of the right channel copy 180°.  That achieves the same 180° relationship between the two Side channel copies yet in a way that is phase-symmetrical so the bass doesn't all go left.  It's how basic 4:1 LCRS matrix surround encoding worked still works to fold the surround channel information into LtRt

[kuba e]

Sorry for little different topic. I recently encountered a little trouble when I was fine-tuning OMT mix.

A simple example - when I try to add e.g. +3db of center mic to the mix, the overall volume increases. Then it is difficult to compare new and old version because it is not the same volume. I tried to solve it in DAW by creating a group of two tracks. One with + 3dB center mic and the second with reverse polarity of spaced omni. I set the volume of reverse polarity omni track to compensate the add of +3dB center mic. When I turn this group on or off, the overall volume is the same, but the center and spaced mics ratio is different.

It's a complicated, maybe it's useless. But I had the feeling that it helped me to make decisions. I wonder if anyone has easy solution for this when mixing OMT.

[Gutbucket]

Most of the time I tend to just make quick ball-park fader adjustments of all channels in question when determining the most appropriate balances, combined with liberal use of the mute and solo buttons.  But your reverse polarity trick to lower the level of the pair is clever!  I've occasionally done something similar by just making a straight (non-inverted) copy of whatever channels I want to adjust simultaneously and unmuting that group when the corresponding group is muted.  However, your method allows for a more accurately adjusted comparison with a single button push.. as well as allowing for the fine tuning of that balance by adjusting the level of the linked group.

The general idea behind this kind of "maintained energy balance across all channels while making an adjustment" approach is something I've wanted implemented in multichannel editors for years.  In addition to tweaking level balances that way, I'd like to be able to make fine EQ and other filtering adjustments this way as well.   An EQ change in the target channel(s) would impart a corresponding "energy balanced" inverse change in all non-target channels, such that the overall frequency balance of the entire mix remains the same as the EQ bias is shifted between sources.  Not only do I feel this would be super beneficial in the 'tweaking stage' once a good rough balance is in place, making the process of finding an optimal balance considerably quicker and easier,  I believe it would open up powerful new approaches to sweetening the mix.  I do some of this kind of balanced trade-off tweaking now, but its a hassle to try and keep the overall balance unchanged while playing with the variations.

[kuba e]

However, your method allows for a more accurately adjusted comparison with a single button push.. as well as allowing for the fine tuning of that balance by adjusting the level of the linked group.

Yes, I need a single button push for comparison. This is helping me a lot because my listening skills are low.

The general idea behind this kind of "maintained energy balance across all channels while making an adjustment" approach is something I've wanted implemented in multichannel editors for years.  In addition to tweaking level balances that way, I'd like to be able to make fine EQ and other filtering adjustments this way as well.   An EQ change in the target channel(s) would impart a corresponding "energy balanced" inverse change in all non-target channels, such that the overall frequency balance of the entire mix remains the same as the EQ bias is shifted between sources.  Not only do I feel this would be super beneficial in the 'tweaking stage' once a good rough balance is in place, making the process of finding an optimal balance considerably quicker and easier,  I believe it would open up powerful new approaches to sweetening the mix.  I do some of this kind of balanced trade-off tweaking now, but its a hassle to try and keep the overall balance unchanged while playing with the variations.

You mentioned this in some previous posts and you inspired me. I have been using a lot reversed tilt eq. It shows me how individual microphone pairs contribute to the overall mix and what to improve. And I usually use this reverse tilt eq (+- 1dB) in final mix too.  It is possible to make reversed classic eq, mid/side manipulations or any plugin in Reaper. But tilt eq is easy and very effective for me.

For those who use Reaper - there is possibility to link all plugins parameters and set their relationship, e.g. inverse change. The procedure is very simply.
1. prepare reverse plugin - Load two plugins, link the parameters and save these plugins as FX chain.
2. use reverse plugin - Load this chain, bundle the first plugin with channels 1, 2 (e.g. center xy pair) and the second plugin with channels 3, 4 (e.g. spaced pair). Now, the change in one plugin will appear in the other mirrored opposite.

[Gutbucket]

I dig tilt type EQ.  It's a "superwide Q totally tonal" type of thing.   Sounds good and reveals basic frequency balance relationships quickly. Surprised its not seen more often.

Thanks for outlining the inverse-change plugin method for Reaper. Very cool that reverse-plugin is available there.  Maybe I should look at Reaper again, but I don't really want to relearn my way around another editor.

[kuba e]

I understand this very well, I don't want to learn another DAW too. I think there is a good chance that other software developers will also add the ability to link parameters. It shouldn't be difficult and this function is useful.

[rocksuitcase]

Sorry for little different topic. I recently encountered a little trouble when I was fine-tuning OMT mix.

A simple example - when I try to add e.g. +3db of center mic to the mix, the overall volume increases. Then it is difficult to compare new and old version because it is not the same volume. I tried to solve it in DAW by creating a group of two tracks. One with + 3dB center mic and the second with reverse polarity of spaced omni. I set the volume of reverse polarity omni track to compensate the add of +3dB center mic. When I turn this group on or off, the overall volume is the same, but the center and spaced mics ratio is different.

It's a complicated, maybe it's useless. But I had the feeling that it helped me to make decisions. I wonder if anyone has easy solution for this when mixing OMT.

kuba, I understand this. Using Audacity, I am mostly like Gutbucket in that I generally throw on the 4 or 6 channels, check for overall level and balance the stereo pairs (or mono channels if fwd|rear), then adjust each channel's sliders and then use mute and solo to isolate what I like or don't and maybe go in and change a slider up or down a few times. once settled on something which sounds solid, with good centering in both headphones and my basic desktop speakers and export to a stereo mix, naming this roughmixXX.
SO I am not doing the type of additions you are, however, I DO understand your issue. It seems like doing a single button inverted polarity is a good way to go.
On this note, I am thinking of having to add Izotope for at least some mastering tools, and am a bit tentative over the learning curve.

[kuba e]

Kyle, I am doing mix like you and Gutbucket. I had two cases when I was totally unsure and I needed fine tuning. One case was too little spaced omnis. I wanted to mix the center cardiod in, but it didn't sound good. I was trying to add just little bit and also to stereoize the center mic. But in the end the best variation was plain omni. The second case, when I'm not sure at times, is mixing the rear channel. It seemed to me that the rear channel is adding great third dimension but also we may sometimes lose a little bit of clarity. These cases are only about details and are not worth the effort. The only positive is that I practice listening.

On this note, I am thinking of having to add Izotope for at least some mastering tools, and am a bit tentative over the learning curve.

I have some cheap pack of Izotope basic plugins. From what I see it is very user friendly. The best is to try it before you buy it, they offer one moth free. They'll probably offer you a discount for purchase too.

[Gutbucket]

One case was too little spaced omnis. I wanted to mix the center cardiod in, but it didn't sound good. I was trying to add just little bit and also to stereoize the center mic. But in the end the best variation was plain omni.

This makes sense.  You could try increasing stereo width of the omni pair to "make room for the center cardioid" in the mix.  Doing so increases the difference information between omni channels and decreases the monophonic sum information shared between channels.  It's that shared mono-sum info that is in conflict with the center cardioid.  Probably don't want to push that too far, but worth a try, if only to hear what it does.

The second case, when I'm not sure at times, is mixing the rear channel. It seemed to me that the rear channel is adding great third dimension but also we may sometimes lose a little bit of clarity.

This is a good description of the general trade off when determining the appropriate contribution from the rear-facing mic(s) in a stereo mix.

[Gutbucket]

Was re-reading Hyunkook Lee's comments on MARRS, which we've discussed in the past elsewhere at TS along with other stereo imaging visualization tools leveraging Stereo Zoom and similar research, and the following jumped out at me (note- my bold, italics and underlining, and ellipsis below) :

MARRS (Microphone Array Recording and Reproduction Simulator) is an interactive and intelligent tool for stereo microphone technique simulation. It allows the user to predict the perceived positions of multiple sound sources for a given microphone configuration. The tool can also automatically configure suitable microphone arrays for the user's desired spatial scene in reproduction.
[jump]
The app can recommend you a range of correct microphone configurations for a desired overall stereo width of the ensemble you are recording... MARRS will then calculate the correct angle and spacing of a microphone array that you need to apply for the desired stereo width... Note that you still have a control over "what type of array" you want it to be. The "XY/AB" section in the configuration view allows you to determine the ratio between 100% coincident (XY) and 100% spaced (AB) depending on the desired spatial characteristics of the stereo image. The more XY you apply, the image will tend to get more localizable but less spacious. The more AB you apply, the image will tend to sound more spacious but less sharp and localizable. Near-coincident techniques such as ORTF and NOS benefit from both localizability and spaciousness. They are popular "preset" techniques with a fixed XY/AB ratio, but you can flexibly choose the ratio, thus designing your own near-coincident technique, depending on your purpose using MARRS.

This echos the statements I frequently make at TS about near-spaced stereo microphone pair techniques representing various "preset, hopefully best compromise" stereo arrangements.  We are able to bias that compromise one way or another by how we set up, informed by experience and tools such as SZ, the Improved-PAS table, MARRS, etc, yet at that point the compromise becomes fixed, prior to recording.  One of the advantages of OMT arrangements which combine a spaced pair and a coincident center pair (with or without additional near-spaced channels) is that we record both and gain control over the blend of stereo characteristics after the recording has been made.   But more than that, when done well I find I can achieve a combination of otherwise disparate stereo aspects which is unachievable by near-spaced stereo "compromise" arrangements on their own.

[EmRR]

If I ever saw this info, I'd forgotten it:

The ear uses mainly two techniques for determining directional information in a horizontal plane: differences in level and differences in time of arrival/phase between the two eardrums. When listening to complex waveforms, both methods help in localization. With respect to frequency, however, the application of one of these two techniques usually excludes the other.

The following paragraphs on localization are informed by the excellent text on psychoacoustics by J. Blauert:

In the low frequency range up to about 800 Hz, the human brain depends mostly on phase information for localization purposes. This figure of 800 Hz coincides to a wavelength of approximately 1.25 feet. Not coincidentally, this length of 1.25 feet corresponds to about twice the greatest length between a pair of human eardrums. In other words, for frequencies below 800 Hz, the phase of the impending sound at the nearer ear always leads the phase of the sound at the farther ear by less than 180 degrees. Such a close phase relationship between the sounds heard at both eardrums creates an unambiguous method for localization. The phase difference is linear with respect to frequency, thus also aiding with the ear's attempts at discerning the spectrum of an incoming sound.

For frequencies above 1.6 kHz (the high frequency area), phase relationships between the eardrums become meaningless. Since the period of such frequencies is less than the amount of time sound takes to travel between the ears, phase relationships between sounds at the ears will be misleading, shifting through a full 360 degrees for every octave. In this range, therefore, the brain must depend on level and time differences between the ears for localization purposes. This method relies somewhat on the diffraction and absorption qualities of the head itself. Off center high frequencies will hit the nearer eardrum with no loss in level, but upon encountering the head as an acoustic obstacle, can fall several decibels before reaching the following eardrum. In certain combinations of direction and frequency, this attenuation can exceed 20 decibels!

As of yet unmentioned has been the frequency range from 800 Hz to 1.6 kHz, the middle area for localization. Perhaps middle area is the wrong term; gray area may be more appropriate. In this region, level differences between the ears are not very effective since the wavelengths are very long in comparison to the size of the head. Also, phase differences are misleading because they now exceed 180 degrees between the ears; the closer ear, therefore, now lags in phase behind the further ear. The brain must thus use a combination of these two methods (phase and level) to localize sound. In practice, however, the localizing ability within this frequency range is basically not good.

The discussion of localization so far has focused on frequency ranges and thus relates to periodic sine wave sources. Such information, however, only allows for an estimation of the ear's true abilities of perception for aperiodic signals like music, speech, or any other real world sounds. The sustained portion of sound from musical instruments can be characterized as sinusoidal, and much of the ear's methods for localizing such portions of sound are as previously described. The attack portion of sound from instruments, especially percussion instruments, however, is most often impulse based. For such quick transients, pure time of arrival differences between the ears are used to localize the source. In other words, direction is determined solely by whichever ear first hears the impulse.

[thunderbolt]

I picked up a used Pearl MSH-10.  I have not used it in a concert setting yet but am going to try it next month for grins with a primary pair of 41s.  New venue for me (Sherman Th., Stroudsburg, PA).  I'll post a clip after.