M-S (mid-side) recording in a small room? Matrix?

[666666]

Never tried M-S recording before but I'm interested. Thing is, the next thing I'll be recording will be in a very small room. If I set up mics for M-S, the figure 8 mic will be only about 5 feet from the side walls. I am wondering if this will simply just not work out... perhaps the figure 8 mic might pick up too much junk from the walls...???

Also, I do have figure 8 mics and cardioids... and have some good pres... but how do you achieve the "M-S matrix"? Is there a special "matrix unit" that needs to be purchased? I think I remember reading that the old Amek 9098 dual pre unit had an M-S matrix built in... any other mic pre units have this? How do I achieve the matrix?

Thanks

[hv_]

Some mic pres like the True Systems P2 and P8 have it built in but its pretty easy to rig a mixer to do it. Just use 3 faders and route the figure-8 to 2 of them panned left/right, and flip the phase switch on one of the 2. Then route the mid mic to the third. You can do basically the same thing in a daw. Or use a plugin... Voxengo makes one called MSED that's free.

You might want to put up some temporary room treatment. Hang blankets, mattresses, sofa cushions, etc. Or pick up some baffles like Aurelex and others make.

Howard

[Mike Brown]

Not sure what you mean by "the matrix" but to decode M-S without any special devices do this:

Take your mid mic and route it normally...
Take your side (the mic in Fig-8) mic and pan it left, then take a copy of the same side mic, invert the polarity, and pan it right.

You can make the copy with half-normals on the patchbay, set multiple outs of your DAW, mults, etc etc etc...

[DrDeltaM]

Originally Posted by Mayor999

Not sure what you mean by "the matrix" but to decode M-S without any special devices do this:

Take your mid mic and route it normally...
Take your side (the mic in Fig-8) mic and pan it left, then take a copy of the same side mic, invert the polarity, and pan it right.

You can make the copy with half-normals on the patchbay, set multiple outs of your DAW, mults, etc etc etc...

Or do it with a plugin in your DAW which can give you additional control like my DrMS plugin: mathewlane.com - DrMS spatial processor plugin

[Mike Brown]

Originally Posted by DrDeltaM

Or do it with a plugin in your DAW which can give you additional control like my DrMS plugin: mathewlane.com - DrMS spatial processor plugin

Does the additional control include anything you couldn't do with MORE control by processing your MS or LR busses with L+R and L-R summation and using seperate plugs?

What does this plug offer that makes it an instant buy other than a very reasonable price?

Looks like a cool plug so sell me on it! Its been too long since i've spent some money :(

[DrDeltaM]

Originally Posted by Mayor999

Does the additional control include anything you couldn't do with MORE control by processing your M-S or L-R busses with L+R and L-R summation and using seperate plugs?

What does this plug offer that makes it an instant buy other than a very reasonable price?

Looks like a cool plug so sell me on it!

The Mid and Side sections are pretty 'standard', but the Focus and Field sections are where the money is! (tho it IS very affordable )
With Focus and Field you can work on the 3D image and depth of the sound-space of stereo-material.

There's a demo to download, so you can try it out yourself

[666666]

So, to simplify the recording process, can I just do the following while recording:

Record the "mid" mic to one track, and record the figure 8 mic to another track... period... and then deal with the "M-S decoding" later in the mix?

In the mix, I'd simply have to duplicate the figure 8 track, then flip the phase on one of them... then I'd have the three tracks needed for M-S stereo manipulation in the mix, correct?

I ask this because I'd like to at least TRY this s few times before I start obtaining specific M-S items. I do not record through a console, nor do I like sending tracks through DAW busses etc while recording. I'd much prefer to record just the two mics directly to a track each (as I normally would), then deal with the "decoding" later.

The only problem with doing this... I would not be able to hear / monitor the actual "M-S" decoded sound while positioning the mics / recording... that is a downside for sure, but I can live with it during the experimental stage. If it turns out that I really dig M-S recording in my small room situation then I can look into the plug-ins or a mixer etc.

Am I on the right track?

Has anyone ever run M-S in a small room? I'd certainly treat the heck out of the side walls, but I still wonder if there might be issues. After all, the figure 8 mic is technically focused right AT both walls... so it'll pick up the walls just as much as it'll pick up the source. I guess with extremely heavy wall treatment, ZERO reflections, it might be ok. Experiences?

[Mike Brown]

Originally Posted by 666666

So, to simplify the recording process, can I just do the following while recording:

Record the "mid" mic to one track, and record the figure 8 mic to another track... period... and then deal with the "M-S decoding" later in the mix?

Has anyone ever run M-S in a small room? I'd certainly treat the heck out of the side walls, but I still wonder if there might be issues. After all, the figure 8 mic is technically focused right AT both walls... so it'll pick up the walls just as much as it'll pick up the source. I guess with extremely heavy wall treatment, ZERO reflections, it might be ok. Experiences?

Yes you can just record the two and decode later.... I would print about 5 seconds of whatever it is and quickly do the whole duplicate and copy thing and listen to the 5 seconds to make sure you are happy. THEN track the song.


Small rooms can work just as well as large ones. Try positioning the performer in a corner as well as against a wall/in the center... that way the walls will be at 45 degree angles instead of 90 degree angles. Move the mics and the performer around til you are happy.

Don't try to kill all the reflections... they are what makes it sound cool!

[666666]

Another thought...

In general, is M-S appropriate to capture an 8 foot wide source at a distance of say 5 feet away? I started thinking that my whole situation (both room and source) might be too "small" to be "appropriate" for M-S technique. Thoughts? Sorry if these are silly questions, but I have zero experience with M-S, I don't think I've ever even used a figure 8 pattern for that matter... this is all a bit new to me. I may not have too much extra time to experiment with this, so I'd like to get an idea in advance if it's worth pursuing at all in my "small" situation. Thanks!

[666666]

Originally Posted by Mayor999

Yes you can just record the two and decode later.... I would print about 5 seconds of whatever it is and quickly do the whole duplicate and copy thing and listen to the 5 seconds to make sure you are happy. THEN track the song.

Small rooms can work just as well as large ones. Try positioning the performer in a corner as well as against a wall/in the center... that way the walls will be at 45 degree angles instead of 90 degree angles. Move the mics and the performer around til you are happy.

Don't try to kill all the reflections... they are what makes it sound cool!

Cool. Thanks!

[PVAUDIO]

I dont think anyone mention this but should'nt you listen in mono and bring the two figure 8 mics up together until they cancel eachother out? Ounce you do that, increase or decreasing the two figure 8 channel will widen or narrow your stereo image...

what mics to people use.

I like a u87 and a 121 or a m-160

[recall]

It depends on what kind of setup you are recording.

I tracked in a similar room last summer. It was a folk setup:

acoustic guitar
male vox

female vox

I miced them up as normal with them kind of 65 degrees facing each other

stood back a few feet and placed my m/s setup where I thought it sounded best to capture the natual acoustic of the room.

It was a small living room with marble floor but plenty of diffusion.

I got a really cool close and intimate natural recording using this method.

[platypipros]

Originally Posted by PVAUDIO

I dont think anyone mention this but should'nt you listen in mono and bring the two figure 8 mics up together

Are you talking about before tracking or during mix....correct me if i'm wrong but m-s only has 1 figure-8 mic for the sides and one mid mic(cardiod, hyper-card, etc.)

[666666]

Seems you guys are digging the room ambience / reflections captured by M-S... it would appear that the room ambience / reflections are an important part of the M-S technique.

In my case, I do NOT wish to capture ANY room reflections, the room is very small, too small to sound good so it's heavily damped... I do not wish to pick up the smallness of the room... my main goal is to capture the source ITSELF only (drums / percussion array) with good natural stereo. The instruments span about 8 feet wide and I'll likely need to place the mics somewhere between 4 and 6 feet away from the front of the instrument array. The room is only about 12 feet wide.

I guess the answer is that I just TRY it and see what happens. But, again, just curious if anyone ever tried exactly what I'm trying to do and discovered that it was a total failure... or perhaps worked just fine. The super small room and lack of ambience / reflections makes this a special and perhaps doomed case with respect to M-S.

[666666]

Originally Posted by platypipros

....correct me if i'm wrong but m-s only has 1 figure-8 mic for the sides and one mid mic(cardiod, hyper-card, etc.)

Yes, one mid mic and one figure 8 mic... but the figure 8 mic track gets duplicated so there is TWO tracks of this one mic. This is how I understand it.

[PVAUDIO]

Originally Posted by platypipros

Are you talking about before tracking or during mix....correct me if i'm wrong but m-s only has 1 figure-8 mic for the sides and one mid mic(cardiod, hyper-card, etc.)

The figure 8 mic gets copied. (I use a malt if working with analog console, if ITB, i just copy the mic to another audio track.

I've done M-S micing both ways... just recording the two mixes and decoding during mixing, or decoding it pre-tape (or the virtual tape). In both cases I always listen in MONO to make sure the two cancel out. This is the way i was taught to do it, I'm pretty sure i was told why, but i forgot... can anyone give some input on the reason for this.

[666666]

Another concern...

When recording a stereo source using M-S, typically how will the M-S signals blend with OTHER mics in the room? I may be running a pair of XY and/or AB in addition to the M-S on the same source (in different positions). Being that M-S uses a phase flipped pair of tracks (post decoded), I'm wondering if this will cause havoc with respect to an XY pair that might also be in the room.

So, can anyone share experiences using M-S along with other patterns in the room at the same time? Does M-S play well with other mics in the room? The phase flip aspect of M-S is what has me concerned.

Thanks

[PVAUDIO]

In my experience with doing both XY and M-S, the M-S is more for the stereo image of the drums. The XY more for cymbals. I think the two together sound great.

Of course you wanan check the phase between the two. But flipping the phase on the 2nd figure 8 channel isn't going to effect the relationship between the two miking tech.

[Mike Brown]

Originally Posted by 666666

Another concern...

When recording a stereo source using M-S, typically how will the M-S signals blend with OTHER mics in the room? I may be running a pair of XY and/or AB in addition to the M-S on the same source (in different positions). Being that M-S uses a phase flipped pair of tracks (post decoded), I'm wondering if this will cause havoc with respect to an XY pair that might also be in the room.

So, can anyone share experiences using M-S along with other patterns in the room at the same time? Does M-S play well with other mics in the room? The phase flip aspect of M-S is what has me concerned.

Thanks

It either plays well right off the bat, or will play well once you invert the polarity of a couple things.

Its the same as anything else... here is the test: Push button (or turn dial) make audio sound good? If yes... keep button pushed.

[andychamp]

Originally Posted by 666666

In my case, I do NOT wish to capture ANY room reflections(...)

Then I suggest you record in an anechoic chamber.

Seriously, one point of M/S stereo IS to capture room/ambience information, and the room often plays a big role in drum sounds.

The M(iddle) mic, depending on distance to the source, and whether it's omni, cardioid, hypercardioid or even fig.-8 will give you a more or less direct/focussed drumsound.
The decoded S(ide) tracks will give you the room sound.

By varying the M to S ratio, you will change the liveness of the recording.
Increasing the percieved size of the room through compression also works with M/S.
Different compression and EQ schemes for M and S can give interesting results, too.

[andychamp]

Note to self: must try M/S drums with 1 Omni and 1 fig.-8 placed close to the drummer (like just over his head, or between him and the snare, or between snare and toms, or...)

[666666]

Originally Posted by andychamp

...Seriously, one point of M/S stereo IS to capture room/ambience information, and the room often plays a big role in drum sounds...

Yes, that was my concern. My problem here is that I will not have a good room to use. The room that is to be used is so small that it simply cannot be made to sound good, nothing but super close reflections, the type that are so close they just about cause phase issues. Wish I had a larger room for this. Maybe next time.

I am hoping that the "side" mic will actually capture some actual stereo information from the 8 foot wide array of drums. In other words, I'd like to think that the left side of the figure 8 would pick up extra ride cymbal and floor tom, and the right side of the figure 8 would pick up extra hit-hat and snare in order to help reproduce a true stereo picture of the drumkit itself (not just room ambience). So, perhaps M-S is not for me? I wonder how well the figure 8 will pick up the drums themselves in stereo.

By the way I DID already experiment with XY and AB (spaced pair)... I had issues with both.

XY yielded a nice tight focused accurate sound of the drums, but very little stereo width (remember, this is a small dry room and the mics can't be more than about 5 feet away from the source). THe mics were pointing at the kit at less than a 90 degree angle, but if I went any wider, the mics would have been pointing towards the walls. I pointed the mics at the far outer ends of the kit... and the angle of the mics may have been around say 75 degrees apart. Just wasn't enough width for me.... but i very much liked the tightness and accuracy.

The spaced pair yielded a SUPER exaggerated ultra wide stereo field, but at the expense of a somewhat blurry phasey inaccurate sound. The perceived width was tremendous, total "ear candy", but no focus, lack of attack, etc.

So, I was hoping that perhaps M-S would possibly give me best of both worlds... tight accurate focus and ALSO decent stereo width, but from the drums themselves, not from room reflections.

Fire away!

[Hammer Mark]

I'm doing M/S with two C414s. The resulting stereo recording is much louder on the left than the right. The M+S is louder than either the M or S and the M-S is much quieter. I realize there's some positive and negative interference going on with the sum and difference. Is there any way to avoid this or limit its effect?

[DrDeltaM]

Originally Posted by Hammer Mark

I'm doing M/S with two C414s. The resulting stereo recording is much louder on the left than the right. The M+S is louder than either the M or S and the M-S is much quieter. I realize there's some positive and negative interference going on with the sum and difference. Is there any way to avoid this or limit its effect?

It is normal that your M-signal is much louder then S signal. Some M/S encoders compensate this, but you need a matching decoder obviously. I chose to implement the standard M = L + R and S = L - R in my DrMS plugin. There's plenty of resolution in 24bit recordings that it's no problem to have the M and S level difference.

[Hammer Mark]

Originally Posted by DrDeltaM

It is normal that your M-signal is much louder then S signal. Some M/S encoders compensate this, but you need a matching decoder obviously. I chose to implement the standard M = L + R and S = L - R in my DrMS plugin. There's plenty of resolution in 24bit recordings that it's no problem to have the M and S level difference.

I'm not talking about a difference between M and S, but between M+S and M-S.

[kludge]

I've used M/S in a small, less than perfect room and LOVED it. I decoded by recording the two mics as a stereo track, then using the free Kelly Industries StereoTools VST. I love that VST; use it all the time. Besides M/S decoding, I strap it across the output buss for mono and phase reversal checks a mouse click away.

I've found two good uses for it in my world... acoustic guitars, and group background vox.

[DrDeltaM]

Originally Posted by Hammer Mark

I'm not talking about a difference between M and S, but between M+S and M-S.

Sorry had read your post too fast it seems...

It is VERY weird that you get a stereoshift on decoding as:

M + S = L + R + L - R = 2L
M - S = L + R - (L - R) = L + R - L + R = 2R

So if your M and S signal is correct, you can't get a shift.

Now, in the case the level of M and S are incorrect, the outcome will still be a correct stereo signal, just smaller or wider, but not out of balance.

So what you record self must be out of balance, or your mic-placement is not correct.

[Corran]

Originally Posted by Hammer Mark

I'm doing M/S with two C414s. The resulting stereo recording is much louder on the left than the right. The M+S is louder than either the M or S and the M-S is much quieter. I realize there's some positive and negative interference going on with the sum and difference. Is there any way to avoid this or limit its effect?

This might seem obvious, but are you sure that the S mic is actually sideways? Since the 414 has switchable patterns it might not be immediately apparent that it must be at a 90 degree angle to the other 414. This is also not meant to be insulting, as assuming you recorded it correctly and matrixed it right, there shouldn't be a large level discrepancy assuming you weren't recording something on the far left/right side!

[Hammer Mark]

Originally Posted by Corran

This might seem obvious, but are you sure that the S mic is actually sideways? Since the 414 has switchable patterns it might not be immediately apparent that it must be at a 90 degree angle to the other 414. This is also not meant to be insulting, as assuming you recorded it correctly and matrixed it right, there shouldn't be a large level discrepancy assuming you weren't recording something on the far left/right side!

I don't think I did the matrix correctly. I put M+S on the left and M-S on the right. Should it be +S on the left, -S on the right and M up the centre?

[Corran]

Yes, and then what results is M+S on the left and M-S on the right. So you had doubled the M mic as well and panned them hard left/right?