Topic: Portable Preamp with M/S ?

[boa]

I've been looking for a Sound Devices Mixpre-D to do M/S on the fly.

Are there any other options I should be considering?

[morst]

I've been looking for a Sound Devices Mixpre-D to do M/S on the fly.

Are there any other options I should be considering?

Consider mixing it later!!!!

[yug du nord]

Schoeps VMS
Grace Lunatec V3 with M/S mod

[morst]

Consider mixing it later!!!!

Sorry for my glib reply!
Perhaps you'd get added benefit from a Mixpre3 which I am pretty sure can record the ISO tracks and also deliver a stereo mix / live stream?

[Gutbucket]

^ This.  It really is best to be able to dial in the optimal Side to Mid ratio while listening later. 

With most multichannel recorders you can produce a live L/R monitor output while recording as Mid and Side channels, preserving the capability of dialing in a more optimal M/S mix later.  I've used the following recorders that way: Edirol R44/R88, TascamDR680/DR680MKII, Zoom F8 (I assume F6 too but have not used it), and the SD Mixpre recorders (I have not used them, but they have the capability), among others. 

With recorders that provide the capability of simultaneously recording the internal stereo mix along with the isolated tracks you can even produce file outputs in both the M/S and (live mix ratio) L/R versions at the same time if you want.  Recorders with that ability include all of the above except the R44.

And yes, if necessary you can always convert from M/S to L/R ahead of the recorder and record as L/R, and then convert back to M/S on the computer later to tweak the ratio by ear, before finally converting back to L/R again, but best to mix directly from the isolated Mid and Side signals if you have them recorded that way.

[tgos3]

used Sound Devices MP-2.  Similar to original MixPre, but M/S instead of L-C-R mike assignment.   Lundahl transformer inputs.  All analog. can monitor and output encoded or decoded M/S.  Can monitor decoded while outputting encoded.
uses AA batteries or external DC (oldest units with barrel/concentric ext. port, later ones Hirose)

mine has worked without probs for ~20 years.  I use eneloops or DIY modified wall wart.

https://www.sounddevices.com/product/mp-2/

sorry if this is a double post -- I don't see my earlier attempt.

[DSatz]

joining the chorus, judging over headphones how a recording is going to sound over loudspeakers is completely impossible when it comes to the spatial aspect of things. Yes, for every M/S setup there is an exactly corresponding X/Y setup, but to make adjustments by ear with an X/Y setup you can (for example) spread the axes of the mikes farther apart, but with the equivalent M/S setup, the corresponding adjustment is exactly what? It would involve a combination of changing the M mike's pattern and tweaking the ratio of S to M gain in the matrix that you're using. And that's just an example (of an adjustment that I make all the time when recording X/Y or closely-spaced).

The point being, it makes little sense to commit to any particular M-to-S ratio during a live recording unless you have monitoring facilities that most of us don't get to have on location most of the time. So record the M and S channels directly, monitor them through a matrix so that you hear L and R, but matrix the actual recording only in post. Otherwise you have to matrix it twice (once to re-derive M and S from L and R, then once to go back to L/R stereo once you've decided what sounds best).

Also I just want to say that double M/S (where a third, backward-facing directional microphone is added to a traditional M/S pickup, and fancier matrixing is applied) gives a much wider range of possibilities than traditional M/S. In my experience with traditional M/S, for any given listening setup, there was only one M-to-S gain relationship that gave a plausible balance between the width of the stereo image and the amount of reverberation in the finished recording. Unfortunately that one control (the gain on the S channel going into the matrix) controls BOTH of the only settable parameters of the recording, interdependently. Double M/S breaks that dependency by, in effect, making the pattern of the forward-facing microphone variable "after the fact". It does require three recorder channels and a software plug-in (or a hardware solution would be possible, though the software approach is more versatile and it's $free).

[voltronic]

joining the chorus, judging over headphones how a recording is going to sound over loudspeakers is completely impossible when it comes to the spatial aspect of things. Yes, for every M/S setup there is an exactly corresponding X/Y setup, but to make adjustments by ear with an X/Y setup you can (for example) spread the axes of the mikes farther apart, but with the equivalent M/S setup, the corresponding adjustment is exactly what? It would involve a combination of changing the M mike's pattern and tweaking the ratio of S to M gain in the matrix that you're using. And that's just an example (of an adjustment that I make all the time when recording X/Y or closely-spaced).

The point being, it makes little sense to commit to any particular M-to-S ratio during a live recording unless you have monitoring facilities that most of us don't get to have on location most of the time. So record the M and S channels directly, monitor them through a matrix so that you hear L and R, but matrix the actual recording only in post. Otherwise you have to matrix it twice (once to re-derive M and S from L and R, then once to go back to L/R stereo once you've decided what sounds best).

+1 on all of this, and to what others have said above. To me, the biggest reason for using M/S in the first place is the ability to tweak the direct / ambient ratio in post. I also can confirm that you get a VERY different picture of this when using monitor speakers vs. headphones. My advice is actually to use both: Go with the monitors first to get your balance where you want it, but then confirm that your mix translates well via headphones.

Also I just want to say that double M/S (where a third, backward-facing directional microphone is added to a traditional M/S pickup, and fancier matrixing is applied) gives a much wider range of possibilities than traditional M/S. In my experience with traditional M/S, for any given listening setup, there was only one M-to-S gain relationship that gave a plausible balance between the width of the stereo image and the amount of reverberation in the finished recording. Unfortunately that one control (the gain on the S channel going into the matrix) controls BOTH of the only settable parameters of the recording, interdependently. Double M/S breaks that dependency by, in effect, making the pattern of the forward-facing microphone variable "after the fact". It does require three recorder channels and a software plug-in (or a hardware solution would be possible, though the software approach is more versatile and it's $free).

I have always wanted to try this. If I understand correctly, the two directional mics function as a variable bipole, whereas in traditional M/S or Blumlein, the mid element is a variable dipole? The "singing guitarist" scenario described here is a pretty compelling argument for the flexibility of double-M/S.
https://www.soundonsound.com/techniques/double-mid-sides-array

[Gutbucket]

Best to conceive of Dual Mid/Side as providing a "variable pattern Mid" I think.  "Variable bipole" is something of an oxymoron - if you vary it, it then becomes some other pattern than bipolar, AKA figure-of-8.

The key take-away with Dual Mid/Side is that it allows for independent control of pattern and X/Y angle. Whereas with standard Mid/Side those two things are always permanently linked - unfortunately with a relationship that is the opposite of the way we would prefer them to be linked.. which is why there is generally only one obvious point where the ratio between Mid:Side produces good stereo.

Dual Mid/Side solves that problem, at the cost of one additional recording channel and cardioid microphone feeding it.  Not a bad trade for the ability to choose any pickup pattern for each microphone AND any angle between them with complete independence.  "Any angle" really does mean that one can point the virtual microphones in any horizontal direction, backward as well as forwards. One can even point each side independently rather than having to keep everything symmetrical, providing a virtual rotation of the stereo array.  Point the whole thing farther to the left or right as necessary to best balance the stereo image.

Lots of options afterwards! Maybe too many for some.

To me, the biggest reason for using M/S in the first place is the ability to tweak the direct / ambient ratio in post.

Except you can do that with ANY stereo recording, regardless of microphone configuration used.

Edit to clarify (see following posts)- Actually, you can tweak the Mid:Side ratio of any recording in post, regardless of microphone configuration used..  Just be aware that some recordings will allow for more adjustment than others without problems.  But even with coincident-stereo recordings which avoid those potential problems, changing the Mid:Side ratio modifies several stereo aspects at the same time, and because of that there may not be much range of adjustment where all aspects work well.  More fundamentally though, recording using Mid/Side alone won't really give you much useful control over direct /ambient ratio afterward.  Dual Mid/Side however can give you some control, essentially allowing you to increase ambience to taste (not so much the other way) based on the direct:ambient relationship present at the recording position in the room, as can other configurations with additional microphones that are dedicated to the pickup of ambient sound with attenuated pickup of direct-arriving sound from the front.

It's just that if you record Mid and Side separately you can do so in a better way without the additional steps required to get back to Mid and Side again.  And it tends to work best when applied to material recorded with a coincident or relativity close spaced near-spaced microphone configuration (rather, those configurations tend to allow a wider range of Mid/Side stereo re-adjustment before other problems arise).  It can certainly be tried on any recording and minor adjustments might be beneficial even with wide-spaced configs, perhaps to correct a hole in the middle.  Just keep your ears open for other problems arising while dialing it in.

[voltronic]

^ Thanks for setting me straight on those points, GB.

I get what you mean about "variable bipole" being a poor choice of words. I was only saying that to differentiate between fig-8 in the sense that the front/rear directional mics in double-M/S are in phase, rather than out of phase as they are on a fig-8 mic, which is a dipole.

Has anyone seen double-M/S used for acoustic ensemble recording before? It seems like there would be a lot of benefits to it. Someday I hope to purchase a good fig-8 mic so I can try it out.

[Gutbucket]

Yes, it is the manipulation of the level and polarity of the forward and rearward facing cardioids prior to their summing that creates the variable Mid pattern and determines if it is facing forward or rearward.

Used with the same polarity the resulting Mid pattern ranges between cardioid and omni.  Used in opposed polarity the resulting Mid pattern ranges between cardioid and fig-8.

[DSatz]

Voltronic wrote:

> To me, the biggest reason for using M/S in the first place is the ability to tweak the direct / ambient ratio in post.

to which Gutbucket replied:

> Except you can do that with ANY stereo recording, regardless of microphone configuration used.

Well, yes, kinda sorta. You can certainly go through the same motions--deriving L+R and L-R signals, then rematrixing them in some other proportion. And that will vary the ratio of direct to reverberant sound in the recording. But if the recording wasn't made with a coincident main pair, that kind of adjustment will create comb-filter effects in both channels and screw up the whole high-frequency and upper-mid half of your recording, depending on how far apart the main microphones were and on how much you change the original proportions of your derived M vs. your derived S when you rematrix to L/R stereo.

The only way to make this kind of processing work without destructive interference between the channels is to start with a coincident main pair, whether M/S or X/Y. You have to decide in advance what your priorities are for a given recording. Do you want a groovy spacious feel in your recording? Then space your microphones apart at least somewhat. Or do you want the post-processing flexibility? Then use a coincident main pair (and record any spot or fill-in microphone signals onto separate recorder channels, so that you can blend them in after you've chosen your settings for the main pair).

But the sad fact is, you can't really have both at the same time, at least not in the way that I've always wished for. Double M/S, like Ambisonics, comes closer to giving you real post-processing options that work on the esthetic/feeling level--still not unlimited ones, but certainly less constrained than traditional M/S.

--best regards

[Teen Wolf Blitzer]

I have a V2 with ms mod.  It's #13.  From Lucas Sound.  I would trade for a regular V2 (+) as I never use it.

[Gutbucket]

Thanks for pointing out the potential problems.  To clarify a bit further..

I don't find a Mid/Side ratio adjustment of an existing stereo recording can "tweak the direct / ambient ratio in post" in any meaningful way.  However I do find it can be useful to make small adjustments in stereo image width and feel that can be beneficial.  In the end, its a tool that can be abused like any other, best approached with an ear open for potential problems as well as the potential benefit.  Notably, the same potential problems can arise with any other form of stereo post processing applied in a Mid/Side rather than Left/Right manor.  And yet such use is relatively common, as a mastering EQ technique for example.  But in that case the adjustment is typically minor and and comb filtering artifacts from it will be constrained to the targeted frequency bands.  As always, tread carefully and trust your ears.  Its post-work so you can always undo.

[Gutbucket]

Here's the thing. Dual-M/S and ambisonics won't allow us to "tweak the direct / ambient ratio in post" in any meaningful way either.  Although I've not recorded using Dual-Mid/Side, from my experience recording with an ambisonic microphone I don't think either of those techniques provide a realistic way of doing that even though the separate control over pattern and angle they provide are very useful in other ways.  I did find I could often derive two acceptable stereo outputs for any one recording: typically a narrower angled pair of super/hypercardioids and a wider angled pair of subcardioids.  The two present significant stereo differences, and one might think they might provide significantly different direct/ambient ratios, but they don't.  All stereo combinations that work well end up with the same direct:ambient ratio.

This strikes me as somewhat analogous to standard Mid/Side where there tends to be only one matrix ratio that provides good balanced stereo qualities.  Similarly, with Dual-M/S and ambisionics there can be now several combinations of angle and pattern that provide good stereo attributes, but they will all tend to have the same direct/ambient ratio.

That's because control over the necessary third degree of freedom beyond pattern and angle is still missing, and that is the spacing between the microphones.  If we have control over spacing as well we can then trade angle for spacing to some degree while retaining a similar stereo image.  That does provide some control over direct/ambient ratio without overly compromising other important stereo aspects.

Pattern, angle, and spacing.  Dual-M/S and ambisonics provide for separate manipulation of the first two, but we need all three to apply the Stereo Zoom techniques and unlink imaging and other stereo qualities from direct/ambient ratio, so as to gain some useful degree of control over it, and even then the useful degree of control is limited. 

Extending the analogy drawn earlier even further..  We really need control over pattern, angle, spacing, and position, as recording position will always be the primary determinant of the direct/ambient ratio. 

TL:DR- Control over direct/ambient ratio is best achieved by choice of recording position.  If constrained to recording from further away than what would be ideal, using directional microphones and less angle between them can help somewhat, but then one needs to compensate with more spacing between them to retain good stereo qualities.  Control over pattern and angle alone in post won't do it.