Topic: Oddball microphone technique (OMT) - part 3

[Gutbucket]

Fig-8 center-

I've not done it but expect Fig-8 to work well in the center position in the right situation.  Virtues are it's tight forward pattern as EmRR mentions, plus the general naturalness of sound quality from that pattern.  I find a good fig-8 is sort of like an omni in this way, both generally sound more natural and open to me than the intermediate patterns between these two basic patterns.  The challenge is the "right situation" part.

The potential problem is no control over front/back pickup balance, and getting that right is important.  The demarcation between too much and too little from the back is easy to hear when mixing and most of the time you'll want considerably less from the back than from the front.   That depends greatly on room and placement, but also on program material, with the total overall range of level from the rear somewhere between none to equal to that from the front. For sparser numbers and quiet sections, or between numbers, approximately equal sensitivity can sound the most natural - same for ambient environmental recording where equal sensitivity in all directions sounds most natural and is preferred unless additional focus is needed toward the direction of interest, or reduced sensitivity is needed in other directions.  For loud, dense material, less contribution from the rear is desirable, and of course if the rear direction is dominated by distracted chatter or other noise which overly distracts from the intended focus of the recording, you'd want less of that. 

I'd have to check a few mixes, but generally I start mixing with the rear channel(s) around -6dB down in comparison to the other "directions", and it may end up being -12dB or more.  Complicating that, I'm often raising level of the center channel in comparison to the other directions (sometimes substantially), so the front/back differential actually ends up greater than that.

So pattern tightness and naturalness from a fig-8 are good, but the lack of balance control means it may only be really good in some situations.  I think it could work well outdoors.  Indoors, one could maybe try placing a small baffle like a jecklin disk or something behind a forward facing center figure-8, but it would require trial and error and only provide a limited range of rear-sensitivity "tuning" determined prior to recording. 

Another strategy could be making the L/R pair more forward-directional to compensate for lack of forward directionality of the center microphone.   Maybe substitute forward-facing subcards or cardioids for the omnis.

[Gutbucket]

BTW, any recording made with Blumlein crossed fig-8s between spaced omnis can be mixed as if it were a 3-ch recording made with a single forward-facing 8 between omnis - by summing the L/R Blumlein channels together in equal amounts before mixing with the omnis.  This is equivalent to a Mid/Side manipulation which discards the Side channel information, leaving only a virtual figure-8 Mid. 

Rocksuitcase and Kindms recordings incorporating both OMT and AKG C426 in Blumlein from the same recording location come to mind.

[EmRR]

Grabbed an Impact LS-3SAP pole, seems unique in that it has a 2 way socket that'll give you a right angle, or remove the stud to used it as an extension.  I haven't noticed any other mention of this hardware around here. 

https://www.bhphotovideo.com/c/product/1339819-REG/impact_ls_3sap_adjustable_pole_with_socket.html

That's the most recent one I picked up a while back.  Works well into a superclamp because it allows me to orient the extension either vertically or horizontally from the clamp. I most frequently use it in combination with one of these folding feet to create a short stand:

https://www.bhphotovideo.com/c/product/1119713-REG/impact_ls_3sb_backlight_stand_base.html

A few weeks ago I played around with finding the best way to record a friends jazz trio at their regular outside patio gig.  No PA, and no real good place to put the rig without it being either overly distant or in the way directly in front of the band, so I tried placing the mic array over the band looking directly downward.  To do that I placed my big stand next to the drummer with the LS-3SAP extension extending vertically so that I could use the 90 degree oriented stud socket at the top to mount the mic array facing straight down.  That fit nicely and got the array in good proximity with a good stereo perspective, without it being a nuisance.  Only problem was the guitar and keys amps placed on the ground were then 90-degrees off axis and came across overly diffuse.  Made for a sweet recording of the drums though!

Cool, I like OH recordings many times. 

[rocksuitcase]

BTW, any recording made with Blumlein crossed fig-8s between spaced omnis can be mixed as if it were a 3-ch recording made with a single forward-facing 8 between omnis - by summing the L/R Blumlein channels together in equal amounts before mixing with the omnis.  This is equivalent to a Mid/Side manipulation which discards the Side channel information, leaving only a virtual figure-8 Mid. 

Rocksuitcase and Kindms recordings incorporating both OMT and AKG C426 in Blumlein from the same recording location come to mind.

https://taperssection.com/index.php?topic=192098.0              kickdown thread for the OMT 6 using the c426 in Blumlein.
https://taperssection.com/index.php?topic=192034.0              for rig pics                       

[EmRR]

BTW, any recording made with Blumlein crossed fig-8s between spaced omnis can be mixed as if it were a 3-ch recording made with a single forward-facing 8 between omnis - by summing the L/R Blumlein channels together in equal amounts before mixing with the omnis.  This is equivalent to a Mid/Side manipulation which discards the Side channel information, leaving only a virtual figure-8 Mid. 

Yep.  Worth re-mentioning I've noticed you lose some top end info summing Blumlein, it really seems to work best as hard-panned information at least to my ears.  Contrast with turning Blumlein 45º so one is pointing mid, the other pointing side, and treating it as mid-side.  Then you are just picking side quantity.   

Either can be the better option, depending on placement/situation.  Blumlein might be advantageously PAS, or MS might be advantageously stage-centric. 

[Gutbucket]

^ Yep.  Which gets back to my speculation concerning M/S or X/Y being the optimal coincident configuration in the center microphone position.  I image the optimal answer is PAS X/Y + stereo width adjustment as needed, while the more practical answer might be M/S + stereo width adjustment.. such as using a shotgun Mid, or in my case where the same recording array is intended for both 2ch and surround output, and there are additional L/R microphone pair(s) feeding the L and R speakers. 

That's because I only use center coincident-stereo in a 2ch mix, where I really have grown to like what it does.  For surround playback through a front L/C/R speaker arrangement, I use the Center Mid channel as a discrete Center speaker feed and the Center Side channel goes unused.  The irony is not lost on me that I actually end up recording one additional channel for the 2ch stereo mix that I am not using for 5.0 or higher-count surround output.  Granted that in a 2ch mix the levels of the the surround/ambient mic channels are significantly lower than what is used for surround through discrete speakers.

Thinking about it, I'd probably use the Center Side channel in any mix which uses two speakers for the primary playback stage across front- so I'd use it in 4ch quad mix if I ever did that, while for 3.0, 5.0, 7.0 or whatever where there is a center speaker in the playback arrangement, that Center Side channel goes used. 

..at least currently. Someday I'll play around with surround front stage mixing more thoroughly to determine if there is any benefit of using some of it to the L/R speakers of an L/C/R setup to improve seamless image blend across the front as it does in 2ch, or if that only adds unwanted cross-talk and combing.

[Gutbucket]

BTW, any recording made with Blumlein crossed fig-8s between spaced omnis can be mixed as if it were a 3-ch recording made with a single forward-facing 8 between omnis - by summing the L/R Blumlein channels together in equal amounts before mixing with the omnis.  This is equivalent to a Mid/Side manipulation which discards the Side channel information, leaving only a virtual figure-8 Mid. 

Rocksuitcase and Kindms recordings incorporating both OMT and AKG C426 in Blumlein from the same recording location come to mind.

https://taperssection.com/index.php?topic=192098.0              kickdown thread for the OMT 6 using the c426 in Blumlein.
https://taperssection.com/index.php?topic=192034.0              for rig pics                       

Thanks Kyle!  Mike's Blumlein sounds great on my cheap headphones at the office.  I'll try and give a listen to the OMT mix (and the Blumlein again) tonight or tomorrow at home.  A few questions on how you mixed the OMT-

ch1|2     AKGck22 omni spread 6 feet apart > Grace V2 >Tascam DR680|SD
ch3|4     AKG ck8 shotgun fwd 0'|AKGck61 cardioid rear 180' >Tascam DR680|SD
ch5|6     AKG c426 (Blumlein:fig8, 90) >V3 >HDP2 (24/48) >Tascam DR680|SD

Did you start with chs 1-4 and balance that first (as you normally would) before bringing in 5/6? Or did you start with the Blumlein and add the other channels to that?  Or some other mixing strategy?

What level relationship did you end up with between the fwd/rear pair and the Blumlein pair?
When adding the c426 to the mix, did you find you needed to adjust the balance between the front ck8 and the back ck61?
When adding the c426 to the mix, did you find you needed to adjust the balance between the center mics and omnis?
How was the c426 + omnis without the front/back pair?

And curious about any thoughts in general about the inclusion of Blumlein to the existing OCT4 arrangement you've been using and are familiar with, partly because I've really grown to love the pseudo-OCT near-spaced L/R supercards (now generally angled forward somewhat) and wonder if the Blumlein center does something similar.

[prepschoolalumniblues]

Fig-8 center-

I've not done it but expect Fig-8 to work well in the center position in the right situation.  Virtues are it's tight forward pattern as EmRR mentions, plus the general naturalness of sound quality from that pattern.  I find a good fig-8 is sort of like an omni in this way, both generally sound more natural and open to me than the intermediate patterns between these two basic patterns.  The challenge is the "right situation" part.

The potential problem is no control over front/back pickup balance, and getting that right is important.  The demarcation between too much and too little from the back is easy to hear when mixing and most of the time you'll want considerably less from the back than from the front.   That depends greatly on room and placement, but also on program material, with the total overall range of level from the rear somewhere between none to equal to that from the front. For sparser numbers and quiet sections, or between numbers, approximately equal sensitivity can sound the most natural - same for ambient environmental recording where equal sensitivity in all directions sounds most natural and is preferred unless additional focus is needed toward the direction of interest, or reduced sensitivity is needed in other directions.  For loud, dense material, less contribution from the rear is desirable, and of course if the rear direction is dominated by distracted chatter or other noise which overly distracts from the intended focus of the recording, you'd want less of that. 

I'd have to check a few mixes, but generally I start mixing with the rear channel(s) around -6dB down in comparison to the other "directions", and it may end up being -12dB or more.  Complicating that, I'm often raising level of the center channel in comparison to the other directions (sometimes substantially), so the front/back differential actually ends up greater than that.

So pattern tightness and naturalness from a fig-8 are good, but the lack of balance control means it may only be really good in some situations.  I think it could work well outdoors.  Indoors, one could maybe try placing a small baffle like a jecklin disk or something behind a forward facing center figure-8, but it would require trial and error and only provide a limited range of rear-sensitivity "tuning" determined prior to recording. 

Another strategy could be making the L/R pair more forward-directional to compensate for lack of forward directionality of the center microphone.   Maybe substitute forward-facing subcards or cardioids for the omnis.

Thanks for the input. Backing up a bit, the reason this was on my mind is because I just bought a used Audio-Technica AT-4050 multi-pattern mic from the Yard Sale (omni, cardioid, figure-eight) thinking I would use it as a center cardioid between two omnis (all three going into a MixPre-3). I've never used a figure-eight before, though, so after looking it up, I started wondering about possibilities. I'll probably try out that configuration on a low-priority show just to see what it does (and will obviously report back).

And now I'm definitely curious about how a fig-8 between a directional stereo pair would sound too... maybe angling cardioids farther out than would normally make sense as a standalone pair, and then having a fig-8 low in the mix to provide some presence in the center (from both forward and rear perspectives)? I'm still trying to digest that Stereo Zoom paper, so my understanding of angles and spacing remains unhelpfully rudimentary...

[rocksuitcase]

Did you start with chs 1-4 and balance that first (as you normally would) before bringing in 5/6? Or did you start with the Blumlein and add the other channels to that?  Or some other mixing strategy?

For this show I actually mixed all six channels together. Typically, I would do a four or 5 channel mix using the center mic as my base point, but in this case with two centers available and given my crazy personal schedule I opted to do a "down and dirty mix". In the past, using all AKG mics, I've noticed mixing these specific set of sources to be very smooth compared to other sources we've used.  soooooooooo, in this specific instance:
I took all six channels in Audacity. In the master take channels 1|2 were at -2.5 dB (all levels approximate), channels 5|6 about -3.5 dB . channels 3|4 were at -8 dB or so.
                                                      It took three mixes to get what I found a decent compromise.

                                                    the final mix went like this (first, level adjust each pair; mono ch 3|4)
                                                      ch 1|2  were -2dB from master level =-4.5 dB
                                                      ch 3     was upped 4dB = -4dB
                                                      ch 4     was brought down -12dB = -20 dB
                                                      ch 5|6 down -1dB = -4.5 dB

What level relationship did you end up with between the fwd/rear pair and the Blumlein pair?

So I kept the mid ck8 in the final mix a bit lower than the Blumlein

When adding the c426 to the mix, did you find you needed to adjust the balance between the front ck8 and the back ck61?

the mix without c426 had a bit less "airiness" or natural room verb as the 426 added some accurate "envelopness" to my ears.

When adding the c426 to the mix, did you find you needed to adjust the balance between the center mics and omnis?
How was the c426 + omnis without the front/back pair?

That was a large amount of my listening before deciding to go with the down n dirty OMT6. I did two main listens:
1] I muted the c426 and listened to a different mix of omni vs fwd/rear.
2] Then muted the fwd|rear and listened to only fig 8 crossed plus omni.
                           The latter is a mix which should be tried out. As I said, this mix added some accurate "envelopness", some high frequency "pizazz" and a bit less in your face kick drum than the ck8/ck61 plus omni. (this ck8/ck61 fwd/rear has a bunch of kick drum and bass punch to my ears)

I should add that I could probably offer the raw files up via WT if any of you OMT folks re interested. Also that in discussing the mixing aspects with Lee over time, he knows and I agree that the possibilities and combinations of pairs, sub mixes and such can get overwhelming. In order to come out with a listenable 2 channel mix it is usually necessary to have a pallette of known good ideas/combinations and then let your ears guide you.  In this instance I did use all six channels in various levels. Often though one or more channels are not in the mix at all.

All critiques or listening opinions welcome for sure     

[Gutbucket]

Thanks for your mixing impressions.  I just downloaded and plan to give a listen tonight.

When adding the c426 to the mix, did you find you needed to adjust the balance between the front ck8 and the back ck61?

the mix without c426 had a bit less "airiness" or natural room verb as the 426 added some accurate "envelopness" to my ears.

This is interesting and makes sense.  I'm attempting to make a mental comparison to what the pseudo-OCT arranged L/R supercard pair adds to a 2ch mix in my OMT6 implementation, and I might describe that as "an increased sense of proximity to the source, presence enhancement and listening excitement".

[Gutbucket]

Thanks for the input. Backing up a bit, the reason this was on my mind is because I just bought a used Audio-Technica AT-4050 multi-pattern mic from the Yard Sale (omni, cardioid, figure-eight) thinking I would use it as a center cardioid between two omnis (all three going into a MixPre-3). I've never used a figure-eight before, though, so after looking it up, I started wondering about possibilities. I'll probably try out that configuration on a low-priority show just to see what it does (and will obviously report back).

And now I'm definitely curious about how a fig-8 between a directional stereo pair would sound too... maybe angling cardioids farther out than would normally make sense as a standalone pair, and then having a fig-8 low in the mix to provide some presence in the center (from both forward and rear perspectives)? I'm still trying to digest that Stereo Zoom paper, so my understanding of angles and spacing remains unhelpfully rudimentary...

Try all three patterns to get a feel for them. Listen to that channel solo'd to get an idea of how the 8 and omni patterns in some ways sound more similar to each other than they do to the cardioid, yet sound different from each other in other ways.  Listen to the sound of what the mic is pointed at, but also listen to the sound of all the stuff the mic is pointed far away from. For me it helps to think of different pickup patterns not really so much as being tighter focused or more widely focused toward what they are pointed at, but rather how different patterns are less sensitive to sound arriving from various directions other than where the mic is pointed.  That may seem like word-play, but even the most directional microphone has a forward focus "window" at least 90 degree wide for the most part.

All three patterns can work between spaced omnis, or spaced cardioids.  We can talk about the differences between them here, but it will be most useful to simply hear the difference for yourself, in your own recordings.  Best to keep the omni spacing the same (3' apart is good if you can do that, say 2' minimum) while changing the center pattern.  If you can't space them as much as you'd like, use of the 8 in the center should help somewhat, as will switching the omnis to cardioids and pointing them outwards somewhat (PAS at minimum, +/- 45 degrees typical, on up to +/-180 degrees max).


The key idea underpinning all the Stereo Zoom stuff is the inverse relationship between microphone angle and spacing.  If we want to keep the intended pickup window (SRA or Stereo Recording Angle in SZ speak) the same, as spacing between a stereo pair of microphones is increased, the angle between them needs to be decreased.  And vice-versa.  As the spacing is reduced, the angle between microphones needs to be increased.

If we change the angle without making the inverse adjustment to spacing, or change spacing without changing the angle, the stereo pickup angle for the pair is changed.

In applying that to a 3 channel mic setup, it sometimes helps to think of it as two separate stereo pairs sharing one microphone in the center.  Each pair faces somewhat off-center so that their shared edge of their pickup angle joins up in the middle as long as the spacing/angle relationship between the two pairs is arranged correctly.  That may makes it easier to get a intuitive feeling for how the outside microphones of a 3-mic set need to be angled more widely than a stereo pair when the center shared microphone is pointing directly forward.. or alternately, the spacing between the outside mics and the center one needs to be increased.. or some combination of the two, in comparison to a stereo pair with the same pickup angle. 

[Gutbucket]

https://taperssection.com/index.php?topic=192098.0              kickdown thread for the OMT 6 using the c426 in Blumlein.
https://taperssection.com/index.php?topic=192034.0              for rig pics                       

Oh yeah!

I totally dig Herring's projects along this line and lost myself in this last night listening on some good headphones.  Good stuff.

I pulled up both Mike's straight c426 Blumlein and Kyle's OMT6 incorporating the same, and went back and forth between the two.  At first I EQ'd as I felt necessary (partly headphone response correction), and was interested to compare how different the EQ curves might be that I ended up with for each, before switching my attention to more subtle aspects.  That varied from song to song, as well as between recordings, yet in the end the curves converged for the most part.  The two sound different, but I didn't really end up with an overall preference for one over the other.  I'd convince myself one was better, then switch and feel the same about the other.  Granted the the 426 blumlein pair in both sources made this less of a contrasting comparison and more of a straight Blumlein versus Blumlein + other channels type comparison.

One of the things that came to mind switching to the straight Blumlien was a feeling of "lushness yet uncomplicated purity".  It really does stand strong on its own.  The OMT mix including it excels in bass extension and envelopment, and in its slightly more diffuse portrayal of ambience and audience, almost hinting at another dimension. I sort of expected to hear sharper front imaging in the straight Blumlein, but I think it's inclusion in in OMT mix as well limited that potential difference between the two.

How was the c426 + omnis without the front/back pair?

The latter is a mix which should be tried out. As I said, this mix added some accurate "envelopness", some high frequency "pizazz" and a bit less in your face kick drum than the ck8/ck61 plus omni. (this ck8/ck61 fwd/rear has a bunch of kick drum and bass punch to my ears)

Thinking out loud..

I speculate c426 + wide omnis alone might be an ideal 4ch recipe for that location in the Egg, and think it may be best to narrow the 90° Blumlein angle somewhat in that combination (this could be emulated to some degree with the existing raw files by making a M/S ratio adjustment to the blumlein pair). This would bring the fig-8's closer to on-axis with the front sources to help fill the role the OMT center mic plays, possibly making that channel unnecessary (unless you miss the the kick-drum and bass qualities you mention it contributing).  I expect that would also increase energy in the center and tighten up imaging across the middle, attributes I expect would be made better still combined with a slight pattern tweak to the 426.   Given the Blumlein's rear-lobe sensitivity, the rear-facing center mic is probably even less needed than it's front-facing partner.  Yet then the front/back ratio is "baked-in" by the polar pattern chosen for the 426 and it's hard to say how much shift in pattern incrementally toward hyper/super would get that balance just right (guessing one or two clicks on the pattern box away from 8 ).  You all have recorded there enough to probably develop a good feeling about that, and your OMT mix experience in dialing in an appropriate front/back balance at that recording position in that room is a key data point.

If you guys had to split for some inexplicable reason leaving me to set up your gear,      , I'd probably select hyper patterns one click away from fig-8 toward cardioid, narrow the included angle somewhat, and run that between the omnis. Idea being to get a bit more front focus from the 426 and let the omnis shoulder the ambient room/audience part a bit more, while still retaining plenty of rear lobe from the 426.  If I had a pair of supercards as well, I'd place them between the 426 and the omnis angled +/- 45° and narrow the 426 included angle further down to PAS, and we have coincident-center OMT6.

Heynow, thanks for letting me play with your gear, and so happy you both were able to make it back before showtime in my thought experiment! 

Hmm, maybe I should finally get around to sending the Peluso P-Stereo coincident LD mic back to John P for repair.

Thanks for the great listen last night!  That's good medicine for a taper's soul.

[heathen]

Fig-8 center-

I've not done it but expect Fig-8 to work well in the center position in the right situation.  Virtues are it's tight forward pattern as EmRR mentions, plus the general naturalness of sound quality from that pattern.  I find a good fig-8 is sort of like an omni in this way, both generally sound more natural and open to me than the intermediate patterns between these two basic patterns.  The challenge is the "right situation" part.

The potential problem is no control over front/back pickup balance, and getting that right is important.  The demarcation between too much and too little from the back is easy to hear when mixing and most of the time you'll want considerably less from the back than from the front.   That depends greatly on room and placement, but also on program material, with the total overall range of level from the rear somewhere between none to equal to that from the front. For sparser numbers and quiet sections, or between numbers, approximately equal sensitivity can sound the most natural - same for ambient environmental recording where equal sensitivity in all directions sounds most natural and is preferred unless additional focus is needed toward the direction of interest, or reduced sensitivity is needed in other directions.  For loud, dense material, less contribution from the rear is desirable, and of course if the rear direction is dominated by distracted chatter or other noise which overly distracts from the intended focus of the recording, you'd want less of that. 

I'd have to check a few mixes, but generally I start mixing with the rear channel(s) around -6dB down in comparison to the other "directions", and it may end up being -12dB or more.  Complicating that, I'm often raising level of the center channel in comparison to the other directions (sometimes substantially), so the front/back differential actually ends up greater than that.

So pattern tightness and naturalness from a fig-8 are good, but the lack of balance control means it may only be really good in some situations.  I think it could work well outdoors.  Indoors, one could maybe try placing a small baffle like a jecklin disk or something behind a forward facing center figure-8, but it would require trial and error and only provide a limited range of rear-sensitivity "tuning" determined prior to recording. 

Another strategy could be making the L/R pair more forward-directional to compensate for lack of forward directionality of the center microphone.   Maybe substitute forward-facing subcards or cardioids for the omnis.

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?

[rocksuitcase]

I speculate c426 + wide omnis alone might be an ideal 4ch recipe for that location in the Egg, and think it may be best to narrow the 90° Blumlein angle somewhat in that combination (this could be emulated to some degree with the existing raw files by making a M/S ratio adjustment to the blumlein pair).

Thanks for the positive reviews! To address this specific mention briefly: Iagree ideally for OMT techniques that a center Blumlein pair should eb narrowed to at least PAS if not 50-60 degrees. The physics of it would offer a much more solid center tonality plus allowing the omnis to have a bit more decorrelation between them and center.

The main reason we do not do this recently is that we are also making a "Safety" or distinctly different recording with the 2 channel 426. I was going to suggest this to kindms before we went into the theater but sort of pushed it aside and went with the safety theory aspect. Also, If channels or space were restricted, I would definitely do only the 426 center with omnis outside (DPA 4060's if space was that tight). I also agree that in this instance I could have left off the ck8 center, kept the ck61 rear and the spaciousness would be there with a bit less defined center image.

[Gutbucket]

The main reason we do not do this recently is that we are also making a "Safety" or distinctly different recording with the 2 channel 426. I was going to suggest this to kindms before we went into the theater but sort of pushed it aside and went with the safety theory aspect.

A wise choice I think, and one I assumed fit the MO of the everyone involved.  And great to have that straight Blumlein pull as a baseline for comparison. We've talked in these threads before about the additional leap of faith and potential dangers involved with abandoning the "safety" to go all out.

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.