At 9:08 AM -0700 8/23/10, James Shatto wrote:
>.. if you were to mix M and S to L and R, then
>save as. Then load the new file and derive M
>and S and derive -S, that S and -S wouldn't
>completely cancel each other out.
Hi James--
With phase cancelling they do. You are right that
they are NOT a perfect match, sample to sample,
as I hurriedly concluded from phase cancellation.
Here are some statistics showing some
differences: http://tinyurl.com/3658x4c
>...If you're effectively adding a noticeable dB
>of junk per edit (even if it's not audible),
>enough to motivate someone to do the matrix in
>post and not in the field. ...And results
>probably vary greatly between software packages.
Could it be that the two files cancel out with
inverted phase because the differences are
smaller than rounding-off that is going on
somewhere during routine playback? I got phase
cancellation with three audio apps.
> .. But it depends on what one considers good enough.
That's always worth considering. We can slice
numbers ad infinitum and sound is the experience
of a physical condition. I'm guessing that
neither of us could ID which file was which using
any field recordings and using any real world
playback means. :-)
It seems to me that mics and speakers can have a
very profound effect on "qualities" but the gear
and digital processes in between are less subject
to quality losses. Some nature recordists create
high demand by recording in low sound level
locations where high gain is needed and where
reflected sounds make-up a large percentage of
the sound. I suspect this is why mic pre input
noise often crops up next as a quality factor.
That said, folks who mostly record loud
vocalizations with sensitive mics don't need very
low noise pres. Rob D.
>
>- James
>
>--- On Sun, 8/22/10, Rob Danielson
><<type%40uwm.edu>> wrote:
>
>From: Rob Danielson <<type%40uwm.edu>>
>Subject: Re: [Nature Recordists] M/S recording setup
>To:
><naturerecordists%40yahoogroups.com>=
m
>Date: Sunday, August 22, 2010, 11:07 PM
>
>
>
>Hi James--
>
>It worked, sample for sample using your E and C
>
>chords/files. Here is the Reaper Session I used
>
>with +Matrix plugs and with all the file
>
>mentioned below:
>
><https://pantherfile.uwm.edu/type/public/media/James_MS_TEST.zip>https://p=
antherfile.uwm.edu/type/public/media/James_MS_TEST.zip
>(8mb)
>
>I took your E chord and made that Mid
>
>I took your C chord and made that Side
>
>I made a stereo file with Left=3DMid and Right =3D Side
>
>I decoded the above file as a recorder would and
>
>created this file: MS_Decoded.wav
>
>I sent that through MS ->L/R Processing (at
>
>unity) and then L/R ->MS Processing (at unity)
>
>and produced this file: MS-LR-MS.wav
>
>MS_Decoded.wav and MS-LR-MS.wav are the same.
>
>Rob D.
>
>=3D =3D =3D =3D
>
>At 11:59 AM -0700 8/22/10, James Shatto wrote:
>
>>
>
>>
>
>>It's a bash script. That runs sox a lot of
>
>>times. AKA borne again shell and sound
>
>>exchange. I've uploaded the generated 30 second
>
>>wave files. For those without technical prowess
>
>>to run the script. My former ISPs webspace
>
>>limit is 10MB which these files exceed. Not
>
>>including the individual pitches since those
>
>>triple the archive size.
>
>>
>
>><<http://www.sendspace.com/file/yfn71i>http://www.sendspace.com/file/yfn7=
1i><http://www.sendspace.com/file/yfn71i>http://www.sendspace.com/file/yfn7=
1i
>
>>
>
>>- James
>
>>
>
>>--- On Sun, 8/22/10, Rob Danielson
>
>><<type%40uwm.edu><type%40uwm.edu>> wrote:
>
>>
>
>>From: Rob Danielson
>><<type%40uwm.edu><type%40uwm.edu>>
>
>>Subject: Re: [Nature Recordists] M/S recording setup
>
>>To:
>
>><naturerecordists%40yahoogroups.com><naturerecordists%40yah=
oogroups.com>
>
>>Date: Sunday, August 22, 2010, 10:35 AM
>
>>
>
>>
>
>>
>
>>At 1:16 AM -0700 8/22/10, James Shatto wrote:
>
>>
>
>>>I guess it depends on how you do the math. Not
>
>>
>
>>>a real world test by any means, but I
>
>>
>
>>>synthesized some stuff to better clarify my line
>
>>
>
>>>of thinking. Not that you'll have conditions
>
>>
>
>>>where M and S have matching phase + amp +
>
>>
>
>>>frequency (much) in practical applications.
>
>>
>
>>>But...
>
>>
>
>>>
>
>>
>
>>><<<http://home.earthlink.net/~shadow_7/MS_TEST.sh>http://home.earthlink.=
net/~shadow_7/MS_TEST.sh><http://home.earthlink.net/~shadow_7/MS_TEST.sh>ht=
tp://home.earthlink.net/~shadow_7/MS_TEST.sh><<http://home.earthlink.net/~s=
hadow_7/MS_TEST.sh>http://home.earthlink.net/~shadow_7/MS_TEST.sh><http://h=
ome.earthlink.net/~shadow_7/MS_TEST.sh>http://home.earthlink.net/~shadow_7/=
MS_TEST.sh
>
>>
>
>>incomplete extension on the uploaded file. .sh?
>
>>
>
>>>
>
>>
>
>>>It uses bash and sox to synth some pitches
>
>>
>
>>>across a broad range. Kept the root and 5th of
>
>>
>
>>>a chord unique to M and S. Mixed a common tone
>
>>
>
>>>(the third of a chord) between the two tracks
>
>>
>
>>>and the results were interesting. -S for the
>
>>
>
>>>right channel completely cancels out the 3rd of
>
>>
>
>>>the chord on the right channel. When going from
>
>>
>
>>>M and S to L and R.
>
>>
>
>>>
>
>>
>
>>>Going from the generated L and R back to M and
>
>>
>
>>>S. And +S and -S don't cancel each other out
>
>>
>
>>>anymore. And to make matters worse the 5th
>
>>
>
>>>which was unique to M (and NOT S) is the result
>
>>
>
>>>of trying to cancel S out. Not that it's a real
>
>>
>
>>>world practical test. And I probably didn't get
>
>>
>
>>>the math just right. But it seems a bit
>
>>
>
>>>interesting to me. Assuming that it's not some
>
>>
>
>>>software quirk with sox or my math.
>
>>
>
>>Hi James--
>
>>
>
>>Should work with any two, single channel sound
>
>>
>
>>files. Seems like your patch has issues.
>
>>
>
>>Has anyone done this experiment with a PC
>
>>
>
>>compatible plug-in in that is simple (e.g. having
>
>>
>
>>only mid and side gain adjustments with a unity
>
>>
>
>>position) like "+Matrix" is for Mac?
>
>>
>
>>Maybe try the experiment using the suggested simple MS matrix plug-in? Ro=
b D.
>
>>
>
>>>
>
>>
>
>>>I was only expecting to find some minute
>
>>
>
>>>differences in the waveforms of the derived
>
>>
>
>>>versus original tracks when I started out. And
>
>>
>
>>>maybe some loss of frequencies in the upper
>
>>
>
>>>bands. Needless to say that the quirks I found
>
>>
>
>>>were VERY audible. The timing and other
>
>>
>
>>>variances of actual M/S would likely not yeild
>
>>
>
>>>the same results. Especially if M and S are NOT
>
>>
>
>>>mixed at unity gain. But it should lend some
>
>>
>
>>>credibility to recording the M and S tracks and
>
>>
>
>>>not the L and R tracks in the field. If you
>
>>
>
>>>intend on doing any editing IMO.
>
>>
>
>>>
>
>>
>
>>>- James
>
>>
>
>>>
>
>>
>
>>>--- On Fri, 8/20/10, Marinos Koutsomichalis
>
>>
>
>>><<marinos%40agxivatein.com><marinos%40agxivatein.com><mail=
to:marinos%40agxivatein.com>>
>
>>
>
>>>wrote:
>
>>
>
>>>
>
>>
>
>>>From: Marinos Koutsomichalis
>
>>
>
>>><<marinos%40agxivatein.com><marinos%40agxivatein.com><mail=
to:marinos%40agxivatein.com>>
>
>>
>
>>>Subject: Re: [Nature Recordists] M/S recording setup
>
>>
>
>>>To:
>
>>
>
>>><naturerecordists%40yahoogroups.com><naturerecordists%40ya=
hoogroups.com><naturerecordists%40yahoogroups.com>=
ahoogroups.com
>
>>
>
>>>Date: Friday, August 20, 2010, 2:04 AM
>
>>
>
>>>
>
>>
>
>>>
>
>>
>
>>>
>
>>
>
>>>I still think that if the maths are done
>
>>
>
>>>correctly there will be no loss at all.. some
>
>>
>
>>>plugins may have 'strange' algorithms for doing
>
>>
>
>>>this, but if you find the right one, or better
>
>>
>
>>>if you program this yourself in some platform
>
>>
>
>>>then it should be totally transparent..
>
>>
>
>>>
>
>>
>
>>>m
>
>>
>
>>>
>
>>
>
>>>On 20 =C9=FC=C9"=C9=A1 2010, at 7:09 1=CE4.=C9 ., Dan Dugan wrote:
>
>>
>
>>>
>
>>
>
>>>> > Yes the sample rate remains the same. Yes
>
>>
>
>>>>the file size probably remains the same. And
>
>>
>
>>>>yes the duration of the recording stays the
>
>>
>
>>>>same. But I find it hard to believe that a
>
>>
>
>>>>twice edited digital file (L/R -> M/S -> L/R)
>
>>
>
>>>>is a bit for bit exact copy of the
>
>>
>
>>>>original(even if you exclude the headers from
>
>>
>
>>>>the comparison).
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> I wouldn't expect it would be, but the
>
>>
>
>>>>"generation loss" would be down in the noise
>
>>
>
>>>>bits and insignificant. MS equalization is
>
>>
>
>>>>quite commonly done in mastering suites, where
>
>>
>
>>>>you find the best monitors and ears in the
>
>>
>
>>>>business.
>
>>
>
>>>
>
>>
>
>
>>
>
>>>
>
>>
>
>
>>
>
>>>
>
>>
>
>>>
>
>>
>
>>--
>
>>
>
>
>>
>
>
>>
>
>>
>
>--
>
>
>
>
--
|