You are here

Mid Sides recording with OC818

All about the tools and techniques involved in capturing sound, in the studio or on location.

Mid Sides recording with OC818

Postby jodaki » Fri Feb 21, 2020 4:49 pm

Is it overkill to record both sides of the OC818 when used as a side mic, or would you just record the main feed and double it in the DAW (as with a normal fig 8 mic)?
jodaki
Regular
Posts: 113
Joined: Thu Feb 21, 2019 1:03 pm
Location: Somerset

Re: Mid Sides recording with OC818

Postby Sam Spoons » Fri Feb 21, 2020 5:05 pm

I can't see any advantage if it is for M/S. There benefit of recording the two separately is that you can alter the effective polar pattern in post production. For M/S you need a fig 8 for the sides component, nothing else will do so you would not be changing the virtual polar after the fact just the mix of M and S as with a 'normal' M/S recording.
User avatar
Sam Spoons
Jedi Poster
Posts: 11513
Joined: Thu Jan 23, 2003 1:00 am
Location: Manchester UK
Finally taking this recording lark seriously (and recording my Gypsy Jazz CD)........

Re: Mid Sides recording with OC818

Postby Hugh Robjohns » Fri Feb 21, 2020 5:10 pm

To use the OC818 as a Side mic in an MS array it has to have a sideways-facing fig-8 polar response -- nothing else will do -- and you can only get that by combining it's two cardioid capsules together at equal level and in opposite polarities. No other combination is acceptable in that role.

So, seeing as the mic can do that combination job perfectly for itself... why waste tracks and effort in recording the capsules separately just to recombine them to allow use as a fig-8 sides mic? It's not like you could do anything else with the two outputs in that application!

So yes... it would be overkill!

And I wouldn't bother with all that duplication and polarity inversion nonsense anymore. There's absolutely no point and no need when you can do it quicker and easier with a VST plugin -- and Voxengo's MSED does it all perfectly and for free:

https://www.voxengo.com/product/msed/

MS array into a stereo track (mid on the left, sides on the right), drop in MSED, select the decode mode, adjust S level to taste. Job done.
User avatar
Hugh Robjohns
Moderator
Posts: 26278
Joined: Fri Jul 25, 2003 12:00 am
Location: Worcestershire, UK
Technical Editor, Sound On Sound

Re: Mid Sides recording with OC818

Postby jodaki » Fri Feb 21, 2020 6:00 pm

Hugh Robjohns wrote:To use the OC818 as a Side mic in an MS array it has to have a sideways-facing fig-8 polar response -- nothing else will do -- and you can only get that by combining it's two cardioid capsules together at equal level and in opposite polarities. No other combination is acceptable in that role.

So, seeing as the mic can do that combination job perfectly for itself... why waste tracks and effort in recording the capsules separately just to recombine them to allow use as a fig-8 sides mic? It's not like you could do anything else with the two outputs in that application!

Thanks Hugh, Im all for keeping it simple - I've downloaded MSAD and will have a play with it. I have a further question though:

If the 2 opposite cardiod signals listening left and right were kept separate and each recorded to a separate track, wouldn't that give a better stereo field than summing them to mono and duplicating that track (or using a software decoder), seeing as in the latter case each side channel would contain the same information?

I get that this wouldn't be MS recording anymore but wouldn't the effect be a sort of improved MS? Or is it that the sides tracks wouldn't benefit enough from being genuinely stereo to go to the effort?
jodaki
Regular
Posts: 113
Joined: Thu Feb 21, 2019 1:03 pm
Location: Somerset

Re: Mid Sides recording with OC818

Postby jodaki » Fri Feb 21, 2020 6:06 pm

Sam Spoons wrote:I can't see any advantage if it is for M/S. There benefit of recording the two separately is that you can alter the effective polar pattern in post production. For M/S you need a fig 8 for the sides component, nothing else will do so you would not be changing the virtual polar after the fact just the mix of M and S as with a 'normal' M/S recording.

Sorry Sam only just seen your post. Yes I get now my example isn't really MS as Id not be recording with a fig 8 sides.
jodaki
Regular
Posts: 113
Joined: Thu Feb 21, 2019 1:03 pm
Location: Somerset

Re: Mid Sides recording with OC818

Postby Hugh Robjohns » Fri Feb 21, 2020 6:25 pm

jodaki wrote:If the 2 opposite cardiod signals listening left and right were kept separate and each recorded to a separate track, wouldn't that give a better stereo field than summing them to mono and duplicating that track (or using a software decoder), seeing as in the latter case each side channel would contain the same information?

Well, you can't have a 'better stereo field' when the standard MS matrix concept already works perfectly! There is nothing better than perfect, as far as I am aware*... ;-)

If you really wanted to, you could record each cardioid output separately, and then add the left one to the Mid mic to generate the decoded left output, and the right one (at exactly the same level) to the Mid mic to generate the decoded right output.

But if you do the maths, you'll see it works out exactly the same as doing the conventional MS matrix decode with a fig-8 Side mic (see below).

The only practical difference is that you'd need to record three source tracks instead of two -- which isn't very convenient -- and you'd need to set up your own decode matrix using three channels in your DAW -- which isn't very convenient -- and you won't be able to enjoy the simplicity and flexibility of using the MSED plugin to perform the decoding for you... which isn't very convenient!

In contrast, setting the Side mic to provide a (composite) fig-8 polar pattern, recording that alongside the Mid mic as two channels, passing those channels through a stereo DAW channel, and dropping the MSED plugin to handle the decode is amazingly convenient... and simple...and has perfect stereo imaging**... ;-)

I get that this wouldn't be MS recording anymore but wouldn't the effect be a sort of improved MS?

Actually, it would still be an MS recording... but in 'kit form' ...and there would be no improvement over standard MS whatsoever. Just inconvenience and more scope for error!

This diagram which I provided for Matt's article on MS in this month's magazine illustrates the decode matrix maths in a pictorial form. It shows a fig-8 for the Mid mic for simplicity.

Image

To change the Mid mic into a cardioid, you just need to draw an Omni pattern in addition to the forward-facing fig-8, with the same size as the fig-8 and the same polarity as the front of the fig-8.

MS decode cardioids.png


... and if you then work through the matrix, adding each component in the same way you'll see the decoded virtual mics become cardioids too facing 45 degrees left/right.

Now... if you do the same kind of thing, but changing the Side fig-8 for a pair of (back-to-back) cardioids, you'll soon realise that the omni-elements of those two cardioids facing hard left/right instantly cancel out because they have opposite polarities... leaving just the fig-8 element to sum with the Mid mic (in opposite polarities for each side).

...so why go to all the bother of recording separate cardioid elements that will be instantly discarded anyway, when all you need is a sideways fig-8? :lol:

H


*Unless it's recorded on analogue tape obvs... :crazy: :lolno:

**Or as perfect as it can be with the practical realities of the given mics.
User avatar
Hugh Robjohns
Moderator
Posts: 26278
Joined: Fri Jul 25, 2003 12:00 am
Location: Worcestershire, UK
Technical Editor, Sound On Sound

Re: Mid Sides recording with OC818

Postby jodaki » Fri Feb 21, 2020 6:40 pm

Ok many thanks Hugh. I will keep it simple :)
jodaki
Regular
Posts: 113
Joined: Thu Feb 21, 2019 1:03 pm
Location: Somerset

Re: Mid Sides recording with OC818

Postby Hugh Robjohns » Fri Feb 21, 2020 6:42 pm

:thumbup: Always the best approach in my experience! :lol:
User avatar
Hugh Robjohns
Moderator
Posts: 26278
Joined: Fri Jul 25, 2003 12:00 am
Location: Worcestershire, UK
Technical Editor, Sound On Sound

Re: Mid Sides recording with OC818

Postby jodaki » Fri Feb 21, 2020 11:07 pm

Im really sorry Hugh, you've answered my question already and I accept what you've said as being the right way to do it but I still don't understand one thing...

If the fig 8 mic is capturing a mono signal, how can that be expanded through decoding to equate to being exactly the same as two mics recording left and right separately?

If a cello is on my right and the bass on my left. How can a summed mono fig 8 mic reflect the balance differences coming from the right and left from the bass and cello? They'll just get summed by the mic and split and panned but the relative (volume and ambient) differences get lost.

Im struggling to see how a summed mono signal can be decoded to equate to recording both sides independently, because the summed mono signal is ignorant of the differences between the left and right signals.
jodaki
Regular
Posts: 113
Joined: Thu Feb 21, 2019 1:03 pm
Location: Somerset

Re: Mid Sides recording with OC818

Postby blinddrew » Fri Feb 21, 2020 11:38 pm

It can't on its own, it needs the mid signal as well.
User avatar
blinddrew
Jedi Poster
Posts: 9442
Joined: Sun Jul 05, 2015 12:00 am
Location: York
Ignore the post count, I have no idea what I'm doing...

Re: Mid Sides recording with OC818

Postby Hugh Robjohns » Sat Feb 22, 2020 12:01 am

jodaki wrote:If the fig 8 mic is capturing a mono signal, how can that be expanded through decoding to equate to being exactly the same as two mics recording left and right separately?

Every individual microphone captures a mono signal!

The clever bit in a Mid-Side array is the matrix decoding process which combines the two mics -- the Mid AND the Side mic -- in a sum and difference process which takes advantage of the fact that the two lobes of a fig-8 mic have different polarities. That's the unique and specific facet that allows the sum of the Mid and Sides mics to create a virtual mic facing left, and the difference creates a virtual mic facing right.

If a cello is on my right and the bass on my left. How can a summed mono fig 8 mic...

Er... There is no 'summed mono' in the Side mic... It's just a mic... Facing sideways... Capturing the sounds to the left and right of the sound stage.

However, sound sources on the left are in the same polarity as the Mid mic, while sounds to the right are in the opposite polarity with respect to the Mid mic. The decoder uses these characteristics to derive the appropriate left and right signals.

They'll just get summed by the mic and split and panned but the relative (volume and ambient) differences get lost.

You appear to be forgetting the importance of the opposite polarity lobes, and how they work in concert with the decoder's sum and difference processes.

That diagram I included above, showing how the process works with a fig-8 Mid mic, makes the concept about as simple as I can make it.

Consider the decoder's left output, which is the sum (M+S) of the Mid and Side mics. The two in phase lobes are on the top and left, so adding them together gives a virtual positive lobe facing up and 45 degrees left. The two negative lobes face bottom and right, so adding them gives a virtual negative lobe facing Down and 45 degrees right. In the two opposite corners you have one positive and one negative lobe which cancel... So the resulting virtual polar pattern is another fig-8, facing top left.

The decoder's right output is the difference of the Mid and Side mics -- meaning M-S...or, more easily understood, M+(-S). So it's the exact same process, but with the Side mic polarity inverted, so the positive lobe is effectively now facing right rather than left as far as the decoder is concerned. This results in a virtual fig-8 mic facing top right.

Hence, a correctly decoded MS array produces left and right outputs which are the exact equivalent of a conventional XY array of left/right facing mics.
User avatar
Hugh Robjohns
Moderator
Posts: 26278
Joined: Fri Jul 25, 2003 12:00 am
Location: Worcestershire, UK
Technical Editor, Sound On Sound

Re: Mid Sides recording with OC818

Postby John Willett » Sat Feb 22, 2020 11:51 am

Hugh Robjohns wrote: -- and Voxengo's MSED does it all perfectly and for free:

https://www.voxengo.com/product/msed/

MS array into a stereo track (mid on the left, sides on the right), drop in MSED, select the decode mode, adjust S level to taste. Job done.

:thumbup: This is the only plug-in I use. :thumbup:
User avatar
John Willett
Jedi Poster
Posts: 6806
Joined: Tue Mar 07, 2000 1:00 am
Location: Oxfordshire UK
John
Sound-Link ProAudio
Circle Sound Services
Sound-Link are UK Distributors for: Microtech Gefell, ME-Geithain, AETA, HUM, Håkan, Meyer Turtle

Re: Mid Sides recording with OC818

Postby John Willett » Sat Feb 22, 2020 11:57 am

jodaki wrote:Im really sorry Hugh, you've answered my question already and I accept what you've said as being the right way to do it but I still don't understand one thing...

If the fig 8 mic is capturing a mono signal, how can that be expanded through decoding to equate to being exactly the same as two mics recording left and right separately?

If a cello is on my right and the bass on my left. How can a summed mono fig 8 mic reflect the balance differences coming from the right and left from the bass and cello? They'll just get summed by the mic and split and panned but the relative (volume and ambient) differences get lost.

Im struggling to see how a summed mono signal can be decoded to equate to recording both sides independently, because the summed mono signal is ignorant of the differences between the left and right signals.

I think you need to read this paper:-
M-S Stereo: A Powerful Technique for Working in Stereo

It's by Wes Dooley and Ron Streicher and is a really excellent paper (with good diagrams) and very readable. :thumbup:
User avatar
John Willett
Jedi Poster
Posts: 6806
Joined: Tue Mar 07, 2000 1:00 am
Location: Oxfordshire UK
John
Sound-Link ProAudio
Circle Sound Services
Sound-Link are UK Distributors for: Microtech Gefell, ME-Geithain, AETA, HUM, Håkan, Meyer Turtle

Re: Mid Sides recording with OC818

Postby jodaki » Sat Feb 22, 2020 12:00 pm

Hugh Robjohns wrote:You appear to be forgetting the importance of the opposite polarity lobes...

Very generous, it was more a lack of understanding than remembering. Thanks for your patience, I get it now. Fascinating and ingenious... J
jodaki
Regular
Posts: 113
Joined: Thu Feb 21, 2019 1:03 pm
Location: Somerset

Re: Mid Sides recording with OC818

Postby John Willett » Sat Feb 22, 2020 12:03 pm

jodaki wrote:
Hugh Robjohns wrote:You appear to be forgetting the importance of the opposite polarity lobes...

Very generous, it was more a lack of understanding than remembering. Thanks for your patience, I get it now. Fascinating and ingenious... J

The Dooley and Streicher paper (link above in my previous post) explains it all - and includes excellent diagrams that show the resultant left/right stereo pattens with different mid mics at different gain levels - excellent. :thumbup:
User avatar
John Willett
Jedi Poster
Posts: 6806
Joined: Tue Mar 07, 2000 1:00 am
Location: Oxfordshire UK
John
Sound-Link ProAudio
Circle Sound Services
Sound-Link are UK Distributors for: Microtech Gefell, ME-Geithain, AETA, HUM, Håkan, Meyer Turtle

Next