Become a Site Supporter and Never see Ads again!

Author Topic: M/S Processing for non m/s sources  (Read 2952 times)

0 Members and 1 Guest are viewing this topic.

Offline noahbickart

  • phishrabbi
  • Site Supporter
  • Trade Count: (31)
  • Taperssection All-Star
  • *
  • Posts: 1893
  • Gender: Male
  • So now I wander over grounds of light...
M/S Processing for non m/s sources
« on: October 15, 2018, 10:52:56 AM »
Hi everyone:

As I've been thinking about modified PAS and OMT techniques, and doing more multi-microphone recordings, one of the things I'm thinking more and more about is the degree to which a good tape depends on the right mix of direct and reflected sound.

We set up our microphones arrays in such a way to provide a good balance between the “dry" direct sound from the PA and "wet" sound of the room. If done well, yay!! If not, well try again.

But there's an easy way to adjust this stuff in post, by using a M/S plug in on a Stereo L/R signal. With the Magic of modern DAW plugins, there's no need to convert to M/S first.

So, too much ambience/audience chatter? Decrease the S component and/or bump up the M. Recording too Monoish? Decrease the M and/or increase the S.

Here are some great resources:
https://klanghelm.com/contents/products/VUMTdeluxe/VUMTdeluxe.php (This is what I use, as the channel knobs can be L/R or M/S)
https://www.voxengo.com/product/msed/
https://hofa-plugins.de/en/plugins/4u/

**edit to add more resources:**
Here's a good free parametric eq which allows for different EQ setting for M and S:
https://www.ikmultimedia.com/products/trclasseq/

If you prefer a free graphic, the same folks who do MSED make this:
https://www.voxengo.com/product/marvelgeq/

Here's a free VST which does M/S level, eq & dynamics in a single plugin:
https://www.soundonsound.com/news/internet-co-release-free-ms-eq-comp-plug

I find the toneboosters eq to be really easy to work with and is only ~$35:
https://www.toneboosters.com/tb_equalizer_v4.html
« Last Edit: October 25, 2018, 09:20:52 AM by noahbickart »
Recording:
Capsules: Schoeps mk41v (x2), mk4v (x2), mk22 (x2), mk3 (x2), mk21 & mk8
Cables: 2x nbob KCY, 1 pair nbob actives, Darktrain 2 and 4 channel KCY extensions:
Preamps:    Schoeps VMS 02iub, Naiant Littlebox, Naiant IPA, Sound Devices Mixpre6
Recorders: Sound Devices Mixpre6, Sony PCM m10

Home Playback: Mytek DSD 192> McIntosh MC162> Eminent Tech LFT-16; Musical Fidelity xCan v2> Hifiman HE-4XX / Beyerdynamic DT880

Office Playback: Grace m903> AKG k701 / Hifiman HE-400

Offline Gutbucket

  • record > listen > revise technique
  • Trade Count: (13)
  • Needs to get out more...
  • *****
  • Posts: 12607
  • Gender: Male
Re: M/S Processing for non m/s sources
« Reply #1 on: October 15, 2018, 12:19:20 PM »
Any stereo file can have it's balance re-adjusted in this way. IME lots of live recordings made with narrowly spaced and angled microphones can benefit from some adjustment.  Just listen carefully to determine if it is doing what you want without introducing other problems. 

It can sometimes be helpful to think of this in terms of manipulating the basic Sum/Difference balance between the two channels rather than Mid/Side.  Mid/Side representing the higher-order categorization of the sum and difference components.

And if you want to get really fancy, you can apply inverse EQ curves to the Sum and Difference channels so as to adjust stereo width in different amounts by frequency, without changing the overall frequency balance.  Probably easiest to do that by adjusting EQ of the difference channel to obtain the width result you want (typically by emphasising the low frequencies in the Side channel), then EQing the re-combined output to whatever sounds timbrally appropriate overall.  Otherwise, there are plugins which do that for you, providing controls for varying width by frequency.
musical volition > vibrations > voltages > numeric values > voltages > vibrations> virtual teleportation time-machine experience
Better recording made easy - >>Improved PAS table<< | Made excellent- >>click here to download the Oddball Microphone Technique illustrated PDF booklet<<

Offline noahbickart

  • phishrabbi
  • Site Supporter
  • Trade Count: (31)
  • Taperssection All-Star
  • *
  • Posts: 1893
  • Gender: Male
  • So now I wander over grounds of light...
Re: M/S Processing for non m/s sources
« Reply #2 on: October 15, 2018, 01:07:52 PM »
Isn’t another way of saying "mid," "sum" and side another way of saying "side," "difference?"
« Last Edit: October 15, 2018, 02:17:14 PM by noahbickart »
Recording:
Capsules: Schoeps mk41v (x2), mk4v (x2), mk22 (x2), mk3 (x2), mk21 & mk8
Cables: 2x nbob KCY, 1 pair nbob actives, Darktrain 2 and 4 channel KCY extensions:
Preamps:    Schoeps VMS 02iub, Naiant Littlebox, Naiant IPA, Sound Devices Mixpre6
Recorders: Sound Devices Mixpre6, Sony PCM m10

Home Playback: Mytek DSD 192> McIntosh MC162> Eminent Tech LFT-16; Musical Fidelity xCan v2> Hifiman HE-4XX / Beyerdynamic DT880

Office Playback: Grace m903> AKG k701 / Hifiman HE-400

Offline Gutbucket

  • record > listen > revise technique
  • Trade Count: (13)
  • Needs to get out more...
  • *****
  • Posts: 12607
  • Gender: Male
Re: M/S Processing for non m/s sources
« Reply #3 on: October 15, 2018, 01:25:13 PM »
Yes, just different terms for the same thing:
Sum = Mid = L+R
Difference = Side = L-R (and/or R-L)

This past weekend I was considering running a setup using two sideways-oriented Mid/Side pairs, which would have decoded as Front/Back rather than Left/Right.  In talking about this with Nick D over the phone it became confusing describing a sideways-facing Mid and forward-facing Side. 
musical volition > vibrations > voltages > numeric values > voltages > vibrations> virtual teleportation time-machine experience
Better recording made easy - >>Improved PAS table<< | Made excellent- >>click here to download the Oddball Microphone Technique illustrated PDF booklet<<

Offline wforwumbo

  • Trade Count: (6)
  • Taperssection Regular
  • **
  • Posts: 92
Re: M/S Processing for non m/s sources
« Reply #4 on: October 15, 2018, 06:14:15 PM »
All phenomenal information in this thread so far.

First thing I have to say: don't trust that "Nick D" character. He has no idea what he's talking about.

Mid/side processing is also useful for equalization, especially on stereo equalization. Some EQ plugins - I'm pretty sure the default Logic EQ plugin will do it, and I want to say Fab Filter Pro-Q 2 will as well - will also encode/decode and mix in M/S for you as well. Particularly useful if you want to address mid/side differently, and still have the information presented in traditional L/R.

What's interesting to me about M/S, has to do with coherence. The "mid" channel is the amount of "coherent" signal between left and right, while "side" is about "incoherent" or "decorrelated" sound. Practically speaking, this is similar enough (if not identical) to using "sum/difference" terminology - where the waveform aligns in sum it reinforces, and where it does not it cancels.
2x Schoeps mk21 (matched pair) -> nbob KCY -> Naiant PFA -> Sound Devices Mixpre 6

Offline Gutbucket

  • record > listen > revise technique
  • Trade Count: (13)
  • Needs to get out more...
  • *****
  • Posts: 12607
  • Gender: Male
Re: M/S Processing for non m/s sources
« Reply #5 on: October 16, 2018, 09:30:25 AM »
What's interesting to me about M/S, has to do with coherence. The "mid" channel is the amount of "coherent" signal between left and right, while "side" is about "incoherent" or "decorrelated" sound. Practically speaking, this is similar enough (if not identical) to using "sum/difference" terminology - where the waveform aligns in sum it reinforces, and where it does not it cancels.
^
This! 

And loosely correlates with direct / reverberant, for the sources of primary interest in our recordings.  Which loops back to Noah's initial dry/wet observation.

coherent/decorrelated
direct/reverberant
sum/difference
front/all-other-directions

^
To me, these concepts generally represent more useful ways of thinking about the nature of recording and acoustics than Left/Right
musical volition > vibrations > voltages > numeric values > voltages > vibrations> virtual teleportation time-machine experience
Better recording made easy - >>Improved PAS table<< | Made excellent- >>click here to download the Oddball Microphone Technique illustrated PDF booklet<<

Offline noahbickart

  • phishrabbi
  • Site Supporter
  • Trade Count: (31)
  • Taperssection All-Star
  • *
  • Posts: 1893
  • Gender: Male
  • So now I wander over grounds of light...
Re: M/S Processing for non m/s sources
« Reply #6 on: October 16, 2018, 10:17:55 AM »
To me, these concepts generally represent more useful ways of thinking about the nature of recording and acoustics than Left/Right

Indeed. And that's why I'm so surprised that while most tapers probably do (or at least know how to do) L/R adjustment in post, so few are willing to use a plugin to adjust the M/S ratio.
Recording:
Capsules: Schoeps mk41v (x2), mk4v (x2), mk22 (x2), mk3 (x2), mk21 & mk8
Cables: 2x nbob KCY, 1 pair nbob actives, Darktrain 2 and 4 channel KCY extensions:
Preamps:    Schoeps VMS 02iub, Naiant Littlebox, Naiant IPA, Sound Devices Mixpre6
Recorders: Sound Devices Mixpre6, Sony PCM m10

Home Playback: Mytek DSD 192> McIntosh MC162> Eminent Tech LFT-16; Musical Fidelity xCan v2> Hifiman HE-4XX / Beyerdynamic DT880

Office Playback: Grace m903> AKG k701 / Hifiman HE-400

Offline Sloan Simpson

  • Trade Count: (1)
  • Needs to get out more...
  • *****
  • Posts: 3541
  • Gender: Male
    • Southern Shelter
Re: M/S Processing for non m/s sources
« Reply #7 on: October 16, 2018, 10:21:59 AM »
Sometimes I’ll high pass or low shelf cut the sides on an audience tape in a muddy room.
Neumann KM-184> Tascam DR-680

Offline capnhook

  • All your llamas are belong to us....
  • Site Supporter
  • Trade Count: (20)
  • Needs to get out more...
  • *
  • Posts: 3675
  • All your llamas are belong to us....
Re: M/S Processing for non m/s sources
« Reply #8 on: October 16, 2018, 02:03:24 PM »
Sometimes I’ll high pass or low shelf cut the sides on an audience tape in a muddy room.

Great tip, never considered that.  Thanks.

Then MSED....
Proud member of the reality-based community

BSCS-L->JB-mod [NAK CM-300 (CP-3) and/or (CP-1)]->LSD2->CA CAFS-Omni->Sony ECM-907**Apogee MiniMe Rev. C->CA Ugly II->**Edirol OCM R-44->Tascam DR-22WL->Sony TCD-D8


"Don't ever take an all or nothing attitude when it comes to making a difference
and being beautiful and making the world a beautiful place through your actions.
Every little bit is registered.  Every little bit.  So be as beautiful as you can as often as you can"

"It'll never be over, 'till we learn."
 
"My dream is to get a bus and get the band and just go coast to coast. Just about everything else except music, is anti-musical.  That's it.  Music's the thing." - Jeb Puryear

Offline Gutbucket

  • record > listen > revise technique
  • Trade Count: (13)
  • Needs to get out more...
  • *****
  • Posts: 12607
  • Gender: Male
Re: M/S Processing for non m/s sources
« Reply #9 on: October 16, 2018, 02:54:57 PM »
Whatever works.  If the center bass sum is tighter and you don't want to make a reduction in overall bass level but only wish to tighten things up you can boost the lows in the Mid channel to compensate for the cut you make in the Side channel.

However, in a good sounding situation it's generally advantageous to go the opposite route, boosting the low frequencies of the Side channel.  Two reasons for that- one is that typical stereo mic'ing often leads to weaker than optimal low frequency difference information.  The other is that increased difference at low frequencies is advantageous in taking advantage of the perceptual nature of human hearing, increasing the sense of being there, immersed in the recording space, which is enhanced by sufficinent low frequency difference info.

So try it both ways, cutting the lows in the Side/Difference versus cutting them in the Mid/Sum before going with whatever works better.


Doubtless, folks will find online a lot of info and plugins specifically designed to sum or mono-ise low frequency content, which is a frequency-specific Mid/Side manipulation.  That may have a place in some live music recordings, but mostly comes from other areas which don't apply as strongly to taping-

Vinyl records cannot support strong difference information, especially at the lowest frequencies.. and can't go super low regardless.  Difference information is conveyed via the vertical modulation of the record groove, and a particularly strong difference signal can cause the needle to jump out of the groove.  Also, increasing the playtime per side requires further reduction of vertical modulation depth to squeeze the grooves more tightly together.

EDM and other low synth heavy music typically monoize the low bass content as a way of maximizing low frequency loudness by keeping the woofers in-phase at those frequencies, increasing power transfer though the impedance mismatch between the driver cone area and the air it is pushing, and radio and other broadcast services do the same for somewhat different reasons.
musical volition > vibrations > voltages > numeric values > voltages > vibrations> virtual teleportation time-machine experience
Better recording made easy - >>Improved PAS table<< | Made excellent- >>click here to download the Oddball Microphone Technique illustrated PDF booklet<<

Offline Gutbucket

  • record > listen > revise technique
  • Trade Count: (13)
  • Needs to get out more...
  • *****
  • Posts: 12607
  • Gender: Male
Re: M/S Processing for non m/s sources
« Reply #10 on: October 16, 2018, 03:37:28 PM »
Here's a fun thing to play around with for those of you running two (or more) pairs of microphones and/or mixing multiple sources-

Do an initial mix the way you typically would, then set that aside for later comparison. In listening to the two sources when making that mix, determine which has better clear, direct sound versus which conveys the ambient/diffuse material in a better way. Make a second mix which includes narrowing the width of the more clear/direct source and widening the ambient/diffuse source.  Play around with the width of both in combination to see if you can improve the mix this way, both in terms of imaging clarity and "you are there" feel.

This can sometimes work really well by emphasizing the better qualities of each source, even to the extent that such a manipulation would be too much when either pair is listened to in isolation, yet when combined the two balance each other and blend better with less conflict.. or it might just produce a smeared mess.  Tread carefully and use your ears to find the most appropriate blend of the two.  Generally I'd suggest narrowing the width (increasing the sum) of your closer-spaced/more-narrowly-angled/more-directional source, and increasing the width (increasing the difference) of the wider-spaced/wider-angled/less-directional source.

An opposite approach to that is to listen specifically to the Left/Right stereo imaging locations of each source in isolation, adjusting the width of one so as to match the perceptual sense of the direction to each sound source in the other as closely as possible.  This is more about matching clear and precise specific imaging locations between sources without smearing those apparent locations.  However, in doing this we increase the degree of overlap between the two sources such that other conflicts may arise- congested imaging, timbral issues, less overall sense of envelopment, etc. 

Personally, I find the prior approach more useful most of the time, which in a way is the application of Mid/Side to mixing in a general sense.  But do whatever sounds right, take a break, then compare against your initial straight-up mix to see if what you came up with is actually an improvement or not.
« Last Edit: October 16, 2018, 03:43:38 PM by Gutbucket »
musical volition > vibrations > voltages > numeric values > voltages > vibrations> virtual teleportation time-machine experience
Better recording made easy - >>Improved PAS table<< | Made excellent- >>click here to download the Oddball Microphone Technique illustrated PDF booklet<<

Offline noahbickart

  • phishrabbi
  • Site Supporter
  • Trade Count: (31)
  • Taperssection All-Star
  • *
  • Posts: 1893
  • Gender: Male
  • So now I wander over grounds of light...
Re: M/S Processing for non m/s sources
« Reply #11 on: October 17, 2018, 06:30:58 PM »
Here's a great example.  wforwumbo made a great recording of Phish in Albany last night, you can find it here: http://bt.etree.org/details.php?torrentId=602184

To my ears, there's just a touch too much room in the recording. So I ran it through the Klanghelm VUMT Deluxe (https://klanghelm.com/contents/products/VUMTdeluxe/VUMTdeluxe.php), and dropped the S signal by 1.5db, and then compensated for the overall decrease in volume by switching the same plugin back to L/R and raising both L and R. It's subtle, but definitely cleans it up a little.

I love being able to use plugins for playback, and that klanghelm one is *always* in the mix
« Last Edit: October 25, 2018, 09:25:20 AM by noahbickart »
Recording:
Capsules: Schoeps mk41v (x2), mk4v (x2), mk22 (x2), mk3 (x2), mk21 & mk8
Cables: 2x nbob KCY, 1 pair nbob actives, Darktrain 2 and 4 channel KCY extensions:
Preamps:    Schoeps VMS 02iub, Naiant Littlebox, Naiant IPA, Sound Devices Mixpre6
Recorders: Sound Devices Mixpre6, Sony PCM m10

Home Playback: Mytek DSD 192> McIntosh MC162> Eminent Tech LFT-16; Musical Fidelity xCan v2> Hifiman HE-4XX / Beyerdynamic DT880

Office Playback: Grace m903> AKG k701 / Hifiman HE-400

Online kuba e

  • Site Supporter
  • Trade Count: (1)
  • Taperssection Regular
  • *
  • Posts: 198
  • Gender: Male
Re: M/S Processing for non m/s sources
« Reply #12 on: October 18, 2018, 05:28:00 PM »
This is interesting thread. I would like to ask about phase cancellation when doing M/S equalization for spaced pair of mics:

- If I equalize M/S for lows, the phase cancellation can appear only in the low region (the region that was touched by eq)? Or the phase cancellation can appear in all regions (included regions that was not touched by eq)?

- Is the lower probability of phase cancellation when eq bass than eq height?

Offline Gutbucket

  • record > listen > revise technique
  • Trade Count: (13)
  • Needs to get out more...
  • *****
  • Posts: 12607
  • Gender: Male
Re: M/S Processing for non m/s sources
« Reply #13 on: October 18, 2018, 07:31:57 PM »
Not sure I follow. Is this something you notice this when doing Left/Right equalization?  If not, it's not something you will notice when doing Mid/Side (sum/difference) equalization.  The best practical answer I can give is to listen carefully when you make the adjustment to determine if it is doing anything other than what you want.

Standard (minimum-phase) EQ causes some degree of phase shift which is confined to the EQ'd region.  And as I understand it, the amount of phase shift depends on the gain and Q of the filter.  However, you are unlikely to hear that as cancellation per se.


With regards to spaced microphones and the phase-relationship between channels-
For direct sound arriving along the plane perpendicular to an imaginary line between the microphones, the content in both channels will be coherent and in-phase across the full frequency range. For direct sound arriving from off-axis, and for diffuse/reverberant sound which arrives essentially from all directions equally, the content is in-phase at the lowest frequencies and transitions to increasing degrees phase shift as frequency increases, becoming essentially randomized once the degree of phase shift exceeds 180 degrees.  The frequency at which this trend begins to occur is determined by the spacing between the microphones and the angle of arrival.  This applies to any spaced pair, a near-spaced directional pair of microphones as well as a wide-spaced pair of omnis, the difference between the two being the corner frequency at which that begins to occur, which will be lower for wide-spaced omnis, and considerably higher for a near-spaced pair.  Either way, the sum of the two channels (Mid) favors the in-phase components and the difference of the two channels (Side) favors the randomized phase components.
musical volition > vibrations > voltages > numeric values > voltages > vibrations> virtual teleportation time-machine experience
Better recording made easy - >>Improved PAS table<< | Made excellent- >>click here to download the Oddball Microphone Technique illustrated PDF booklet<<

Offline ycoop

  • Trade Count: (2)
  • Taperssection Member
  • ***
  • Posts: 335
  • Gender: Male
Re: M/S Processing for non m/s sources
« Reply #14 on: October 19, 2018, 01:02:05 AM »
^If I’m understanding that correctly, that explains why my first attempts at pseudo-MS processing showed peaks of around +3 dB from the low end (maybe to around 200 Hz) and the side around 0 dB (though now that I'm writing this I'm not sure what the reference dB is).
« Last Edit: October 19, 2018, 03:44:02 AM by ycoop »
Mics: Avantone CK-1s, CA-14 omnis
Pres: CA9100
Recorders: DR-60d mkII, DR-22wl, DR-07

 

RSS | Mobile
Page created in 0.263 seconds with 42 queries.
© 2002-2018 Taperssection.com
Powered by SMF