Monitor Mid-Side While Recording?

[AdrianW]

I normally do this by recording three tracks at once... One center track, two tracks from figure 8 mic, one panned left, one right, phase switched on one. While recording I can monitor the stereo effect and make sure its balanced by adjusting my position. Is there a way to do the same thing using a M/S plugin or is that something used only after recording?? If it can be done where should the plugin go in the chain and how should it be used? Again, I want to hear the actual midside effect while recording. Thank You.

[endorka]

I've done this with one of the Reaper MS decoder plugins. Think it works on a single 'stereo' (in fact MS as recorded) track. Set track monitoring on, make sure you are not listening to the MS signal from zero latency mixer on your audio interface. Job done.

[AdrianW]

Thanks. I have the plugins, just haven't been able to figure out where to put it while recording. The manual method works alright but I figured the plugin might make it easier. They make sense after the tracks are mixed to stereo but I wanted to hear it real time. I have the hardest time ending up with a mid-side recording that's evenly balanced and fixing it after the fact, while it looks good on the meters, just sounds wrong.

[Philbo King]

3 tracks? From 3 mics? True MS is only two channels. Try inverting polarity on the right mic and mixing it to mono with left mic.

The'S' is L-R. If you're doing it right the S track will end up 6 to 10 dB weaker than the M track for a 'very' stereo source, and will get really weak for a mono source in the center.

Monitoring is simple: monitor the 'M' track, which is mono. 'M' is L+R.

[richie43]

Also remember that recording in a true mid-side format is a very different thing than applying a mid-side plugin to a mono or stereo source. In fact, some plugin developers prefer to call the plugin process as "sum and difference", since that's what it really is.

[AdrianW]

Originally Posted by Philbo King

3 tracks? From 3 mics? True MS is only two channels. Try inverting polarity on the right mic and mixing it to mono with left mic.

The'S' is L-R. If you're doing it right the S track will end up 6 to 10 dB weaker than the M track for a 'very' stereo source, and will get really weak for a mono source in the center.

Monitoring is simple: monitor the 'M' track, which is mono. 'M' is L+R.

No,THREE tracks from TWO mics. One center, two from figure 8 panned left and right with reversed phase on one. In other words I just record the two figure 8 tracks at once instead of making a copy later. It's the same result but this way I can monitor in stereo while recording and correct the mic position as needed. The RENDER, of course, is to two channel stereo. I haven't found a way to monitor in stereo using two tracks and a plugin which was my original question.

[SonicAxiom]

Originally Posted by AdrianW

I normally do this by recording three tracks at once... One center track, two tracks from figure 8 mic, one panned left, one right, phase switched on one. While recording I can monitor the stereo effect and make sure its balanced by adjusting my position. Is there a way to do the same thing using a M/S plugin or is that something used only after recording?? If it can be done where should the plugin go in the chain and how should it be used? Again, I want to hear the actual midside effect while recording. Thank You.

In the digital realm, converting L/R stereo to MS stereo and vica-versa is a lossless process. Mid channel is simply the L signal summed with R signal which is then attenuated by 6 dB: M=(L+R)/2). Side signal is left channel summed with polarity-flipped right channel: S=L+(-R), thus the difference signal between L and R.

Most convenient way of recording an MS source signal originating from one mic set up being the M channel (having virtually any directivity) and one mic for the S channel (has to be figure 8!) is to configure a Reaper track to record the two signals via a stereo input (like inputs 1/2 of your interface) onto a single "stereo" track. You do not need to record three signals! M signal will be on the left channel, S signal on the right channel. The left channel (M signal) will usually show a considerably higher volume than the S signal on the right. If you wanna listen to the decoded LR signal, put an MS decoding plugin into the track's fx chain.

There are lots of MS decoding plugins available. I coded my own "Stereo Wizard" plugin which can not only encode or decode signals between LR and MS but which can also do some other useful things. You virtually have the full control over the stereo signal, regardless of what stereo mode the incoming signal is in. You can also independently toggle the stereo output mode of the plugin between LR and MS. The plugin is a Windows VST and can be downloaded and used for free.


download: http://www.audioworld.de/VST/SonicAx...reo-Wizard.zip

I also coded a stripped down version:


download: http://www.audioworld.de/VST/SonicAx...-%20simple.zip

As LR <-> MS conversion is lossless, you can do as many of these conversions as you like at any desired point during the production. Manipulating a 2-channel file ("stereo" recording) is much easier than handling signals on 3 separate tracks plus the required enconding/decoding configuration which may easily get messed up by accident.

If you place the plugins into the input fx chain of the track, the state of the plugin will get recorded. I wouldn't recommend to do this, though.

Very important: ALWAYS put some kind of stereo scope / goniometer / stereo correlation plugin right behind any ms processing to visually control the mono-compatability to avoid any phase issues! I'd recommend using the free FLUX Stereo Tool 3 (https://fluxhome.com/project/stereo-tool-v3/).

[G-Sun]

Depending on your interface latency,
it's a matter of monitoring through Reaper (1), or just with your interface mixer (2).

If 1, then just record to a stereo-track, put the MS decoder either on input or regular fx

If 2, then yes, you might need 3 tracks as described in your first post. The recording however is the same, 1 stereo-track.

[G-Sun]

Originally Posted by SonicAxiom

Very important: ALWAYS put some kind of stereo scope / goniometer / stereo correlation plugin right behind any ms processing to visually control the mono-compatability to avoid any phase issues!

MS has no mono-phase issues as it translates perfectly to mono (one of the big benefits of using MS rec. tech.)

[SonicAxiom]

Originally Posted by G-Sun

MS has no mono-phase issues as it translates perfectly to mono (one of the big benefits of using MS rec. tech.)

yes, but you still wanna avoid creating extremely wide stereo fx that are not perceivable in mono but will be sounding "phasy" in stereo.

[G-Sun]

Originally Posted by SonicAxiom

yes, but you still wanna avoid creating extremely wide stereo fx that are not perceivable in mono but will be sounding "phasy" in stereo.

Yes, true

[AdrianW]

SonicAxiom: This is great! Thanks so much, does exactly what I want and I can do it with a single stereo track. It's also easy to edit later if needed. The only question I have is setting the gain on each mic to get an optimum level. I understand that one side will be lower and that it can be adjusted later with the plugin but I'm wondering if there is a basic starting point for the levels of each mic. On a single mono track I usually look for a peak of -12 when recording acoustic guitar. Not sure how that might equate to mid-side. Thanks again.

[SonicAxiom]

Originally Posted by AdrianW

SonicAxiom: This is great! Thanks so much, does exactly what I want and I can do it with a single stereo track. It's also easy to edit later if needed. The only question I have is setting the gain on each mic to get an optimum level. I understand that one side will be lower and that it can be adjusted later with the plugin but I'm wondering if there is a basic starting point for the levels of each mic. On a single mono track I usually look for a peak of -12 when recording acoustic guitar. Not sure how that might equate to mid-side. Thanks again.

Glad it's working for you, Adrian.

I'd say that with a decent interface you may set the gains for both mics identically (with the M mic picking up more volume usually; -12 dB peak seems fine). This will give you more or less the original stereo width of the performance during playback (not taking into account a possible difference in sensitivity between the two mics). You could also set the S mic's gain a few dB higher than that of the M mic to take the usually lower volume on the side channel into account and get a better signal/noise ratio for the S channel. This will produce a slightly exaggerated stereo width in the recorded file but you can always reduce it to taste during mixing.

Just never set a gain level that might eventually risk clipping one of the signals if suprisingly high levels are produced during the performance! A headroom of 12 dB should be enough.

[AdrianW]

Thanks to all. Finally and at the risk of sounding incredibly stupid I have one more question. When I did midside recording in Reaper manually I always knew which track was the side and which was the mid. I'd just duplicate the side, change the phase and pan each left and right and then render to stereo. I never record to a single stereo track directly, always to two mono tracks and pan as needed. When recording to a single stereo track from two mics and then decoding the result to midside with a plugin wouldn't you have to be sure which side each mic is panned to? For example, is it standard that the mid should always be on the left by default so the decoder works properly?

[SonicAxiom]

Originally Posted by AdrianW

Thanks to all. Finally and at the risk of sounding incredibly stupid I have one more question. When I did midside recording in Reaper manually I always knew which track was the side and which was the mid. I'd just duplicate the side, change the phase and pan each left and right and then render to stereo. I never record to a single stereo track directly, always to two mono tracks and pan as needed. When recording to a single stereo track from two mics and then decoding the result to midside with a plugin wouldn't you have to be sure which side each mic is panned to? For example, is it standard that the mid should always be on the left by default so the decoder works properly?

yes, mid signal (L+R) always goes to the left channel, side signal (L+(-R)) to the right channel by convention. If you ever encounter a swapped MS signal for some reason (S on the left, M on the right), you can use my "Stereo Channel Selector" plugin or my "Stereo Wizard" plugin to swap channels back to normal. A swapped MS signal has no use.

In an MS stereo mic setup, the S mic (which has to be figure 8 polar pattern) captures all the directional information and determins the stereo image (LR or RL) by its own orientation: It should be positioned so that its front points to the left of the sound source when looking from behind the mics towards the sound source. If you recorded something in MS with the figure 8 mic facing in the wrong direction (front of mic pointing to the right side of the sound source) you'll simply get a swapped LR stereo image with the left signal appearing on the right channel and the right signal on the left channel. This may not even get noticed by anybody and is usable as long as it is not mixed with other sources that do respect the original stereo orientation.

MS is not rocket science

.

[G-Sun]

Originally Posted by AdrianW

is it standard that the mid should always be on the left

Yes.

[AdrianW]

I was sure the left/right routing mattered but couldn't find any reference to it for any of the matrix plugins, which seemed odd. Many plugins allow flipping the channels but some don't. When I did this manually it was easy to know which was which. This explains it. Thanks to all.