naturerecordists
[Top] [All Lists]

Re: X-Y -> M-S

Subject: Re: X-Y -> M-S
From: "maxfrick78" maxfrick78
Date: Mon Mar 12, 2007 5:36 am ((PDT))
Hello MS-talkers,
John, you're making some excellent points here. Let me tune in to try takin=
g away some
part of the mythical aura of MS..
Although the way the recording setup works remains a bit magical  (how can =
that ever
become a good stereo-signal?), the decoding process to L-R is very simple. =
That's why you
can do it with just 3 tracks and a polarity-inversion. (How to set it up ha=
s been described
many times here so I won't do that again).

Here's my point: The MS to LR decoding can VERY EASILY be reversed, you don=
't need a
special mambo-jumbo plugin for that! Why? Because the setup for the convers=
ion "MS to
LR" is exactly the same as "LR to MS", except that you loose 6 dB of gain i=
n the latter. Just
try it. Take an MS recording, insert a standard MS-decoder, insert a second=
 one just
behind it and boost the level 6 dB. You will end up with exactly the same s=
ignal that you
started out with (except the lowest bits, I know, but if you record at 24 b=
its that's
somewhere around -100 dBFS which should be irrelevant if it's a proper reco=
rding. It's so
much more important to make a good recording!)

greets,
Max



--- In  "John Lundsten" <=
> wrote:
>
> From: Rob Danielson
> Hi Andy,
> Yes, several of the MS plugs including +matrix have the math ability
> to reverse the process. I believe we determined in an earlier
> discussion on this list, however, that one cannot reclaim the exact
> same mid and side signals/tracks by this means. So, save your field
> M-S original files; they are unique.
>
> Not true, The process is 100% accurate, 100% reversible  (if your doing t=
his
> digitally that is - kinda close with analog electronics) My demo "party
> piece" to demonstrate this to my students is to take 2 signals; some musi=
c
> in mono and something nasty sounding like White noise or Timecode. Then S=
um
> & difference these (IE MS encode them) then decode back to 2 very separat=
e &
> clean signals.  If you listen to either just the M or the S  you get load=
s
> of noise and barely audible music I tend to crank up the noise to emphasi=
se
> how effective this all works.
>
> BTW, Personally I tend to keep every generation, so I do save my MS
> originals.
>
> John Hartog Wrote
> "How is MS considered mono compatible, if when you sum the stereo L&R
> you are left with only the information from the original M channel? Is
> that really true? "
>
> The S chan is a record of what is DIFFERENT between the Left & Right.
> If you mono any stereo recording any diffeence is lost (its Mono yeh)
> In the case of MS this results in all the S info going, leaving you with=

> just a mic (the M) pointing on axis at the subject just as you would do i=
f
> you if you had originally made a mono recording.
> But
> In the case of XY, ORTF etc  the sum of L&R will be from 2 mics that can=

> never be tottally coincident, hense there will be some comb filtering
> cancellantion because there will be some time delay of the same sound tha=
t
> is picked up by 2 mics spaced apart.
>
> "Seems like the S channel that gets lost would normally contain a
> majority of the information (like 2/3 or something considering a
> cardioid M) of the original L&R mix (talk about phase cancellations!).
> If so that would not really be mono compatible at all, would it?"
>
> No the M picks up ALL the sound (it's just there is no info as to what
> direction(s) the sounds came from.
> The S only picks up what is DIFFERENT between L&R. A Fig 8 mic in a MS ri=
g,
> does a rather good job of not hearing sounds from straight ahead in the
> middle. IE for a sound that arrives at the same time & at the same level =
in
> both channels there clearly is no Difference, hence it doesn't get record=
ed.
> But the S does record sound other than from straight ahead, the more they=

> get to either the extreme Left or Right the louder will be their
> contribution to the S signal.
> hope that helps
> regards
> JL
>






<Prev in Thread] Current Thread [Next in Thread>
Admin

The University of NSW School of Computer and Engineering takes no responsibility for the contents of this archive. It is purely a compilation of material sent by many people to the naturerecordists mailing list. It has not been checked for accuracy nor its content verified in any way. If you wish to get material removed from the archive or have other queries about the archive e-mail Andrew Taylor at this address: andrewt@cse.unsw.EDU.AU