Double M/S setup - Any Experience here?

[boojum]

I am getting set up to run a pair of Schoeps Mk4's 180 apart with the diaphragms aligned and above that a Schoeps Mk8 for a double M/S setup. Schoeps has a plugin on their site to be downloaded (I did it) to decode this matrix. The advantage seems to be that you can vary reverb and field width independently. Maybe there is some other smoke and mirror stuff in the plugin, too. Anyway, has anyone run this setup, Schoeps or not, and what were their experiences? I am hoping to cut down the error curve. The learning curve will be another tale.

I plan to do field recording of small to medium acoustic groups with this setup. Guidance welcomed!


Thanks.

[Corran]

Sounds like it's pretty simple...just do it and if you don't like it you can not use the mic facing backwards.

So does the mic facing the crowd get phase reversed? If so then that combined with the mic facing forward should just become what a figure-8 mic would be in that position.

If you don't flip the phase then I guess you could actually do double M-S with just two figure-8 mics and split them / change the phase to do the same thing you are doing. I'll have to try that one day.

[boojum]

Originally Posted by Corran

Sounds like it's pretty simple..<snip>

So does the mic facing the crowd get phase reversed? If so then that combined with the mic facing forward should just become what a figure-8 mic would be in that position.

<snip>

OK, here is a link to a 38 page paper on the subject. http://schoeps.de/E-2004/PDFs/Schoep...leMS_Paper.pdf
I so love the German thoroughness, and wish more had followed down the genealogical lines to me, but it hasn't. ;o) The dematrixing software is run as a plugin in to SAM9 and higher and other DAW packages. I don't know how it handles the back, or reveberent, field. But it offers the option of standard M/S and 5.0. The link for the short blurb explaining the dematrixer is here: dmsplugin1_E.

I know it works; the Germans leave little or nothing to chance. I have the parts to make it go. I am just hoping to avoid a few of the beginner's pitfalls on starting this. Maybe I will not be able to.

I do know I am very pleased with the Mk4 - Mk8 M/S so far. I have only used it a couple of times but it has been just great. I do plan to record the tracks raw to disk and manipulate them in post with a Voxengo plugin for M/S decoding. I do a lot of on the fly setups that do not allow the time to do a practiced setup, level tests and so on. This M/S has been the kindest setup I have used. I worked my way through ORTF, NOS, Din, DINa and the Williams arrays with omni and card setups. Now I am trying the M/S setup for my on the fly setups. So far, so good. What will happen with the double M/S will be pretty exciting. Yup.

[Westmalle]

Can someone compare this setup to the Soundfield microphone?

[JEGG]

Originally Posted by Corran

So does the mic facing the crowd get phase reversed?

Polarity inverted? I don't think so. It would not sum correctly. The mids are additive.

I haven't used the plugin as there's no Mac version. Too bad. Others I know have, and they love it, including people who have the Soundfield. I haven't heard any comparative comments from them. And it isn't the same as the Soundfield approach.

BTW, if anyone is curious, there are some LD mics where the outputs of the dual diaphragms are available separately-Pearl makes one or more.

Other SD mics come to mind as well: The Schoeps dual cardioid meant for stereo use would be perfect(ly easy) for this, with the two cardioids set at 180 degrees. There's a Sanken stereo dual cardioid SD mic that might work well too, if the capsules will go to the 180 position.

Other note: The Schoeps figure 8 is not absolutely symetrical. Each half has a slightly different frequency response/polar pattern.

Other note: The Schoeps plugin is intended specifically for Schoeps mics. It has built in EQ (for the figure 8) and level compensation for the Schoeps mics-maybe they compensated here with EQ as well for the non symetrical nature of the figure 8 mic, but I don't know.

Those I know who use the plugin do not use Schoeps mics with it. They have no problem getting things matched.

[Corran]

Hmm, I might try this next orchestra concert with my pair of oktava 012s and a oktava ribbon for the figure-8...

Sounds interesting!

[boojum]

I have been reading about this in Barttlett and Eargle, etc., and it does seem the right lobe of the figure has its polarity inverted. I cannot give the citation here, but can later if someone wishes.

Schoeps, in their paper (link posted above), claims their technique is more stable that Soundfield so long as you are not looking for a vertical dimension. In the horizontal plane they feel their technique is superior. This is not too surprising and may be true. I do not have Soundfield hardware so I cannot make the appropriate tests.

As soon as I get the hardware I will start playing with, umm, testing, the array.

[Westmalle]

Maybe we should organise a comparison? I live in Belgium and have a Soundfield.
I am VERY curious to hear the Double MS of Schoeps.
There is also a nice Josephson double MS mike

[Shadowgames]

this looks interesting.

i don't have access to a schoeps mk8, so i was thinking of trying it with other mics.

apart form the obvious mounting issues, is there any reason why this technique wouldn't work well with say an AKG 414 for the fig.8 and two Gefell M300's for the cardiods?

has anyone else tried this?

[boojum]

If there is that much other interest I would suggest reading the very dry and technical report from Schoeps which is downloadable from their site: http://schoeps.de/E-2004/PDFs/Schoep...leMS_Paper.pdf. I am sure it will work with other mics, and I am equally sure that Schoeps is not go to push the idea. And all the testing they have done is with their standard Mk4 - Mk8 setup plus the second Mk4. Regardless, the physics is the same. The decoder will work on any rig, but is set up for Schoeps. It can be tweaked so it should work for any rig.

It seems a very interesting avenue. I am excited that I stumbled upon it and have the hardware to do it. Well, I am only two-track capable now but have a used SD 722 in the mail to yoke to my current SD 722. That will do it, and allow for a spot mic, too, which Schoeps also talks about in the setup, IIRC.

Williams, Barttlett and all the rest make it clear, as does Schoeps in their paper, that there is no one perfect setup. But this setup seems like a way to go, is simpler than the Soundfield and way cheaper and may yield similar results.

I will find out from the Schoeps end. Westmalle will find out from the other end. ;o) I will keep you folks posted. My biggest problem right now is getting a mounting system to work with the Schoeps Collete mics. Schoeps in their experiments use just the caps with those high-priced cables connecting them to the hardware. I am going to get it done with the regular setup. It may not be optimum, but it will work and will show the system.

I will keep you posted with the simple bluegrass and folk music from around here and any chamber music that happens. I hope I never grow up. ;o)

[pkautzsch]

If you sum the Front M and the Rear M, you'll basically get ONE M signal with omni pattern as a result. That's the same principle as in dual-diaphragm LDCs. If you phase reverse the Rear M, you'll get ONE fig-8 signal; if phase reverse Rear M and attenuate it by, say, 6 dB, you'll get a hypercardioid pattern as a result.
So far, you're just doing what any dual-diaphragm mic can do - but you're doing the pattern selection in the mixer/DAW instead of the mic.
Matrixing this Summed M (Front + Rear) with the S will, of course, yield different SRAs depending on the resulting M pattern (as can be read in William's Stereophonic Zoom paper), as the level differences depend on the M pattern.

What makes double MS so interesting is that it's a surround setup which is very small and very controllable. Using it for surround means: Front M + S = L, Front M - S = R, Rear M + S = LS, Rear M - S = RS. A center channel can be derived from Front M directly, a .1 LFE channel can come from whatever you like, even from the sum of all mics, as there are no phase differences (well, except that Rear M would appear inverted when summed with Front M). So you can get a 5.1 signal with just 3 mics. Using the small CCM mics Schoeps suggests allows you to have the whole setup on a single microphone boom, eg. for documentaries.

[Corran]

I just read the paper, and it's informative but I little wordy

However I think this could be pretty neat as a more tweak-able M-S rig. I am recording an orchestra next weekend and I'm going to try this out as I said. The paper also gave a suggestion that I've done before as well: since cardioids are a little weak in the LF range, he suggested adding an omni and low pass it to add in that frequency (or you could use it for the .1 signal!).

[Yannick]

We've been doing that with a double MS array with Sennheiser MKH mics or with a Pearl DS60 mic - for surround.

The Schoeps plugin looks nice, but it crashes in our Soundscape soft - which means its VST implementation is dodgy ...

We always augment the rear with a AB couple of omni's, a few meters behind the main array.

[danijel]

Originally Posted by pkautzsch

Using it for surround means: Front M + S = L, Front M - S = R, Rear M + S = LS, Rear M - S = RS. A center channel can be derived from Front M directly

(OT) pkautzsch, i've found this kind of matrix gives too much M, so i tend to separate them simply like this: S = L, -S = R, M = C

[Corran]

That's just M-S, not double.

Someone mentioned earlier that two cardioids facing opposite each other would basically just become an omni. I usually use an omni for my M mic to begin with, so would doing the Double M-S yield similar results (and therefore be pointless)?

[David Rick]

Double M/S has been widely taught in seminars for public radio engineers. In that context, it was intended as a stereo technique, with the rear pickup folded to the front. I think the idea was to come up with something that would produce acceptable results despite limited setup time and having to monitor on headphones. With M/S, you can tweak the stereo spread after the fact, and with Double M/S, you can tweak the amount of hall sound independently. Mono compatibility is a given.

But like Yannick, I prefer to augment a single-point array with some kind of spaced pair to get some lateral decorrelation. Given four channels and adequate rigging time, I'd probably use a standard stereo array (maybe M/S, but often not) for the front, and a spaced pair of rear-facing cardioids for the surround feed. Let's not be dogmatic though -- if it sounds good, it is good.

David L. Rick
Seventh String Recording

[MBishopSFX]

I've used double M/S for surround orchestral sessions many times over the years. Each M/S pair faces somewhat to the sides of the hall and each pair is panned as follows: Pair 1 LF/LS, Pair 2 RF/RS. For fun you add a third M/S pair facing the ceiling for the vertical dimension. Decoding is a matrix-ing circus! None of the M/S pair is the main pickup for the ensemble on-stage.

.

[JEGG]

Originally Posted by Corran

That's just M-S, not double.

Someone mentioned earlier that two cardioids facing opposite each other would basically just become an omni. I usually use an omni for my M mic to begin with, so would doing the Double M-S yield similar results (and therefore be pointless)?

Not pointless because you can't independently control front and rear.

[JEGG]

Originally Posted by pkautzsch

So you can get a 5.1 signal with just 3 mics. Using the small CCM mics Schoeps suggests allows you to have the whole setup on a single microphone boom, eg. for documentaries.

Actually with only two mics, if you count the Schoeps dual cardioid that goes 180 degrees as one mic. Same with Sanken but I don't know if it goes as far as 180. And Pearl, too, if you count the independently available diaphragms in one housing as one mic.

[boojum]

Originally Posted by pkautzsch

If you sum the Front M and the Rear M, you'll basically get ONE M signal with omni pattern as a result. That's the same principle as in dual-diaphragm LDCs. If you phase reverse the Rear M, you'll get ONE fig-8 signal; if phase reverse Rear M and attenuate it by, say, 6 dB, you'll get a hypercardioid pattern as a result.
So far, you're just doing what any dual-diaphragm mic can do - but you're doing the pattern selection in the mixer/DAW instead of the mic.
Matrixing this Summed M (Front + Rear) with the S will, of course, yield different SRAs depending on the resulting M pattern (as can be read in William's Stereophonic Zoom paper), as the level differences depend on the M pattern.

What makes double MS so interesting is that it's a surround setup which is very small and very controllable. Using it for surround means: Front M + S = L, Front M - S = R, Rear M + S = LS, Rear M - S = RS. A center channel can be derived from Front M directly, a .1 LFE channel can come from whatever you like, even from the sum of all mics, as there are no phase differences (well, except that Rear M would appear inverted when summed with Front M). So you can get a 5.1 signal with just 3 mics. Using the small CCM mics Schoeps suggests allows you to have the whole setup on a single microphone boom, eg. for documentaries.

Peter, I do not know enough to agree or disagree. If this were true, though, I wonder why Schoeps did not approach the problem in this manner, essentially a Blumlein rotated horizontally 45 degrees. I am sure they talked about it, and their paper refers to this. But the paper is so technical and I am so untechnical that I will have to re-read it a few times. I am also mildly dyslexic so that makes it more difficult with technical papers. I will persevere.

This technique does sound promising. And for what I want, sweet stereo, it may be a great tool. David Rick alluded to this.

I also plan to run either a Jecklin or a spaced array of omni's with this setup just to see what I can do with the mix. I had not planned to get into sophisticated mixing but it seems I will.

Thanks, all, so far for your ideas and support. I have a 722 now, have another arriving this week, and a 788T as soon as they are shipped, so I will be able to do the spaced omnis/Jecklin + 2M/S when the 788T arrives.

This is so much more fun than canasta with the old farts in the senior center.

[Shadowgames]

Originally Posted by MBishopSFX

None of the M/S pair is the main pickup for the ensemble on-stage.

so you're just using the m/s pairs to generate some sort of surround image in addition to a 'standard' stereo rig on the ensemble?

[JEGG]

Originally Posted by boojum

If this were true, though, I wonder why Schoeps did not approach the problem in this manner, essentially a Blumlein rotated horizontally 45 degrees.

The double MS as called for in their software plugin requires two cardioids front and rear-you cannot derive a front and rear independently controllable (nor even not independently controllable) cardioids. This is definitely not a Blumlein array turned 45 degrees and matrixed MS.

Also, it's not the same as an omni in practice. Besides the front and rear independence issues, you would have to deal with the HF directivity of the omni-at the "back" there would be as much as a -12dB drop off. An omni with a HF boost could be used pointed straight up, but then you'd get what you're pointing at. Yes, a noisy tiny omni would work, but you'd still not have the front and rear independence.

[MBishopSFX]

Originally Posted by Shadowgames

so you're just using the m/s pairs to generate some sort of surround image in addition to a 'standard' stereo rig on the ensemble?

My double M/S setup is specifically for surround pickup in conjunction with the front mics. I use it for the rear channels and to help bridge front to rear imaging.

.

[Kewl]

I mentioned this elsewhere today, but I made sort of a copy of the Schoeps Double MS plug-in, for Mac OS X, VST and Audio Unit.
Ambisonic Studio | DMS2Five Plug-in

- Daniel

[boojum]

Daniel, I am impressed. I saved a link to your page on 2MS for future reference.

I got my second 722, but it was sick and is off to the factory for a small repair. I have the third CMC6 and a hokey setup on a Sabra mic rack so I am ready to go. This is going to be fun. And a long learning exercise. I am hopeful I will be able to make standard stereo have some good ambiance.

[recordinghopkins]

I have a flute/sax/piano trio recording coming up and I thought I would revive this thread to get some slutz input... I am thinking about implementing a double MS technique for my main array, and could use some suggestions.


The budget is relatively small, so please don't suggest that I go out and get such-and-such mic. The Mics I have for this session are:

2 Shure SM80 or 2 modified Shure SM81 capsules
2 Groovetubes gt55 Cardioid LDCs
1 Sterling audio st59 multipattern LDC
1 Royer SF-12 Stereo Ribbon
2 Coles 4038 Ribbons

I am considering the two gt55's and st59 for the double MS, perhaps flanked by the omni sm80's.
SF-12 for piano spot and the two 4038's for flute/sax spots.

This may or may not be mixed surround, but I would like to have the option.

Maybe you would do it differently with the mics I have? Let's hear it.

[aracu]

An older "double ms" formula is to use one near and one far ms array. Easy to set up with two stereo mics.

[9sbean]

I did 2 different double MS setup.

Standard CMIT5 double MS, incorporating a CCM4, and CCM8. Very good result.

Another double MS: CCM4, CCM2, and CCM8:

CCM4 facing forward, as M
CCM8, certainly S
CCM2, this one play double role: Combined with phase reversed CCM4 as rear facing M. And LPF for subwoofer for the superior distant LF performance.

However, I cannot use Schoeps plug-in for this is not a standard setup to decode.

For 2 channel stereo, I still perfer ORTF or AB than MS.

[boojum]

I just got two of the plastic pieces used to carry the CCM4 over the CMC68 for M/S. I am hoping to be able to rig them in some way to carry a rear-facing Mk4 attached to the KCY cable and likewise a Mk4 w/KCY cable facing forward. I will have to cut the smaller clip off the bigger clip and attach it to a piece of plastic or carbon fiber for the rear-facing mic. I plan to take the plastic piece to a local jeweler to make the cut. They have fine saws and can do it better than I. This plastic pieces are not cheap. And I think I got the last two Jerry at Posthorn had. It is not a high-volume item.

[Yannick]

Just to react to the pic above:

double MS should be a point-source setup, not a spaced setup. That will never work ...