Topic: Audacity midside help (why is my L channel louder then my R)

[page]

I have a given audio file in raw/undecoded M-S format with the mid on the left (as intended). I then split the stereo into two mono tracks in audacity (mid and side), then duplicate the S track. Now I have a mid and two sides. I then invert the second S (or track 3). I can pan the two S tracks to the center and I come back with nothing (so I know the phase is reversed and it comes down to a null mono).

So why when I have the mid in the center, and each S track on the sides, without any gain applied or removed, do I end up with a left channel that is approximately +4db difference from the right?

Someone else here asked this question and I can't find the thread after an hour of searching, so I apologize if this is an easy gimme. Any information is greatly appreciated.

PS: I'm using OSX, and Audacity's VST bridge crashes, so I'm down to AUs and manual.

[JackHenry]

Don't you need to pan the two side tracks left and right??

[page]

Don't you need to pan the two side tracks left and right??

Second paragraph, sorry I wasn't clearer on that.

[boojum]

You need to decode the tracks,.  This can be done manually, ask someone else how to, or with the Voxengo plug-in which is free.  I use the plug-in.

Cheers

[page]

You need to decode the tracks,.  This can be done manually, ask someone else how to, or with the Voxengo plug-in which is free.  I use the plug-in.

Cheers

I thought thats what the inversion of the side track did. Did I miss something?

As for the plugin, the voxengo MSED was only packaged in windows dll. I havn't been able to get mac vst plugins to work thanks to the crappy vst bridge that audacity uses either.

[hummat]

A couple of Mac M/S plugins:

Have not used this one ( just found it recently)
http://www.brainworx-music.de/index.php?nav=26&um=2&lang=en


Have used this one and been happy with it.
http://www.soundhack.com/freeware.php

-jay

[Scooter]

you are doing it properly.  I don't know why you are getting the weird levels though.  I always matrix my M/S stuff like you do, i don't like using plugins.  i'd just bounce the thing down to 2 trks, then fix the levels discrepancy and see how it sounds.  were you off to the side or something??

[page]

A couple of Mac M/S plugins:

Have not used this one ( just found it recently)
http://www.brainworx-music.de/index.php?nav=26&um=2&lang=en


Have used this one and been happy with it.
http://www.soundhack.com/freeware.php

-jay

the Soundhack doesn't load on my machine for some reason, and the BX one causes audacity to crash when it's loading (vst bridge sucks). On my wife's machine, I havn't tried soundhack's yet, and the BX one doesn't show up when audacity loads. Will investigate it more later this evening.

you are doing it properly.  I don't know why you are getting the weird levels though.  I always matrix my M/S stuff like you do, i don't like using plugins.  i'd just bounce the thing down to 2 trks, then fix the levels discrepancy and see how it sounds.  were you off to the side or something??

I don't think that was it, I was real close to center (sonically). Maybe a foot off or so. Even then, it seems to happen on all three recordings, and one of them I *know* I was DFC for cause I had to sit slightly to one side otherwise it sounded strange being at the center...

If you look at the BSCS-L thread, the Omni/8 recording is one of the tapes that exhibits this. I'd be curious if you decode it what you get for results (level wise).

[Scooter]

if you can get a sample to me, i'll gladly give it a shot...

are you for sure that you didn't have any real-time M/S decoding going on with the preamp??

[page]

if you can get a sample to me, i'll gladly give it a shot...

are you for sure that you didn't have any real-time M/S decoding going on with the preamp??

Not on the UA-5 I didn't. Plus, you can hear that the left channel is the mid upon playback.

[SmokinJoe]

I think your left is mid+side, and your right is only side???  That would make it about 4db hotter.
I can't envision how this works with 3 tracks.

I do this kind of like you do, but a little differently...
- Take M/S raw (as stereo)
- duplicate this whole thing
- split both tracks so now the order from top to bottom is L R L R = M S M S
- move 3rd track up, so now it's M M S S
- invert 3rd track, so now it's M M -S S
- Combine M M into a stereo track
- Combine -S S into a stereo track

[page]

I think your left is mid+side, and your right is only side???  That would make it about 4db hotter.
I can't envision how this works with 3 tracks.

I do this kind of like you do, but a little differently...
- Take M/S raw (as stereo)
- duplicate this whole thing
- split both tracks so now the order from top to bottom is L R L R = M S M S
- move 3rd track up, so now it's M M S S
- invert 3rd track, so now it's M M -S S
- Combine M M into a stereo track
- Combine -S S into a stereo track

Still, left channel is 4db hot. If I run the positive lobe on the right side, the right side is hot. I could see it if one recording was like that (oriented wrong), anyone can screw up. But 3 recordings in a row? That takes talent... Especially for one which I was anal about my placement. On one of them the negative lobe should be higher due to placement (I was ROC by about 4-5ft).

[SmokinJoe]

What are you using for a mic? and any chance you got things mixed up, like you had the figure 8 pointed forward and the card pointed right?  I've got an LSD2 and I've done stupid things like that...

[Brian Skalinder]

I do this kind of like you do, but a little differently...
- Take M/S raw (as stereo)
- duplicate this whole thing
- split both tracks so now the order from top to bottom is L R L R = M S M S
- move 3rd track up, so now it's M M S S
- invert 3rd track, so now it's M M -S S
- Combine M M into a stereo track
- Combine -S S into a stereo track

Wouldn't you need an additional step to mix the resulting stereo tracks down to a single stereo track?  The M M stereo track is really just dual-mono.  You'd want to mix the M/M stereo track with the -S/S stereo track.  That way, after combining the two stereo tracks, you'd have a single stereo track remaining:  one that's M + (-S) and one that's M + S.

any chance you got things mixed up, like you had the figure 8 pointed forward and the card pointed right?  I've got an LSD2 and I've done stupid things like that...

:raises hand:  Me, too.  :smacks forehead:

[page]

I thought about that, especially with my blumlein. Blew an hour flipping the channel polarity on that one to see if it would effect it much.

It's a possibility on the Card/8, but I've got an Omni/8 that it does the same thing (left is hotter after the matrix). All were taped with a Busman BSCS-L > UA-5.

I've only heard from one other person who decoded any of them and that was a card/8 and he said it sounded nice, so I'm trumping this to either something dumb I'm doing or audacity being weird. Has anyone else tried decoding the samples? Both should be mid-left side-right.

Omni/8
http://www.sendspace.com/file/ucxp4v

Card/8
http://www.sendspace.com/file/yvph0a

30sec flacs, 3mb each.

edit: +Ts around, at least your giving it a shot, I appreciate the help. I'll try the soundhack again on my wife's ibook and see if it will load in audacity. If I kill another night at this, I'm going to crack and buy a better program for this... I'd rather pay the cash and get something to keep M/S an option then give up.

[Scooter]

I got the same results you did.  L chan 4dB hotter.  it sounded fine when i just lowered the L chan 4dB, then adjusted the S's with that relationship.  quite odd...

[page]

I got the same results you did.  L chan 4dB hotter.  it sounded fine when i just lowered the L chan 4dB, then adjusted the S's with that relationship.  quite odd...

Did you use audacity and/or a plugin? I wonder if there isn't something in the mic that favors the positive lobe.

[Will_S]

I got the same results you did.  L chan 4dB hotter.  it sounded fine when i just lowered the L chan 4dB, then adjusted the S's with that relationship.  quite odd...

Did you use audacity and/or a plugin? I wonder if there isn't something in the mic that favors the positive lobe.

It's not unheard of for there to be a mismatch between the two lobes of a figure 8 mic.  See this thread for instance:  http://taperssection.com/index.php/topic,80553.0/all.html

A 4dB imbalance in the mixdown seems a bit much though.

[SmokinJoe]

Wouldn't you need an additional step to mix the resulting stereo tracks down to a single stereo track?  The M M stereo track is really just dual-mono.  You'd want to mix the M/M stereo track with the -S/S stereo track.  That way, after combining the two stereo tracks, you'd have a single stereo track remaining:  one that's M + (-S) and one that's M + S.

Yes,  that happens in the end when I "export as wav".  It just says "I'm going to mix this down to a single stereo track OK?", and I let it.

I don't usually do a specific combine.  After my above steps I generally spend time playing with the levels in an effort to achieve my desired mix/side mix/width, and getting levels I'm happy with.  Then I export for tracking, and it's mixed down.

[Scooter]

I got the same results you did.  L chan 4dB hotter.  it sounded fine when i just lowered the L chan 4dB, then adjusted the S's with that relationship.  quite odd...

Did you use audacity and/or a plugin? I wonder if there isn't something in the mic that favors the positive lobe.

no.  i used nuendo, the old fashioned way, clone and invert the 3rd chan.  very strange indeed.

[page]

no.  i used nuendo, the old fashioned way, clone and invert the 3rd chan.  very strange indeed.

Ok, then it's something with either my setup, or the mic. I'll try recording again tonight and see what I get.

Thanks folks, I appreciate the help.

[page]

Hmm, I'm going to chalk it up to user error, as I didn't have that problem tonight for Mike Gordon. I'll post a sample tomorrow.

[Scooter]