I understand more or less how mid/side works, or at least it's fundamental difference from left/right, but what advantage does it offer over left/right stereo? And what sort of things factor into choosing between the two?

Does a specific m/s plug in such as Voxengo MSED work in a similar way to a typical stereo widener/spread/enhancer etc? I noticed that Logic's Direction Mixer has the option of switching between L and R, and M/S...

Perhaps i should get some experience making decent L/R stereo mixes before i start looking at mid/sign. Or maybe not?

M/S widening is inherently mono compatible. Which is great for mixes that will be played on radio, TV etc, where weaker signals are collapsed to mono, even if there's more than one speaker. It's also a useful means of stereo-linking (in a manner of speaking) two mono compressors. It enables you to process central elements (eg bass, kick, vox...) with one setting, and wide-panned stuff with another. So if that's what you want to do, this is the way to go.

There is no 'typical' widener. They work in different ways (M/S balance, delays, phase shifts etc), and sometimes combine these different processes.

But... what is it you're trying to do? That's probably the best question to answer first.

Quote PeasantPunch:

---

...what advantage does it offer over left/right stereo? And what sort of things factor into choosing between the two?

---

It allows you to process central sounds with a greater degree of independence from those sounds panned out towards the edges (and vice versa) which can be very useful. For example, to gain better dynamic control of the kick drum or vocals without squasing some wide spacious guitars or synth pads... It also allows you to adjust the apparent stereo width very easily.

Quote:

---

Does a specific m/s plug in such as Voxengo MSED work in a similar way to a typical stereo widener/spread/enhancer etc?

---

Yes, simple stereo wideners and enhancers work by converting the LR signal to MS, adjusting the relative balance of M and S to achieve the altered stereo width, and then converting back to LR. (more elaborate ones mess about more comprehensively with phase, delays asnd EQ... but that's another story!)

Quote:

---

Perhaps i should get some experience making decent L/R stereo mixes before i start looking at mid/sign. Or maybe not?

---

It's a tool like all the others, and worthwhile getting to understand when it can be usefully applied.

Hugh

--------------------
Technical Editor, Sound On Sound

Quote Mixedup:

---

M/S widening is inherently mono compatible.

But... what is it you're trying to do? That's probably the best question to answer first.

---

Do you mean to say there is more likelihood of problems in the mono mix if i'm using a combination of L/R panning, stereo delay and stereo widener plug ins?

I'm fairly new to mixing, i've read as much as i can, heard a lot of stuff that i don't understand or problematises what i thought i knew, and i'm trying to lay some basic ground rules/codes of practice for a nice wide (and hopefully not too messy) stereo mix.

I've got a lot of soft synth instruments in the mix, and i'm basically trying to test the water and see if there is something else i should know before going in there and trying to pull off a good stereo sound by:

1) Keeping kick, bass (lower end at least) mono
2) Widening mono soft synth sounds that are important in the mix with a simple stereo spread tool, whilst panning more decorative mono or stereo sounds left and right.
3) Leaving in built in stereo fx on certain soft synths like the Nexus2 that i like, but don't feel i can necessarily scrap and reproduce myself, and hoping that they won't cause problems in the mix.

I guess with judicious use of a stereo imager (i have a copy of Waves' S1) on tracks and the master bus (also have a copy of Izotope Ozone) i'll know if i'm doing something that stinks, whether i'm pushing it too far or whether the stereo needs boosting here and there. Again though, i'm kind of concerned as to whether, having got the mix sounding as good as i can, i can then go and start using a multi band stereo imager to widen out certain frequencies that may well already have stereo delay fx etc going on. As i said, because i'm new to this i feel i need be clearer on what i should and shouldn't be doing, so things don't get messy.

I won't be doubling any stereo tracks for Haas stereo movement (i'll probably use some stereo delay plugins, or double mono and put a chorus on one), so hopefully i should be getting any phase cancellation when i mono the L/R stereo mix.

Quote Hugh Robjohns:

---

It allows you to process central sounds with a greater degree of independence from those sounds panned out towards the edges (and vice versa) which can be very useful. For example, to gain better dynamic control of the kick drum or vocals without squasing some wide spacious guitars or synth pads... It also allows you to adjust the apparent stereo width very easily.

Hugh

---

I presume it makes it easier to achieve apparent stereo width because the "sides" are all grouped together and can be widened together, as opposed to adjusting the stereo of individual tracks? However, is it not just as straightforward to widen the sound on the master bus by taking those same things that belong at the "sides" (i.e. certain mids to uppers) further with a multi band stereo imager, as you would do with an m/s plug in? I guess the advantage of the m/s plug in is that it already recognises what is at the "sides" and will only widen those things that are already there, whereas the mutliband stereo will widen ANYTHING that occupies a certain frequency range and will widen sounds that might possibly need to stay centred?

Also, does the m/s widening of sides from the mid not cause gaps in between?

Quote PeasantPunch:

---

Do you mean to say there is more likelihood of problems in the mono mix if i'm using a combination of L/R panning, stereo delay and stereo widener plug ins?

---

What I mean is that some widening tools work by introducing delays and phase shifts. What happens then is that when the signal is collapsed to mono some frequencies in the L and R channels will cancel out, and you'll notice a difference in tonality. Whether that's a problem is down to your ears — and there are plenty of great-sounding hits that sound very different in mono versus stereo. The mono versus stereo mix is always a trade off in any case, but such processes can make finding the right balance more problematic. Altering the balance of the M and S signals will give the effect of widening or narrowing the stereo image without introducing any such problems.

Whatever you use, it makes sense to keep checking the mix in mono as you go along, just to make sure you're getting the balance broadly right and that there are no surprises. A good stereo phase meter can help too, as it makes it easy to see how far you can push things without causing significant problems. I really like Brainworx bx_digital v2 for this, but there are lots of useful options available, including a freeware one by Flux.

Have a read of this SOS article if you haven't already done so.

Mid/side processing is definitely an advanced technique.
And yes you should be able to make good sounding mixes before you even consider it.
Mid/side processing is not going to fix your mix. It is kinda of the icing on the cake at mastering stage.
The first benefit from it IMHO would be to tighten the bass in the center.

Here is a video where Fab shows how to use Mid/Side processing:
Mastering with the Dangerous Master

Edited by chavernac (23/12/11 12:39 AM)

Quote chavernac:

---

Mid/side processing is definitely an advanced technique.

---

Really? I find it very simple and easy.

--------------------
John - Sound-Link ProAudio
President - Federation Internationale des Chasseurs de Sons

That's probably because you've been doing it for thirty years, John!

Most people do struggle at first to get their head around the concept and then the practical implementation and application.

Hugh

--------------------
Technical Editor, Sound On Sound

Yes it is a difficult concept at first. Be careful using it at mastering. If possible try to get a recall of the mix and fix the issues there. Yes, I know that is not always possible, you might be treading on toes! But don't go putting MS processing across your own mix. Better to use it on certain elements, occasionally, as a special effect.

--------------------
madtheory creations
Synths and pianos for Kontakt

Quote Hugh Robjohns:

---

That's probably because you've been doing it for thirty years, John!

Most people do struggle at first to get their head around the concept and then the practical implementation and application.

Hugh

---

I actually found it pretty easy from the start.

But I did buy the excellent AES publication on Stereo Microphone Techniques (still in print) and I remember the Wes Dooley and Ron Streicher papers were very good.

--------------------
John - Sound-Link ProAudio
President - Federation Internationale des Chasseurs de Sons

Quote John Willett:

---

I actually found it pretty easy from the start.

---

I thought I'd have to join the Stone Cutters (Simpsons reference) to understand this esoteric technique, but after the first time I did it I found it utterly obvious.

The key is doing it. I've found some very technically minded people (I do not include myself in that group!) experience glazed over eyes when contemplating the mysteries of MS, but once done it is learned forever.

BTW, my first mic's were multi-pattern specifically for getting all the techniques down. Not exactly top shelf, but I still get use from them, and they were an invaluable education.

Quote chavernac:

---

Mid/side processing is definitely an advanced technique.
And yes you should be able to make good sounding mixes before you even consider it.
Mid/side processing is not going to fix your mix. It is kinda of the icing on the cake at mastering stage.
The first benefit from it IMHO would be to tighten the bass in the center.

Here is a video where Fab shows how to use Mid/Side processing:
Mastering with the Dangerous Master

---

Ok, so m/s allows you to process mid and side separately, so you can widen the sides whilst keeping the mid dead centred, and the compression, for example, can be applied to the mid signal to further tighten things up.

Is there a way of feeding effects into the mid and side separately on the Voxengo MSED then, because i'm left wondering how you can do this with just a gain control for the side and the mid? Also, it does seem strange that simply increasing the volume of the sides actually has the apparent effect of widening. I suppose its a trick of the mind..

On an aside, I put a multi band stereo imager on a master out and was amazed at the general transformation with just a subtle tweak to the top two upper bands. But just by widening the top two bands the bass guitar seemed to lose its place at the rock centre of the mix, and to lose a tiny bit of punch. However, the main vocals also also seemed to gain presence by some subtle multi banding, even though the text book says they're meant to be left at the centre. I guess there is a certain amount of leeway for widening vocals before they start to thin? Before i used the multi bander the vocals were in mono with a small amount of stereo reverb, and perhaps this is too narrow within a stereo mix? All a matter of taste i guess...

I kind of want to keep the stereo multi band (ozone) on the master because i like what it has done to the mix in general. I suppose a solution to keeping the bass rock centre in this scenario is to group all the instruments that benefit most from some upper widening, leave out the bass, and then re-apply to multi band. However, it seems like quite a long winded way of doing it. ??

If i could get the same stereo results with the voxengo mid/side as i have done with the ozone multi band then i'd go ahead and use that given the advantages people have mentioned, but if i can sort out the fact the bass is still getting widened by the top two frequency bands the ozone multi band has impressed me more, given the instant results.

Is there somewhere i could look to improve my technique with the vox MSED, or another m/s plug in? I'm playing a novices game in terms of technique, but i have good ears!

Quote PeasantPunch:

---

Is there a way of feeding effects into the mid and side separately on the Voxengo MSED......

---

That's what the plugin does............

By chance, did you miss the little box on the interface which toggles among "in line," "decode," and "encode?"

Quote PeasantPunch:

---

I kind of want to keep the stereo multi band (ozone) on the master because i like what it has done to the mix in general. I suppose a solution to keeping the bass rock centre in this scenario is to group all the instruments that benefit most from some upper widening, leave out the bass, and then re-apply to multi band. However, it seems like quite a long winded way of doing it. ??

If i could get the same stereo results with the voxengo mid/side as i have done with the ozone multi band then i'd go ahead and use that given the advantages people have mentioned, but if i can sort out the fact the bass is still getting widened by the top two frequency bands the ozone multi band has impressed me more, given the instant results.

Is there somewhere i could look to improve my technique with the vox MSED, or another m/s plug in? I'm playing a novices game in terms of technique, but i have good ears!

---

Think about narrowing the LFs as much as about widening other bits. Plenty of plugins offer a bass-to-mono facility with control of the turnover frequency. This will ensure equal distribution of LF power across the speakers and avoid those frequencies .being affected by any MS widening treatment. Alternatively, male sure any sources you don't want affected by widening are mono and panned centrally. Impressive-sounding bass presets with lots of FX to create width will inevitably suffer when widening... but you can always use that track's stereo pan pots to compensate.

Quote Jeraldo:

---

Quote PeasantPunch:

---

Is there a way of feeding effects into the mid and side separately on the Voxengo MSED......

---

That's what the plugin does............

By chance, did you miss the little box on the interface which toggles among "in line," "decode," and "encode?"

---

I must be completely missing the point here, or i'm failing to read in between the lines of advice people have kindly offered. I can see the "in line," "decode," and "encode" options, but i still don't understand how this mid/side plug in is used to allow you to apply effects to the mid/side separately.

Like Hugh said earlier, mid/side "allows you to process central sounds with a greater degree of independence from those sounds panned out towards the edges"...

All i can do so far is adjust the mid/side gain separately.

You need to use a plug-in to convert the stereo signal to M/S, process each of the resulting (mono) signals as necessary, then convert back from M/S to stereo again.

In my case I often use M/S to send out to a valve equaliser, so my insert chain might be:
MSED (convert from stereo to M/S)
External plug-in (to my dual mono EQ - left processing Mid, right processing Side)
MSED (convert from M/S to stereo)

HTH!

--------------------
An Eagle for an Emperor, A Kestrel for a Knave.

Quote Mixedup:

---

Plenty of plugins offer a bass-to-mono facility with control of the turnover frequency.

---

You mean a stereo multi band with a bass to mono/stereo narrowing facility? Any recommendations?

Quote PeasantPunch:

---

Quote Mixedup:

---

Plenty of plugins offer a bass-to-mono facility with control of the turnover frequency.

---

You mean a stereo multi band with a bass to mono/stereo narrowing facility? Any recommendations?

---

Yes. They're not really doing complex maths so any plugin that offers this should do the trick. I like Brainworx bx_digital v2 for widening/narrowing. There was a freebie Basslane by Otium FX. I think they're called Tone Projects now, and not sure if they still do it. Sounds like you might also enjoy Melda's multiband stereo widener. DDMF Metaplugin is worth a look too, as it shops with encode/decode and multiband splitter plugins and you can therefore create complex M/S chains quite easily.

I like the brainworks too!

Quote The Elf:

---

You need to use a plug-in to convert the stereo signal to M/S, process each of the resulting (mono) signals as necessary, then convert back from M/S to stereo again.

In my case I often use M/S to send out to a valve equaliser, so my insert chain might be:
MSED (convert from stereo to M/S)
External plug-in (to my dual mono EQ - left processing Mid, right processing Side)
MSED (convert from M/S to stereo)

HTH!

---

So i presume by simply creating an insert chain on the master out of:

MSED "encode"
Logic Match EQ (left channel selected)
Logic Match EQ (right channel selected)
MSED "decode"

...i'd be able to to eq mid and side separately.

However, i'm struggling to find a compressor with the same option for choosing between L and R. Logic doesn't seem to have one, nor does Waves..

How would i go about applying compression to the mid and side separately in a chain like the above?

Just use a compressor in dual mono (ie unlinked) mode in between the encode and decode plugins. Not sure if Logic has only stereo comps or offers dual/multi channel mono, but there are plenty of third party ones that do. If it doesn't, as I said before DDMF Metaplugin is great for this sort of routing.